U.S. Department of Commerce Volume 111 Number 3 July 2013 Fishery Bulletin U.S. Department of Commerce Penelope S. Pritzker Secretary of Commerce National Oceanic and Atmospheric Administration Kathryn D. Sullivan Acting Administrator National Marine Fisheries Service Samuel D. Rauch III Acting Assistant Administrator for Fisheries /f'X ★ Scientific Editor Bruce C. Mundy Associate Editor Kathryn Dennis National Marine Fisheries Service Pacific Islands Fisheries Science Center Aiea Heights Research Facility 99-193 Aiea Heights Drive, Suite 417 Aiea, Hawaii 96701-3911 Managing Editor Sharyn Matriotti National Marine Fisheries Service Scientific Publications Office 7600 Sand Point Way NE Seattle, Washington 98115-0070 The Fishery Bulletin (ISSN 0090-0656) is published quarterly by the Scientific Publications Office, National Marine Fish- eries Service, NOAA, 7600 Sand Point Way NE, Seattle, WA 98115-0070. Although the contents of this publica- tion have not been copyrighted and may be reprinted entirely, reference to source is appreciated. The Secretary of Commerce has deter- mined that the publication of this peri- odical is necessary according to law for the transaction of public business of this Department. Use of funds for printing of this periodical has been approved by the Director of the Office of Management and Budget. For sale by the Superintendent of Documents, U.S. Government Printing Office, Washington, DC 20402. Subscrip- tion price per year: $32.00 domestic and $44.80 foreign. Cost per single issue: $19.00 domestic and $26.60 foreign. See back for order form. Editorial Committee Richard Brodeur John Carlson Kevin Craig Jeff Leis Rich McBride Rick Methot Adam Moles Frank Parrish Dave Somerton Ed Trippel Mary Yoklavich National Marine Fisheries Service, Newport, Oregon National Marine Fisheries Service, Panama City, Florida National Marine Fisheries Service, Beaufort, North Carolina Australian Museum, Sydney, New South Wales, Australia National Marine Fisheries Service, Woods Hole, Massachusetts National Marine Fisheries Service, Seattle, Washington National Marine Fisheries Service, Auke Bay, Alaska National Marine Fisheries Service, Honolulu, Hawaii National Marine Fisheries Service, Seattle, Washington Department of Fisheries and Oceans, St. Andrews, New Brunswick, Canada National Marine Fisheries Service, Santa Cruz, California Fishery Bulletin web site: www.fisherybulletin.noaa.gov The Fishery Bulletin carries original research reports and technical notes on investigations in fishery science, engineering, and economics. It began as the Bulletin of the United States Fish Commission in 1881; it became the Bulletin of the Bureau of Fisheries in 1904 and the Fishery Bulletin of the Fish and Wildlife Service in 1941. Separates were issued as documents through volume 46; the last document was No. 1103. Beginning with volume 47 in 1931 and continuing through volume 62 in 1963, each separate appeared as a numbered bulletin. A new system began in 1963 with volume 63 in which papers are bound together in a single issue of the bulletin. Beginning with volume 70, number 1, January 1972, the Fishery Bulletin became a periodical, issued quarterly. In this form, it is available by subscription from the Superintendent of Documents, U.S. Government Printing Office, Washington, DC 20402. It is also available free in limited numbers to libraries, research institutions. State and Federal agencies, and in exchange for other scientific publications. U.S. Department of Commerce Seattle, Washington Volume 111 Number 3 July 2013 Fishery Bulletin Contents Articles 205-217 Essington, Timothy E„ Kathryn Dodd, and Thomas P. Quinn Shifts in the estuarine demersal fish community after a fishery closure in Puget Sound, Washington 218-232 Bacheler, Nathan M„ Valerio Bartolino, and Marcel J. M. Reichert Influence of soak time and fish accumulation on catches of reef fishes in a multispecies trap survey 233-251 Moiseev, Sergey I., Svetlana A. Moiseeva, Tatyana V. Ryazanova, and Anna M. Lapteva Effects of pot fishing on the physical condition of snow crab ( Chionoecetes opilio) and southern Tanner crab ( Chionoecetes bairdi) The National Marine Fisheries Service (NMFS) does not approve, recommend, or endorse any proprie- tary product or proprietary materia! mentioned in this publication. No reference shall be made to NMFS, or to this publication furnished by NMFS, in any advertising or sales promotion which would indicate or imply that NMFS approves, rec- ommends, or endorses any propri- etary product or proprietary mate- rial mentioned herein, or which has as its purpose an intent to cause directly or indirectly the advertised product to be used or purchased be- cause of this NMFS publication. The NMFS Scientific Publications Office is not responsible for the contents of the articles or for the standard of English used in them. 252-264 Zudaire, Iker, Hilario Murua, Maitane Grande, and Nathalie Bodin Reproductive potential of Yellowfin Tuna ( Thunnus olbacares) in the western Indian Ocean 265-278 Peemoeller, Bhae-Jin, and Bradley G. Stevens Age, size, and sexual maturity of channeled whelk ( Busycotypus canaliculatus ) in Buzzards Bay, Massachusetts 279-287 Wood, Anthony D., and Steven X. Cadrin Mortality and movement of Yellowtail Flounder ( Limanda ferruginea ) tagged off New England 288 Best Paper Awards for 2012 289 Guidelines for authors 205 Shifts in the estuarine demersal fish community after a fishery closure in Puget Sound, Washington Abstract— Puget Sound is one of the largest and most ecologically signifi- cant estuaries in the United States, but the status and trends of many of its biological components are not well known. We analyzed a 21- year time series of data from stan- dardized bottom trawl sampling at a single study area to provide the first assessment of population trends of Puget Sound groundfishes after the closure of bottom trawl fisheries. The expected increase in abundance was observed for only 3 of 14 species after this closure, and catch rates of most (10) of the abun- dant species declined through time. Many of these changes were step- wise (abrupt) rather than gradual, and many stocks exhibited changes in catch rate during the 3-year pe- riod from 1997 through 2000. No detectable change was recorded for either temperature or surface salin- ity over the entire sampling period. The abrupt density reductions that were observed likely do not reflect changes in demographic rates but may instead represent distributional shifts within Puget Sound. Manuscript submitted 27 August 2012. Manuscript accepted 3 April 2013. Fish. Bull. 111:205-217 (2013). doi 10.7755/FB. 11 1.3.1 The views and opinions expressed or implied in this article are those of the author (or authors) and do not necessar- ily reflect the position of the National Marine Fisheries Service, NOAA. Timothy E. Essington (contact author) Kathryn Dodd Thomas P. Quinn Email address for contact author: essing@uw.edu School of Aquatic and Fishery Sciences University of Washington UW Box 355020 Seattle, Washington 98195 Estuaries support diverse marine communities and act as nursery ar- eas for many coastal populations (Beck et ai., 2001; Armstrong et al., 2003), but they are the most heavily affected marine ecosystems (Lotze et al., 2006; Halpern et al., 2008) be- cause they also commonly serve as ports for shipping, support commer- cial and recreational fisheries, and are used as recreational areas. Be- cause of their close link with terres- trial systems, they are susceptible to coastal eutrophication (Carpenter et al., 1998) that leads to hypoxia (Diaz, 2001; Breitburg et al., 2009), harmful algal blooms (Anderson et al., 2002), and concentrations of contaminants (Nichols et al., 1986). Because of the myriad ecosystem services they pro- vide (Guerry et al., 2012) and the many human activities that may im- pair their delivery, there is a grow- ing effort to protect and restore these ecosystems. However, assessment of the efficacy of protection measures is often hindered by the lack of long- term, standardized data and by con- founding changes in many aspects of the ecosystem, such as fishery man- agement, shoreline protection, and water quality. Puget Sound is one of the larg- est and most ecologically significant estuaries in the United States, sup- porting a rich fauna with more than 200 fish species, 26 marine mam- mals, more than 100 bird species and a high diversity of invertebrates.1 It is the second-largest estuary (2330 km2) in the coterminous United States, and its watershed supports a large and growing human popula- tion. Land alteration and habitat loss (Levings and Thom, 1994), fishing,2 and toxic contaminants (Landahl et al., 1997) have had widespread ef- fects on this ecosystem. Currently, 8 fish species or fish stock in Puget Sound are protected under the U.S. Endangered Species Act, and many others are identified as being at risk (Musick et ah, 2000). Tagging stud- ies, genetic analyses, and differences in toxic contaminant levels all in- dicate that Puget Sound stocks of various fish species are distinct from coastal stocks (Day, 1976; Andrews et al., 2007; West et al., 2008; Andrews and Quinn, 2012). Notwithstand- ing these issues, much of the area 1 Ruckelshaus, M. H., and M. M. McClure. 2007. Sound science: synthesizing eco- logical and socioeconomic information about the Puget Sound ecosystem. U.S. Dep. Commer., NOAA, National Marine Fisheries Service, Northwest Fisher- ies Science Center. Seattle, WA. 93 p. [Available from http://www.nwfsc.noaa. gov/research/shared/sound_science/docu- ments/sound_science_finalweb.pdf. 2 Palsson, W. A., T. J. Northrup, and M. W. Barker. 1998. Puget Sound ground- fish management plan, 48 p. Washing- ton Department of Fish and Wildlife, Olympia, WA. [Available from http:// wdfw.wa. gov/publications/00927/. | 206 Fishery Bulletin 111(3) in Puget Sound is deep because of its glacial origin (Burns, 1985), compared with the much shallower es- tuarine systems on the East Coast of North America. Therefore, the shallow biogenetic habitats that have been altered by humans — a common occurrence in es- tuaries— encompass a relatively small fraction of the total available habitat. The local government is work- ing to assess the status of the Puget Sound ecosystem and to identify and implement measurable restoration goals.3 However, this planning process is hindered by a paucity of long-term data on species and commu- nity trends.4 Without such time series, it is difficult to assess the rate and extent of recovery that may be reached within planning timelines. Here we present our analysis of one of the longest continual and standardized surveys of the groundfish community in Puget Sound (surveys conducted by the University of Washington) to assess the nature and ex- tent of change that has accompanied significant resto- ration measures. Most significantly, the state of Wash- ington progressively prohibited commercial trawling by closing most waters of central and southern Puget Sound in 1989, and then closing all inland marine waters to nontribal bottom trawling in 2010. 2 There- fore, we hypothesized that these survey data — the collection of which began in 1991 — would provide an indication of rates of recovery of exploited fish stocks and community reorganization because not all species were exploited. Our time series is limited in spatial extent but provides a 20-year record of species com- position and abundance of the groundfish community and, therefore, may indicate the effect of commer- cial trawling and enable assessment of the status of recovery. This ecosystem affords a rare opportunity to track the recovery of groundfish populations and communi- ties in response to a commercial fishery closure. Typi- cally, information about fisheries effects has come from tracking changes in “no-take” marine reserves (Russ and Alcala, 1996; Babcock et al., 1999; Halpern, 2003). Although such spatial closures provide important in- formation for identifying restoration targets, no-take areas are often smaller in area than the range of popu- lations affected by fishing and, therefore, may not re- veal the full extent of fishing effects (Claudet et ah, 2008). In contrast, the commercial trawl-fishing clo- sure in Puget Sound covered a large area that closely matches the distribution scales of resident populations. Moreover, because more than 2 decades have passed 3 Puget Sound Partnership. 2009. Puget Sound Action Agenda : Protecting and restoring the Puget Sound ecosys- tem by 2020, 213 p. Pugest Sound Partnership, Olympia, WA. [Available from http://www.psp.wa.gov/downloads/ AA2009/Action_Agenda_FINAL_063009.pdf. 4 Essington, T. E., T. Klinger, T. Conway-Cranos, J. Buchanan, A. James, J. Kerschner, I. Logan, and J. West. 2011. The bio- physical condition of Puget Sound. In Puget Sound Science Update, p. 205-423. Puget Sound Partnership, Tacoma, WA. [Available from http://www.psp.wa.gov/scienceupdate.php.] since the closure, we have the potential to describe not only the extent but also the trajectory of population recovery. Therefore, we can ask whether recovery was monotonic as predicted by simple population models or whether, instead, it was characterized by abrupt and sustained shifts in abundance and composition that would indicate either nonlinear population or commu- nity dynamics (Doak et al., 2008; McClanahan et ah, 2011) or decadal-scale environmental drivers (Mantua et al., 1997; Anderson and Piatt, 1999). Our specific objectives were to determine 1) whether catch rates of resident Puget Sound groundfishes gen- erally increased through time following the closure of bottom trawl fisheries, 2) the extent to which shifts in this time series may represent population fluctua- tions or instead represent local effects that result from distribution shifts, 3) whether dynamics are best rep- resented by smooth trends through time or instead though more abrupt state-changes, and 4) whether ob- served trends in catch rates for resident Puget Sound groundfish populations may be linked to changes in environmental conditions reflected in oceanographic monitoring data. Materials and methods Study location and design Catch data were derived from bottom trawl surveys conducted in Port Madison, a large bay on the west side of central Puget Sound, north of Bainbridge Island (Fig. 1; see also Andrews and Quinn [2012] for specific sampling sites) as part of a Fisheries Ecology course of the University of Washington. The study area is lo- cated in the central basin of Puget Sound, the largest of the 4 main basins that compose the inland marine waters of Washington State. The area is not industri- alized, and the shoreline is primarily a natural bluff- beach formation typical of central Puget Sound with some armoring around private residences. All sampling was conducted on the third weekend in May, beginning in 1991 and continuing until 2012; sampling did not occur in 1992 and 1998. Bottom trawl surveys consisted of single tows of ~5 min conducted at 4 fixed index sites at discrete depths (10, 25, 50 and 70 m) over 5 diel time periods: after- noon (-15:00-18:00 h), evening (-20:00-23:00 h), night (-01:00-04:00 h), morning (-06:00-09:00), and mid-day (-11:00-14:00 h). This survey design was intended to capture and account for diel shifts in onshore-offshore distribution of key species (Andrews and Quinn, 2012). Trawl paths did not overlap within sampling years but were staggered slightly. All sampling was conducted from RV Kittiwake. Each tow covered 0.37 km at 0.5 m/s, with a standard Southern California Coastal Wa- ter Research Project bottom trawl that had a footrope Essington et al.: Shifts in the estuarine demersal fish community after a fishery closure in Puget Sound, Washington 207 Figure t Map of the locations of the study area (rectangle at the center) where bottom trawl surveys were conducted in Port Madison from 1991 to 2012 and of 2 nearby monitored sites (labeled West point and Jefferson Head) where time series data on environmental conditions, such as temperature and surface salinity, were recorded. Map inset in upper-right corner shows location of Puget Sound, Washington, along the U.S. Pacific coast, and map inset in lower-left corner shows detailed view of the Port Madison study area and locations of 4 sampling sites with corresponding depths listed. of 5 m and net width of 3.5 m during fishing.5 The bot- tom trawl was fitted with a 3.8-cm body mesh and 3.2- cm codend mesh with a 0.4-cm codend liner. The net primarily targets flatfishes but also catches small de- mersal fishes, such as gadids and some elasmobranchs. Fish were identified to species on deck with the aid of dichotomous keys (Hart, 1973), but a few individu- als were retained for examination in the laboratory. We measured fork length for all species except length of Spotted Ratfish ( HycLrolagus colliei), for which precau- dal length (tip of snout to second dorsal fin; Anderson and Quinn, 2012) was measured; all length measure- ments were made to the nearest millimeter. Consisten- cy in field identification was facilitated by the presence of one of us (T. Quinn) for virtually every tow in the entire time series. J Eaton, C. M., and P. A. Dinnel. 1993. Development of a trawl-based criteria for assessment of demersal fauna (mac- roinvertebrates and fishes): pilot study in Puget Sound, Washington, 87 p. Final report to the U.S. Environmental Protection Agency. Bio-Marine Enterprises, Seattle ,WA. Environmental data We obtained data from 2 monitoring sites on water- column characteristics (temperature and salinity pro- files) for March, April, and May. The King County Water and Land Resources Division samples a loca- tion 4.4 km northeast of Port Madison called Jeffer- son Head, and the Washington State Department of Ecology samples a location 8.5 km southeast of Port Madison called West Point — both on a monthly basis (Fig. 1). We used data from both of these sampling pro- grams (1990-2008 for West Point, 1992-2008 for Jef- ferson Head) to identify years and time periods with unusual environmental conditions on the basis of sub- mixed-layer temperature and surface salinity. Surface salinity gives a measure of seasonal runoff and, there- fore, indicates seasonal weather events (years with high precipitation have low surface salinity). Sub- mixed-layer temperature is indicative of the thermal habitat experienced by groundfishes. Sub-mixed-layer temperature was used instead of bottom temperature because the latter was not always sampled. When bot- 208 Fishery Bulletin 111(3) tom temperature was sampled, it was generally within 0.7°C of sub-mixed-layer temperature (depth=20 m) for March, April, and May. We focused on data from these months because they include the time periods immediately before bottom trawl sampling and, there- fore, could best indicate changes in environmental conditions that might affect catch rates. Moore et al. (2008) demonstrated strong intra-annual coherence of oceanographic properties within Puget Sound basins; therefore these data are likely representative of intra- annual environmental conditions throughout the cen- tral Puget Sound basin. Analysis For most years, all 20 depthxtime combinations were successfully sampled, but gear malfunction and other events resulted in missing sets for some sampling sites. These missing sets constituted only 5% of the to- tal sample design, but we wanted to account for them in deriving annual catch levels. We first ascertained whether these differences can alter annual estimates of catch rates by fitting an analysis of variance (ANO- VA) for each of our study species with year, depth, time, and a depthxtime interaction term. All but one species, the Shiner Perch ( Cymatogaster aggregata), showed either a significant effect of depth, depth+time, or a depthxtime interaction term. We used a simple approach to account for the small numbers of missing sets. Rather than fitting general- ized linear models to calculate a statistical “year ef- fect,” we instead calculated an annual average catch anomaly for each year on the basis of expected catches for each timexdepth combination. This approach is equivalent to fitting a generalized linear model with a time+depth+timexdepth interaction term, but it has a straightforward interpretation and permitted a paral- lel calculation for the trawl and environmental data. We calculated the mean catch rate (number of fish/tow) for each depthxtime combination for each species with data from the entire sampling period. We then calcu- lated the catch anomaly as the difference between ob- served species-specific catch and the expected (mean) catch rate given the depth and time of sampling. The annual abundance index for each species was equal to the average catch anomaly over all samples conducted within a year. We used the same approach to gener- ate temperature and salinity anomalies for each year. For each month and monitoring site, we calculated the mean temperature and salinity values from all avail- able data, generated anomalies for each year, month, and site, and averaged these across months to derive a yearly anomaly value. We generally tracked abundances at the species level, but, in some cases, we aggregated closely relat- ed species. Rock soles were allocated to a single spe- cies when the survey began, but subsequent genetic work indicated that the rock sole genus ( Lepidopsetta ) consists of 3 species (Orr and Matarese, 2000), 2 of which occur in Puget Sound: Rock Sole ( Lepidopsetta bilineata ) and Northern Rock Sole ( Lepidopsetta po- ly xystra). We conducted our analysis at the scale of an aggregated species group because the 2 Puget Sound species are not readily distinguished in the field and we wanted to maintain consistency throughout the time series. Further, Speckled Sanddab (Citharichthys stigmaeus ) and Pacific Sanddab ( Citharichthys sordi- dus) are morphologically similar as juveniles; for this reason, species-level identifications were not reliable. We, therefore, combined all individuals identified as either species into a species group termed “sanddab; ( Citha ri ch thys ) . ” We focused analysis on the most common species and species groups encountered with the sampling gear so that we had sufficient statistical power to detect changes in abundance through time. We set an arbi- trary threshold of 200 sampled individuals over the en- tire time period for species to be included in the analy- sis. This use of a threshold eliminated species so rarely encountered that trends would not be reliable, species for which the gear was not appropriate, and samples for which species identity could not be determined (e.g., samples in very early juvenile stages). For each species, we asked whether abundance changed through time, and, if so, whether it was best described by a continu- ous linear increase or decrease or a discontinuous shift in the mean catch rate. The latter is consistent with re- gime shifts as reflected by rapid and persistent chang- es in population densities (Rodionov and Overland, 2005). For each time series, we used Akaike’s informa- tion criteria adjusted for a small sample size (AICc) to choose between 3 models: constant, linear, or change point. For each model, we assumed normally distrib- uted residuals. We used the changepoint package (vers. 0.6; Killick and Eckley, 2011) in R software (vers. 2.13; R Development Core Team, 2011) to assess discontinu- ous shifts in the mean catch rate. We required that the best fitting change-point model consist of time periods spanning at least 4 years of data. In other words, es- timated change points that broke the time series into increments shorter than 4 years were discarded, thus preventing the model from placing change points at the beginning or end of time series. Because we found evidence of change points for many flatfishes, we explored the data for flatfish spe- cies in more detail. The gear captures individuals across a wide size range and range of life history stag- es; therefore we evaluated whether changes in catch rate could be attributed to changes in recruitment pat- terns. If changes in densities were driven by changes in recruitment, we would expect to see time trends of abundance for small size classes to lead trends for larger size classes. For each flatfish species, we calcu- lated catch anomalies separately for small and large size classes (individuals below the 33rd percentile and above the 66th percentile of the cumulative length- frequency distribution, respectively). Size-at-age data are not available for most species, but for English Essington et al.: Shifts in the estuarine demersal fish community after a fishery closure in Puget Sound, Washington 209 Table t Summary of total catches for species or species groups that were commonly collect- ed during bottom trawl surveys conducted from 1991 to 2012 in Port Madison, Puget Sound, Washington. A full listing of all species collected is presented in the appendix table. Catch Species (number of fish) Percentage of total Blackbelly Eelpout ( Lycodes pacificus) 3398 9.3 Dover Sole (Microstomus pacificus ) 366 1.0 English Sole (Parophrys vetulus ) 10,427 28.6 Flathead Sole ( Hippoglossoides elassodon ) 1665 4.6 Pacific Hake (Merluccius productus ) 958 2.6 Pacific Herring ( Clupea pcillasii) 478 1.3 Pacific Tomcod ( Microgadus proximus) 2677 7.3 Plainfin Midshipman ( Porichthys notatus ) 497 1.4 Rock soles ( Lepidopsetta bilineata and L. polyxystra) 773 2.1 Sanddabs ( Citharichthys sordidus and C. stigmaeus ) 1075 3.0 Sand Sole ( Psettichthys melanostictus) 680 1.9 Shiner Perch ( Cymatogaster aggregata) 1139 3.1 Slender Sole ( Lyopsetta exilis) 2242 6.2 Spotted Ratfish ( Hydrolagus colliei) 4068 11.2 Sole ( Parophrys vetulus), the most common species, available aging data (senior author, unpubl. data) in- dicate that this procedure effectively separates age-1 individuals from those individuals aged 3 years and older. Results During the 20-year survey 65 fish species were sam- pled, and the 14 species that were sampled frequently enough (>200 individuals) to evaluate time trends ac- counted for more than 85% of the total catch (Table 1). Notably, 7 of these species were flatfishes (Pleuronec- tidae and Paralichthyidae). English Sole and Spotted Ratfish were by far the most common species, collec- tively, contributing more than 40% of all individuals sampled. Other common species included Blackbelly Eelpout ( Lycodes pad ficus ) and Pacific Tomcod (Micro- gadus proximus). Time series of catch anomalies were nonstationary for most species (Fig. 2). Spotted Ratfish was the only species for which no trend or apparent change in abun- dance over the sampling period was observed (Table 2; Fig. 2). As for changes in abundance values that were seen, a change point was identified for 9 species and a continuous linear trend was found for only 3 species. In all cases where the change-point model provided the best fit to the data, the relationship indicated a reduction in the mean catch rate in the later portion of the time series. These cases included the one for the most abundant species, English Sole, for which the mean catch anomaly shifted from +27/tow before 1998 to —7/tow afterward. In contrast, increases in the mean catch anomaly were observed for all 3 species for which abundance trends were best described by the linear model. Trends in total catch (unstandardized) summed across all species mirrored the trends of English Sole (Fig. 2). Many species exhibited changes in catch rates at similar time periods. Of the 9 species whose dynamics were best described by a change-point, 5 species had es- timated change points between 1997 and 1999 (catches were not sampled in 1998). Three species had change points between 1999 or 2000 and 2000 or 2001. There- fore, there was evidence of a change in catch rates be- tween 1997 and 2001 reflected by several species. Analysis of catch-anomaly trends among different size classes of flatfishes did not support the hypoth- esis that trends were driven by changes in recruitment (Fig. 3). For most species, anomalies for small- and large-size fishes were synchronous with no apparent lag. On the basis of the length-frequency distribution of each species or species group, fishes were catego- rized as small if they fell in the 33rd percentile or lower and as large if they were assigned to the 67th percentile or higher. For instance, the catch anomalies for English Sole were nearly identical between small (age 1) and large (age 3+) size classes. Catch-anomaly trends for the rock sole species group were more con- sistent with recruitment shifts because catch anom- alies of small rock soles declined steeply after 1997 but catch anomalies for large rock soles had a less sudden and delayed decline. For Dover Sole (Mi- crostomus pacificus ) and sanddabs, nearly all the variation in catch anomaly was attributed to large-size individuals; catches of small individuals changed little. 210 Fishery Bulletin 111(3) Blackbelly Eelpout Dover Sole English Sole Flathead Sole Pacific Hake Pacific Herring Pacific Tomcod Plainfin Midshipman > ra E o c J - 1 6 - 4 - 1 30 - 20 - ,1 2 - 1 - i ,/ \ A k nT 1 1 1 1 1 M O .J W i i — i — i — 10 - 0 - j: 0 - -1 - JU i i — i — i — 1990 2000 2010 Rock soles 1990 2000 2010 Sanddabs 1990 2000 2010 Sand Sole 1990 2000 2010 Shiner Perch Slender Sole Spotted Ratfish Figure 2 Time series of annual catch anomalies for 14 species or species groups commonly captured during bottom trawl surveys conducted during 1991-2012 in Port Madison, Puget Sound, Washington. Gray lines depict instances where the linear or change-point model provided the best fit to the data. Graph in the bottom right corner depicts total survey catch (unstandardized). Catch anomalies are the yearly averaged differences between each trawl tow and the average catch rate (number of fish) for all tows collected at the same depth and time of day. Time series of anomalies in sub-mixed-layer tempera- ture and surface salinity were generally consistent with each other between the 2 monitored sites (Fig. 4). The data indicated exceptionally warm years in 1992, 1994, 1998, 2003, and 2004 and cool years in 1993, 1999-2002, and 2008. For each of the 2 sites, temperature time se- ries indicated neither a distinct shift near 1997-1999 nor any other change point or linear trend (Table 3). The time series of surface salinity for the West Point site was generally more variable than the time series for the Jefferson Head site, yet both time series showed the same years as exceptionally high or low. At both sites, 1992, 2001, 2004, and 2008 were high-salinity years and 1991 and 1997 were low-salinity years. There was no indication of a linear trend or change point in the sur- face salinity at West Point (Table 3), but the surface sa- linity at Jefferson Head showed a positive linear trend (change of roughly 0.04 ppt/year). Essington et al.: Shifts in the estuarine demersal fish community after a fishery closure in Puget Sound, Washington 211 Table 2 Comparison of 3 models of changes in catch rates of species or species groups commonly collected during bottom trawl surveys conducted from 1991 to at Port Madison, Puget Sound, Washington. Akaike’s information criteria adjusted for small sample size (AICc) was used to choose between the models compared: constant (no change), linear (change through time), or change point (abrupt change at a single point in time, with no temporal change elsewhere). Values indicate a difference in AICc for each species from lowest AICc among all species. No result is given for species for which the change- point model estimated a breakpoint in the first 4 or last 4 years of the time series and, therefore, these species could not be considered in change-point model comparisons. Species Constant AAICc Linear Change point Blackbelly Eelpout ( Lycodes pacificus) 1.98 0.00 - Dover Sole (Microstomus pacificus ) 31.07 12.95 0.00 English Sole (. Parophrys vetulus) 65.77 26.91 0.00 Flathead Sole ( Hippoglossoides elassodon) 28.70 0.00 - Pacific Hake ( Merluccius productus) 21.37 16.74 0.00 Pacific Herring ( Clupea pallasii ) 9.62 0.00 - Pacific Tomcod iMicrogadus proximus ) 30.74 19.15 0.00 Plainfin Midshipman ( Porichthys notatus) 19.94 0.00 - Rock soles ( Lepidopsetta bilineata and L. polyxystra) 154.96 82.95 0.00 Sanddabs ( Citharichthys sordidus and C. stigmaeus ) 30.06 18.88 0.00 Sand Sole (Psettichthys melanostictus) 28.72 23.23 0.00 Shiner Perch ( Cymatogaster aggregata) 15.22 11.95 0.00 Slender Sole ( Lyopsetta exilis) 82.67 38.21 0.00 Spotted Ratfish (. Hydrolagus colliei) 0.00 0.23 - Discussion We hypothesized that the data from Port Madison would reveal trends of increasing abundance in resi- dent groundfish populations in Puget Sound after the cessation of commercial bottom trawling and, thereby, would indicate rates and magnitudes of recovery. Be- fore the ban on commercial trawling in the central ba- sin of Puget Sound, commercial catches ranged from 224 metric tons (t)/year to more than 500 t/year and, therefore, likely represented a significant source of mortality for many targeted species.6 Commercial fish catches through other methods (set nets, purse seines, or set lines) also have been reduced sharply.5 However, most species exhibited nonlinear patterns of abun- dance characterized by abrupt and sustained changes in relative abundance indices during the 21-year time period that the survey spanned. These abrupt abun- dance shifts were notable because they were most com- monly in the opposite direction from our expectation and appeared to be synchronous among different com- mon groundfish species. Moreover, these shifts did not appear to be related to demographic changes indicative of recruitment shifts, and they were not linked to tem- poral patterns in local water temperature and salinity. 6 Schmitt, C. S., S. Quinnell, M. Rickey, and M. Stanley. 1991. Groundfish statistics from commercial fisheries in Puget Sound, 1970-1988. Progress Report No. 285, 315 p. Wash- ington Department of Fisheries, Olympia, WA. Currently, no recruitment time series are available for demersal fishes in Puget Sound. There are several possible explanations for the syn- chronous reduction in catch rates of groundfish species that occurred in the late 1990s to early 2000s. The first is loss or impairment of habitat that resulted in emi- gration out of the survey area. Most of these groundfish species reside on soft-bottom habitats (sand or mud) and do not rely on biogenic habitats, such as eelgrass beds, that are particularly vulnerable. However, Nich- ols (2003) reported an increase in abundance of com- mon prey items of English Sole in Port Madison and in nearby areas from the early 1960s to the early 1990s. It is possible that this trend reversed after this time period, although direct data are needed to evaluate this hypothesis. Alternatively, trawling itself may have altered physical habitat and benthic infaunal commu- nities (Auster et al., 1996); cessation of this activity may have promoted a community of less-preferred prey for these fish predators. Little information, however, is available on bottom habitat or infaunal community dy- namics to test any of these hypotheses. Alternatively, the second explanation is that changes in catch rates in Port Madison may reflect expansions and contraction of population ranges, possibly as a con- sequence of changes in population densities (MacCall, 1990). However, the sharp decreases in abundance that we witnessed suggest a decline in densities throughout Puget Sound and a contraction to other habitats. This implication is not supported by data from bottom trawl 212 Fishery Bulletin 111(3) Dover Sole English Sole o 6 S. _>' ra E o c Time series of annual catch anomalies for small (solid circles) and large (open circles) size classes for 7 species or species groups of flatfishes collected during bottom trawl surveys conducted during 1991-2012 in Port Madison, Puget Sound, Washington. Small and large-size individuals were categorized as those fishes in the 33rd percentile or lower and those fishes in the 67th percentile or higher, respectively, on the basis of the length-frequency distribution of each species or species group. surveys that cover a wider area and that generally in- dicate greater groundfish densities through the 1990s.7 Third, decreases in catch rates could be explained by changes in predator abundance in this region that may 7 Palsson, W. 2010. Personal commun. NOAA Fisheries, Alaska Fisheries Science Center, Seattle, WA. 98115. impose increased mortality or result in distributional shifts of prey species (Heithaus et al., 2008). Demersal fish species in Puget Sound are consumed by elasmo- branchs, such as Spiny Dogfish ( Squalus acanthias ) and Bluntnose Sixgill Shark (Hexanchus griseus), and marine mammals, such as the harbor seal (Phoca vitu- lina\ Bromaghin et al., 2013; Howard et al., 2013) and Essington et al.: Shifts in the estuarine demersal fish community after a fishery closure in Puget Sound, Washington 213 California sea lion (, Zalophus californianus), but we are unaware of any abrupt changes in predator densities in Port Madison to explain these patterns. Jeffries et al. (2003) reported a monotonic increase in harbor seal abundance in Puget Sound from the late 1970s to the 1980s that was followed by little change in abundance during the mid- to late 1990s. Fourth, groundfish densities can be sensitive to water quality, especially to oxygen concentrations at the seafloor that result in distributional shifts to nor- moxic conditions (Breitburg et al., 2009; Essington and Paulsen, 2010), and chronic hypoxia exposure could diminish productivity of groundfish prey (Diaz and Rosenberg, 1995). There is no consistent sampling for dissolved oxygen in Port Madison to evaluate this hy- pothesis, but the exposure of this area to strong tidal currents and subsequent high mixing likely mean that this area is not particularly prone to low dis- solved oxygen (Nichols, 2003). Moore et al. (2008) did not detect a change in water temperatures throughout Puget Sound from 1993 to 2002; there- fore, changes are unlikely a result of a shift in temperature on a scale larger than Puget Sound. Several important limitations of our data war- rant specific discussion. First, our bottom trawl sampling — although highly standardized in time, space, and method — may not be representative of the entire Puget Sound. Indeed, one of our main conclusions is that shifts in densities of demersal fish species more likely were indicative of distri- butional shifts than of population shifts. Also, the opening of the bottom trawl was small and, there- fore, likely had low selectivity for large-size ground- fishes (e.g., >50 cm). Additionally, because sampling was restricted to a standardized and limited time of year, the data cannot account for seasonal changes (Reum and Essington, 2011) and may not reflect trends apparent in different seasons. Our environmental data were collected from monitoring sites near the study area for bottom trawl surveys, and bottom temperature was not al- ways recorded. We used sub-mixed-layer tempera- ture as a proxy for bottom temperature, which ap- peared to be robust for years in which bottom data were available. Bottom water temperature may de- viate from temperature of the shallower sub-mixed layer because of water exchange between Admi- ralty Inlet, the Strait of Juan de Fuca, and the coastal Pacific Ocean. However, because deepwater dynamics reach equilibrium over time scales of months, they reflect local, seasonal environmental conditions (e.g., air temperature, freshwater runoff) (Ebbesmeyer and Barnes, 1980) and, therefore, are useful for interannual comparisons. Despite these limitations, this study presents the first long-term standardized assessment of the groundfish commu- nity in Puget Sound and, therefore, can provide a baseline for expanded sampling efforts. A large body of research on estuarine fishes fo- cuses on the roles of estuaries as nursery habitats, the value of protecting specific critical habitats, and de- scriptions of patterns of juvenile survival and growth. Estuaries are often viewed as critical habitats that support coastal fish populations (Beck et al., 2001), and nearshore habitat features, such as eelgrass beds, are commonly identified as key features of estuarine habi- tats (Levin and Stunz, 2005). Although loss of eelgrass beds has been identified as a threat in Puget Sound, their importance to the groundfish species examined here is unknown. In well-studied estuarine ecosys- tems, extensive time series of fish abundance indices have permitted exploration of the roles of density de- pendence, overwinter survival, predation, and growth- dependent mortality on year-class strength of fishes (Hurst and Conover, 1998; Buckel et al., 1999; Kim- 214 Fishery Bulletin 111(3) Table 3 Comparison of 3 models for changes in environmental time series data collected at 2 monitoring sites — West Point and Jefferson Head — in Port Madison, Puget Sound, Washing- ton, near sampling sites at which bottom trawl surveys were conducted from 1991 to 2012. Akaike’s information criteria adjusted for small sample size (AICc) was used to choose between the models: no change (“constant”), linear change (“linear”); or abrupt change (“change point”). Separate models were run for each monitoring site. Constant Linear Change point Temperature West Point 0.00 0.62 2.85 Jefferson Head 0.00 2.41 2.85 Surface salinity West Point 0.00 2.41 2.85 Jefferson Head 7.89 0.00 10.81 merer et al., 2000; Taylor et al., 2009). The processes that regulate juvenile survivorship and fish population dynamics in Puget Sound are not easily discerned be- cause of a paucity of long-term monitoring data. Conclusions Catch rates of resident groundfishes from a study area in Puget Sound indicated that a synchronous and abrupt decline in densities occurred in the late 1990s, counter to expectations formed on the basis of the ces- sation of commercial bottom trawling that preceded our sampling. Available evidence suggests that these declines may have resulted from a distributional shift rather than a demographic shift, although an analy- sis of data sets that span a spatial extent wider than our study area in Port Madison is needed to test this hypothesis. Therefore, considerable additional analy- ses are needed to address the response of species and food webs to fishing and to determine how localized closures, such as marine protected areas, may promote recovery of species. Further, there is a need to relate density shifts to environmental and biological changes (e.g., climatic drivers, human-induced habitat shifts, or trophodynamics). Finally, the unexpected shifts in localized catch rates in this study indicate a need for caution when time series are used in evaluating long- term shifts in population and community structure without consideration of whether the data are repre- sentative of entire populations. Acknowledgments We thank Charlie Eaton, the owner and operator of the RV Kittiwake, for the 2 decades of careful vessel op- eration and logistic assistance. Sampling took place as field research for the Fisheries Ecology class at the School of Aquatic and Fishery Sciences, Uni- versity of Washington (UW), and funding for ves- sel charter was provided by the teaching program. We thank the numerous teaching assistants and even more numerous students in the class over the years for their help with sorting and measur- ing fishes. Additional funding for the data analy- sis was provided by the SeaDoc Society, Lowell Wakefield Endowment, the UW Climate Impacts Group, and the Puget Sound Gatekeepers Alli- ance. We thank Wayne Palsson and 3 anonymous reviewers for helpful comments on the manu- script. We thank Chantel Wetzel for conducting preliminary analysis. Literature cited Anderson, D. M., P. M. Glibert, and J. M. Burkholder. 2002. Harmful algal blooms and eutrophication: nutrient sources, composition, and consequenc- es. Estuaries 25:704-726. Anderson, P. J., and J. F. Piatt. 1999. Community reorganization in the Gulf of Alaska following ocean climate regime shift. Mar. Ecol. Prog. Ser. 189:117-123. Andrews, K. S., P. S. Levin, S. L. Katz, D. Farrer, V. F. Gallucci, and G. Bargmann. 2007. Acoustic monitoring of sixgill shark movements in Puget Sound: evidence for localized movement. Can. J. Zool. 85:1136-1143. Andrews, K. S., and T. P. Quinn. 2012. Combining fishing and acoustic monitoring data to evaluate the distribution and movements of spotted ratfish Hydrolagus colliei. Mar. Biol. 159:769-782. Armstrong, D. A., C. Rooper, and D. Gunderson. 2003. Estuarine production of juvenile Dungeness crab (Cancer magister) and contribution to the Oregon- Washington coastal fishery. Estuaries 26:1174-1188. Auster, P. J., R. J. Malatesta, R. W. Langton, L. Watling, P. C. Valentine, C. L. S. Donaldson, E. W. Langton, A. N. Shepard, and I. G. Babb. 1996. The impacts of mobile fishing gear on seafloor habitats in the Gulf of Maine (northwest Atlantic): im- plications for conservation of fish populations. Rev. Fish. Sci. 4:185-202. Babcock, R. C., S. Kelly, N. T. Shears, J. W. Walker, and T. J. Willis. 1999. Changes in community structure in temperate ma- rine reserves. Mar. Ecol. Prog. Ser. 189:125-134. Beck, M. W., K. L. Heck, K. W. Able, D. L. Childers, D. B. Egg- leston, B. M. Gillanders, B. Halpern, C. G. Hays, K. Hoshino, T. J. Minello, R. J. Orth, P. F. Sheridan, and M. P. Weinstein. 2001. The identification, conservation, and management of estuarine and marine nurseries for fish and inverte- brates. Bioscience 51:633-641. Breitburg, D., D. W. Hondorp, L. A. Davias, and R. J. Diaz. 2009. Hypoxia, nitrogen, and fisheries: integrating ef- fects across local and global landscapes. Annu. Rev. Mar. Sci. 1:329-349. Bromaghin, J. F., M. M. Lance, E. W. Elliott, S. J. Jeffries, A. Acevedo-Gutierrez, and J. M. Kennish. Essington et al.: Shifts in the estuarine demersal fish community after a fishery closure in Puget Sound, Washington 215 2013. New insights into the diets of harbor seals (Phoca vitulina) in the Salish Sea revealed by analysis of fatty acid signatures. Fish. Bull. 111:12-26. Buckel, J. A., D. O. Conover, N. D. Steinberg, and K. A. McKown. 1999. Impact of age-0 bluefish ( Pomatomus saltatrix) predation on age-0 fishes in the Hudson River estuary: evidence for density-dependent loss of juvenile striped bass (. Morone saxatilis). Can. J. Fish. Aquat. Sci. 56:275-287. Burns, R. 1985. The shape and form of Puget Sound, 100 p. Wash- ington Sea Grant Program, Univ. Washington Press, Se- attle, WA. Carpenter, S. R., N. F. Caraco, D. L. Correll, R. W. Howarth, A. N. Sharpley, and V. H. Smith. 1998. Nonpoint pollution of surface waters with phos- phorus and nitrogen. Ecol. Appl. 8:559-568. Claudet, J., C. W. Osenberg, L. Benedetti-Cecchi, P. Domenici, J.-A. Garcfa-Charton, A. Perez-Ruzafa, F. Badalamenti, J. Bayle-Sempere, A. Brito, F. Bulleri, J.-M. Culioli, M. Dimech, J. M. Falcon, I. Guala, M. Milazzo, J. Sanchez-Meca, P. J. Somerfield, B. Stobart, F. Vandeperre, C. Valle, and S. Planes. 2008. Marine reserves: size and age do matter. Ecol. Lett. 11:481-489. Day, D. E. 1976. Homing behavior and population stratification in the central Puget Sound English sole (Parophrys vetu- lus). J. Fish. Res. Board Can. 33:278-282. Diaz, R. J. 2001. Overview of hypoxia around the world. J. Envi- ron. Qual. 30:275-281. Diaz, R. J., and R. Rosenberg. 1995. Marine benthic hypoxia: a review of its ecological effects and the behavioural responses of benthic macro- fauna. Oceanogr. Mar. Biol. Annu. Rev. 33:245-303. Doak, D. F., J. A. Estes, B. S. Halpern, U. Jacob, D. R. Lind- berg, J. Lovvorn, D. H. Monson, M. T. Tinker, T. M. Williams, J. T. Wootton, I. Carroll, M. Emmerson, F. Micheli, and M. Novak. 2008. Understanding and predicting ecological dynamics: Are major surprises inevitable? Ecology 89:952-961. Ebbesmeyer, C. C., and C. A. Barnes. 1980. Control of a fjord basin’s dynamics by tidal mix- ing in embracing sill zones. Estuar. Coast. Mar. Sci. 11:311-330. Essington, T. E., and C. E. Paulsen. 2010. Quantifying hypoxia impacts on an estuarine de- mersal community using a hierarchical ensemble ap- proach. Ecosystems 13:1035-1048. Guerry, A. D., M. H. Ruckelshaus, K. K. Arkema, J. R. Bern- hardt, G. Guannel, C.-K. Kim, M. Marsik, M. Papenfus, J. E. Toft, G. Verutes, S. A. Wood, M. Beck, F. Chan, K. M. A. Chan, G. Gelfenbaum, B. D. Gold, B. S. Halpern, W. B. La- biosa, S. E. Lester, P. S. Levin, M. McField, M. L. Pinsky, M. Plummer, S. Polasky, P. Ruggiero, D. A. Sutherland, H. Tallis, A. Day, and J. Spencer. 2012. Modeling benefits from nature: using ecosystem services to inform coastal and marine spatial planning. Int. J. Biodivers. Sci. Ecosyst. Serv. Manage. 8:107-121. Halpern, B. S. 2003. The impact of marine reserves: Do reserves work and does reserve size matter? Ecol. Appl. 13:S117-S137. Halpern, B. S., S. Walbridge, K. A. Selkoe, C. V. Kappel, F. Micheli, C. D’Agrosa, J. F. Bruno, K. S. Casey, C. Ebert, H. E. Fox, R. Fujita, D. Heinemann, H. S. Lenihan, E. M. P. Madin, M. T. Perry, E. R. Selig, M. Spalding, R. Steneck, and R. Watson. 2008. A global map of human impact on marine ecosys- tems. Science 319:948-952. Hart, J. L. 1973. Pacific fishes of Canada. Fish. Res. Board Can. Bulletin 180, 740 p. Heithaus, M. R., A. Frid, A. J. Wirsing, and B. Worm. 2008. Predicting ecological consequences of marine top predator declines. Trends Ecol. Evol. 23:202-210. Howard, S. M. S., M. M. Lance, S. J. Jeffries, and A. Acevedo-Gutierrez. 2013. Fish consumption by harbor seals (Phoca vituli- na) in the San Juan Islands, Washington. Fish. Bull. 111:27-41. Hurst, T. P, and D. O. Conover. 1998. Winter mortality of young-of-the-year Hudson Riv- er striped bass ( Morone saxatilis): size-dependent pat- terns and effects on recruitment. Can. J. Fish. Aquat. Sci. 55:1122-1130. Jeffries, S., J. Huber, J. Calambokidis, and J. Laake. 2003. Trends and status of harbor seals in Washington State: 1978-1999. J. Wildl. Manage. 67:207-218. Killick, R., and I. A. Eckley. 2011. Changepoint: an R package for changepoint anal- ysis. R package, vers. 0.6. lAvailable from http:// cran.r-project.org/web/packages/changepoint/, accessed December 2012.1 Kimmerer, W. J., J. H. Cowan, L. W. Miller, and K. A. Rose. 2000. Analysis of an estuarine striped bass ( Morone sax- atillis) population: influence of density-dependent mor- tality between metamorphosis and recruitment. Can. J. Fish. Aquat. Sci. 57:478-486. Landahl, J. T., L. L. Johnson, J. E. Stein, T. K. Collier, and U. Varanasi. 1997. Approaches for determining effects of pollution on fish populations of Puget Sound. Trans. Am. Fish. Soc. 126:519-535. Levin, P. S., and G. W. Stunz. 2005. Habitat triage for exploited fishes: Can we iden- tify essential “Essential Fish Habitat?” Estuar. Coast. Shelf Sci. 64:70-78. Levings, C. D., and R. M. Thom. 1994. Habitat changes in Georgia Basin: implications for resource management and restoration. Can. Tech. Rep. Fish. Aquat. Sci. 1948:330-349. Lotze, H. K., H. S. Lenihan, B. J. Bourque, R. H. Bradbury, R. G. Cooke, M. C. Kay, S. M. Kidwell, M. X. Kirby, C. H. Peter- son, and J. B. C. Jackson. 2006. Depletion, degradation, and recovery potential of estuaries and coastal seas. Science 312:1806-1809. MacCall, A. D. 1990. Dynamic geography of marine fish populations, 153 p. Univ. Washington Press, Seattle, WA. Mantua, N. J., S. R. Hare, Y. Zhang, J. M. Wallace, and R. C. Francis. 1997. A Pacific interdecadal climate oscillation with im- pacts on salmon production. Bull. Am. Meteorol. Soc. 78:1069-1079. McClanahan, T. R., A. J. Graham, M. A. MacNeil, N. A. Muthi- ga, J. E. Cinner, J. H. Bruggemann, and S. K. Wilson. 216 Fishery Bulletin 111(3) 2011. Critical thresholds and tangible targets for eco- system-based management of coral reef fisheries. Proc. Natl. Acad. Sci. USA 108:17230-17233. Moore, S. K., N. J. Mantua, J. P. Kellogg, and J. A. Newton. 2008. Local and large-scale climate forcing of Puget Sound oceanographic properties on seasonal to inter- decadal timescales. Limnol. Oceanogr. 53:1764-1758. Musick, J. A., M. M. Harbin, S. A. Berkeley, G. H. Burgess, A. M. Eklund, L. Findley, R. G. Gilmore, J. T. Golden, D. S. Ha, G. R. Huntsman, J. C. McGovern, S. J. Parker, S. G. Poss, E. Sala, T. W. Schmidt, G. R. Sedberry, H. Weeks, and S. G. Wright. 2000. Marine, estuarine, and diadromous fish stocks at risk of extinction in North America (exclusive of Pacific salmonids). Fisheries 25:6-30. Nichols, F. H. 2003. Interdecadal change in the deep Puget Sound ben- thos. Hydrobiologia 493:95-114. Nichols, F. H., J. E. Cloern, S. N. Luoma, and D. H. Peterson. 1986. The modification of an estuary. Science 231:567-573. Orr, J. W., and A. C. Matarese. 2000. Revision of the genus Lepidopsetta Gill, 1862 (Teleostei: Pleuronectidae) based on larval and adult morphology, with a description of a new species from the North Pacific Ocean and Bering Sea. Fish. Bull. 98:539-582. R Development Core Team. 2011. R: a language and environment for statistical com- puting. R Foundation for Statistical Computing, Vien- na, Austria. [Available from http://www.R-project.org.] Reum, J. C. P, and T. E. Essington. 2011. Season- and depth-dependent variability of a de- mersal fish assemblage in a large fjord estuary (Puget Sound, Washington). Fish. Bull. 109:186-197. Rodionov, S., and J. Overland. 2005. Application of a sequential regime shift detection method to the Bering Sea ecosystem. ICES J. Mar. Sci. 62:328-332. Russ, G. R., and A. C. Alcala. 1996. Marine reserves: rates and patterns of recov- ery and decline of large predatory fish. Ecol. Appl. 6:947-961. Taylor, J. C., W. A. Mitchell, J. A. Buckel, H. J. Walsh, K. W. Shertzer, G. B. Martin, and J. A. Hare. 2009. Relationships between larval and juvenile abun- dance of winter-spawned fishes in North Carolina, USA. Mar. Coast. Fish. 1:12-21. West, J. E„ S. M. O’Neill, and G. M. Ylitalo. 2008. Spatial extent, magnitude, and patterns of per- sistent organochlorine pollutants in Pacific herring (Clupea pallasi ) populations in the Puget Sound (USA) and Strait of Georgia (Canada). Sci. Total Environ. 394:369-378. Appendix table Complete list of species or taxonomic groups collected in Port Madison, Puget Sound, Washington during the period of 1991-2001, with total number of specimens collected, length range, and number of years that each was collected. * indicates that the maximum length could not be determined because of ambiguous identification of morphologically similar species. All lengths are fork length, except for Spotted Ratfish (precaudal length was measured for this species). Min. length Max. length Number Species Total number (mm) (mm) of years English Sole ( Parophrys vetulus) 10,427 11 470 20 Spotted Ratfish (Hydrolagus colliei) 4067 80 550 20 Blackbelly Eelpout ( Lycodes pcicificus) 3398 33 997 20 Pacific Tomcod (Microgadus proximus ) 2677 16 480 20 Slender Sole (Lyopsetta exilis ) 2242 20 340 20 Flathead Sole ( Hippoglossoides elassodon) 1665 60 342 20 Shiner Perch ( Cymatogaster aggregata) 1139 10 * 16 Pacific Hake ( Merluccius productus) 958 69 565 19 Speckled Sanddab ( Citharichthys stigmaeus ) 853 55 340 19 Rock soles (Lepidopsetta bilineata and L. polyxystra) 773 74 415 20 Sand Sole ( Psettichthys melanostictus ) 680 76 450 20 Plainfin Midshipman ( Porichthys notatus ) 497 58 326 17 Pacific Herring (Clupea pallasii ) 478 75 320 17 Dover Sole ( Microstomus pacificus) 366 45 381 19 Pacific Sanddab (Citharichthys sordidus ) 222 48 352 18 Walleye Pollock ( Theragra chalcogramma) 174 22 490 8 Pile Perch (Rhacochilus vacca) 170 95 211 12 Rex Sole (Glyptocephalus zachirus) 142 50 250 16 Snake Prickleback (Lumpenus sagitta) 99 50 395 16 Roughback Sculpin (Chitonotus pugetensis ) 89 50 145 14 Shortfin Eelpout (Lycodes brevipes) 86 29 80 10 Pacific Staghorn Sculpin (Leptocottus armatus) 60 73 258 15 Tubesnout ( Aulorhynchus flauidus) 33 64 165 11 Esssngton et al.: Shifts in the estuarine demersal fish community after a fishery closure in Puget Sound, Washington 217 Appendix table (cont.) Complete list of species or taxonomic groups collected in Port Madison, Puget Sound, Washington during the period of 1991-2001, with total number of specimens collected, length range, and number of years that each was collected. * indicates that the maximum length could not be determined because of ambiguous identification of morphologically similar species. All lengths are fork length, except for Spotted Ratfish (precaudal length was measured for this species). Species Total number Min. length (mm) Max. length (mm) Number of years Brown Rockfish ( Sebastes auriculatus) 33 65 325 12 C-0 Sole (Pleuronichthys coenosus ) 29 110 305 10 Slim Sculpin (Radulinus asprellus ) 22 80 160 9 Starry Flounder (Platichthys stellatus ) 19 110 410 12 Northern Anchovy ( Engraulis mordax ) 19 73 190 8 North Pacific Spiny Dogfish (Squalus suckleyi) 12 290 460 3 Bay Goby (Lepidogobius lepidus) 9 59 100 7 Sturgeon Poacher ( Podothecus accipenserinus ) 8 75 166 5 Longnose Skate ( Raja rhino) 8 365 640 7 Copper Rockfish (Sebastes caurinus ) 8 56 300 2 Sablefish (Anoplopoma fimbria) 6 320 400 1 Northern Spearnose Poacher (Agonopsis vulsa) 6 87 174 4 Eulachon ( Thaleichthys pacificus ) 6 89 150 4 Big Skate (Raja binoculata ) 6 125 1050 3 Soft Sculpin (Psychrolutes sigalutes) 5 25 45 3 Snailfishes (family Liparidae) 6 24 35 5 Chinook Salmon (Oncorhynchus tshawytscha) 5 78 96 2 Blacktip Poacher (Xeneretmus latifrons) 5 140 184 1 Pacific Sand Lance (Ammodytes hexapterus) 4 59 145 2 Butter Sole (Isopsetta isolepis ) 4 75 258 3 Sailfin Sculpin ( Nautichthys oculofasciatus) 3 120 121 3 Quillback Rockfish ( Sebastes maliger) 3 82 230 3 Curlfin Sole (Pleuronichthys decurrens) 3 161 260 2 Red Brotula ( Brosmophycis marginata) 3 220 305 1 Bigfin Eelpout ( Lycodes cortezianus) 3 260 280 1 Surf Smelt ( Hypomesus pretiosus) 2 86 125 2 Pacific Pompano (Peprilus simillimus) 2 110 149 2 Pacific Cod (Gadus macrocephalus) 2 420 610 2 Great Sculpin ( Myoxocephalus polyacanthocephalus) 2 220 395 2 Tadpole Sculpin ( Psychrolutes paradoxus) 1 30 30 1 Pygmy Poacher ( Odontopyxis trispinosa) 1 161 161 1 Pink Salmon ( Oncorhynchus gorbuscha) 1 66 66 1 Padded Sculpin (Artedius fenestralis) 1 125 125 1 Northern Ronquil (Ronquilus jordani) 1 165 165 1 Longspine Combfish (Zaniolepis latipinnis) 1 230 230 1 Cabezon (Scorpaenichthys marmoratus ) 1 410 410 1 Brown Irish Lord (Hemilepidotus spinosus) 1 90 90 1 American Shad (Alosa sapidissima) 1 274 274 1 218 Influence of soak time and fish accumulation on catches of reef fishes in a multispecies trap survey Email address for contact author: nate.bacheler@noaa.gov Abstract— Catch rates from fishery- independent surveys often are as- sumed to vary in proportion to the actual abundance of a population, but this approach assumes that the catchability coefficient (g) is constant. When fish accumulate in a gear, the rate at which the gear catches fish can decline, and, as a result, catch asymptotes and q de- clines with longer fishing times. We used data from long-term trap surveys (1990-2011) in the south- eastern U.S. Atlantic to determine whether traps saturated for 8 reef fish species because of the amount of time traps soaked or the level of fish accumulation (the total num- ber of individuals of all fish species caught in a trap). We used a delta- generalized-additive model to relate the catch of each species to a variety of predictor variables to determine how catch was influenced by soak time and fish accumulation after accounting for variability in catch due to the other predictor variables in the model. We found evidence of trap saturation for all 8 reef fish species examined. Traps became sat- urated for most species across the range of soak times examined, but trap saturation occurred for 3 fish species because of fish accumula- tion levels in the trap. Our results indicate that, to infer relative abun- dance levels from catch data, future studies should standardize catch or catch rates with nonlinear regres- sion models that incorporate soak time, fish accumulation, and any other predictor variable that may ultimately influence catch. Determi- nation of the exact mechanisms that cause trap saturation is a critical need for accurate stock assessment, and our results indicate that these mechanisms may vary considerably among species. Manuscript submitted 27 September 2012. Manuscript accepted 6 May 2013. Fish. Bull. 111:218-232 (2013). doi 10.7755/FB.111.3.2 The views and opinions expressed or implied in this article are those of the author (or authors) and do not necesarily reflect the position of the National Marine Fisheries Service, NOAA. Nathan M. Bacheler (contact author)1 Valerio Bartolino2' 3 Marcel J. M. Reichert4 1 Beaufort Laboratory Southeast Fisheries Science Center National Marine Fisheries Service, NOAA 101 Pivers Island Road Beaufort, North Carolina 28516 2 Swedish University of Agricultural Sciences Department of Aquatic Resources Lysekil, 45330, Sweden 3 Department of Earth Sciences University of Gothenburg Gothenburg, 40530, Sweden 4 Marine Resources Research Institute South Carolina Department of Natural Resources 217 Fort lohnson Road P.O. Box 12559 Charleston, South Carolina 29412 Robust fishery-independent survey data are a critical component of mod- ern fisheries stock assessments (Pen- nington and Stromme, 1998). Catch rates from fishery-independent sur- veys often are assumed to vary in proportion to the actual abundance of a fish population and, therefore, provide a relative measure of an- nual changes in abundance that can be used as a tuning index in a stock assessment (Kimura and Somerton, 2006). The basic assumption of this approach is that the catchability coefficient (q), or the efficiency of a fishery or survey gear, is constant over space, time, and over the range of environmental conditions encoun- tered in a survey (Hilborn and Wal- ters, 1992). It is also typically assumed that q is not influenced by the amount of time a particular fishing gear is fished (Hamley, 1975). When the rate at which a fishing gear catches fish declines as fish accumulate in it, the fishing gear becomes satu- rated and q declines as fishing times increase (Miller, 1979; Olin et al., 2004). Therefore, catch rates tend to increase asymptotically rather than proportionally with fish abundance, and at high levels of abundance, catch rates are an insensitive indica- tor of change (Ricker, 1975). Numer- ous mechanisms have been shown to cause gear saturation, which can be broadly categorized as space limita- tion of gear, increased gear avoid- ance, interspecific competition, bait degradation or consumption of bait, or fishing gear that causes local de- pletion of fish (Kennedy, 1951; Rich- ards et al., 1983; Olin et al., 2004). Depending on the exact mechanism that causes gear saturation, the catch at which a fishing gear be- comes saturated may or may not re- flect actual abundance (Beverton and Holt, 1954). Although saturation in gill nets, longlines, and trawl nets has been well studied (Ragonese et al., 2001; Olin et al., 2004; Rodgveller et al., 2008), there has been a paucity of empirical research on the presence of trap saturation. Traps are widely used, especially in sensitive habi- Bacheler et al.: Influence of soak time and fish accumulation on the catches of reef fishes 219 tats (e.g., seagrass meadows and coral reefs), and are sometimes the only feasible method of sampling in these habitats because of their relatively low effect on substrate and benthic communities (Miller, 1990). Clearly, the catch of fishes or invertebrates in traps cannot continue to increase linearly with soak time because the space inside a trap is finite and will eventually become filled with animals to the point at which no additional individuals can enter (Ben- nett, 1974; Austin, 1977; Miller, 1990). Models have been developed to describe the relationship between catch per trap and soak time, with the intention of using those models to standardize catch for differ- ent soak times (Munro, 1974; Somerton and Mer- ritt, 1986; Zhou and Shirley, 1997). Unfortunately, these approaches to standardization do not account for landscape (e.g., depth) or environmental effects (e.g., water temperature) on catch. Nor do they help us understand how the catch of one species may be influ- enced by the catches of other species. One recent method to examine these drawbacks has been to model the catch of a species as a func- tion of soak time and fish accumulation (i.e., the to- tal number of individuals of all species caught by a fishing gear; Olin et al., 2004) after accounting for other variables that may influence catch. In other words, catch rates can first be standardized by all of the predictor variables in the model building process (Lo et al., 1992; Maunder and Punt, 2004), and then the specific effects of soak time and fish accumulation can be extracted and examined independently of other predictor variables (Li et al., 2011). For example, Li et al. (2011) showed that gill nets became saturated with Yellow Perch ( Perea flavescens ) because of the total number of individuals caught in a gill net but not because of increased soak times. These results in- dicate that Yellow Perch catch rates decline when, for instance, this species sees fish already caught in the gill net, and not for reasons associated purely with increased soak time (i.e., when all the Yellow Perch in an area are caught, which takes some time). In our study, we used a standardized catch approach to examine the influence of soak time and fish accumu- lation on the catches of several reef fish species from long-term fishery-independent, multispecies trap sur- veys occurring in the southeastern U.S. Atlantic (SEUS) from North Carolina to Florida. The inclusion of soak time and fish accumulation separated mechanisms that cause gear saturation into 2 groups: those mechanisms related to fish accumulation (e.g., agonistic behaviors or bait depletion) and those mechanisms related to the length of time a trap soaks (e.g., local depletion of the target species or loss of bait freshness). We developed a delta-generalized-additive model (delta-GAM) that was able to accommodate both nonlinearities between the response and predictor variables and zero-inflation (i.e., a high proportion of zero catches; Martin et al., 2005). This approach allowed us to determine whether trap saturation occured because of either soak time or fish accumulation, or both, and then we used the model to predict relative abundance after accounting for the influence of soak time and fish accumulation. Materials and methods Study area In our study reef fish species associated with hard substrates were sampled on the continental shelf and continental shelf-break in the SEUS. The continental shelf and shelf-break in the SEUS are dominated by sand and mud substrates, within which areas of hard, rocky substrates (“hard bottom”) occur and a highly diverse reef fish assemblage associates. Hard bottom habitats range in complexity from flat limestone pave- ment, sometimes covered with a sand or gravel veneer, to high-relief rocky ledges (Schobernd and Sedberry, 2009; Glasgow, 2010). Hard bottom areas often host di- verse epifauna that can provide food and shelter for reef fishes. The major oceanographic feature of the SEUS is the Gulf Stream, which influences outer sec- tions of the continental shelf as it flows northward. Consistently warm Gulf Stream waters along the outer SEUS shelf allow tropical and subtropical species to inhabit areas at least as far north as North Carolina (Miller and Richards, 1980). For our study, sampling occurred on continental shelf and shelf break habitats from approximately Cape Lookout, North Carolina, to St. Lucie Inlet, Florida (Fig. 1). Sampling approach The Marine Resources Monitoring, Assessment, and Prediction (MARMAP) Program of the South Carolina Department of Natural Resources has used chevron fish traps to index reef fish abundance since the late 1980s. Since 2009, MARMAP funding for reef fish sam- pling has been supplemented by the cooperative South- east Area Monitoring and Assessment Program — South Atlantic (SEAMAP-SA) administered by the National Marine Fisheries Service. We analyzed MARMAP data from 1990 through 2011, during which time sampling with chevron fish traps was conducted in a consistent manner (as described later in this section). We also in- cluded in our analyses 2010-11 data from the South- east Fishery-independent Survey (SEFIS), which the National Marine Fisheries Service created in 2010 to increase fishery-independent sampling in the SELTS, because sampling methods were identical. Hereafter, the 2 sampling programs are referred to as “MARMAP/ SEFIS.” Hard bottom sampling stations included in the anal- yses were selected for sampling in 1 of 3 ways. First, most sites were selected randomly from the MARMAP/ SEFIS sampling frame that consisted of approximately 2000 sampling stations on hard bottom habitat. Second, some stations in the sampling frame were sampled op- 220 Fishery Bulletin 111(3) Figure 1 Spatial distribution of samples of reef fish species (black circles) collected in chevron traps during 2 long-term, fishery-independent survey programs in the southeastern U.S. Atlantic — the Marine Resources Monitoring, Assessment, and Prediction Program (1990- 2011) and the Southeast Fishery-independent Survey (2010-11) — for examination of the relationship between soak time, fish accu- mulation, and catch of 8 reef fish species. Note that black circles overlap in many instances. Gray lines indicate 35- and 70-m depth contours (derived from bathymetry data), and arrows indicate the approximate path of the Gulf Stream. portunistically even though they were not selected ran- domly for sampling in a given year. Third, new hard- bottom stations were added during the study period through the use of information from fishermen, charts, and historical surveys. These new locations were in- vestigated with a vessel echosounder or drop cameras and sampled if hard bottom habitat was present. We assumed that the catch of each species was influenced similarly at all stations by the various predictor vari- ables (described later). All sampling for this study oc- curred during daylight hours between March and Octo- ber and was conducted on 1 of 4 vessels: MARMAP and SEAMAP-SA used the RV Palmetto (1990-2011), and SEFIS used the RV Savannah (2010-11), NOAA Ship Nancy Foster (2010), and NOAA Ship Pisces (2011). Chevron fish traps were deployed at each sta- tion sampled in this study. Chevron traps were constructed from plastic-coated, galvanized 2-mm-diameter wire (mesh size=3.4 cm2) and shaped like an arrowhead that measured 1.7 mxl.5 mx0.6 m, with a total volume of 0.91 m3 (Fig. 2) (Collins, 1990). The mouth openings of traps were shaped like a teardrop and measured approximately 18 cm wide and 45 cm high. Each trap was baited with 24 menhaden ( Brevoor - tia spp.): 16 were attached to freely accessible stringers and 8 were placed loosely inside. Traps typically were deployed in a group of 6. The mini- mum distance between individual traps was ap- proximately 200 m to provide some measure of independence between traps. Because the primary purpose of MARMAP/SE- FIS sampling was to provide standardized catch information for reef fish species in the SEUS, a soak time of 90 min was targeted for each trap. We were not able to soak traps for a wide range of experimentally chosen amounts of time. How- ever, for many different reasons, soak time was somewhat variable, ranging from 9 to 270 min (mean: 97.6 min and standard deviation of the mean [SD] = 12.8) (Fig. 3). All trap deployments that did not fish properly (e.g., traps that dragged in current) were excluded from analysis. Soak times were variable enough to allow an examina- tion of the ways in which fish catch was related to variability in soak time. We included in our analyses the 8 most com- monly caught reef fish species in the MARMAP/ SEFIS trap surveys: Black Sea Bass ( Centropris - tis striata), Tomtate ( Haemulon aurolineatum), Red Porgy ( Pagrus pagrus), Bank Sea Bass ( Cen - tropristis ocyurus), Gray Triggerfish (Batistes ca- priscus), Vermilion Snapper ( Rhomboplites auror- ubens ), Stenotomus spp., and Sand Perch (Diplec- trum formosum) (Table 1). Stenotomus spp. may represent more than one species, but, for the pur- pose of discussion, we will refer to this taxon as a single species and the group of taxa studied as 8 species of fish. Fish coloration, shape, and meristics were used to identify individuals to genus and species levels with field guides (e.g., Robins et al., 1986; Hoese and Moore, 1998; McEachran and Fechhelm, 1998; Car- penter, 2002; Humann and Deloach, 2002; McEachran and Fechhelm, 2005). Black Sea Bass, Red Porgy, Gray Triggerfish, and Vermilion Snapper are targeted by commercial and recreational fishermen in the SEUS. The 8 species included in our analyses were the most common species caught in the traps in the MARMAP/ SEFIS surveys by both frequency of occurrence and mean catch per trap (Table 2). Additional species were not analyzed if their frequency of occurrence was less Bacheler et al.: Influence of soak time and fish accumulation on the catches of reef fishes 221 Figure 2 Schematic of the design of the chevron trap used to sample reef fish species in 1990-2011 by the Marine Resources Monitoring, Assessment, and Prediction Pro- gram and the Southeast Fishery-independent Survey. The gray oval is the mouth opening of the trap (-18 cm wide and 45 cm high), and the total trap volume is 0.91 m3. than 15%, but individuals of all species were included in analyses of fish accumulation. Data analyses We used a generalized additive modeling approach to test the hypothesis that trap catch of reef fishes was influenced by soak time and fish accumulation. GAMs use nonparametric smoothing functions to account for nonlinearities between predictor and response vari- ables (Hastie and Tibshirani, 1990; Bacheler et ah, 2009). GAMs extend traditional additive models by al- lowing for alternative distributions of underlying ran- dom variation, just as generalized linear models allow for alternative distributions in linear models. We developed delta-GAMs because there was a high proportion of zero observations (>50%) for the catch of each species that could not be modeled appropriately with standard statistical distributions. We considered zero-inflated models, but they were challenging to work with because they rarely converged, and when they did, model solutions were often unreasonable. Therefore, we developed a delta-GAM for each of the 8 species. Each delta-GAM contained 2 submodels that estimated the influence of soak time and fish accumulation on catch: 90 110 130 150 soak 50 100 fishacc 150 doy i i i i i i i r 27 28 29 30 31 32 33 34 35 I at temp tod Figure 3 Histograms of predictor variables used in the delta-generalized-additive models of reef fish catch from trap surveys con- ducted by the Marine Resources Monitoring, Assessment, and Prediction Program and the Southeast Fishery-independent Survey in the Atlantic Ocean from North Carolina to Florida during 1990-2011. Variables were soak time (soak; min), fish accumulation (fishacc; total fish per trap), year of sampling (year), day of the year (doy), latitude (lat; °N), depth of sampling (depth; m), bottom water temperature (temp; °C), and time of day (tod; Coordinated Universal Time). 222 Fishery Bulletin 111(3) Table 1 Life history characteristics of the 8 fish species analyzed in our study of soak time and fish accumulation as mechanisms that can cause trap saturation. Characteristics come from data sets of 2 sampling programs in the southeastern U.S. Atlantic: the Marine Resources Monitoring, Assessment, and Prediction Program (1990-2011) and the Southeast Fishery-independent Survey (2010-11). LM=maximum length; LmaturityGength at maturity; Amax=maximum age in years; f=fish; i=invertebrates; PH=protogynous hermaprodite; SH=sequential hermaphro- dite; GO=gonochoristic. All lengths are total lengths in centimeters. Common name Scientific name hoc -^maturity ■^max Diet Reproduction Bank Sea Bass Centropristis ocyurus 33° 14° 9“ f,i“ PH“ Black Sea Bass Centropristis striata 50a 16“ ii“ f,F PH“ Gray Triggerfish Batistes capriscus 48a 19“ 12“ if GO“ Red Porgy Pagrus pagrus 51a 25“ 20“ f,ie PH“ Sand Perch Diplectrum formosum 24* 19* 8* f,i' SH“ Tomtate Haemulon aurolineatum 35“ 18“ 17“ i d GO“ Vermilion Snapper Rhomboplites aurorubens 51a 15s 13“ f,i* GO“ Stenotomus spp. 46' 18' 15> i' GO“ Sources: “ MARMAP (unpubl. data); b Sedberry (1988); c Manooch and Barans (1982); d Sedberry (1985); e Manooch (1977); f Kurz ( 1995); g Zhao et al. ( 1997); h Sedberry and Cuellar ( 1993); ' O’Brien et al. (1993);-) Finkelstein ( 1969); * Bubley and Pashuk (2010); 1 South Atlantic Fishery Management Council (http://www.safmc.net/). one modeling the presence-absence of each species and another modeling the positive catches only (Lo et al., 1992; Pennington, 1996; Stefansson, 1996). The overall effects of a particular predictor variable on catch were then obtained by multiplying the effects from each sub- model (Maunder and Punt, 2004; Murray, 2004; Li et al., 2011). We examined the influence of 8 predictor variables on the catch of 8 reef fish species: soak time, fish ac- cumulation, year, depth, time of day, day of the year, water temperature, and latitude (Fig. 3). In this study, we were particularly interested in how soak time and fish accumulation influenced the catch of each species after accounting for variability in the other 6 predictor variables. To accomplish this goal, we first fitted the delta-GAM, and then we predicted catch for a range of values for soak time and fish accumulation, fixing all other predictor variables to their own mean, ex- cept the variable year, which was fixed at year 2000 (i.e., the midpoint of the time series; results were in- variant to the year chosen). Soak time (soak) was the number of minutes a trap soaked between deployment and retrieval, and fish accumulation ( fshacc ) was the total number of individuals of all fish species caught Table 2 Catch information, mean length, and frequency of occurrence (FO) for each of the 8 most common species of reef fishes in the data sets of 2 sampling programs in the southeastern U.S. Atlantic: the Marine Resources Monitoring, Assessment, and Predic- tion Program (1990-2011) and the Southeast Fishery-independent Survey (2010-11). Mean proportion of catch is the mean proportion of catch in each trap that was com- posed of a single species. Fork length was measured for all species, except Black Sea Bass (Centropristis striata) and Bank Sea Bass (C. ocyurus ), which were measured for total length. SD=standard deviation of the mean. Species FO Mean (SD) catch per trap Mean (SD) proportion of catch Mean (SD) length (cm) Bank Sea Bass 0.301 1.4 (3.9) 0.06 (0.17) 22.5 (3.0) Black Sea Bass 0.408 10.2 (21.8) 0.20 (0.30) 23.6 (4.5) Gray Triggerfish 0.263 0.9 (3.3) 0.06 (0.18) 31.3 (7.0) Red Porgy 0.364 2.0 (4.4) 0.13 (0.26) 27.1 (4.6) Sand Perch 0.184 0.6 (1.8) 0.03 (0.13) 22.8 (1.6) Stenotomus spp. 0.203 6.5 (19.8) 0.09 (0.20) 15.6 (2.0) Tomtate 0.403 9.3 (21.9) 0.18 (0.28) 18.1 (1.9) Vermilion Snapper 0.263 3.0 (9.9) 0.07 (0.17) 23.7 (3.9) Bacheler et al.: Influence of soak time and fish accumulation on the catches of reef fishes 223 in a trap. We limited our analysis to soak between 50 and 150 min and to fishacc less than 200 total fish per trap because of low small sample sizes outside of these ranges (i.e., <2% of all observations). Year was included as a categorical variable (1990-2011), but all other variables were included as continuous vari- ables. Depth (depth) was measured in meters for each trap deployment; depths were recorded in a range of 13-218 m over the course of the surveys used in our study. Because of small sample sizes, samples collected at depths >100 m were excluded from our analyses and remaining depths were log-transformed to achieve normality. Time of day (tod) was measured in Coordi- nated Universal Time, and day of the year (doy) was the day of the year that the trap sample was collect- ed. Water temperature (temp) was bottom water tem- perature measured in degrees Celsius for each group of 6 simultaneously deployed traps, and latitude (lat) was the latitude (degrees north) at which the samples were collected. Longitude was not included because of its statistically significant correlation (PcO.OOl) with depth that occurred because of the north-south orien- tation of our study area. Before the development of models, multicollinearity among predictor variables was examined because its presence can cause erratic model behavior and should be avoided (Zar, 1999). We assessed the severity of multicollinearity among predictor variables through calculation of the variance inflation factor (VIF) for each variable, which measures the amount of variance that is inflated for each variable as a result of its col- linearity with other predictor variables. The VIF for all predictor variables was less than 4.0 — below the level generally acknowledged to be problematic (5-10; Neter et al., 1989) — thus indicating no significant multicollinearity among predictor variables in our data set. We also included fishacc as a predictor variable in our GAMs. Because trap catch often was composed of a mixture of fish species in the multispecies survey, the contribution of a single species to the fishacc variable was generally small, but there were instances when the catch in traps was dominated by a single species. Inclu- sion of samples in which the catch was dominated by a single species may have positively biased the reported deviance of the models for those particular species, but the functional relationship between single species catch and fishacc was not affected. If catch of a single species was influenced entirely by fishacc, the relation- ship between the 2 variables would have been perfectly linear. By definition, then, any deviation from a linear relationship between the 2 variables could not have been the result of a potential lack of independence. Be- cause we were primarily interested in the shape of the relationship between catch and fishacc, we agree with Li et al. (2011) that the inclusion of the fishacc variable is a useful approach to examine trap saturation due to fish accumulation. Initially, a full model was fitted on the presence-ab- sence data for each species. Following Li et al. (2011), we used a binomial GAM submodel to estimate the probability of presence for each species being caught in individual traps (rj), which was assumed to be an independent draw from a binary variable with a prob- ability of success p: E(p) = v~1(r\); (1) N r| = a + g1(soak) + g2(fishacc)+ £ Sj^xj\ j= 3 where E(p) v a soak fishacc Sj N the expectation of p; the logit link function; the intercept; soak time; fish accumulation; are smoothing functions, the number of predictor variables in the model; and the yth remaining explanatory variable. We next coded a positive-catch GAM submodel that related the Gaussian fourth-root transformed catch of each reef fish species when caught to the 8 predic- tor variables. We compared the error structure of log- normal, log-gamma, and Gaussian (with a fourth-root transformation) distributions using the Akaike infor- mation criterion (AIC; Burnham and Anderson, 2002) for each full model: AIC = -21og(z(0|y)) + 2A, (2) where = the log-likelihood; and = the number of parameters of each model . The model with the lowest AIC value was considered the best model in the model set. For all species, the Gaussian distributions with fourth-root transforma- tions had the lowest AIC values and were therefore considered the most parsimonious distributions for the positive-catch submodels. For the positive-catch GAM submodel, we used the following equation: AT y0'25 = a' + h^soak) + h2( fishacc) + £ A.y( JCy ), (3) 7=3 where y = the trap catch of a particular reef fish species; a' = the intercept; hj are smoothing functions; N = the number of predictor variables in the model; and Xj = the yth remaining explanatory variable. For each reef fish species, we then compared the full GAM submodels containing 8 predictor variables to various reduced models that contained fewer predic- tor variables. We compared various binomial GAM sub- models with the unbiased risk estimator (LIBRE) score. 224 Fishery Bulletin 111(3) Table 3 Final generalized additive models (GAMs) for 8 species of reef fishes in our study. Data were obtained from 2 sampling pro- grams in the southeastern U.S. Atlantic: the Marine Resources Monitoring, Assessment, and Prediction Program (1990-2011) and the Southeast Fishery-independent Survey (2010-11). Binomial GAMs were constructed with presence-absence data, and Gaussian GAMs were constructed with only positive catch. The best model for each species was the one with the low- est unbiased risk estimator (binomial GAM) or generalized cross validation (Gaussian GAM) scores (see the Materials and methods section for full descriptions). A/=number of samples from chevron traps that were included in each model. Dev. exp.= the percentage of deviance explained by each model, ex means that the covariate was excluded from the final model, gi... g7 are nonparametric smoothing functions, f= a categorical function, soa&=soak time, fishacc= fish accumulation, year= year, doy= day of the year, fat=latitude, depth= bottom depth, temp =bottom temperature, and fod=time of the day. Estimated degrees of freedom and statistical significance are shown for each term: *=P<0.10, **=P< 0.05, * **=P<0.01. Model and species N Dev. exp. gx (soak) g2 (fishacc) fx (year) g3(doy) gi(lat) g^depth) gs(temp) gfitod) Binomial GAM Bank Sea Bass 8530 28.8 ex 3.0*** 21*** 8.6*** 8.1*** 8.9*** 6.3 8.3*** Black Sea Bass 8530 63.6 ex 3.0*** 21*** 2.6**’ 8.6*** 8.2*** g ^*** 1.0** Gray Triggerfish 8530 20.6 5.5*** 3.0*** 21*** 8.9*** 8.8*** 5.8*** 4.8*** 8.1 Red Porgy 8530 27.0 1.6* 3.0*** 21*** 8.0*** 8.9*** rj g*** 8.5*** 5.1" Sand Perch 8530 37.5 1.7*** 6.8*** 21*** ex 7.0*** rj g*** 6.1 1.0*** Stenotomus spp. 8530 61.4 3.3*** 7.4*** 2!*** ex 5.9*** 6.0*** rj q*** 8.1*“ Tomtate 8530 46.9 5.6*** 8.2*** 2 2*** 2.8’** 8.9*** 8.3*** 7.1**’ 2.0* Vermilion Snapper 8530 38.0 1.6 8.6*** 21*** 1.0*** g 2*** rj ^ *** 7.!*** 8.2*** Gaussian GAM Bank Sea Bass 2571 22.1 ex 2.9*** 2!*** 8.9*** 8.2*** 8.5*’* 6.8*** 3.9*** Black Sea Bass 3476 64.4 ex 8.1*** 21‘** rj q*** 8.6*** 7.2*** 2 g*** 4.8“* Gray Triggerfish 2244 18.9 2.3** 2.9*** 21*** 1.0*** 7.5*** 7.4*** 1.0 7.7*** Red Porgy 3104 21.4 3.4’ 2.0*** 21*‘* 2.1 8.8*** 8.8*** 2.2 1.0*’ Sand Perch 1568 26.5 ex 2.8*** 21*** 5.3 8.9*** 2.7*** 3.4* ex Stenotomus spp. 1733 48.6 4.0* 4.0*** 21*** 1.0 8.8*** 7.9*** 5.4** 4.8** Tomtate 3437 51.2 ex 7.0*** 21*** 6.2** 8.8*** 8.8*** 7.7*** 4.4** Vermilion Snapper 2240 36.1 ex 6.4*** 2!*** ex 8.0*** 5.4*** 1.0 ex which is effectively a rescaled AIC approach that is well suited for binomial models (Wahba, 1990). For the positive-catch GAM submodels, we used generalized cross validation (GCV; a measure of the out-of-sample prediction mean squared error) to select the most par- simonious combination of predictor variables. For each approach, the model for each species with the smallest UBRE or GCV score was selected as the best model in that particular model set. In addition, we evaluated the model diagnostics for each final model selected by UBRE or GCV. In all cases, residuals in final models met assumptions of normality and constant variance. All models were coded and analyzed in R,1 vers. 2.14.1 (R Development Core Team, 2011) with the mgcv li- brary, vers. 1.7-13 (Wood, 2008). We used 2 methods to test for the presence or ab- sence of spatial autocorrelation, which is the situation where samples near one another are often more simi- lar than 2 samples farther apart. First, we developed generalized additive mixed models (GAMMs) for each species with the same covariates as the GAM models 1 Mention of trade names or commercial companies is for identifica- tion purposes only and does not imply endorsement by the National Marine Fisheries Service, NOAA. presented previously. GAMMs are spatially explicit regression models that allow for spatially correlated error distributions (Venables and Ripley, 2002). Us- ing positive-catch data, we found that the coefficient of multiple determination ( R 2) and parameter signifi- cance values from the GAMMs were nearly identical to GAM model results for all 8 species. Binomial GAMMs built on presence-absence data never converged for any species. Second, we developed semivariograms for each species for each year using the R package geoR, vers. 1.7-2 (R Development Core Team, 2011). There were no consistent patterns in the relationship between the semivariance of the model residuals and distance be- tween sampling points, indicating negligible spatial autocorrelation in the residuals. The overall influence of soak or fishacc on reef fish catch was calculated as the product of the binomial and positive-catch submodels, and the variance of the overall model was estimated with a bootstrapping ap- proach. We resampled the predictions (N= 10,000) for both submodels at average values of all other predictor variables according to the pointwise estimates of error that were assumed to be distributed normally. For the combined (overall) predictions, we multiplied the simu- lated point estimates of error for each submodel. The Bacheler et al.: Influence of soak time and fish accumulation on the catches of reef fishes 225 95% confidence interval was estimated as the 0.025 and 0.975 quantiles of the 10,000 point estimates. Results The assumption of independence of predictor variables was met in our study. There were no statistically sig- nificant relationships between any of the predictor variables included in our analyses (VIF<4), and, in particular, soak and fshacc variables were unrelated (r2<0.001). Fishacc was the total number of individuals of all species caught in a trap, whereas the response variable (i.e., catch) was the total number of individu- als of each species caught in a trap. Moreover, no single species composed more than 20% of total trap catch (Table 2). Overall, 8530 samples collected from chevron traps in the 22-year span of MARMAP/SEFIS surveys were included in our analyses (annual mean: 388 [SD 179]). Sampling occurred between March and October, and there were no obvious changes in the seasonality of sampling across years. The spatial coverage of the surveys, in contrast, expanded southward in the mid- 1990s to include sampling sites in central Florida. Of the 8 reef fish species included in our analyses, Black Sea Bass had the highest frequency of occurrence (0.408) and Sand Perch had the lowest (0.184; Table 2) . Unstandardized catch per trap ranged from 0.6 for Sand Perch to 10.2 for Black Sea Bass. Mean lengths ranged from 15.6 cm fork length for Stenotomus spp. to 31.3 cm fork length for Gray Triggerfish (Table 2). The binomial GAMs explained 20.6-63.6% of the deviance in presence-absence patterns of the 8 reef fish species (Table 3). Models that explained the least deviance were the ones for Gray Triggerfish (20.6%) and Red Porgy (28.8%), and models for Black Sea Bass (63.6%) and Stenotomus spp. (61.4%) explained the most deviance. All 8 predictor variables were selected in the binomial model for each species on the basis of UBRE scores, with the exceptions of soak for Black Sea Bass and Bank Sea Bass and doy for Stenotomus spp. and Sand Perch (Table 3). The fishacc variable was re- tained in the binomial models for all species. The Gaussian GAMs explained 18.9-64.4% of the deviance in the positive-catch values of the 8 reef fish species (Table 3). The most deviance was explained by the models for Black Sea Bass (64.4%) and Tomtate (51.2%), and models for Gray Triggerfish (18.9%) and Red Porgy (21.4%) explained the least deviance (Table 3) . On the basis of GCV scores, soak was excluded from the models for Black Sea Bass, Bank Sea Bass, Tom- tate, Vermilion Snapper, and Sand Perch. Moreover, cloy was excluded from the Vermilion Snapper model, and tod was excluded from the models for Vermilion Snap- per and Sand Perch. The fishacc variable was included in all 8 models. Over the range of soak values examined, predicted overall catch was invariant to soak for 3 species (Black Sea Bass, Bank Sea Bass, and Vermilion Snapper), posi- tively related to soak for 2 species (Red Porgy and Gray Triggerfish), and negatively related to soak for the re- maining 3 species (Tomtate, Stenotomus spp., and Sand Perch; Fig. 4). For most species, the influence of soak on the probability of obtaining nonzero catch was very similar to its influence on the estimated overall catch when present (Fig. 4). With a doubling of soak from 60 to 120 minutes, estimated catch approximately doubled for Red Porgy (106% increase) and Gray Triggerfish (95%) but increased little for Black Sea Bass (20%), Bank Sea Bass (9%), and Vermilion Snapper (8%; Table 4). The estimated overall catch of Sand Perch (—71%), Tomtate (-32%), and Sand Perch (-26%) declined when soak doubled (Table 4). Note that confidence intervals were larger for Gray Triggerfish and Stenotomus spp. than for the other 6 species. The relationship between the overall catch of the 8 reef fish species and fishacc displayed one of 3 patterns (Fig. 5). The overall catch of Black Sea Bass, Tomtate, Vermilion Snapper, and Sand Perch generally was re- lated linearly to fishacc, indicating that the rate of catch of these species was not strongly influenced by the catch of individuals of all species in the trap. Alter- natively, the overall catch of Red Porgy, Bank Sea Bass, and Gray Triggerfish reached an asymptote at fishacc values between 50 and 100, indicating that individu- als of these 3 species were much less likely to enter a trap once 50 to 100 total individuals of all species were caught in it. The last pattern was displayed by Stenoto- mus spp., the overall catch of which was exponentially related to fishacc, indicating that Stenotomus spp. were more likely to enter a trap once large numbers of indi- vidual of all species were caught (Fig. 5). All 8 species had relatively narrow 95% confidence intervals sur- rounding overall mean estimates. Discussion Fishery-independent survey data form a critical com- ponent of modern stock assessments because they pro- duce indices of abundance that are assumed to vary in proportion to the actual abundance of a population (Pennington and Stromme, 1998). Whether or not indi- ces of abundance track actual abundance is a complex topic that has been the subject of much research (Har- ley et al., 2001; Kimura and Somerton, 2006). The basic assumption of this approach is that q does not change over space, time, or environmental conditions (Hilborn and Walters, 1992). In our study, we tested whether q was influenced by 2 additional variables, the length of time the trap was soaked and fish accumulation. For the species in our study, there was no clear relation- ship between life history traits and mechanisms of trap saturation (Table 1). We found evidence of trap satura- tion for most of the reef fish species examined, but the responses were species-specific. Trap saturation was observed for some reef fishes across the range of soak Probability of obtaining nonzero catch 226 Fishery Bulletin 111(3) Probability of presence T 1 1 T 90 110 130 150 Positive catch only Black Sea Bass Tomtate Red Porgy Bank Sea Bass o ro o ■o 0) ro E Vermilion Snapper i 1 1 1 1 r 50 70 90 110 130 150 Overall Soak time (min) Figure 4 Relationship between soak time and the probability of obtaining a nonzero catch (probability of presence) (left column), estimated catch when present (middle column), and estimated overall catch (right column) for 8 spe- cies of reef fishes on the basis of data from trap surveys conducted in the southeastern U.S. Atlantic by the Marine Resources Monitoring, Assessment, and Prediction Program (1990-2011) and the Southeast Fishery- independent Survey (2010-11). Binomial generalized additive models (GAMs) were used to estimate probabil- ity of obtaining nonzero catch. Gaussian GAMs were used to estimate catch when present. Overall catch was determined through the combination of binomial and Gaussian GAM estimates. Dashed lines indicate 95% confidence intervals. Bacheler et al.: Influence of soak time and fish accumulation on the catches of reef fishes 227 Table 4 Percent change in estimates of overall catch per trap for 8 reef fish spe- cies when soak time doubled (from 60 to 120 min) on the basis of data from 2 sampling programs in the southeastern U.S. Atlantic: the Marine Resources Monitoring, Assessment, and Prediction Program (1990-2011) and the Southeast Fishery-independent Survey (2010-11). Catch was es- timated with a delta-generalized-additive model where mean values of all other predictor variables were used. Species are arranged from high- est to lowest percent change in catch per trap. Species Estimated catch at 60 min Estimated catch at 120 min Percent change Red Porgy 1.25 2.57 106 Gray Triggerfish 0.80 1.56 95 Black Sea Bass 6.45 7.71 20 Bank Sea Bass 3.51 3.83 9 Vermilion Snapper 2.79 3.02 8 Sand Perch 1.67 1.24 -26 Tomtate 6.78 4.58 -32 Stenotomus spp. 0.99 0.29 -71 times examined, and, for others, be- cause of fish accumulation in the trap. A number of methods have been de- veloped to determine if gear saturation is occurring. Addison and Bell (1997) used a simulation approach to show that the relationship between lobster catch and abundance was asymptotic, a prob- lematic result because models would predict an even spatial distribution of lobster catches across a study area de- spite a true underlying aggregated dis- tribution. Some researchers have docu- mented gear saturation by the fact that the cumulative catch of individuals in traps that are periodically emptied is of- ten much higher than the catch in traps that were hauled and redeployed with- out being emptied (Miller, 1979; Robert- son, 1989). Alternatively, Li et al. (2011) developed a delta-GAM to quantify the relationship between catch of Walleye (Sander vitreus) and Yellow Perch and soak time or fish accumulation at average values of all other covariates in the model. We used the Li et al. (2011) modeling approach to show that chevron traps became saturated for all 8 reef fish species examined across a range of values for soak time or fish accumula- tion. There are 3 major benefits of this approach. First, it is possible to test for trap saturation through the use of long-term survey data, as long as there has been suf- ficient variation in soak time. Second, the relationship between catch and soak time or fish accumulation can be quantified after accounting for variation due to the other predictor variables in the model. Last, zero-in- flation, the situation where a large number of zero ob- servations in a data set cannot be properly accounted for with traditional statistical distributions, can be properly accounted for through the use of a delta model. Our study was improved by the inclusion of the fish- acc variable in the models. If the rate at which a spe- cies entered the trap was unaffected by the number of individuals (of all species) already caught in a trap, then one would expect a positive, linear relationship between the catch of a species and the fishacc variable (Li et al., 2011). We showed that catches of Red Porgy, Bank Sea Bass, and Gray Triggerfish plateaued once a moderate number of total individuals were already caught in a trap, indicating that these species are more sensitive to species interactions and, therefore, much less likely to enter a trap once it began filling up. These results are consistent with previous work that has shown that behavioral interactions in and around traps can strongly influence the catch of target species (Addison and Bell, 1997; Jury et al., 2001; authors, per- sonal observ. ). In contrast and, perhaps, more surpris- ingly, Black Sea Bass, Tomtate, Vermilion Snapper, and Sand Perch continued to enter a trap at about the same rate no matter how many total individuals of all spe- cies were caught in it. A primary benefit of inclusion of a predictor variable for fish accumulation in our model was that it allowed us to distinguish between species that saturated the gear because of fish accumulating in a trap from the species that appeared to saturate the gear because of the amount of time a trap soaked. In addition, the inclusion of the fishacc variable stan- dardized the catch of each of the 8 reef fish species to a common total catch of all species in the trap. In other words, we were able to remove variations in the catch of each reef fish species that were attributable to species-specific responses to fish accumulation. The inclusion of a variable for fish accumulation in a stan- dardization model is one straightforward approach that can be used to account for some species interactions. Our results indicate that catch per trap, not catch per trap minute, should be the response variable used in future standardization models for all the species we examined. If catch is invariant to soak time and catch per unit of effort (CPUE) is used as the response in a catch standardization model, then CPUE on average would be lower in traps with longer soak times than in traps with shorter soak times. Soak-time-dependent CPUE could become a serious problem if soak time for traps was longer in some years than in others because real changes in the relative abundance of a species would be confounded with the effects on CPUE due to changes in soak time. Instead, we recommend the use of a model-based approach with catch as the response variable and soak time and fish accumulation as pre- dictor variables to properly account for any variation in catch due to these 2 factors. Whether catch data from trap surveys can be used to index reef fish abundance ultimately depends on the underlying mechanisms responsible for the observed Probability of obtaining nonzero catch 228 Fishery Bulletin 111(3) Probability of presence Positive catch only Overall CD Q. ■O "5 E Figure 5 Relationship between fish accumulation (total fish per trap) and the probability of obtaining a nonzero catch (probability of presence) (left column), estimated catch when present (middle column), and estimated overall catch (right column) for 8 species of reef fishes on the basis of data from trap surveys conducted in the south- eastern U.S. Atlantic by the Marine Resources Monitoring, Assessment, and Prediction Program (1990-2011) and the Southeast Fishery-independent Survey (2010-11). Binomial generalized additive models (GAMs) were used to estimate probability of obtaining nonzero catch. Gaussian GAMs were used to estimate catch when present. Overall catch was determined through the combination of binomial and Gaussian GAM estimates. Dashed lines indicate 95% confidence intervals. Bacheler et al.: Influence of soak time and fish accumulation on the catches of reef fishes 229 trap saturation. For instance, if catch reaches an as- ymptote because most or all of the individuals in a lo- cal area have been caught, then catch is likely a good index of abundance. But catch may not reflect actual abundance if trap saturation occurs because of space limitation in the gear, negative species interactions, increasing avoidance of the gear due to individuals al- ready being caught, handling time limitations, or bait deterioration or consumption of bait (Kennedy, 1951; Munro, 1974; Olin et ah, 2004). If traps become satu- rated at a level of catch unrelated to actual abundance, then statistical models for censored data may be useful (Bagdonavicus et ah, 2011). Another factor that may affect trap saturation is changes in feeding motivation of fishes with time of day or light levels. In our study, changes in feeding with time of day were not related to catch because all trapping was done during daylight hours. Light levels, therefore, were driven by water clarity more than by time of day. However, preliminary occupancy and W-mixture modeling for a few species has shown that water clarity does not appear to influ- ence the detection probability of traps.2 In our study, we could not identify the exact mechanisms respon- sible for trap saturation; therefore, this topic clearly requires more research. Catches of Tomtate, Stenotomus spp., and Sand Perch declined with increasing soak times, indicating that at least some individuals of these 3 species may have es- caped from the trap. These results are consistent with our own observations and a growing body of literature that indicates that some fish, crab, and lobster species frequently escape from traps and pots (Jury et al., 2001; Cole et al., 2004; Sturdivant and Clark, 2011). Tomtate, Stenotomus spp., and Sand Perch were among the 3 smallest fish species examined in our study, and their small size may have allowed them to escape through the narrow trap entrance more easily than could spe- cies of larger size. Size was the only life-history trait (Table 1) that influenced catch. An alternative explana- tion for the decreased catch of these 3 species during soak times over 100 min is that they had a longer time over which they may have been exposed to and eaten by predatory fish species caught in the same trap. We consider this explanation less likely because the diets of predatory fishes caught in traps only occasionally contain freshly consumed Tomtate and Stenotomus spp. (Goldman3), but the 2 explanations are not mutually exclusive. Experimental work should be conducted with underwater video to quantify entry and exit rates of reef fishes in fish traps — research that can provide significant insights into the catch dynamics and spe- cies interactions of marine organisms (e.g., Jury et ah, 2001; Cole et al., 2004; Sturdivant and Clark, 2011). 2 Bacheler, N., and L. Coggins. 2012. Unpubl. data. Beau- fort Laboratory, Southeast Fisheries Science Center, National Marine Fisheries Service, NOAA, Beaufort, NC 28516. 3 Goldman, S. 2012. Personal commun. Marine Resources Monitoring, Assessment, and Prediction Program, South Caro- lina Department of Natural Resources, Charleston, SC 29422. Delta-GAMs provided a convenient analytical ap- proach that helped us understand the influence of soak time and fish accumulation on the catch of reef fish species in a multispecies trap survey. The primary ben- efit of a delta-GAM is that the effects of soak time and fish accumulation can be understood after accounting for variation in total fish catch that might be due to all the other predictor variables in a model (Li et al., 2011). By accounting for soak time, fish accumulation, and other predictor variables, we found an improve- ment over previous (primarily gill net) studies that focused on only those predictor variables that were di- rectly related to the gear saturation process itself (e.g., Minns and Hurley, 1988; Hansen et ah, 1998; Akiyama et ah, 2007). A secondary benefit of delta-GAMs is that they can account for zero-inflation. It is important to note that delta-GAMs, which are composed of separate submodels that must be combined, have been criticized as less elegant than the recently developed zero-inflat- ed GAMs to account for zero-inflation (Chiogna and Gaetan, 2007; Liu and Chan, 2011). In our study, zero- inflated models were challenging to work with because they rarely converged, and, when they did, model solu- tions were often unreasonable. There were some potential drawbacks of our experi- mental design and analyses. First, the range of soak times used in our study (50-150 min) was relatively narrow, and broader insights into the catch dynamics of species in traps would have been possible if large numbers of traps had been soaked for much shorter or longer periods of time. Second, the fishacc variable was made up partially of the catch of each individual species (Olin et ah, 2004; Li et al., 2011), but we do not consider this aspect of our study to be a problem because we were interested primarily in the shape of the relationship between catch and fish accumulation, not necessarily in the significance of fishacc in the delta-GAMs. Third, the fishacc variable did not distin- guish large, predatory species from smaller, nonpreda- tory species caught in the trap; future analyses could separate the catch of potential predators from smaller species. Conclusions We showed that the rate at which reef fish species entered traps in long-term programs of fishery-inde- pendent surveys decreased either over a range of soak times or over a range of fish accumulation levels. Trap saturation occurred for all 8 reef fish species that we examined; therefore, we recommend that future stud- ies use catch standardization on raw catch or CPUE data (Maunder and Punt, 2004). It is also extremely important to understand the exact mechanisms that cause fish to saturate fishing gears, and our results in- dicate that these mechanisms may vary considerably among species. Ultimately, whether catch or CPUE can be used to index abundance will depend on a clear- 230 Fishery Bulletin 111(3) er understanding of the mechanisms that cause gear saturation. Acknowledgments We thank the captains and crews of the RV Palmetto , RV Savannah, NOAA Ship Nancy Foster, and NOAA Ship Pisces, the MARMAP and SEFIS staffs, and the numerous volunteers for making our field work pos- sible. We benefited greatly from discussions with D. Berrane, J. Buckel, L. Coggins, K. Gross, Y. Jaio, T. Kel- lison, Y. Li, W. Mitchell, K. Shertzer, C. Schobernd, Z. Schobernd, T. Smart, and E. Williams. We also thank P. Marraro, C. Schobernd, K. Shertzer, and 4 anony- mous reviewers for providing comments on earlier ver- sions of this manuscript. This work was supported in part by funds provided by the National Marine Fish- eries Service (award numbers NA06NMF4540093 and NA06NMF4350021). This article is contribution 701 of the South Carolina Marine Resources Center. Literature cited Addison, J. T., and M. C. Bell. 1997. Simulation modelling of capture processes in trap fisheries for clawed lobsters. Mar. Freshw. Res. 48:1035-1044. Austin, C. B. 1977. Incorporating soak time into measurement of fish- ing effort in trap fisheries. Fish. Bull. 75:213-218. Akiyama, S., E. Saito, and T. Watanabe. 2007. Relationship between soak time and number of en- meshed animals in experimentally lost gill nets. Fish. Sci. 73:881-888. Bacheler, N. M., L. M. Paramore, J. A. Buckel, and J. E. Hightower. 2009. Abiotic and biotic factors influence the habitat use of an estuarine fish. Mar. Ecol. Prog. Ser. 377:263—277. Bagdonavicus, V., J. Kruopis, and M. Nikulin. 2011. Nonparametric tests for censored data, 23 p. Wiley-ISTE, London. Bennett, D. B. 1974. The effects of pot immersion time on catches of crabs. Cancer pagurus L. and lobsters, Homarus gam- marus (L.). J. Cons. int. Explor. Mer 35:332-336. Beverton, R. J. H., and S. J. Holt. 1954. On the dynamics of exploited fish populations, 515 p. Chapman & Hall, London. Bubley, W. J., and O. Pashuk. 2010. Life history of the simultaneously hermaphroditic fish, Diplectrum formosum. J. Fish Biol. 77:676-691. Burnham, K. P, and D. R. Anderson. 2002. Model selection and multimodal inference: a prac- tical information-theoretic approach, 2nd ed., 488 p. Springer- Verlag, New York. Carpenter, K. E. 2002. The living marine resources of the western central Atlantic, vol. 1, 600 p. FAO, Rome. Chiogna, M., and C. Gaetan. 2007. Semiparametric zero-inflated Poisson models with application to animal abundance studies. Environmet- rics 18:303-314. Cole, R. G., N. K. Alcock, A. Tovey, and S. J. Handley. 2004. Measuring efficiency and predicting optimal set durations of pots for blue cod Parapercis colias. Fish. Res. 67:163-170. Collins, M. R. 1990. A comparison of 3 fish trap designs. Fish. Res. 9:325-332. Finkelstein, S. L. 1969. Age and growth of scup in the waters of eastern Long Island. N.Y. Fish Game J. 16:84-110. Glasgow, D. M. 2010. Photographic evidence of temporal and spatial variation in hardbottom habitat and associated biota of the southeastern U.S. Atlantic continental shelf. M.S. thesis, 120 p. College of Charleston, Charleston, SC. Hamley, J. 1975. Review of gill-net selectivity. J. Fish. Res. Board Can. 32:1944-1969. Hansen, M., R. Schorfhaar, and J. Selgeby. 1998. Gill-net saturation by lake trout in Michigan waters of Lake Superior. N. Am. J. Fish. Manage. 18:847-853. Harley, S. J., R. A. Myers, and A. Dunn. 2001. Is catch-per-unit-effort proportional to abundance? Can. J. Fish. Aquat. Sci. 58:1760-1772. Hastie, T. J., and R. J. Tibshirani. 1990. Generalized additive models, 352 p. Chapman & Hall/CRC, London. Hilborn, R., and C. J. Walters. 1992. Quantitative fisheries stock assessment: choice, dynamics, and uncertainty, 570 p. Kluwer Academic Pubis., Boston, MA. Hoese, H. D., and R. H. Moore. 1998. Fishes of the Gulf of Mexico: Texas, Louisiana, and adjacent waters, 2nd ed., 416 p. Texas A&M Univ. Press, College Station, TX. Humann, P., and N. Deloach. 2002. Reef fish identification: Florida, Caribbean, Baha- mas, 3rd ed., 512 p. New World Pubis., Jacksonville, FL. Jury, S. H., H. Howell, D. F. O’Grady, and W. H. Watson. 2001. Lobster trap video: in situ video surveillance of the behavior of Homarus americanus in and around traps. Mar. Freshw. Res. 52:1125-1132. Kennedy, W. 1951. The relationship of fishing effort by gill nets to the interval between lifts. J. Fish. Res. Board Can. 8:264-274. Kimura, D. K., and D. A. Somerton. 2006. Review of statistical aspects of survey sampling for marine fisheries. Rev. Fish. Sci. 14:245-283. Kurz, R. C. 1995. Predator-prey interactions between gray trigger- fish (Balistes capriscus Gmelin) and a guild of sand dol- lars around artificial reefs in the northeastern Gulf of Mexico. Bull. Mar. Sci. 56:150-160. Li, Y., Y. Jaio, and K. Reid. 2011. Gill-net saturation in Lake Erie: effects of soak time and fish accumulation on catch per unit effort of walleye and yellow perch. N. Am. J. Fish. Manage. 31:280-290. Bacheler et al.: Influence of soak time and fish accumulation on the catches of reef fishes 231 Liu, H., and K. S. Chan. 2011. Generalized additive models for zero-inflated data with partial constraints. Scand. J. Stat. 38:650-665. Lo, N., L. Jacobson, and J. Squire. 1992. Indices of relative abundance from fish spotter data based on delta-lognormal models Can. J. Fish. Aquat. Sci. 49:2515-2526. Manooch, C. S. III. 1977. Foods of the red porgy, Pagrus pagrus Linnaeus (Pisces: Sparidae), from North Carolina and South Car- olina. Bull. Mar. Sci. 27:776-787. Manooch, C. S. Ill, and C. A. Barans. 1982. Distribution, abundance, and age and growth of tomtate, Haemulon aurolineatum , along the southeast- ern United States. Fish. Bull. 80:1-19. Martin, T. G., B. A. Wintle, J. R. Rhodes, P. M. Kuhnert, S. A. Field, S. J. Low-Choy, A. J. Tyre, and H. P Possingham. 2005. Zero tolerance ecology: improving ecological infer- ence by modelling the source of zero observations. Ecol. Lett. 8:1235-1246. Maunder, M., and A. Punt. 2004. Standardizing catch and effort data: a review of recent approaches. Fish. Res. 70:141-159. McEachran, J., and J. D. Fechhelm. 1998. Fishes of the Gulf of Mexico, vol. 1, 1120 p. Univ. Texas Press, Austin, TX. 2005. Fishes of the Gulf of Mexico, vol. 2, 1014 p. Univ. Texas Press, Austin, TX. Miller, G. C„ and W. J. Richards. 1980. Reef fish habitat, fauna! assemblages, and factors determining distributions in the South Atlantic Bight. Proc. Gulf Carib. Fish. Inst. 32:114-130. Miller, R. J. 1979. Saturation of crab traps: reduced entry and es- capement. J. Cons. int. Explor. Mer 38:338-345. 1990. Effectiveness of crab and lobster traps. Can. J. Fish. Aquat. Sci. 47:1228-1251. Minns, C., and D. Hurley. 1988. Effects of net length and set time on fish catches in gill nets. N. Am. J. Fish. Manage. 8:216-223. Munro, J. L. 1974. The mode of operation of Antillean fish traps and the relationships between ingress, escapement, catch and soak. J. Cons. int. Explor. Mer 35:337-350. Murray, K. 2004. Magnitude and distribution of sea turtle bycatch in the sea scallop ( Placopecten magellanicus) dredge fishery in two areas of the northwestern Atlantic Ocean, 2001-2002. Fish. Bull. 102:671-681. Neter, J., W. Wasserman, and M. H. Kutner. 1989. Applied linear regression models, 2nd ed., 688 p. Irwin, Homewood, IL. O’Brien, L., J. Burnett, and R. K. Mayo. 1993. Maturation of nineteen species of finfish off the northeastern coast of the United States, 1985-1990. NOAA Tech. Rep. NMFS 113, 66 p. Olin, M., M. Kurkilahti, P. Peitola, and J. Ruuhijarvi. 2004. The effects of fish accumulation on the catchabil- ity of multimesh gillet. Fish. Res. 68:135-147. Pennington, M. 1996. Estimating the mean and variance from highly skewed marine data. Fish. Bull. 94:498-505. Pennington, M., and T. Stromme. 1998. Surveys as a research tool for managing dynamic stocks. Fish. Res. 37:97-106. R Development Core Team. 2011. R: a language and environment for statistical com- puting. R Foundation for Statistical Computing, Vien- na, Austria. [Available from http://www.R-project.org/, accessed December, 2012.] Ragonese, S., M. Zagra, L. Di Stefano, and M. L. Bianchini. 2001. Effect of codend mesh size on the performance of the deep-water bottom trawl used in the red shrimp fishery in the Strait of Sicily (Mediterranean Sea). Hy- drobiologia 449:279-291. Richards, R. A., J. S. Cobb, and M. J. Fogarty. 1983. Effects of behavioral interactions on the catchabil- ity of American lobster ( Homcirus americanus) and two species of Cancer crabs. Fish. Bull. 81:51-60. Ricker, W. E. 1975. Computation and interpretation of biological sta- tistics of fish populations, 382 p. Dep. Environ. Fish. Mar. Serv., Ottawa, Canada. Robertson, W. D. 1989. Factors affecting catches of the crab Scylla ser- rata (Forskal) (Decapoda: Portunidae) in baited traps: soak time, time of day and accessibility to bait. Estuar. Coast. Shelf Sci. 29:161-170. Robins, C. R., G. C. Ray, and J. Douglas. 1986. A field guide to Atlantic coast fishes, 356 p. Houghton Mifflin Co., Boston. Rodgveller, C. J., C. R. Lunsford, and J. T. Fujioka. 2008. Evidence of hook competition in longline studies. Fish. Bull. 106:364-374. Schobernd, C. M., and G. R. Sedberry. 2009. Shelf-edge and upper-slope reef fish assemblages in the South Atlantic Bight: habitat characteristics, spa- tial variation, and reproductive behavior. Bull. Mar. Sci. 84:67-92. Sedberry, G. R. 1985. Food and feeding of the tomtate, Haemulon auto- lineatum (Pisces, Haemulidae), in the South Atlantic Bight. Fish. Bull. 83:461-466. 1988. Food and feeding of black sea bass, Centropristis striata, in live bottom habitats in the South Atlantic Bight. J. Elisha Mitchell Sci. Soc. 104:35-50. Sedberry, G. R., and N. Cuellar. 1993. Planktonic and benthic feeding by the reef- associated vermilion snapper, Rhomboplites aurorubens. Fish. Bull. 91:699-709. Somerton, D. A., and M. F. Merritt. 1986. Method of adjusting crab catch per pot for differ- ences in soak time and its application to Alaskan Tan- ner crab ( Chionoecetes bairdi ) catches. N. Am. J. Fish. Manage. 6:586-591. Stefansson, G. 1996. Analysis of groundfish survey abundance data: combining the GLM and delta approaches. ICES J. Mar. Sci. 53:577-581. Sturdivant, S. K., and K. L. Clark. 2011. An evaluation of the effects of blue crab ( Calli - nectes sapidus) behavior on the efficacy of crab pots as a tool for estimating population abundance. Fish. Bull. 109:48-55. Venables, W. N., and B. D. Ripley. 2002. Modern applied statistics with S, 495 p. Springer, New York. Wahba, G. 1990. Spline models for observational data, 180 p. So- 232 Fishery Bulletin 111(3) ciety for Industrial and Applied Mathematics, Philadel- phia, PA. Wood, S. N. 2008. Fast stable direct fitting and smoothness selection for generalized additive models. J. Roy. Stat. Soc. (Ser. B) 70:495-518. Zar, J. H. 1999. Biostatistical analysis, 4th ed., 663 p. Prentice Hall, Upper Saddle River, NJ. Zhao, B., J. C. McGovern, and P. J. Harris. 1997. Age, growth, and temporal change in size at age of the vermilion snapper from the South Atlantic Bight. Trans. Am. Fish. Soc. 126:181-193. Zhou, S„ and T. C. Shirley. 1997. A model expressing the relationship between catch and soak time for trap fisheries. N. Am. J. Fish. Man- age. 17:482-487. 233 Abstract— The effects of commercial fishing with crab pots on the physical condition of the snow crab (Chion- oecetes opilio ) and southern Tanner crab (C. bairdi) were investigated in the Bering Sea and in Russian wa- ters of the Sea of Okhotsk. In crabs that were subjected to pot hauling, the presence of gas embolism and the deformation of gill lamellae were found in histopathological in- vestigations. Crab vitality, which was characterized subjectively through observation of behavioral responses, depended on not only the number of pot hauls but also the time between hauls. Immediately after repeated pot hauls at short time intervals (<3 days), we observed a rapid de- cline in vitality of crabs. When haul- ing intervals were increased to >3 days, the condition of crabs did not significantly change. After repeated pot hauls, concentration of the re- spiratory pigment hemocyanin ([He]) was often lower in the hemolymph of crabs than in the hemolymph of freshly caught animals. Our research indicated that changes in [He] in crabs after repeated pot hauls were caused by the effects of decompres- sion and not by starvation of crabs in pots or exposure of crabs to air. We suggest that the decrease in [He] in hemolymph of snow and southern Tanner crabs was a response to the adverse effects of decompression and air-bubble disease. The decrease in [He] in affected crabs may be a re- sult of mechanisms that regulate in- ternal pressure in damaged gills to optimize respiratory circulation. Manuscript submitted 20 April 2012. Manuscript accepted 15 May 2013. Fish. Bull. 111:233-251 (2013). doi 10.7755/FB.111.3.3 The views and opinions expressed or implied in this article are those of the author (or authors) and do not necesarily reflect the position of the National Marine Fisheries Service, NOAA. Effects of pot fishing on the physical condition of snow crab iChionoecetes opilio ) and southern Tanner crab iChionoecetes bairdi ) Sergey I. Moiseev (contact author)1 Svetlana A. Moiseeva2 Tatyana V. Ryazanova3 Anna M. Lapteva4 Email address for contact author: moiseev@vniro.ru 1 Department of Marine Commercial Invertebrates and Algae Russian Federal Research Institute of Fisheries and Oceanography 17 Krasnoselskaya Upper Street Moscow, 107140 Russia 2 Institute of Cell Biophysics Russian Academy of Science 3 Institutskaya Street Pushchino, 142290 Russia 3 Kamchatka Research Institute of Fisheries and Oceanography 18 Naberezhnaia Street Petropavlovsk-Kamchatsky, 683000 Russia 4 Polar Research Institute of Marine Fisheries and Oceanography 6 Knipovich Street Murmansk, 183038 Russia Commercial fishing for crabs in most countries is currently carried out with the use of baited crab pots. With this method of catching crabs, it is possible to reduce the effects of fish- ing on crab populations by sorting the catch, because females, sublegal- size males, and crabs with missing legs are returned to the sea alive. However, some of the discards die from body injuries and physiological stress associated with temperature changes and exposure to air. In sur- viving crabs, physiological imbalance and altered behavior can affect their ecological functions, such as predator avoidance, feeding, migration, and reproduction. Crabs are discarded in high proportions in relation to land- ings, and comprehensive information on the fate of the discards and their survival rate is required to make management decisions in a fishery. Prediction of discard survivability un- der a wide range of fishing conditions requires fundamental knowledge of the effects of catch-related stressors on crab physical condition. To date, many studies have exam- ined the effects of stressors, such as trauma from the hauling of pots, air exposure, and adverse temperatures, that are associated with capture, handling, and discarding of crabs (van Tamelen, 2005; Tallack, 2007; Stoner, 2009; Darnell et al., 2010). Decompression due to fast transport through the water column is one of the stressors that affect crabs dur- ing fishing with pots. Decompression results in air-bubble disease, the for- mation of gas bubbles in the blood and body fluids of crabs (Ryazanova, 2009). Gas bubbles form in the hemal system, organs, and tissues of crabs, causing hemal stasis, disruption and displacement of tissues. Although decompression is a constant and un- avoidable adverse factor associated with pot fishing, relatively little at- tention has been paid to its effects on crab condition. Traditionally, de- compression is not considered to be a substantial source of discard mortal- ity (Stoner, 2012). However, gas-bub- ble disease caused by decompression could have prolonged deleterious ef- fects on crab condition because air bubbles may persist in the body of a crab for a long time (Johnson, 1976). 234 Fishery Bulletin 111(3) The subjects of this study were snow crab ( Chionoece - tes opilio) and southern Tanner crab (C. bairdi). Both species are heavily fished along the continental shelf of the Bering Sea and the Sea of Okhtosk of the Rus- sian Federation. Snow and southern Tanner crabs are shallow-water species. They generally live from subtidal areas to a depth of -450 m, although snow crab has also been found in much deeper water (Jadamec et ah, 1999; Mihailov et ah, 2003; Slizkin, 2010). In Russian waters, the snow crab inhabits environments with a tempera- ture range from -1.8°C to 7°C. The southern Tanner crab is usually found in temperatures of 2-3°C, avoiding bottom areas where a cold intermediate layer persists for a long time (Slizkin et ah, 2001; Slizkin, 2010). In the area of overlap where the 2 species occur together, the abundance ratio of these crabs is strongly variable. Unlike many other crustaceans, snow and southern Tanner crabs undergo a terminal molt before they be- come mature adults (Conan and Comeau, 1986; Comeau and Conan, 1992; Sainte-Marie et ah, 1995). After their terminal molt, adult males of these species are avail- able to pot fishing for only 0. 5-3.0 years because their carapaces become progressively fouled and ultimately deteriorate. Also after their terminal molt, crabs can no longer grow and regenerate lost limbs. Adult males that lack a limb are returned to the sea after a catch is sorted and because discarded crabs may be re-injured, the number of individuals that lack limbs increases in heavily fished populations. High-density aggregations of female snow and southern Tanner crabs are found to occur in different areas from those where males are caught. Therefore, female snow and southern Tanner crabs are rarely encountered in catches. The percent- age of undersize males (carapace width <100 mm) usu- ally does not exceed 5-10% of the total catch because of selectivity characteristics of trap gear. Therefore, dam- aged and “dirty” male snow and southern Tanner crabs may often compose a significant portion of the discards that occur in pot fishing. Although these crabs do not have market value, they may be involved in reproduc- tion. It is probable that these crabs may be recaptured multiple times throughout their life and, therefore, may be exposed potentially to cumulative effects of fishing procedures. This study was designed to investigate the effects of decompression and other stressors associated with pot hauling on the physical condition of adult male snow and southern Tanner crabs that had completed a terminal molt between 1.5 years and a few months previously. We conducted our experiments in the field. Crabs taken from the commercial catch were placed in commercial pots. The pots were then sunk and hauled back to the surface at different frequencies. With this approach, multiple factors in addition to decompression can affect the survivability of animals. These other fac- tors include significant changes in environmental condi- tions during hauls, interaction of crabs with the gear, and exposure to the air. Therefore, to specifically assess the effects of decompression on crab physical condi- tion, histopathological changes in organs and tissues of crab caused by air-bubble disease were studied. Our approach was derived from the histological evidence of Johnson (1976), who found that the gills, heart, and an- tennal gland were the organs of blue crab ( Callinectes sapidus) most severely affected by exposure to water supersaturated with air to simulate decompression ef- fects. Bubbles persisted in the gills longer (more than a month) than in other organs of blue crab. We assumed that gill dysfunction caused by decompression would greatly affect the survivability of crabs after their re- lease. Because of the logistical constraints of direct ex- perimental research on gill function in the field, we in- vestigated physiological parameters that could be used to estimate gill function disorders. We selected hemocyanin (He) as the indicator for the assessment of the effects of pot fishing on the physical condition of the snow crab and southern Tanner crab. He is a copper-containing respiratory pigment that in- creases solubility of oxygen (O2) by a factor of 2-4 and accounts for from 70% to more than 90% of the total protein concentration in the hemolymph of crustaceans (Truchot, 1992). Structural and functional variability of crustacean He has been shown to be of great impor- tance for the adaptive condition of these organisms in response to metabolic changes and environmental stim- uli (Bridges, 2001; Giomi and Beltramini, 2007). There- fore, we hypothesized that physiological alterations caused by impairment of respiratory function of the gills of snow and southern Tanner crabs could be correlated with changes in the structure and concentration of He. Plasma ion concentrations — for sodium ion (Na+), potas- sium ion (K+), chlorine ion (Cl-), calcium ion (Ca2+), and magnesium ion (Mg2+) — also were measured to examine ionoregulatory changes at the gills of affected crabs. During our experiments, factors other than decompres- sion had an effect on crabs, including starvation and air exposure. Therefore, we also conducted experiments to assess the separate effects of starvation and air expo- sure on He. We had the following objectives: 1) the investiga- tion of histopathological deterioration of organs and tis- sues in snow and southern Tanner crabs subjected to pot hauling; 2) an analysis of the relationship between hauling intensity (the number and frequency of hauling events) and animal stress responses, including changes in behavior and biochemical parameters of hemolymph; and 3) an analysis of the utility of biochemical assays of He for prediction of survivability of snow and south- ern Tanner crabs subjected to hauling and discarding processes. Materials and methods Study areas Snow crab and southern Tanner crab were collected during 5 fishing voyages in 2006, 2008, and 2010 in Moiseev et a!: Effects of pot fishing on the physical condition of Chionoecetes opilio and Chionoecetes bairdi 235 both the Bering Sea (Fig. 1) and the Sea of Okhtosk (Fig. 2). In the areas where we conducted our investiga- tions, surface temperatures typically vary from 4°C to 12°C in summer and from -1.5°C to 3°C in winter. Sur- face salinities generally vary between 32.2%e and 33.0%e. In area I of the Bering Sea (Fig. 1) and in area I of the Sea of Okhtosk (Fig. 2), the 2 species occur together. At crab habitat depths of 50-350 m, water temperatures range from -0.5 to 5.5°C through- out the year and salinity varies from 33.2%c to 34%c. Crab capture and handling Collection of crabs and experimental work were carried out on a crab boat. The fishing gears that we used in this study were square pots of the “Ameri- can” type and conical pots of the “Japa- nese” type (Moiseev, 2003). Both types of pots were covered with nylon-twine webbing (mesh size: 60-70 mm). Crabs caught with commercial pots were maintained in large tanks supplied with natural seawater. Water temperatures during the holding period ranged from 4°C to 10°C. These tanks were covered with tarpaulins to reduce stress effects of light on the crabs. All individuals were tagged with a piece of rubberized fabric marked with a unique number; a tag was attached to a crab claw with a plastic clip-strip. Because of the ten- dency for autotomy of limbs in southern Tanner and snow crabs, all manipula- tions were carried out quickly with the greatest possible care. During holding periods, no crabs died or showed signs of stress in the form of impaired reflex- es (Moiseev et al., 2012). For this study, experiments were limited to crabs that met the following criteria: 1) morphometrically mature males had completed a terminal molt between 1.5 years and a few months previously, according to criteria in Co- meau and Conan (1992) and Sainte- Maria et al. (1995); 2) carapace width for snow crab was 110-150 mm, and carapace width for southern Tanner crab was 120—155 mm; 3) crabs were without apparent physical injuries or limb loss; and 4) crabs were in perfect condition as revealed with a vitality index (VI). 168°E 172°E 1 76°E 18CPE/W 178°W Figure 1 Map of areas where experiments and sample collection were conducted in the western Bering Sea in 2006-10 for our study of effects of commercial fishing with crab pots on the physical condition of the snow crab (Chionoecetes opilio) and southern Tanner crab (C. bairdi). Area 1 was near the Koryak Coast, and area II was the shelf and slope near Cape Navarin. 140°E 1 44°E 1 48°E 152°E 156°E 160°E Figure 2 Map of areas where experiments and sample collections were conducted in the Sea of Okhtosk in 2006-10 for our study of effects of commercial fish- ing with crab pots on the physical condition of the snow crab ( Chionoecetes opilio) and southern Tanner crab (C. bairdi). Area I was on part of the Western Kamchatka shelf, area II was on the shelf and continental slope on the beam of the Babushkin Gulf, and area III was in the central area of the northern Sea of Okhtosk. 236 Fishery Bulletin 111(3) Moiseev et al: Effects of pot fishing on the physical condition of Chionoecetes opilio and Chionoecetes bairdi 237 Field experiments Effects of repeated pot hauling on crabs Snow crab and southern Tanner crab were placed in commercial crab pots. In experiments 1, 3, 5, 6, and 7 we used a single pot of the American-type (Table 1). Before deployment of the pots, the entrance tunnels of pots were sewn shut with webbing to prevent individuals in our experiment from escaping and other crabs and big animals (e.g., mammals, big fishes) from entering the pots from the environment. During our experiments, crabs were not fed. The pots were sunk and hauled back to the sur- face at different frequencies. During the preparation of pots for the next deployment, crabs were maintained in tanks with running seawater for periods from 10-15 min to 1-2 h. Experiments that involved repeated pot hauls were divided into 2 sets, depending on the hauling intervals: 1) short time intervals (<3 days) — experiments 5, 6, and 7 — or 2) long time intervals (>3 days) — experi- ments 1 and 3 (Table 1). Both control animals and those animals that were collected for experiments had been lifted to the water surface during capture. Hereafter, the number of pot lifts we specify includes only the lifts made during the experiment, unless otherwise stated. Upon retrieval of pots, mortality was noted and the condition of each crab was monitored with the vital- ity index. Criteria for the index incorporated reflexes, spontaneous movements of appendages, and righting behavior (i.e., an animal’s ability to turn from a ven- trum-up position to normal orientation): 0 = no signs of life: no movement of legs or mouthparts; 1 = weak, slow movement of legs or maxillipeds, at- tempts at righting behavior, third pair of maxilli- peds droop open and retract to cover small mouth parts when touched by pencil, chelae grab any ob- ject slightly and not for long (up to 30 s); 2 = moderate movement of legs, righting time of 10-60 s in water bath, fast movement of maxillipeds, chelae grab any object strongly and for long periods (up to 1.5 min); 3 = fast, active movement of legs, righting time of 5-40 s in water bath or on the sorting table, fast move- ment of maxillipeds, chelae grab any object strong- ly and for long periods (> 2 min), threat display. Effects of long-term starvation in pots on crabs In ex- periments 2, 4, 8, and 9, to investigate the effects of starvation, pots containing snow crab or southern Tan- ner crab were soaked (i.e., pots were deployed in the sea) for extended periods of time (14-55 days) (Table 1). Experiment 2 was combined with experiment 1. In the course of experiment 1, after the second lift of the pot, we added 5 freshly caught snow crab and 6 south- ern Tanner crab. The pot was lifted to the surface 16 days later; ocean conditions caused this long hauling interval (Table 1). At the end of all these experiments, the amount of meat in crab limbs was assessed by vi- sual comparison of the slices of merus after boiling versus schematic representation of slices of merus in crab with varying degrees of meat content (Borisov et ah, 2003). Effects of long-term exposure to air on crabs Experiment 10 was designed to investigate the effects of long-term exposure to air on snow crab. It was replicated 3 times (Table 1). Snow crabs were kept out of the water on the deck from 6 to 8 h after capture. Animals were covered with a tarpaulin, which periodically was irrigated with seawater to protect the crabs from desiccation. Surface water temperatures were 6-7°C, and air temperatures were 7-8°C. Thereafter, 10-15 individuals were placed in conical pots of the Japanese type, and these pots were deployed in the sea. In each repetition of the ex- periment we used a single pot which was hauled 2-3 times at intervals of 2 days. Sampling of hemolymph Sampling of hemolymph in crabs from commercial catches was completed within 30 min after a pot had appeared at the water surface, the length of time re- ferred to hereafter as immediately after capture. Sam- ples of hemolymph taken from crabs immediately af- ter capture served as control samples. Most crabs from which control hemolymph samples were taken were dis- carded, but, in experiments 1 and 3, some crabs were used in the experiments after they had control samples of blood taken from them. In experiment 9, control sam- ples of hemolymph were taken from crabs immediately after capture, as well as from the crabs exposed to air for 6-8 h before deployment of a pot. In most cases, crabs used for control samples of blood and crabs for our experiments were collected from the same commer- cial catches. In some cases, because of ocean conditions, crabs used for control samples of blood were collected from other commercial catches, but they always were collected from catches in the same fishing area at a dis- tance <0.9 km from the experimental pots and always during the periods of our experiments. Upon retrieval of pots, during our experiments that involved repeated pot hauls, hemolymph was sampled from some of the crabs used in experiments; these individuals were discarded after blood was sampled. Only in experiments 1 and 3, crabs from which hemolymph samples were taken re- main in the pot. In experiment 1, in some individuals, blood was sampled repeatedly (3 or 4 times). Hemolymph (3-5 mL) was quickly sampled from in- dividual crabs with a syringe through the arthrodial membrane at the base of the fourth or fifth pair of legs. All manipulations were carried out quickly and careful- ly (the duration of crab exposure to air was not more than 10 min). The hemolymph samples were frozen and stored below -30° while on board. For transport to the biochemistry laboratory, they were kept on ice. Before analysis, the samples of hemolymph were thawed at 4°C and centrifuged for 15 min at 12,000 rpm, and the supernatant was collected. In snow and southern Tan- 238 Fishery Bulletin 111(3) Table 2 Dates, locations, depths, hauling conditions, and sampling for histology during repeated pot hauls in experiments conducted in 2010 to study the effects of commercial pot fishing on the physical condition of snow and southern Tanner crabs ( Chionoecetes opilio and C. bairdi ) in the Bering Sea and in Russian waters of the Sea of Okhotsk. Total number of crabs sampled for histology Experiment number Year Date Areas2 Depth (m) °C at depth Hauling intervals min/max (day) Duration of experiment (day) No. of pot lifts Crab species Imme- diately after capture During repeated pot hauls 1 2010 4-11 Sept Bering Sea 170-260 2.5-3. 5 2/3 7 3 C. opilio 3 5 area II C. bairdi 3 5 2 2010 28 Oct- Sea of Okhtosk 230-340 1.6-2. 5 2/3 5 2 C. opilio 3 5 2 Nov area III 1 See Figures 1 and 2. ner crabs, unlike in many other crabs, flocculent pre- cipitate is formed during blood coagulation and can be removed easily by centrifugation. Histological examinations To examine the histopathological condition of organs and tissues in crabs that were subjected to repeated pot hauls, tissue samples were taken from both crab species during repeated pot hauls at time intervals of 2-3 days (Table 2). In each experiment, samples were taken from 5 individuals used in that experiment and from 3 freshly caught individuals that served as control animals. Frag- ments of internal organs were fixed for 24-40 h with Davidson’s fluid (Bell and Lightner, 1988) prepared in seawater. After fixation, the samples were transferred to 70% alcohol, in which they were kept in the labora- tory. This material was further treated with standard histological techniques: samples were dehydrated in an alcohol series of increasing concentrations, clarified with chloroform, and embedded in paraffin. Sections were stained with Meyer’s hematoxylin-eosin (Johnson, 1980) and slides were examined under an Olympus Al-2 light microscope (Olympus Corp., Tokyo, Japan1). Ionic composition of hemolymph Colorimetric tests (DiaSys Diagnostic Systems GmbH, Holzheim, Germany) were used to measure concentra- tions of Ca2+ ([Ca2+]), Mg2+ ([Mg2+]), and Cl- ([Cl-]) in hemolymph. Concentration of K+ ([K+]) and Na+ ([Na+]) were determined by flame photometry (FLAPHO 4, Carl Zeiss, Jena, Germany). 1 Mention of trade names or commercial companies is for iden- tification purposes only and does not imply endorsement by the National Marine Fisheries Service, NOAA. Hemocyanin concentration The [He] in the hemolymph of individual crabs was estimated with an established spectrophotometric method (Nickerson and Van Holde, 1971; deFur et ah, 1990). An aliquot of serum was diluted by 50 mmol L-1 Tris HC1, containing 10 mmol L-1 EDTA, pH 8.9 (1:39). The absorbance of the resultant mixture was measured at 338 nm. The [He] was calculated with the extinction coefficient given by Nickerson and van Holde (1971), by assuming an average molecular weight of 75 kDa for subunits of hemocyanins (He’s) from snow crab and southern Tanner crab. Electrophoresis To estimate the molecular weight of the different sub- units of the He’s from snow crab and southern Tan- ner crab, sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS/PAGE) was carried out on whole hemolymph by using the discontinuous buffer system of Laemmli (1970) on slab gels. The stacking gel was made of 4.0% and 2.7% acrylamide and bis-acrylamide, respectively, with a pH of 6.8, and the resolving gel was made of 10.0% and 2.7%, respectively, with a pH of 8.8. The samples (10 pg) were run at 4°C and 20 mA with a tris-glycine buffer (pH 8.3). The gels were stained with Coomassie Brilliant Blue (Bio-Rad Labo- ratories, Hercules, CA). Copper content in the hepatopancreas In experiments 6 and 7, which involved repeated pot hauls at short time intervals (<3 days), samples of he- patopancreas were dissected, after blood was sampled (Table 1). Samples of hepatopancreas also were dis- sected from southern Tanner crab after starvation in Moiseev et al: Effects of pot fishing on the physical condition of Chionoecetes opilio and Chionoecetes bairdi 239 the pot for 55 days in experiment 9 (Table 1). Samples of hepatopancreas were frozen in plastic bags, stored, and transported in the same manner as that used for the samples of hemolymph. The hepatopancreas copper content of individual crabs was determined by atomic absorption spectrophotometry (Shimadzu AA-6800, Shimadzu Corp., Kyoto, Japan) with the method of Kirichenko et al. (2005). In experiments 6 and 7, copper content was determined in samples of hepatopancreas from only those individual snow crab and southern Tan- ner crab in which [He] was twice as low as the mean [He] in the crabs immediately after capture. Results Dependence of crab vitality on haul frequency Our experiments, in which we examined the physiologi- cal capacity of snow crab and southern Tanner crab to survive the variable environment associated with haul and return to the sea revealed that the condition of an animal was strongly dependent on the time interval between pot lifts. During repeated pot hauls at short time intervals (<3 days) in experiments 5, 6, and 7, we observed a rapid reduction in reflex responses and lo- comotor activity in crabs. After 2-3 lifts in pots, the vitality of crabs was close to the minimum (VI=0-1). In experiment 7, the total number of pot lifts was 9. After the first 2 lifts, VI= 1—2; after the third to sixth lifts, VI=0-1; and after the last 2 lifts, VI approached 0. During repeated pot hauls at long time intervals (>3 days) in experiments 1 and 3, the condition of the crabs did not significantly change or was only moderately suppressed (VI=2-3). At time intervals of more than a week between pot hauls (experiments 2, 4, 8, and 9), the condition of most crabs was not significantly differ- ent from the condition of freshly caught animals. Changes in hemocyanin concentration depending on hauling interval Long intervals 03 days) In experiment 1, 3 pot lifts were made: the first lift on the fifth day, the second lift on the ninth day, and the third lift only on the 25th day because of ocean conditions (Table 1). We observed a sig- nificant decrease in [He] in the hemolymph of snow crab and southern Tanner crab during this experiment: both in crabs in which blood was sampled only once (Fig. 3, A and B) and in crabs in which blood was sampled more than once (Fig. 4, A and B). The main decrease in [He] occurred during the time that elapsed between the cap- ture of crabs and the second pot lift. During the next 16 days and up to the 25th day of the experiment, further decreases in [He] were observed but were not significant (Fig. 3, A and B). In experiment 3, mean [He] in the hemolymph of snow crab used in our experiments decreased by 51%, after the first pot lift on the seventh day, compared A B Days Figure 3 Changes in mean hemocyanin concentration [He] in (A) southern Tanner crab (Chionoecetes bairdi ) and (B) snow crab (C. opilio) sampled for hemolymph only once during repeated pot hauls at long time intervals (>3 days) in experiments conducted on 18 May-12 June 2006 in area I of the Bering Sea to study the effects of commercial pot fishing on the physical condition of these species (see Fig. 1 for location of area I). Open bars indicate mean [ He] observed in crabs sampled im- mediately after capture (18 individuals for southern Tanner crab and 15 for snow crab), and shaded bars indicate mean [He] observed in crabs sampled after repeated pot hauls (for both species of crabs, 4-7 in- dividuals were sampled after each pot haul). Asterisks (*) indicate values significantly different by Student’s t-test (PcO.Ol), compared with mean [He] observed in crabs sampled immediately after capture. Error bars indicate ±1 standard error of the mean. with the mean [He] found in control samples collected immediately after capture, and the difference was signif- icant (Student’s Atest, PcO.Ol). After the second pot lift on the 14th day, large variations in [He] in the hemo- lymph of snow crab were observed, and the difference between experimental and control animals, therefore, was not significant (data not shown). 240 Fishery Bulletin 111(3) Short time intervals (<3 days) In experiment 7, the planned time intervals between lifts were 1-2 days. However, because of ocean conditions, they ranged from 1 to 4 days. The total number of pot lifts was 9, and hemolymph was not sampled after the third, fifth, and seventh lifts (Table 1). Large variations in [He] in the hemolymph of snow crab were observed after the sec- ond pot lift until the end of the experiment (Fig. 5). In one-third of experimental animals, [He] was markedly lower, by 50% or more, than mean [He] observed in con- trol crabs, although, in the remaining animals used in the experiment, [He] did not differ significantly from He levels in control samples. Experiments 5 and 6 were similar to but not as pro- longed as experiment 7. In both of these experiments, large variations in [He] in snow and southern Tanner crabs were observed. In experiment 5, in approximately Figure 5 Changes in mean hemocyanin concentration [He] in snow crab (Chionoecetes opilio) during repeated pot hauls at short time intervals (<3 days) in experiments conducted on 17 October-3 November 2008 in area II of the Sea of Okhtosk to study the effects of commer- cial pot fishing on the physical condition of these spe- cies (see Fig. 2 for location of area II). In experiment 7, ocean conditions extended time intervals to 1-4 days rather than the planned 1-2 days. Open bar indicate mean [He] observed in crabs sampled immediately af- ter capture (number of individuals was 12), and shaded bars indicate mean [He] observed in crabs sampled af- ter repeated pot hauls (the number of individuals sam- pled after each pot haul was 3-5). Error bars indicate ±1 standard error of the mean. Arrows indicate third, fifth, and seventh pot lifts when sampling of blood was not carried out. one-half of individual snow crab, [He] was markedly lower, by 80% or more, than mean [He] in control crabs. In experiment 6, in approximately one-half of individual southern Tanner crab, [He] was markedly lower, by 70% or more, than mean [He] in control crabs. In both of these experiments, in one-half of experimental animals, [He] did not differ significantly from He levels in con- trol samples (data not shown). Effects of long-term starvation on crabs in pots In experiment 2 and 4, the duration of pot soaking was 16 and 14 days, respectively. At the end of these ex- periments, mean [He] in hemolymph of snow crab and southern Tanner crab did not significantly differ from mean [He] in hemolymph of control crabs immediately after capture (data not shown). In experiment 8, the duration of pot soaking was 25 days. At the end of this experiment, mean [He] in hemolymph of southern Tanner crab was slightly lower than [He] in control crab immediately after capture, but the mean difference between experimental and control ani- mals was not significant (Student’s t-test, P>0.05) (Fig. 6A). Moiseev et ai: Effects of pot fishing on the physical condition of Chionoecetes opilio and Chionoecetes bairdi 241 A C Figure 6 Changes in mean hemocyanin concentration [He] in snow crab ( Chionoecetes opilio ) and southern Tanner crab (C. bairdi) after starva- tion in crab pots. Open bars indicate mean [He] observed in crabs sampled immediately after capture (9 individuals for snow crab and 5 for southern Tanner crab for each experiment), and shaded bars indicate mean [He] observed in crabs sampled after starvation (5 individuals for snow crab and 9 and 5 for southern Tanner crab for experiments with pot soaking time of 25 days and 55 days, respectively): (A) southern Tanner crab, pot soaking time=25 days, Bering Sea (area I), 7 July-2 August 2008; (B) south- ern Tanner crab, pot soaking time=55 days, 19 September-12 November 2008, Sea of Okhtosk (area I); and (C) snow crab, pot soaking time=55 days, 19 September-12 November 2008, Sea of Okhtosk (area I). An asterisk (*) indicates val- ues were significantly different by Student’s t-test (PcO.001), compared with mean [He] ob- served in crabs immediately after capture. Error bars indicate ±1 standard error of the mean. In experiment 9, pots soaked for 55 days. At the end of this experiment, mean [He] in southern Tanner crab decreased by 44% compared with mean [He] in control crab immediately after capture (Fig. 6B), but the differ- ence between the experimental and control animals was not significant (Student’s t-test, P>0.05). In snow crab, mean [He] decreased by 67% compared with mean [He] in control crab (Fig. 6C), and the difference was signifi- cant (Student’s t-test, P<0.001). During experiment 4, 2 crab died. At the end of the remaining experiments, all experimental animals were alive and their vitality was high (VI=2-3). At the end of all experiments, the weight of experimental crabs and the amount of meat in crab limbs did not change significantly. Effects of long-term exposure to air on snow crab In experiment 10, after long-term exposure to air, the vitality of snow crab was low ( VI=0 — 1) and mean [He] in these animals was comparable with the mean [He] in control crabs immediately after capture. After long- term exposure to air, most of the crab that were sub- jected to the pot-hauling process were dead after the first pot-hauling event. Some individuals were able to survive 2-3 pot hauls (Table 1). Samples of hemolymph could be obtained from crab only after the first pot haul. [He] in hemolymph of those animals was compa- rable with the mean [He] in control crab immediately after capture. Histological investigations In our experiments both control and experimental ani- mals were hauled from the water in the pot at least once and, therefore, gas embolism was observed in hearts and gills of both control and experimental ani- mals during gross examination of dissected crabs (Fig. 7). Histopathological investigations found intact and de- formed gill lamellae in both experimental and control individuals of snow crab and southern Tanner crab (Fig. 8, A and B; Fig. 9, A-C). Hemolymph spaces of deformed gill lamellae were greatly expanded, sometimes in the form of vesicular swellings, or, conversely, the opposite walls of lamellae were stuck together (Fig. 9A). Often the connections of pillar cells in dilated parts of lamel- lae were torn (Fig. 9B). Marginal canals had irregular shapes or were swollen and filled with a large number of hemocytes. Opposite walls of lamellae under the mar- ginal canals, as a rule, were stuck together (Fig. 9C). Gas bubbles and aggregations of hemocytes were found in the myocardia of individuals of both experimental and control snow crab and southern Tanner crab. Simi- lar changes often were observed in antennal glands, in skeletal muscles, and in connective tissues of the stom- achs and guts of crabs. Infiltrations of hemocytes and different-size, melanized inclusions often were found in the gill lamellae of both control and experimental crabs (Fig. 9D). We were unable to find or show dependence 242 Fishery Bulletin 111(3) Figure 7 Photographic image showing air-gas bubbles (T) found in the median shaft of the gill of a southern Tanner crab ( Chionoecetes bairdi) during gross examination of dissected crabs as part of our study of the effects of pot fishing on the physical condition of Chionoecetes crabs in the Bering Sea and Russian waters of the Sea of Okhotsk. of the degree of damage in gills on the number of lifts (Table 2). Copper content in the hepatopancreas A significant difference in hepatopancreas copper lev- els between control and experimental individuals (Stu- dent’s Gtest, P<0.05) was observed in experiment 9, where long-term starvation of southern Tanner crab in a crab pot was investigated (Table 3). There were no significant differences in hepatopancreas copper levels between control and experimental snow crab and south- ern Tanner crab in experiments 6 and 7, which involved frequent lifts of crab pots. Ionic composition of hemolymph The ionic composition of hemolymph from snow and southern Tanner crabs under different experimental conditions was investigated (Table 4). In experiment 1, a significant decrease in [K+] was observed in hemo- lymph from snow crab (Student’s Gtest, P<0.01) and southern Tanner crab (Student’s Gtest, P<0.05) after 3 pot hauls. In experiment 9, a significant decline in [K+] was observed in hemolymph from snow crab after star- vation in a pot for 55 days (Student’s f-test, P<0.05). Subunit composition of hemocyanins In the Hcs from snow crab and southern Tanner crab, 4 and 3 differently migrating polypeptide chains, respec- tively, were found in the SDS/PAGE experiment. In our investigations, the patterns of bands obtained by SDS/ PAGE in both crab species did not change under differ- ent experimental conditions (data not shown). Discussion Effects of pot fishing on crab vitality In most studies on the effects of pot fishing on crabs, the main focus was mortality of animals. Experiments in other studies that involved repeated pot hauls showed that mortality of deepwater beni-zuwai crab ( Chionoecetes japonicus ) and triangle Tanner crab ( Chionoecetes angulatus) depended on molt stage, and mortality of new-shell crabs was greater than mortality of old-shell animals (Koblikov, 2004; Vasilyev and Kli- Moiseev et al: Effects of pot fishing on the physical condition of Chionoecetes opilio and Chionoecetes bairdi 243 Figure 8 Light micrographs of normal gills in snow crab (Chionoecetes opilio) and south- ern Tanner crab (C. bairdi) from our study of the effects of pot fishing on the physical condition of these species: (A) distal part of lamellae in a southern Tanner crab and (B) apical part of lamellae in a snow crab. Hematoxylin and eosin staining was used. Abbreviations: pc=pillar cells; mc=marginal canal. Scale bar=100 pm. nushkin, 2011). Tallack (2007) investigated mortality of red deepsea crab ( Chaceon quinquedens) in different hauling conditions (i.e., crabs hauled every 24 h, after intervals of 4 or 8 days) and found that crab hauled to the surface every 24 h showed significantly greater mortality than crab retrieved after 4 or 8 days. In our study, experiments on snow and southern Tanner crabs, like all experiments described in the previous paragraph, were conducted as actual field operations with minimal control of external variables, including air exposure, changes in temperature, light, and physical injuries. In our experiments, mortality of crabs was induced by blood sampling. For example, in experiment 1, mortality was 20% for snow crab and 24% for southern Tanner crab, and all the dead crabs that could be identified from their remains were crabs from which samples of blood were taken 1 or 2 times. This mortality most likely resulted from predation of amphipods (e.g., Orchomenella affinis ) that were at- tracted to the wounds caused by sampling. However, in experiment 3, mortality rates did not differ signifi- cantly between crabs from which samples of blood were taken (17%) and crab from which no blood samples were taken (15%). Another substantial source of crab mortality in our experiments was trauma associated with gear encoun- ters. Externally visible injuries to the carapace were found often in dead animals that had encountered gear. In all dead crabs, soft tissues were quickly eaten away by predatory amphipods. It can be assumed that, within the size range of crabs kept in the pots in our experiments (see the Materi- als and methods section), aggres- sive interaction between crabs and cannibalism were not significant factors in the mortality of ani- mals. Cannibalism in snow crab occurs only between individuals of significantly different sizes (Dutil et ah, 1997; Lovrich and Sainte- Marie, 1997). Our research on the attractiveness of dead crabs as bait showed that they did not at- tract individuals of their own spe- cies (senior author, unpubl. data). Therefore, mortality of crabs in our experiments was not a reliable indicator of the physiological dis- orders in crabs. Reversible physi- ological disorders did not result in death of the animals, and mor- tality of crabs often was caused by trauma from gear. During the course of our experiments in the field, the vitality of each crab was monitored with a VI. Vitality, or general health status of crabs, can be characterized subjectively by observing or testing behavioral or whole-animal re- sponses (reflex actions). The severity of physiological disorders in crabs is related directly to changes in vi- tality of crabs (Stoner, 2009, 2012). The haul and return of pots to the seabed caused shifts in the homeostasis of crabs, leading to changes in health status of the animals. In turn, the health sta- tus of the animals influenced their physical responses to stress. We used adult male snow crab and southern Tanner crab that had completed a terminal molt be- tween 1.5 years and a few months previously. There- fore, the vulnerability of animals in our research to external factors was less influenced by intrinsic vari- ables, such as processes of growth, maturation, or molt stage, than it would have been if we had used imma- ture or egg-producing female crabs. A temporary shift of physiological homeostasis in crabs caused by the process of pot hauling may be cor- rected by the natural regulatory capacity of an organism or by adaptive physiological responses. The investiga- tions on the mortality of snow crab in crab pots carried out by Ivanov and Karpinski (2003) revealed that crab weakened by frequent pot hauls could restore natural physiological conditions (i.e., locomotor activity and righting behavior) if the time intervals between suc- cessive pot lifts were increased. In their experiments, it was found that after 2 pot lifts at time intervals of 2 days, the surviving crabs were greatly weakened. How- ever, the condition of animals improved significantly af- 244 Fishery Bulletin 111(3) ter the pot had soaked for 4 days. After an additional 8 days of pot soaking, crabs recovered completely: all ani- mals were alive and their condition did not significantly differ from the condition of freshly caught animals. In our experiments, comparisons of the different hauling conditions also revealed that the vitality of In our experiments, [He] was selected as an objective mea- surement for the assessment of the effects of pot fishing on the physical condition of snow crab and southern Tanner crab. In addition to serving as a practi- cal and simple indicator, it can be easily related to respiratory function, which can be impaired because of air exposure and gill disorders that result from fish- ing and handling operations. The [He] in crustacean he- molymph is determined by the balance between anabolism and catabolism of the protein. The turnover rate of crustacean He is thought to be slow in usual steady-state conditions. Half- times for clearance of 125I-labeled He have been reported to be 25.5 and 36 days in American lobster ( Homarus americanus ) and Ca- ribbean spiny lobster ( Panulirus argus) (Senkbeil and Wriston, 1981). A more rapid synthesis of He can be induced by vari- ous environmental factors, such as changes of ambient salin- ity (Gilles, 1977; Pequeux et al., 1979) and by moderate chronic hypoxia (Hagerman and Baden, 1988; deFur et ah, 1990; Spicer and Baden, 2001). A decrease in [He] in hemo- lymph of crustaceans may be caused by starvation. It has long been noted that [He] is related to the nutritional state of crustaceans under natural conditions (Dali, 1974; and references therein), leading some researches to suggest that, in times of plenty, He serves not only as an 0-2 carrier but also as a storage protein. Terwilliger (1998) noted, “This may occur, although storing ex- Figure 9 Light micrographs of damages to gills in snow crab ( Chionoecetes opilio) and south- ern Tanner crab (C. bairdi) from our study of the effects of pot fishing on the physical condition of these species: (A) deformation of lamellae; note the irregu- lar shape of the marginal channels and vesicular swellings; (B) torn connections of pillar cells in the dilated parts of the lamellae; (C) swollen marginal canals filled with hemocytes and collapsed lamellae walls (marked by arrow); and (D) melanized inclusions in lamellae (arrows); note the deformation of lamellae and torn connections of pillar cells. Hematoxylin and eosin stains were used. Ab- breviations: mc=marginal canal; sl=swelling of lamellae; pc=normal pillar cells; t=torn pillar cells; and h=hemocytes. Scale bar=100 pm. animals was strongly dependent not only on the number of but also on the length of time be- tween pot lifts. During repeated pot hauls at short intervals (<3 days), we observed a rapid de- cline in vitality of crabs. When we increased hauling intervals to >3 days, the condition of crabs did not change significantly or was suppressed only moderately. Effects of pot fishing on hemocya- nin concentration Moiseev et al: Effects of pot fishing on the physical condition of Chionoecetes opilio and Chionoecetes bairdi 245 Table 3 Copper concentration, [Cu], measured in micrograms (g wet weight-1), in samples of the hepatopancreas of snow crab (Chi- onoecetes opilio) and southern Tanner crab (C. bairdi ) under different experimental conditions. [Cu] values are means (±1 standard error [SE] ). C. bairdi1 C. opilio2 Experimental conditions [Cul No. of samples [Cu] No. of samples Immediately after capture Starvation in a crab pot within a period of 55 days Repeated pot hauls at short time intervals (<3 days) 5.5(1.3) 7 37.3 (18.5)* 5 17.0(12.1) 6 0.92(0.07) 7 2.6 0.8) 6 1 19 September-12 November 2008, 6-11 October 2008, Sea of Okhtosk (area I). 2 17 October-3 November 2008, Sea of Okhtosk (area II). * Significantly different from that of individuals sampled immediately after capture (P<0.05), Student’s t-test. cess nutrients as circulating oligomeric proteins seems a bit expensive” (p. 1088). Nonetheless, studies have shown that concentrations of total protein and He in hemolymph steadily decrease in starved animals under laboratory conditions (Hagerman, 1983; and references therein). However, it is probable that some of these effects were a result not of reduction in the absolute amount of protein but of dilution of the protein. Dali (1974) showed that blood protein concentration in the longlegged spiny lobster ( Panulirus longipes ) decreased with starvation within 4 weeks, but this decrease was actually due to the increase in blood volume as solid tis- sues were metabolized. Djangmah (1970) observed that blood protein decreased from 7.9 g L_1 to 2.55 g L_1 and that there was a progressive accumulation of copper in the hepatopancreas of the common shrimp ( Crangon crangoti ) after 23 days of starvation. He suggested that the stored copper came from catabolism of He, which is, therefore, rendered usable as an energy source in a starving animal. Laboratory studies showed that the changes of [He] elicited by food availability occurred gradually over periods of many days (Djangmah, 1970; Hagerman, 1983). However, Spicer and Stromberg (2002) demonstrated that starvation had a significant effect on [He] in northern krill ( Meganyctiphanes nor- vegica ) over the course of a few hours (<10) under labo- ratory conditions. They used laboratory studies to in- terpret changes in krill [He] that they observed during the diel vertical migration of northern krill. Spicer and Stromberg (2002) attributed such dramatic decreases in [He] in starved individuals to the high metabolic rates of northern krill. During our experiments, crabs were not fed because it was not possible to control the amount of food con- sumed by individuals. Any food placed in crab pots is quickly eaten by predatory amphipods, which also are thought to attack weak or injured crabs (Ivanov and Karpinski, 2003). Because the duration of our experi- ments was up to several weeks, there was a need to determine to what extent the fall in crab [He] was re- lated to the forced starvation of animals during the experiments. In our experiments on the effects of star- vation on crabs, at pot soaking durations of 14, 16, and 25 days, [He] in snow crab and southern Tanner crab did not change significantly compared with [He] in crabs immediately after capture (Fig. 6A). A signifi- cant drop in mean [He] between crabs before and crabs after soaking was observed only in experiment 9, in which crabs soaked in a pot for 55 days (Fig. 6, B and C). We, therefore, assume that starvation had no sig- nificant effect on the changes in the [He] in crabs in all experiments that involved repeated pot hauls be- cause, in these experiments, [He] significantly changed over the course of a few days. A rapid decrease in [He] in crustaceans may be as- sociated with the processes of anaerobic metabolism, which is one of the major physiological responses to O2 deprivation. Increases of lactate and bicarbonate-carbon- ic acid during anaerobic metabolism result in increased hydrogen ion concentration and a drop in pH in crusta- cean hemolymph (Barrento et al., 2009; and references therein). Crustaceans have compensation mechanisms that act in response to acidosis through mobilization of buffer bases. However, during severe and prolonged hypoxia, the ability of an organism to compensate may be exhausted. When no further metabolic compensa- tion occurs, the pH level will drop below the tolerated range and, as a result, the rates of enzymatic reactions, ionic and osmoregulatory controls, and cell membrane structure will be affected. Poor health, morbidity, and mortality have been reported to be correlated with low pH values in crustaceans (Whiteley and Taylor, 1992; Ridgway et al., 2006). In laboratory experiments, Norway lobster ( Neplirops norvegicus) exposed to water with low concentration of dissolved O2 (<20% 02-saturation) showed a sudden and rapid decrease in [He] (Baden et al., 1990). The length of time before the decline began depended on the O2 concentration. When lobster were exposed to 10% 02- saturation, an immediate reduction of their mean [He] 246 Fishery Bulletin 111(3) Table 4 Ionic composition of hemolymph from snow and southern Tanner crabs under different experimental conditions. Ionic con- centration values are means (±1 standard error [SE]). n=no. crabs sampled. Ion concentration (mmol Lr1) Species Experimental conditions Na+ n K+ n Ca2+ n Mg2+ n ci- n Seawater from Bering Sea 449 - 9.5 — 9.8 — 51.1 — 523 — Seawater from Sea of Okhtosk 453 ~ 9.6 ~ 10.5 ~ 51.6 - 528 - C. opilio Repeated pot hauls at long time intervals (>3 days)7 Immediately after capture 465 (13) 4 12.6 (0.4) 7 8.3 (0.3) 16 36.0 (0.6) 10 482 (4) 10 After 3rd lift 443 (28) 3 9.2 (0.7)** 6 8.0 (0.4) 8 34.0 (1.5) 8 469 (17) 8 Long-term starvation in the crab pot (55 days)2 Immediately after capture 399 (15) 7 10.8 (0.5) 7 8.2 (0.4) 7 35.7 (1.5) 7 482.8 (19) 7 After starvation 374(33) 5 8.7 (0.9)* 5 7.2 (0.6) 5 32.6 (3.8) 5 463 (48) 5 C. bairdi Repeated pot hauls at long time intervals (>3 days)7 Immediately after capture 472 (5) 4 13.0 (0.8) 7 7.7 (0.3) 18 33.8 (0.7) 10 477 (5) 10 After 3rd lift 388 (78) 3 8.2 (2.1)* 6 7.5 (0.5) 7 30.8 (2.2) 7 444 (28) 7 Long-term starvation in the crab pot (25 days) 3 Immediately after capture 406 (25) 3 11.3 (0.6) 4 8.2 (0.3) 5 27.0 (1.1) 5 413 (38) 5 After starvation 386 (28) 3 8.9 (0.1) 3 8.3 (0.4) 9 29.4 (0.8) 9 400 (25) 9 Long-term starvation in the crab pot (55 days)2 Immediately after capture 367 (19) 6 10.0 (0.8) 6 7.7 (0.4) 6 31.1 (2.2) 6 458 (28) 6 After starvation 370 (23) 5 9.5 (1.1) 5 7.7 (0.3) 5 32.4 (2.7) 5 447 (34) 5 1 18 May-12 June 2006, Bering Sea (area I). 2 19 September-12 November 2008, Sea of Okhtosk (area I). 3 7 July-2 August 2008, Bering Sea (area I). * Significantly different from that of individuals sampled immediately after capture (P<0.05), Student’s f-test. ** Significantly different from that of individuals sampled immediately after capture (P<0.01), Student’s t-test. by 65% of the initial value occurred within 4 days. Ani- mals in water with 12% and 15% 02-saturation first increased their [He] for 15 and 36 days, respectively. After those time periods, a rapid decrease in [He] oc- curred at a rate similar to the one recorded for lobster in water with 10% 02-saturation. Mortality of animals during hypoxic experiments ranged from 40% to 100%. Glycogen depletion in muscles and in the hepatopan- creas of lobster also was seen after O2 deficiency, re- flecting a shift to anaerobic metabolism (Baden et al., 1994). However, loss of functional hemocyanin in the hemolymph of lobsters during these hypoxic experi- ments (as indicated by a reduction in oxyhemocyanin concentration measured with a spectrophotometer) did not correlate with loss of blood copper. Baden et al. (1994) suggested that, under conditions of hypoxic ex- periments, the functional integrity of He molecules is damaged, leading to changes of the 02-binding proper- ties of He. Their assumption seems reasonable because severe metabolic acidosis induces an increase in pro- tein degradation because of activation of proteolysis and nonenzymatic hydrolysis of proteins. In our experiments, respiratory acidosis, internal hypoxia, and anaerobic metabolism in crabs may have been caused by impaired gas exchange in damaged gills or exposure to air during handling on the deck. Howev- er, in our experiments, declines in [He] were not related to decreases in animal vitality. The highest decrease in [He] for snow and southern Tanner crabs was observed in the experiments with repeated pot hauls at long time intervals (>3 days), during which the vitality of animals was high. In animals considerably weakened during re- peated pot hauls at short time intervals (<3 days) or during prolonged air exposure, [He] was reduced to a lesser extent or not changed (Figs. 3, 4, and 5). The rapid decline in [He] observed in our experi- ments was, therefore, unlikely to have been due to deg- radation of the protein as a result of respiratory hypoxia and severe metabolic acidosis. Additionally, SDS/PAGE showed no changes in the subunit composition of He’s from snow crab and southern Tanner crab in our exper- iments. In individuals of both species of crabs, differ- ent proportions of the He subunits were sometimes ob- served, but these differences were not correlated with Moiseev et al: Effects of pot fishing on the physical condition of Chionoecetes opilio and Chionoecetes bairdi 247 experimental conditions. Therefore, the structure of the He’s remained unchanged in our experiments. Changes in [He] in our experiments, therefore, were likely caused by physical and functional impairments that resulted from pot hauling. Changes in vitality of animals and [He] depended not only on the number of pot lifts but also on the time intervals between them. Decreases in [He] were related to high vitality of crabs. These observations indicate that the physiological im- balances caused by repeated pot hauls can be corrected if the time interval until the next haul is sufficiently long. It is also likely that the stress response involves changes in [He]. Copper content in the hepatopancreas It has been postulated that copper metabolism in de- capod crustaceans associated with degradation of He (during starvation or before molting) includes the pro- cesses of copper translocation to the hepatopancreas and storage in the form of complexes with glutathione and copper-binding proteins (Djangmah, 1970; Brouwer et ah, 2002). Brouwer et al. (2002) showed that metabo- lism of the copper-binding protein metallothionein in the hepatopancreas of blue crab was related to molt stage. Metallothionein isoform 3 was present in premolt and softshell crab and was absent in hepatopancreas of intermolt crab. Emergence of this protein appears to co- incide with a decrease in He synthesis and increase in He degradation. In experiment 9, after starvation within a period of 55 days, a significant increase in mean copper concen- tration in the hepatopancreas of southern Tanner crab was found. We also observed a significant decrease in mean [He] in these crabs, compared with He levels in control animals immediately after capture. However, in experiments 6 and 7, which involved repeated pot hauls at short time intervals, there were no significant changes in hepatopancreas copper levels in snow and southern Tanner crabs that had reduced [He] (Table 3). Therefore, much of the copper that was released during He degradation caused by pot hauling did not accumu- late in the hepatopancreas and was probably excreted from the body. This result may be explained by low lev- els of copper-binding proteins in the hepatopancreas of adult males of snow and southern Tanner crabs be- cause they do not have periods of active synthesis and degradation of He associated with molt stage. Ionic composition of hemolymph Plasma ion (Na+, K+, Cl-, Ca2+, Mg2+) concentrations were measured to examine ionoregulatory changes at the gills of affected crabs. The snow and southern Tanner crabs are typical of other stenohaline marine crustaceans in that their hemolymph is isosmotic with their environmental seawater but the ionic composition of their hemolymph can differ considerably from the composition of seawater around them (Prosser, 1973). In the hemolymph of snow crab and southern Tanner crab, [K+] is considerably higher and [Mg2+] is consid- erably lower than concentrations in the surrounding seawater. Biochemical analysis of stress responses in crusta- ceans have shown that various stressors can alter ion concentrations in hemolymph and, therefore, that ions may be used as indicators of physiological stress (Ston- er, 2012; and references therein). However, the exact mechanisms and physiological causes of ionic changes are usually difficult to explain. Several studies have reported increased [Ca2+] and concentration of bicarbon- ate ion ( HCO3-) in crustacean hemolymph during aerial exposure, which may indicate the mobilization of cal- cium carbonate (CaC03) from internal sources, such as calcified exoskeleton to compensate for low pH (Taylor and Whiteley, 1989; Lorenzon et al., 2007). [Mg2+] can be elevated in hemolymph of open-wounded crabs because of entry of magnesium-rich seawater (Uhlmann et al., 2009). Changes in [K+], [Na+], and [Cl-] in hemolymph of crustaceans are highly variable with stress (Stoner, 2012; and references therein). In experiment 1, which involved repeated pot hauls at long time intervals, and in experiment 9, which investigated long-term starvation in pots, [K+] in he- molymph of snow crab and southern Tanner crab were significantly lower compared with [K+] in freshly caught crabs (Table 4). It should be noted that, in ex- periments 1 and 9, mean [He] in hemolymph of snow crab and southern Tanner crab were significantly lower compared with mean [He] in control crab immediately after capture. An essential role is played by K+ in the mainte- nance of the difference in electrical potential across the plasma membrane of a cell, typically referred to as the “membrane potential.” That physicochemical regula- tory function enables normal nerve impulse transmis- sion, normal contraction of muscle fibers, and normal heart function. Interestingly, changes in the ionic com- position of the hemolymph of crabs after repeated pot hauls were the same as changes in the ionic composi- tion during long-term starvation, when metabolic rate and physical activity probably were reduced. However, the meaning of these changes needs to be studied. Effects of decompression on crab condition Changes in ambient pressure during lifts of crabs in pots to the water surface are an unavoidable adverse factor that affects the crabs. Studies of the blood circu- lation of brachyuran crabs, including snow and south- ern Tanner crabs, indicate that fluctuations of ambient pressure have the greatest effect on the blood flowing in the gills (Taylor, 1990). The phyllobranchiate gills of brachyuran crabs consist of 2 alternating rows of close- ly spaced, flattened lamellae that extend forward and backward from a median shaft. Each lamella consists of single layer of epithelial cells lining the cuticle. Pillar cells, which are located in the cellular layers in oppo- 248 Fishery Bulletin 111(3) site walls of lamellae are connected to each other half- way in the hemolymph space. As common gill epithelial cells, pillar cells perform respiratory or salt transport- ing function depending on their localization within the gills and also are thought to play a role in stabilizing the lamellae against ambient and internal hydrostatic pressure (Compere et al., 1989; Johnson, 1980). Large fluctuations of ambient pressure can cause damage of the lamellae because the fragile pillar cells are not well adapted to withstand pressure forces. In our experi- ments, we observed disruption in the connections of pil- lar cells and collapse of the lamellae in gills of snow crab and southern Tanner crab that were subjected to repeated pot hauls (Fig 9). The phyllobranchiate gills of brachyuran crabs are supplied with low pressure venous hemolymph. The af- ferent to efferent drop in pressure and mean internal pressure are generally only a few centimeters of water (Blatchford, 1971). During normal forward ventilation of gills by the scaphognathites, the lamellae tend to be inflated by a small positive transmural pressure, which is the difference between the above-ambient pressure in the lamellar hemocoel and subambient mean pressure in the branchial chamber in crab. Under these condi- tions, the resistance to hemolymph flow through the cir- culatory channels in the gills of crabs is low (McMahon and Burnett, 1990). However, any damage to gill struc- ture could lead to a significant increase in gill resistance and consequently to increases in both the drop in pres- sure through the gills and in mean internal pressure (Taylor, 1990). Increased pressure in the injured gills likely would cause further gill damage and even greater impairment of gill function. The main factor that determines the internal pres- sure in the gills of crabs is thought to be the total vol- ume of internal fluid. One of the mechanisms for ad- justment of a total fluid volume is regulation of urine release (Taylor, 1990). He is the major extracellular protein of crab hemolymph; hence, it is [He] that deter- mines colloid osmotic pressure of blood plasma. In inter- molt marine crabs, the small excess of hydrostatic pres- sure of hemolymph over colloid osmotic pressure drives passive filtration, which forms primary urine (Mangum and Johansen, 1975). The decrease in [He] may elevate the driving force for urine formation and, consequently, may lead to a decrease in total volume of internal fluid in crabs. In our experiments, blood sampling from crabs became more difficult after repeated pot hauls because the exhaust velocity of hemolymph from the cut was re- duced. At the same time, the clotting time of hemolymph in experimental crabs was not changed or was increased compared with the clotting time in control animals (data not shown). These data indirectly indicate a decrease in blood pressure in the leg sinus that may be caused by a decrease in total volume of hemolymph. A decrease in total blood volume through a decrease in [He] could not only prevent further gill damage but also reduce cardiovascular workload. However, a de- crease in [He] causes a decline in the 02-carrying capac- ity of hemolymph that, in combination with impaired gas exchange in the damaged gills, could lead to a further decrease in supply of O2 to organs and tissues. Studies of the role of crustacean He in O2 transport have found complete oxygenation of He in postbranchial hemolymph but only partial deoxygenation of He in the tissues of normoxic, routinely active animals. The oxygenation level of He in the prebranchial hemolymph of resting normoxic animals (often called venous reserve) usually amounts to 50% or more of the O2 capacity of the hemo- lymph. Reduced O2 supply to the blood transport system or increased metabolic demand could result in depletion of the venous reserve. In normoxic conditions, the con- tribution of He to O2 transport significantly increases during physical exercises (e.g., locomotor activity) of ani- mals (Truchot, 1992; and references therein). In our previous studies of [He] in the hemolymph of red king crab ( Paralithodes camtschaticus ) in the Bar- ents Sea and in the Sea of Okhtosk, [He] during the molt cycle was closely correlated with the volume of the limbs filled with muscular tissue (Moiseeva and Moi- seev, 2008; Moiseeva and Moiseev, 2011). Those results also indicate that the O2 delivery system with He is required primarily to support high levels of locomotor activity of crabs. Therefore, reduction of [He] and a si- multaneous decrease in locomotor activity of crabs could optimize blood flow in gills damaged by fisheries opera- tions without a significant decrease in supply of O2 to organs and tissues. Decompression, which results from a rapid lift of a pot to the surface of the water, is not a natural envi- ronmental stressor for crabs. However, our experiments showed that crabs quite successfully adapted to the damage of organs and tissues caused by decompres- sion because they took advantage of existing physi- ological mechanisms. The gills of crabs are damaged often under natural conditions. In our experiments, in September-October 2010 in the northern Sea of Okh- tosk, gross examination of tissues and organs revealed areas of melanization and necrosis in the gills of 40% of 50 freshly caught snow crab that had no sign of shell disease. In some crab, parts of some of the gills were absent (data not shown). Histological examination of or- gans and tissues has revealed histopathological changes in gills in 95% of snow crab with signs of shell disease (Ryazanova, 2006). Gills serve an important role in the immune response of crustaceans. Aggregates of hemocytes and bacteria ap- pear in hemolymph in response to microbial pathogens and then become trapped in the narrow hemolymph spaces of gills, where they are melanized and persist for a long time. The formation of hemocyte aggregates in blue crab that were injected with bacteria led to increas- es in vascular resistance across the gills and subsequent significant increases in the drop of hydrostatic pressure across the gill circulation (Burnett et al., 2006). There- fore, impairment of respiratory circulation apparently occurs often during infectious and noninfectious gill disorders of crabs. Crabs would be expected to possess Moiseev et al: Effects of pot fishing on the physical condition of Chionoecetes opilio and Chionoecetes bairdi 249 mechanisms to cope with such impairments. The rapid degradation of He that we observed in our experiments may be one of these mechanisms. Increased degradation of blood protein, especially of He, has been observed in several euryhaline crab species during hyperosmotic stress (Gilles, 1977). In our experiments on snow and southern Tanner crabs, a similar phenomenon was ob- served in stenohaline crabs for the first time. A single lift in a pot to the water surface apparently does not have significant effects on a crab. All of the crabs in our experiments survived at least 1 pot lift, the lift when they were captured. Therefore, experiments 8 and 9 on long-term starvation for 25 and 55 days, re- spectively, also can be considered for study of the de- layed mortality of crabs after decompression. At the end of these experiments, all the animals were alive and had a high vitality. The negative effect of decompression sig- nificantly increases if the same crab is recaptured with- in a few days. However, the probability of such an event is very low. Crabs normally move actively in search of food; however, it is unlikely that discarded crabs will do so while they recover from their first capture. On the basis of our experiments, a crab can survive (without serious harm) a recapture event that occurs more than a few days after its first capture. However, a decrease in [He] due to stress response would pose a challenge for snow and southern Tanner crabs. If its O2 transport system fails to meet all of the demands for energy during locomotion, a crab would not be able to sustain its normal level of activity. The reduction in [He] in crabs after a single lift in a pot obviously would de- pend on the initial condition of the animals and external factors. As shown in our experiments, this reduction in [He] can be very significant. Several studies have shown that the synthesis of He in crustaceans depends on the quality of food eaten (Hagerman, 1983). It is assumed that food, not seawater, is the most important source of copper in crustaceans (Baden, 1990; and references therein). Copper reserves in the hepatopancreas may be used to reconstitute He, but significant accumulation of copper was not found in our experiments in the hepatopancreas of snow and southern Tanner crabs during degradation of He due to decompression. Therefore, the recovery of [He] in affect- ed crabs would depend greatly on the availability and quality composition of the fodder base in their habitat. Conclusions Our research has shown that pot fishing has a signifi- cant effect on the physical condition of snow and south- ern Tanner crabs. Gill damage caused by ambient pres- sure differences and gas-bubble disease could have long- term effects on the vitality of crabs returned to sea from commercial catches after sorting. Both species of crabs have effective mechanisms for physiological adaptation to the adverse effects of pot fishing. The decrease in [He] that we observed in affected crabs is a long-term adjustment of the internal fluid volume of crabs that reduces pressure in damaged gills and optimizes respi- ratory circulation. The vital activity of crabs returned to sea after sorting of catches would depend on their condition before capture and the challenges presented by the specific environmental conditions at the time of their release. Biochemical assays of He have proved to be a useful tool for the investigation of the effects of pot fishing on the physical condition of snow crab and southern Tanner crab. The plasticity of He content constitutes an efficient mechanism for crustaceans to cope with meta- bolic or environmental challenges. In our experiments, the greatest changes in [He] that resulted from stress response to impairment of blood flow in gills occurred in the most viable crabs. Therefore, [He] can be a useful indicator of the health of crabs in conditions that can lead to gill dysfunction. Acknowledgments This research was part of the studies of marine re- sources conducted by Russian Federal Research Insti- tute of Fisheries and Oceanography, Moscow, Russia. We thank the crews of the vessels Sorvind and Evening Star for valuable assistance during experiments. We are especially grateful to the captain of the Shprvind, V. Gubsky, for his unfailing help and support. We ap- preciate very much the critical reading by V. Bizikov of an earlier version of this manuscript. Literature cited Baden, S. R, M. H. Depledge, and L. Hagerman. 1994. Glycogen depletion and altered copper and man- ganese handling in Nephrops norvegicus following star- vation and exposure to hypoxia. Mar. Ecol. Prog. Ser. 103:65-72. Baden, S. P., L. Pihl, and R. Rosenberg. 1990. Effect of oxygen depletion on the ecology, blood physiology and fishery of the Norway lobster Nephrops norvegicus. Mar. Ecol. Prog. Ser. 67:141-155. Barrento, S., A. Marques, P. Vaz-Pires, and M. L. Nunes. 2009. Live shipment of immersed crabs Cancer pagu- rus from England to Portugal and recovery in stocking tanks: stress parameter characterization. ICES J. Mar. Sci. 67:435-443. Bell, T. A., and D. V. Lightner. 1988. A handbook of normal penaeid shrimp histology, 114 p. World Aquacult. Soc.. Baton Rouge, LA. Blatchford, J. G. 1971. Haemodynamics of Carcinus maenas < L. ). Comp. Biochem. Physiol. A: Comp. Physiol. 39:193-202. Borisov, V. V., V. Stepanenko, and V. F. Tolkacheva 2003. Processing technology of crab in the Barents Sea. In Red king crab in the Barents Sea, p. 299-311. PIN- RO Publ., Murmansk, Russia. [In Russian.] 250 Fishery Bulletin 111(3) Bridges, C. R. 2001. Modulation of haemocyanin oxygen affinity: prop- erties and physiological implication in a changing world. J. Exp .Biol. 204:1021-1032. Brouwer, M., R. Syring, and T. Brouwer-Hoexum. 2002. Role of copper-specific metallothionein of the blue crab, Callinectes sapidus, in copper metabolism associ- ated with degradation and synthesis of hemocyanin. J. Inorg. Biochem. 88:228-239. Burnett, L. E., J. D. Holman, D. D. Jorgensen, J. L. Ikerd, and K. G. Burnett. 2006. Immune defense reduces respiratory fitness in Callinectes sapidus, the Atlantic blue crab. Biol. Bull. 211:50-57. Comeau, M., and Conan, G. Y. 1992. Morphometry and gonad maturity of male snow crab, Chionoecetes opilio. Can. J. Fish. Aquat. Sci. 49:2460-2468. Compere, Ph., S. Wanson, A. Pequeux, R. Gilles, and G. Goffinet. 1989. Ultrastructural changes in the gill epithelium of the green crab Carcinus maenas in relation to the ex- ternal salinity. Tissue Cell 21:299-318. Conan, G.Y., and M. Comeau. 1986. Functional maturity and terminal molt of male snow crab, Chionoecetes opilio. Can. J. Fish. Aquat. Sci. 43:1710-1719. Dali, W. 1974. Indices of nutritional state in the western rock lobster, Panulirus longipes (Milne Edwards). 1. Blood and tissue constituents and water content. J. Exp. Mar. Biol. Ecol. 16:167-180. Darnall, M. Z., K. M. Darnell, R. E. McDowell, and D. Rittschof. 2010. Postcapture survival and future reproductive potencial of ovigerous blue crabs Callinectes sapidus caught in the central Noth Carolina pot fishery. Trans. Am. Fish. Soc. 139:1677-1687. deFur, P. L., C. P. Mangum, and J. E. Reese. 1990. Respiratory responses of the blue crab Callinectes sapidus to long-term hypoxia. Biol. Bull. 178:46-54. Djangmah, J. S. 1970. The effects of feeding and starvation on copper in the blood and hepatopancreas, and on blood proteins of Crangon vulgaris (Fabricius). Comp. Biochem. Physiol. 32:709-731. Dutil, J.-D., J. Munro, and M. Peloque. 1997 Laboratory study of the influence of prey size on vulnerability to cannibalism in snow crab (Chionoece- tes opilio O. Fabricius, 1780). J. Exp. Mar. Biol. Ecol. 212:81-94. Gilles, R. 1977. Effects of osmotic stresses on the proteins concen- tration and pattern of Eriocheir sinensis blood. Comp. Biochem. Physiol. A: Comp. Physiol. 56:109-114. Giomi, F., and M. Beltramini. 2007. The molecular heterogeneity of hemocyanin: its role in the adaptive plasticity of Crustacea. Gene 398:192-201. Hagerman, L. 1983. Haemocyanin concentration of juvenile lobsters (Homarus gammarus) in relation to moulting cycle and feeding conditions. Mar. Biol. 77:11-17. Hagerman, L., and S. P. Baden. 1988. Nephrops norvegicus : field study of effects of oxy- gen deficiency on haemocyanin concentration. J. Exp. Mar. Biol. Ecol. 116:135-142. Ivanov, B. G., and M. G. Karpinskiy. 2003. Mortality of crabs trapped in pots: snow crab in the northern Sea of Okhotsk. Problems Fisheries 4:590—607. [In Russian.] Jadamec, L. S., W. E. Donaldson, and P. Cullenberg. 1999. Biological field techniques for Chionoecetes crabs. Alaska Sea Grant College Program Report AK-SG-99-02, 80 p. Univ. Alaska, Fairbanks, AK. doi:10.4027/bftcc. 1999. Johnson, P. T. 1976. Gas-bubble disease in the blue crab, Callinectes sapidus. J. Invertebr. Pathol. 27:247-253. 1980. Histology of the blue crab, Callinectes sapidus: a model for Decapoda, 440 p. Praeger, New York. Kirichenko, S. G., O. A. Galutva, and V. F. Poluyaktov. 2005. Determination of some chemical elements by atomic absorption. In Quality, safety and methods of analysis of products from aquatic organisms. Guidelines for determining the content of toxic substances in prod- ucts from aquatic organisms (L. P. Kopylenko, ed.), p. 3-6. VNIRO Publ., Moscow. [In Russian.) Koblikov, V. N. 2004. On the mortality of Japanese red crab ( Chionoece- tes japonicus) in the crab pots and some aspects of the crab fisheries in the northern Japan Sea. Problems Fisheries 5:458-469. [In Russian.] Laemmli, U. K. 1970. Cleavage of structural proteins during the as- sembly of the head of bacteriophage T4. Nature 227:680-685. Lorenzon, S., P. G. Giulianini, M. Martinis, and E. A. Ferrero. 2007. Stress effect of different temperatures and air ex- posure during transport on physiological profiles in the American lobster, Homarus americanus. Comp. Bio- chem. Physiol., A: Mol. Integr. Physiol. 147:94-102. Lovrich, G. A., and B. Sainte-Marie. 1997. Cannibalism in the snow crab Chionoecetes opilio (O. Fabricius) (Brachyura: Majidae), and its potential importance to recruitment. J. Exp. Mar. Biol. Ecol. 211:225-245. Mangum, C. P, and K. Johansen. 1975. The colloid osmotic pressures of invertebrate body fluids. J. Exp. Biol. 63:661-671. McMahon, B. R., and L. E. Burnett. 1990. The crustacean open circulatory system: a reex- amination. Physiol. Zool. 63:35-71. Mihailov, V. I., K. V Bandurin., A. V. Gornichich, and A. N. Karasev. 2003. Commercial invertebrates of shelf and continental slope of the northern part of the Okhotsk sea. 284 p. Magadan NIRO Publ., Magadan, Russia. [In Russian.] Moiseev, S. I. 2003. A study of effectiveness of different crab pots in coastal zone of the Barents Sea. In Bottom ecosystems of the Barents Sea. VNIRO Proceedings, vol. 142 (V. I. Sokolov, ed.), p. 178-191. VNIRO Publ., Moscow. [In Russian.] Moiseev, S. I., S. A. Moiseeva, and A. M. Lapteva. 2012. Change of hemolymph parameters in Tanner crab due to the stress caused by fishing with crab pots. Problems Fisheries 13:125-144. [In Russian.] Moiseev et al: Effects of pot fishing on the physical condition of Chionoecetes opilio and Chionoecetes bairdi 251 Moiseeva, S.A., and S. I. Moiseev. 2008. Relation between muscular tissue condition in limbs and hemocyanin concentration in the hemo- lymph of red king crab (Pctralithodes camtschaticus) in the Barents Sea. Problems Fisheries 9:200-217. [In Russian.] 2011. Relationship between hemocyanin concentration in the hemolymph and filling of limbs with muscu- lar tissues of the red king crab (Paralithod.es camts- chaticus) during postmolt period. Problems Fisheries 12:332-348 [In Russian.] Nickerson, K. W., and K. E. van Holde. 1971. A comparison of mollusean and arthropod hemo- cyanin— I. Circular dichroism and absorption spectra. Comp. Biochem. Physiol. B: Comp. Biochem. 39:855-872. Pequeux, A., A. C. Vallota, and R. Gilles. 1979. Blood proteins as related to osmoregulation in Crustacea. Comp. Biochem. Physiol. A: Comp. Physiol. 64A:433-435. Prosser, C. L. 1973. Comparative animal physiology, 3rd ed., 1011 p. W. B. Saunders Publ. Co., Philadelphia, PA. Ridgway, I. D., A. C. Taylor, R. J. A. Atkinson, G. D. Stentiford, E. S. Chang, S. A. Chang, and D. M. Neil. 2006. Morbidity and mortality in Norway lobsters, Nephrops norvegicus: physiological, immunological and pathological effects of aerial exposure. J. Exp. Mar. Biol. Ecol. 328:251-264. Ryazanova, T. V. 2006. Histopathological changes among the snow crab (Chionoecetes opilio) on the westkamchatkan shelf of the Okhotsk Sea. In Research of water biological re- sources of Kamchatka and of the northwest part of Pa- cific Ocean (Y. P. Diakov, ed.), vol. 8, p. 207-216. Ka- mchatNIRO Publ., Petropavlovsk-Kamchatsky, Russia. [In Russian.] 2009. Development of bacterial infections and gas bubble disease in crabs owing to lifting in pots. In Research of water biological resources of Kamchatka and of the northwest part of Pacific Ocean (Y. P. Dia- kov, ed.), vol. 13, p. 95-100. KamchatNIRO Publishing, Petropavlovsk-Kamchatsky, Russia. [In Russian.] Sainte-Marie, B., S. Raymond, and J.-C. Brethes. 1995. Growth and maturation of the benthic stages of male snow crab, Chionoecetes opilio (Brachyura: Maji- dae). Can. J. Fish. Aquat. Sci. 52:903-924. Senkbeil, E. G., and J. C. Wriston. 1981. Catabolism of hemocyanin in the American lob- ster, Homarus americanus. Comp. Biochem. Physiol. B: Comp. Biochem. 69:781-790. Slizkin, A.G. 2010. Atlas of Russian Far East crabs and shrimps, 256 p. TINRO-Center Publ., Vladivostok. Russia. [In Russian.] Slizkin, A.G., V.N. Koblikov, and E.R. Shaginyan. 2001. Snow crab, Chionoecetes bairdi Rathbun, of the northwestern Pacific: abundance dynamics, size compo- sition and reproduction fautures. In Study of biology of commercial crustaceans and algae of Russia seas. Collected papers (B. G. Ivanov, ed.), p. 75-91. VNIRO Publ., Moscow [In Russian.] Spicer, J. I., and S. P. Baden. 2001. Environmental hypoxia and haemocyanin between- individual variability in Norway lobster Nephrops nor- vegicus (L). Mar. Biol. 139:727-734. Spicer, J. I., and J. O. Stromberg. 2002. Diel vertical migration and the haemocyanin of krill Meganyctiphanes norvegica. Mar. Ecol. Prog. Ser. 238:153-162. Stoner, A. W. 2009. Prediction of discard mortality for Alaskan crabs after exposure to freezing temperatures, based on a re- flex impairment index. Fish. Bull. 107:451-463. 2012. Assessing stress and predicting mortality in economically significant crustaceans. Rev. Fish. Sci. 20:111-135. Tallack, S. M. L. 2007. Escape ring selectivity, bycatch, and discard sur- vivability in the New England fishery for deep-water red crab, Chaceon quinquedens. ICES J. Mar. Sci. 64:1579-1586. Taylor, H. H. 1990. Pressure-flow characteristics of crab gills: implica- tions for regulation of hemolymph pressure. Physiol. Zool. 63:72-89. Taylor, E. W., and N. M. Whiteley. 1989. Oxygen transport and acid-base balance in the haemolmyph of the lobster, Homarus gammarus, dur- ing aerial exposure and re-submersion. J. Exp. Biol. 144:417-436. Terwilliger, N. B. 1998. Functional adaptations of oxygen-transport pro- teins. J. Exp. Biol. 201:1085-1098. Truchot, J. P. 1992. Respiratory function of arthropod hemocyanins. Adv. Comp. Environ. Physiol. 13:377-410. Uhlmann, S. S., M. K. Broadhurst, B. D. Paterson, D. G. Mayer, P. Butcher, and C. P. Brand. 2009. Mortality and blood loss by blue swimmer crabs ( Portunus pelagicus) after simulated capture and dis- carding from gillnets. ICES J. Mar. Sci. 66:455-461. van Tamelen, P. G. 2005. Estimating handling mortality due to air expo- sure: development and application of termal models for the Bering Sea snow crab fishery. Trans. Am. Fish. Soc. 134:411-429. Vasilyev, A. G., and S. V. Klinushkin. 2011. Data on mortality of the triangle Tanner crab (Chionoecetes angulatus) in fishing traps in the north- ern part of the Sea of Okhotsk. Problems Fisheries 12:566-575. [In Russian. | Whiteley, N. M., and E. W. Taylor. 1992. Oxygen and acid-base disturbances in the hemo- lymph of the lobster Homarus gammarus during com- mercial transport and storage. J. Crust. Biol. 12:19-30. 252 Abstract— The reproductive biology of Yellowfin Tuna ( Thunnus alba- cares) in the western Indian Ocean was investigated from samples col- lected in 2009 and 2010. In our study, 1012 female Yellowfin Tuna were sampled: 320 fish on board a purse seiner and 692 fish at a Sey- chelles cannery. We assessed the main biological parameters that de- scribe reproductive potential: matu- rity, spawning seasonality, fish con- dition, and fecundity. The length at which 50% of the female Yellowfin Tuna population matures (L50) was estimated at 75 cm in fork length (FL) when the maturity threshold was established at the cortical al- veolar stage of oocyte development. To enable comparison with previous studies, L50 also was estimated with maturity set at the vitellogenic stage of oocyte development; this assess- ment resulted in a higher value of L50 at 102 cm FL. The main spawn- ing season, during which asynchrony in reproductive timing among sizes was observed, was November-Feb- ruary and a second peak occurred in June. Smaller females (<100 cm FL) had shorter spawning periods (December to February) than those (November to February and June) of large individuals, and signs of skip- spawning periods were observed among small females. The Yellowfin Tuna followed a “capital-income” breeder strategy during ovarian development, by mobilizing accu- mulated energy while using incom- ing energy from feeding. The mean batch fecundity for females 79-147 cm FL was estimated at 3.1 million oocytes, and the mean relative batch fecundity was 74.4 oocytes per gram of gonad-free weight. Our results, obtained with techniques defined more precisely than techniques used in previous studies in this region, provide an improved understanding of the reproductive cycle of Yellowfin Tuna in the western Indian Ocean. Manuscript submitted 7 September 2012. Manuscript accepted 20 May 2013. Fish. Bull. 111:252-264 (2013). doi 10.7755/FB. 111.3.4 The views and opinions expressed or implied in this article are those of the author (or authors) and do not necesarily reflect the position of the National Marine Fisheries Service, NOAA. Reproductive potential of Yellowfin Tuna ( Thunnus albacares ) in the western Indian Ocean Iker Zudaire (contact author)1 Hilario Murua1 Maitane Grande1 Nathalie Bodin2 Email address for contact author: iker.zuda@gmail.com 1 Marine Research Division AZTI-Tecnalia Herrera Kaia Portu aldea z/g 201 10 Pasaia, Spain 2 CRH UMR 212 EME Institut de Recherche pour le Developpement Av Jean Monnet BP 171 34203 Sete, France Knowledge of reproductive traits is important for understanding popu- lation dynamics, including a popula- tion’s resilience to fishing (Schaefer, 2001; Murua and Motos, 2006; Mor- gan et al., 2009). As alternatives to the traditional spawning stock bio- mass (SSB), reproductive potential indices have been proposed in which basic reproductive parameters are included as important factors that affect population productivity (Trip- pel, 1999; Morgan et ah, 2009). These parameters include sex-ratio, the age and size of females, maturation ogive, fecundity, fish condition, and repro- ductive history. Inclusion of these biological parameters allows integra- tion of fluctuations in a population’s reproductive success into the assess- ment and management processes, in addition to estimation of spawning stock biomass (SSB) (Murua and Saborido-Rey, 2003; Murua et al., 2010). Hence, to improve the assess- ment and management of stocks, it is necessary to increase the quality and quantity of the basic reproduc- tive data used to estimate these re- productive parameters (Korta, 2010). Yellowfin Tuna ( Thunnus alba- cares) is one of the major target spe- cies of the tuna fishery in the Indian Ocean. Total annual catch of Yellow- fin Tuna in the Indian Ocean has in- creased significantly, since the early 1980s, with the advent of the purse- seine fishery. Average annual catch reached 473,896 metric tons (t) be- tween 2003 and 2006 but decreased in 2007 and 2008 to around 320,000 t; in 2011, total annual catch was around 300,000 t (IOTC1). The Yellowfin Tuna is a batch- spawner with asynchronous ovary organization (Schaefer, 2001) and indeterminate fecundity (Zudaire et al., 2013). This species can spawn with a frequency of around 1.5 days (McPherson, 1991; Schaefer, 2001) over a vast area of the tropical zone throughout the year (Itano2; Stequert et ah, 2001). Its spawning events, as with other tuna species, occur in rela- 1 IOTC (Indian Ocean Tuna Commission). 2012. Report of the fifteenth session of the IOTC Scientific Committee. Victoria, Seychelles. IOTC Secretariat, P.O. Box 1011, Victoria, Seychelles. Zudaire et al. : Reproductive potential of Thunnus albacares in the western Indian Ocean 253 Figure t Map of locations where female Yellowfin Tuna (Thunnus albacares ) were sam- pled in the first survey (22 January-23 March 2009), second survey (6 May- 25 July 2009), and third survey (12 January-13 April 2010). The surveys were conducted aboard a commercial purse seiner in the western Indian Ocean to study the reproductive potential of Yellowfin Tuna in this region. tion to sea-surface temperature (>24°C), which seems to regulate spawning activ- ity (Schaefer, 2001). In the Indian Ocean, spawning mainly occurs in the equatorial area (0— 10°S) from December to March, and the main spawning grounds have been observed west of 75°E (IOTC1). However, other authors have reported variations in the spawning season of this species, describing periods between January and June (Zhu et al., 2008). Stequert et al. (2001) related reproduc- tive activity of Yellowfin Tuna with the north monsoon (main spawning period) from November to March and with the south monsoon (less reproductive activ- ity) from June to August. Similarly, dif- ferent estimates of maturity have been published for this species in several stud- ies that used the same method to iden- tify maturation stages. For example, the length at which 50% of females mature (L50) was estimated at around 100 cm in fork length (FL) for Yellowfin Tuna in the Indian Ocean by the Indian Ocean Tuna Commission (IOTC1), at 108 cm FL in the western Pacific Ocean by McPherson (1991), at 92 cm FL in the eastern Pacif- ic Ocean by Schaefer (1998), and at 104 cm FL for the equatorial western Pacific Ocean by Itano.2 In spite of the ecological importance of Yellowfin Tuna and the significant decrease of its catches in various re- gions during recent years, there have been few studies that have focused on the reproductive biology of the Indian Ocean Yellowfin Tuna. This study aims to contribute to the knowledge of various reproductive traits essential for examination of the reproductive potential and population dynamics of Yellowfin Tuna in the western Indian Ocean, with an estimation of the following qualities: 1) length at 50% maturity; 2) spawning season; 3) fecundity; and 4) fish condition during reproduction. Materials and methods Field sampling Sampling was done by scientific observers during 3 sur- veys conducted on board a commercial purse seiner in 2 Itano, D. G. 2000. The reproductive biology of yellowfin tuna (Thunnus albacares ) in Hawaiian waters and the west- ern tropical Pacific Ocean: project summary. Pelagic Fisher- ies Research Program (PRFP), Joint Inst. Mar. Atmospheric Research (JIMAR), Univ. Hawaii (UH), HI. JIMAR contri- bution 00-328, p. 69. JIMAR, MSB 312, 1000 Pope Road, Honolulu, HI 96822. the western Indian Ocean during 2009 and 2010 (Fig. 1). At sea, during these surveys, 320 female Yellowfin Tuna were sampled. In addition, 692 ovaries from fe- male Yellowfin Tuna that were captured by the purse- seine fleet that operated in the western Indian Ocean during the same period were collected at the Seychelles cannery (Table 1). Yellowfin Tuna were identified at sea and in the cannery through the use of characters given in Collette and Nauen (1983). Fork length (in centime- ters), total weight (in kilograms), sex, maturation stage and gonad weight (in grams) were recorded for each in- dividual. Liver weight (in grams) was measured only in females that were sampled at sea. Their ovaries were removed and weighed to the nearest gram. A cross sec- tion of the gonad of 4-5 cm was cut between the middle and end part of the right or left lobe and preserved in 4% buffered formaldehyde immediately after collection until it could be processed in the laboratory. In the laboratory, cross sections (~1 cm) from the preserved portions of 819 ovaries were embedded in resin, sectioned at 3-5 pm, and stained with hematoxi- lyn and eosin (Korta, 2010). The histological classifi- cation of Yellowfin Tuna ovaries followed the criteria 254 Fishery Bulletin 111(3) Table 1 Period of sampling, ranges of fork lengths of tuna sampled, number of females sampled in each survey, and number of fe- males used in different analyses for our study of the reproductive potential of Yellowfin Tuna ( Thunnus albacares ) in the western Indian Ocean. Analyses included the histological analysis of ovaries (Hist.), analysis of fecundity (Fee.), and analysis of the condition indices of gonadosomatic index (GSI), hepatosomatic index (HSI), and condition factor (K). Survey Sampling period Length (cm) Females sampled Hist. Fee. GSI HSI K 1 22/01/2009-23/03/2009 37-158 114 110 6 106 105 106 2 5/06/2009-25/07/2009 30-161 95 95 5 95 77 95 3 03/04/2010-21/05/2010 31-157 114 114 - 95 94 95 Cannery 12/01/2010-13/04/2010 61-147 692 500 31 511 - 511 described in Zudaire et al. (2013). On the basis of ter- minology used in Brown-Petersen et al. (2011) and ap- plied to Yellowfin Tuna (Zudaire et al., 2013), the ova- ries were classified according to the most advanced oocyte stage present in the ovary: immature phase (including the primary growth stage [PG]), developing phase (including the cortical alveolar [CA], primary vitellogenesis [Vtgl], and secondary vitellogenesis [Vtg2] stages), spawning-capable phase (including the tertiary vitellogenesis [Vtg3], germinal vesicle migra- tion [GVM], and hydration stages), and regenerating phase. The different stages of alpha-atresia were clas- sified according to Zudaire et al. (2013). The ovaries collected at the cannery were not included in the analyses of atresia because they had been exposed to brine preservation used on board purse seiners, mak- ing it impossible to quantify alpha-atresia precisely. The ovaries collected at sea and at the cannery were analyzed for the identification of postovulatory fol- licles; however, no postovulatory follicles were found in these ovaries. Length at 50% maturity All females included in this analysis were staged histo- logically. L50 was calculated by fitting the proportion of mature females by 5-cm size classes to a logistic equa- tion (Ashton, 1972; Saborido-Rey and Junquera, 1998): ^mature = e^L/l+ea^L, where Pmature = the predicted proportion of mature females; L = the FL in centimeters; and a and p are the coefficients of the logistic equation. The L50 was estimated as the ratio of the coefficients (-a*p-1). A nonlinear regression (the Marquardt meth- od without restrictions; Marquardt, 1963) was used to fit the logistic equation to the data. The curve of L50 was estimated on the basis of the assumption that fe- males with ovaries at the cortical alveolar stage on- ward were mature (Brown-Peterson et al., 2011). A second criterion was used to enable comparisons of our results with results reported in previous publications (Schaefer, 1998; Itano2; Zhu et al., 2008): females with cortical alveolar oocytes at the most advanced develop- mental stage were considered immature and females with advanced vitellogenic oocytes were considered mature. Condition indices Three condition indices were measured to estimate the condition of females: the gonadosomatic index (GSI), hepatosomatic index (HSI), and condition factor (K). These 3 indices were defined in this manner: GSI = (Wg/W) x 102; HSI=(W[/W) x 102; K=(W / L3) x 102; where Wg = gonad weight; W[ = liver weight; W = fish gonad-free weight in grams; and L - FL in centimeters. The seasonal variations in those indices and their ef- fect on the reproductive cycle were analyzed by month. To identify possible physiological changes during the ontogeny of Yellowfin Tuna, the seasonal develop- ment of the 3 condition indices was analyzed in 2 size groups: fish <100 cm FL and fish >100 cm FL — the Lso’s adopted by the Indian Ocean Tuna Commission [IOTC1]). Tuna sampled at the cannery were not in- cluded in the HSI analysis because the livers of these specimens were not weighed. Fecundity estimation Batch fecundity (BF), the total number of oocytes re- leased per batch, was estimated for 40 ovaries with the gravimetric method (Hunter et al., 1989) by counting the oocytes in the germinal vesicle migration or hy- dration stages. Homogeneity in oocyte density among Zudaire et a!.: Reproductive potential of Thunnus ciibacares in the western Indian Ocean 255 Total length (cm) Figure 2 Proportion of mature female Yellowfin Tuna ( Thunnus albacares ) in the western Indian Ocean at 5-cm length intervals and fitted to a logistic regression curve in our study of the reproduction potential of this species in this region. Circles represent the proportions of females considered mature when their ovaries were at the cortical alveolar and later stages; the gray solid line indicates the logistic regression curve for these females. Crosses represent the propor- tions of females considered mature when their ovaries were at the vitellogenic and later stages; the dark solid line indicates the logis- tic regression curve for these females. The horizontal, dotted line indicates L50, the length at which 50% of the female Yellowfin Tuna population were mature. whole ovaries was assumed on the basis of previous studies on tuna (Stequert and Ramcharrun, 1996). For BF analyses, 3 subsamples of 0.1 g (±0.01) were collected from each ovary. Each subsample was satu- rated with glycerin and oocytes were counted under a stereomicroscope (Schaefer, 1987, 1996, 1998). Batch fecundity was calculated as the weighted mean den- sity of the 3 subsamples multiplied by the total weight of the ovary. A threshold of 10% for the coefficient of variance was applied for the 3 subsamples, and when this threshold was surpassed, more subsamples were counted to reach it. Relative batch fecundity (BFrel) was estimated by dividing the BF by the gonad-free weight of the fish. The relationships between the BF and BFrel with the FL, weight, and condition indices (GSI, HSI, and K) of females were determined. The sea- sonal trend in fecundity was analyzed through estima- tion of monthly mean BF and BFrel. Statistical analyses A nonparametric Kruskal-Wallis test (H- test) was applied to determine differences in the GSI, HSI, and K among months and be- tween maturation stages. The relationships between the BF and BFrel and other bio- logical parameters, such as length, weight, and condition indices (HSI and K), were analyzed through application of simple cor- relation and regressions. Analysis of vari- ance (ANOVA) was applied to analyze the differences in the estimates of mean BF and mean BFrel by month during the spawning season. Results Length at 50% maturity L50 was estimated to be 75 cm FL when females with ovaries at the CA stage and onward were considered mature. This es- timate increased to 102 cm FL when the second criterion was applied (i.e., when the maturity threshold was defined as the pres- ence of advanced vitellogenic oocytes, Fig. 2). In both cases, the proportion of mature females by length provided a good fit to the logistic model (coefficient of determination [r2]=0.89 and r2=0.91, respectively) (Table 2). Reproductive cycle The analysis of the female maturation pro- cess throughout the year showed that 30.1% of the individuals sampled were in the im- mature phase, 44.4% were in the developing phase, 20.3% were in the spawning-capable phase, and 5.12% were in the regenerating phase (Ta- ble 3). Overall, 69.8% of the sampled females were in the mature state, 92.6% of which were reproductively active. The analysis of the maturation process was car- ried out for 2 size groups: individuals >100 cm FL and individuals <100 cm FL. The females >100 cm FL showed ovaries more developed and closer to spawning from November to February and in June than in other months (Fig. 3A). The period between November and January was especially important because more than 90% of the females sampled were in the spawning-ca- pable phase. In contrast, during April and May, there was no spawning activity for this size group (fish >100 cm FL), with 40% and 30% of the ovaries in the regen- erating phase in each month, respectively. The occur- rence of immature phase ovaries increased to 50% and 38% in April and May. 256 Fishery Bulletin 111(3) [SD 0.2]), January (0.52 [SD 0.3]), and February (0.43 [SD 0.3]), when the GSI increased slightly during the main spawning period of large speci- mens of Yellowfin Tuna. There was no increase in the mean GSI in June for the smaller size group (Fig. 4B). The monthly variation in HSI and K of both size groups showed statistically significant dif- ferences between months: HSI for fish <100 cm FL (77154=29.1, P<0.01) and for fish >100 cm FL Cffi22=26.3, P<0.01), and K for fish <100 cm FL (/J669=175.5, P<0.01) and for fish <100 cm FL (7/313=30.6, P<0.01). For females >100 cm FL, K values had a pattern that was opposite the one seen in the GSI. The K of large females was low from November (1.82 [SD 0.1]) to January (1.80 [SD 0.0]) and increased from February (1.82 [SD 0.0]) to July (1.89 [SD 0.2]) (Fig. 4A). Except for January, the mean HSI values showed a similar pattern to the one observed in the GSI. The HSI values of females >100 cm FL decreased from February (1.27 [SD 0.3]) to April (0.61 [SD 0.1]) and then increased slightly from May (0.70 [SD 0.1]) to July (0.72 [SD 0.2]) (Fig. 4A). The data series for the HSI was shorter than the data se- ries for the other 2 condition indices because no liver samples were collected in the cannery. Oocyte developmental stage and condition indices There were significant differences in the GSI (T7275=162.1, P<0.01), HSI (77275=49.9, P<0.01), and K (77275=10.68, P<0.05) between ovarian developmental phases. The GSI showed lowest values at the prima- ry growth stage (0.23 [SD 0.1]) and values increased throughout the vitellogenesis process (1.13 [SD 0.8]) until the maximum values at GVM (2.17 [SD 0.7]) and hydration (1.98 [SD 0.8]) stages were reached. After- ward, the GSI showed a sharp decrease in the regen- erating phase (0.44 [SD 0.3]) (Fig. 5). The HSI showed a decreasing trend from immature phase ovaries (0.94 [SD 0.2]) to ovaries in vitellogenesis (0.77 [SD 0.2]), and then it followed the pattern shown by the GSI, increasing from vitellogenesis to hydration stages (1.33 [SD 0.3]). The HSI had its minimum value in the regen- erating phase (0.66 [SD 0.1]). The K followed the oppo- site trend from that of the GSI and HSI; it decreased from the immature phase (1.94 [SD 0.2]) throughout the maturation process, obtaining minimum values at the hydration stage (1.77 [SD 0.0]). The K increased in the regenerating phase (1.85 [SD 0.1]). Fecundity Estimation The estimated mean BF was 3.07 million oocytes and varied from 0.32 million to 6.91 million oocytes. The estimated mean BFrel was 74.4 oocytes per gram of go- nad-free weight and fluctuated between 9.2 and 180.8 oocytes per gram of gonad-free weight. Batch fecundity Table 2 Summary of logit parameters of the female maturity (L5o=length at which 50% of the population is mature) curve for 2 criteria used in our study of Yellowfin Tuna ( Thunnus albacares ) reproduction in the western Indian Ocean. The symbols a and p represent the coefficients of the logistic equa- tion, and r2 is the coefficient of determination. In criterion 1, the L50 was calculated with the assumption that females with ovaries at the cortical alveolar stage onward were mature. The L50 in the criterion 2 was estimated with the assumption that females with cortical alveolar oocytes as the most advanced developmental stage were immature and those with advanced vitellogenic oocytes were mature. Criterion 1 Criterion 2 Parameters a P a P Estimate Standard error -8.654 1.604 0.113 0.021 -6.965 1.246 0.068 0.012 Estimates Estimates Number of females 423 423 L50(-a/p) 74.7 cm 102.0 cm r2 0.89 0.91 The females <100 cm FL showed a considerable pre- dominance of immature phase ovaries from April to November with levels of 50% or higher (Fig. 3B). These percentages of immature phase ovaries decreased be- low 30% during December, January, and February, and the percentage of ovaries in the spawning-capable phase increased to 4.4%, 5.3%, and 6.9%, respectively, for each month. It is noteworthy that ovaries in the regenerating phase appeared during the spawning sea- son for females <100 cm FL, with values of 8.9% and 10.5% in December and January, but ovaries in this phase for larger Yellowfin Tuna (>100 cm FL) appeared in April and May. Condition indices The monthly variation in the GSI of the previously defined 2 size groups showed statistically signifi- cant differences between months for fish <100 cm FL (77669=136.9, P<0. 01) and fish >100 cm FL (77313 = 206.4, P<0.01). Females >100 cm FL had a peak in mean GSI values that coincided with the highest proportion of females in the spawning-capable phase from No- vember to January (GSI>2.0). Afterward, mean GSI values decreased sharply with a minimum value in April of 0.26 (standard deviation [SD] of 0.0). In June, the mean GSI increased again to 0.98 (SD 0.5) in a second spawning period but with lower reproductive activity compared with the main spawning period (Fig. 4A). In contrast, females <100 cm FL had almost constant mean GSI values except in December (0.44 Zudaire et al.: Reproductive potential of / hunnus albacares in the western Indian Ocean 257 Table 3 Summary of oocyte developmental stages and corresponding reproductive phases by 5-cm size class (fork lengths) in our study of Yellowfin Tuna (Thunnus albacares) reproduction in the western Indian Ocean. The numbers of individuals in our study that were at each developmental stage are shown in the columns organized by histological classification of the ova- ries (Brown-Peterson et al. 2011; Zudaire et al., 2013); immature phase, including the primary growth stage (PG); develop- ing phase, including the cortical alveolar (CA), primary vitellogenesis (Vtgl), and secondary vitellogenesis (Vtg2) stages; spawning-capable phase, including the tertiary vitellogenesis (Vtg3), germinal vesicle migration (GVM), and hydration stages (Hydr.); and regenerating phase. Length (cm) Immature phas PG Mature phases and developmental stages Total P Developing to spawning-capable Regenerating CA Vtgl Vtg 2 Vtg 3 GVM Hydr. 48-53 15 2 17 53-58 45 6 51 58-63 51 3 54 63-68 12 5 17 68-73 15 8 1 1 25 73-78 19 10 5 1 2 1 38 78-83 26 51 8 7 4 1 8 105 83-88 25 42 17 5 6 2 10 107 88-93 6 32 16 3 1 58 93-98 4 28 16 12 4 2 4 70 98-103 7 14 7 3 5 1 1 38 103-108 17 12 6 4 2 10 51 108-113 4 14 1 1 6 26 113-118 1 3 1 1 6 12 118-123 3 13 2 1 19 123-128 1 12 5 1 19 128-133 1 3 20 11 1 36 133-138 1 2 21 16 1 41 138-143 4 1 3 11 5 1 25 143-148 1 1 2 3 1 8 148-153 1 1 2 Total 238 82 44 112 48 6 42 819 247 572 was positively related to GSI, length, and weight; BFrel was related only to GSI; HSI and K showed no signifi- cant relationship with BF or BFrel (Table 4). The GSI showed the best correlation with BF and BFrel (coef- ficient of correlation [r] =0.87 and r=0.89, respectively). In contrast, low values for r2 were found for length and weight (0.25). The intercept of the regression line be- tween the BF and fish weight relationship was signifi- cantly different from 0.00 (P<0.05), a result that could indicate that the number of oocytes produced per gram of female is not linearly related to weight, and, there- fore, larger females produce more oocytes per gram of gonad-free weight than do smaller females (Fig. 6). A decrease in mean BF values from November (3.79 mil- lion) to June (2.26 million) was observed. Neverthe- less, the ANOVA of the BF estimates by month did not reveal statistically significant relationships between months at a 95% confidence level (ANOVA; P(5,42)=l-52, P=0.2081). The BF appeared to be highly variable by month, making it difficult to identify a clear pattern for female fecundity. Discussion Length at 50% maturity Analyses of L50 for female Yellowfin Tuna in the Indian Ocean in other studies resulted in different estimates (Table 5). To our knowledge, our study is only the sec- ond one where the L50 of Yellowfin Tuna in the Indian Ocean has been examined by histological method. Ti- mochina and Romanov3 previously used a histological 3 Timochina O. I., and E. V. Romanov. 1991. Notes on re- productive biology of Yellowfin tuna in the western Indian Ocean. IPTP (Indo-Pacific Tuna Development and Manage- ment Program), Coll. Vol. Work. Doc. TWS/91/32, 60 p. P.O. Box 2004, Colombo, Sri Lanka. 258 Fishery Bulletin 111(3) Nov Dec Jan Feb Mar Apr May Jun Jul 0 -1 "-t— 1 — L- H — L-rJ — — ltj — L- H — L-rJ — ltj — A—1 Nov Dec Jan Feb Mar Apr May Jun Jul I I IP ■■ DP I I SCP RP Figure 3 Proportion of female Yellowfin Tuna ( Thunnus alba- cares ) in different ovarian developmental phases by month and size: (A) >100 cm in fork length and (B) <100 cm in fork length. The phases of ovarian devel- opmental shown in this graph are the immature phase (IP), developing phase (DP), spawning-capable phase (SCP), and regenerating phase (RP). method to estimate the length at which all specimens achieve maturity. In our recent work, L50 was estimated at 75 cm FL when the maturity threshold was defined as ovaries in the CA stage. This length is significantly smaller than the L50 reported for all previous studies in the Indian Ocean (Table 5). In all of those studies, the L50 was estimated on the basis of macroscopic methods that defined the maturity threshold by the presence of advanced vitellogenic oocytes in the ovaries. Besides the natural variability in the length at maturity (Itano2), the main reasons for the difference in L50 between our study and previous studies are the use of different meth- ods and oocyte stages to establish the maturity thresh- old for estimating L50. Standardization of the selected maturity index and accurate estimation are required to enable direct comparisons between estimates and to avoid biases. The cortical alveolar stage is the earliest sign of oocyte maturation (Brown-Peterson et al., 2011). Fe- males in this developmental stage normally continue through vitellogenesis and spawn in the upcoming sea- son (Wright, 2007). On the basis of the annual repro- ductive cycle and batch spawning behavior of Yellowfin Tuna, following the recommendation made by Lowerre- Barbieri et al. (2011), we suggest that Yellowfin Tuna females with ovaries in the cortical alveolar stage should be included in maturity estimates. Estimation of L50 through establishment of maturity in vitello- genic oocytes (Schaefer, 1998; Itano2; Zhu et ah, 2008) has the disadvantage that L50 will be overestimated because maturing individuals (i.e., females with corti- cal alveolar stage oocytes) are categorized as immature (Lowerre-Barbieri et ah, 2011). There is no informa- tion on how much individual growth occurs during the time lag between CA and vitellogenesis. Such informa- tion may improve estimation of L50 because growth be- tween those stages may partly explain the difference in length at first maturity obtained with different oocyte maturation thresholds. Reproductive cycle On the basis of the histological evaluation of ovaries and the assessment of the seasonal variation in the GSI, 2 main reproductive periods were identified in our study. The first period identified occurred from No- vember to February, and the second period occurred in June, with lower reproductive activity than the first period. Similar results were obtained by Stequert et ah (2001), who related the spawning activity of Yellowfin Tuna with the monsoons. They identified that spawn- ing activity was higher in the north monsoon (from November to March) than in the south monsoon (from June to August) — probably a result of the decrease in sea-surface temperature during the south monsoon pe- riod. Other authors identified only a single reproduc- tive period: from January to June (Zhu et ah, 2008), from January to March (Stequert and Marsac, 1989), and from November to April (Nootmorn et al.4). The seasonal peaks in spawning described in this study were 4 Nootmorn, P., A. Yakoh, and K. Kawises. 2005. Reproduc- tive biology of yellowfin tuna in the eastern Indian Ocean. Indian Ocean Tuna Commission (IOTC), Working Party Trop- ical Tuna ( WPTT), IOTC-WPTT-14, 378-385 p. IOTC Secre- tariat, P.O. Box 1011, Victoria, Seychelles. Zudaire et at: Reproductive potential of Thunnus albacares in the western Indian Ocean 259 r 2,4 - 2,2 -2.0* • 1,8 - 1,6 2,8 - 2,6 - 2,4 2,2 * 2,0 - 1,8 - 1,6 Figure 4 Mean monthly variation in the gonadosomatic index (GSI), hepatosomatic index (HSI), and condition factor ( K ) in our study of the reproductive potential of Yel- lowfin Tuna (Thunnus albacares) in the western Indian Ocean for females in two groups: (A) >100 cm in fork length (FL) and (B) <100 cm FL supported by GSI values over 1.5, the value correspond- ing to that fish capable of reproducing (Nootmorn et al.4), observed from November to January. The GSI val- ues for February and June below 1.5 could correspond to a period of lower spawning activity, in which the proportion of active females among the mature popula- tion decreased. The absence of samples from August to October affected the interpretation of the reproductive cycle (e.g., the commencement of the main spawning period). The analysis of the reproductive cycle by size groups showed asynchrony in the reproductive activity be- tween small and large Yellowfin Tuna. Larger and older individuals spawned ear- lier and longer in the season and had a higher activity than small- er individuals. This spawning behavior is in accordance with McPherson (1991) and Schaefer (1998), who described a positive relationship between the spawn- ing fraction and female size. Our results showed asynchrony in the appearance of ovaries in the re- generating phase between small and large fish. Smaller individu- als showed high percentages of ovaries in this phase earlier (December and January) than did large specimens (April and May). Ovaries of smaller individ- uals appeared in the regenerat- ing phase within a period (from November to February) in which females generally exhibit high reproductive activity. This behav- ior can occur more often among young individuals maturing for the first time than among older fish (Murua et al., 2003), and it may be related to the energy balance between somatic growth and reproduction, an energy bal- ance that is size related (Clara- munt et al., 2007). Another hypothesis that may explain the asynchrony in repro- ductive timing among sizes is that young females skip spawn- ing events. The early appearance of ovaries in the regenerating phase occurred during the main peak of reproductive activity of mature individuals, and it is unlikely to be related to the end of the spawning season of young fish (Murua and Saborido-Rey, 2003). Female Yellowfin Tuna that skip spawning can be clas- sified as younger skipping females (Secor, 2007). In these first-maturing females, maturation involves large physiological and behavioral transitions. They may not have the required energy resources and, therefore, fore- go reproduction. By skipping spawning, these females increase their growth rate and their chances of sur- vival, resulting in increased lifetimes and reproductive outputs (Rideout and Tomkiewicz, 2011). However, the extended spawning season of Yellowfin Tuna, as well as the variable spawning period among individuals, com- plicates the identification of skipped spawning seasons in the reproductive cycle of females (Lowerre-Barbieri et al., 2009). Therefore, further research is needed to 260 Fishery Bulletin 111(3) Table 4 Coefficients of determination (r2) and correlation (r) and P -values (P) for the relationship between batch fe- cundity (BF) and relative batch fecundity (BFrel) with different biological parameters fork length, weight, go- nadosomatic index (GSI), hepatosomatic ind ex (HSI), and condition factor ( K) in our study of Yellowfin Tuna (Thunnus albacares) reproduction in the western Indian Ocean. r 2 r P BF Length (cm) 0.2114 0.4598 <0.005 Weight (g) 0.2022 0.4497 <0.005 GSI 0.7579 0.8706 <0.001 HSI 0.0237 0.1541 0.6706 K 0.0611 -0.2472 0.1145 BFrel Length (cm) 0.0377 0.1943 0.2174 Weight (g) 0.0263 0.1622 0.3046 GSI 0.8099 0.8999 <0.001 HSI 0.0696 0.2639 0.4612 K 0.0184 -0.1357 0.3914 determine whether the appearance of regenerating phase ovaries among young individuals indicates that females have skipped the spawning season. Condition indices related to reproduction ability than in the first spawning period (from Novem- ber to February), and no distinct pattern was evident. For the period of higher spawning activity (November- February), condition indices showed a clear pattern of energy mobilization from the muscle to the liver or gonad for reproduction. In contrast, during the second spawning period (June), energy was mobilized at a low- er level because of lower spawning activity. The protracted spawning season and population asynchrony in spawning activity of Yellowfin Tuna could mask temporal variations in energy allocation and the mobilization of the factors that were analyzed (Dominguez-Petit et al., 2010). Therefore, the assess- ment of variation in energy reserves by maturation phases was performed to study the energy cycle in fe- males undergoing ovarian development (Alonso-Fernan- dez and Saborido-Rey, 2012). The results showed a high energy investment in reproduction. The GSI showed an increase in the ovary mass from the immature phase to hydration stage, and the HSI described an increase of liver mass mainly between the vitellogenic process and maturation. This HSI pattern is evidence of the importance of the liver in the energy accumulation and synthesis of those energetic compounds (e.g., lip- ids and vitellogenin) essential for ovarian development (Dominguez-Petit et al., 2010). The decrease of K dur- ing the vitellogenic process and the low values of K at the GVM and hydration stages could indicate the role of the muscle in the mobilization of energy to the gonad or liver to fulfill the energetic requirements of maturation, principally during vitellogenesis (Zabou- kas et al., 2006; Dominguez-Petit et al., 2010). Condition indices are im- portant parameters for tun- ing the estimation of repro- ductive potential (Marshall et al., 1999). Analysis with such indices throughout the spawning season allows the determination of energy al- location during reproduction (Murua and Motos, 2006). In our study, during the peak spawning period (November- February), GSI, HSI, and K reflected a seasonal pattern in the accumulation and de- pletion cycles of energy re- serves. The increase in the GSI and HSI at the expense of K was observed both in smaller (<100 cm FL) and larger (>100 cm FL) females, and the exchange of energy was more pronounced in the larger size group. These 3 condition indices from March to July showed higher vari- Table 5 Estimates of the length at 50% maturity (L50) and the fork length at which all specimens achieve maturity (L100) for female Yellowfin Tuna (Thunnus albacares) reported for previous studies in different areas of the Indian Ocean. The methods ap- plied for the classification of ovaries were macroscopic (Macro.) and microscopic (i.e., histological. Micro.). The criterion for the maturity threshold in all of these studies was the presence of vitellogenic stage oocytes. Studies Estimation type Method Length (cm) Stequert and Marsac, 1989 L50 Macro. 120-140 Hassani and Stequert (see A- 5 m the text) L50 Macro. 110-115 Nootmorn et al. (see A- 4 in the text > L50 Macro. 110 Karpinski and Hallier7 L50 Macro. 104-112 Zhu et al. 2008 L50 Macro. 114 Timochina and Romanov (see A- 3 ln the text > L100 Micro. 120 Maldeniya and Joseph5 L50 Macro. 100 'Karpinski, B., and J. P. Hallier. 1988. Preliminary results on yellowfin spawning in the western Indian Ocean. Indo-Pacific Tuna Development and Management Pro- gram (IPTP) Coll. Vol. Work. Doc.TWS/88/31, 50-59 p. 8 Maldeniya, R., and L. Joseph. 1986. On the distribution and biology of yellowfin tuna (T. albacares ) from the western and southern coastal waters of Sri Lanka. FAO/ IPTP Coll. Vol. Work. Doc. 2: TWS/86/18, 21-32 p. Zudaire et al.: Reproductive potential of Thunnus albacares in the western Indian Ocean 261 I c n r 2,00 1,95 - 1,90 1,85 1,80 1,75 Figure 5 Mean variation in gonadosomatic index (GSI), hepatosomatic index (HSI), and condition factor (K) for Yellowfin Tuna (Thunnus albacares), in our study of the reproductive potential of this species in the western Indian Ocean, by oocyte de- velopmental stage: primary growth (PG), cortical alveolar (CA), primary, second- ary, and tertiary vitellogenesis (Vtg); germinal vesicle migration (GVM); hydra- tion (Hydr.), and regenerating (Reg) stage. The energy allocation strategy varies among organ- isms as an adaptive measure to deal with the fluctua- tions in the availability of energy in marine environ- ments (Alonso-Fernandez and Saborido-Rey, 2012). Subtropical and tropical waters with relatively low fluctuations in environmental conditions (e.g., in food supply, temperature, and photoperiod) are conducive habitats for species to offset the cost of the reproduc- tive process by concurrent energy from feeding, with- out having to rely entirely on stored energy reserves (Alonso-Fernandez and Saborido-Rey, 2012). Yellowfin Tuna continue to feed while they reproduce, and it has been suggested that the high spawning activity of this species depends on prey availability during spawning (Itano2). On the basis of our results, we suggest that Yellowfin Tuna, like other tropical species (Arrington et ah, 2006), requires energy from feeding as well as from stored energy to carry out ovarian development. There- fore, Yellowfin Tuna could be described as a capital- income breeder (Alonso-Fernandez and Saborido-Rey, 2012), in which the energy stored before reproduction is not enough to offset the cost of reproduction, and en- ergy allocation from feeding is necessary for successful reproduction (Henderson and Morgan, 2002). Fecundity estimation Few studies have dealt with the fecundity of Yellowfin Tuna in the Indian Ocean. In our study, the estimated mean BF (3.1 million oocytes) is within the range reported by Hassani and Stequert,5 and in Table 6 by Timochina and Ro- manov2 for the western Indian Ocean and by Sun et al.6 for the western Pacific Ocean. Howev- er, our estimate is larger than the values reported by Schae- fer (1996, 1998) for the eastern Pacific Ocean and by Itano2 for the western Pacific Ocean. It is lower than the values reported by Itano2 for the Hawaii area. Furthermore, the mean BFrel value (74.4 oocytes per gram of gonad-free weight) deter- mined in our study for Yellow- fin Tuna is slightly higher than the values described by Schae- fer (1996, 1998) and Sun et al.6 (Table 6), and it is consid- erably higher than the values reported by Itano2. Besides the geographic differences among studies, intrapopulation vari- ability in fecundity (Schaefer, 1996) could be the main factor that caused the difference be- tween the results of our study and previously published results. The BF increased with female size and weight (P<0.05); however, the BF-length and BF-weight rela- tionships showed low r 2 (<0.25) with high variability — a finding that could result from the asynchrony of the population spawning (i.e., with some individuals at the beginning and others at the end of their individual spawning season). The intercept of the regression line between the BF and weight was statistically different from zero, indicating that larger females spawn more oocytes per gram than smaller females; this dynamic in turn would contribute to increasing the egg production for large females (Dominguez-Petit and Saborido-Rey, 2010). However, the BF of fish of the same size varied greatly, indicating that other factors, such as fish con- dition during spawning, could also drive female produc- tivity (Hassani and Stequert5; Murua and Motos, 2006). Experiments with captive Yellowfin Tuna showed evidence of a positive relationship between the daily food ratio and egg batch production (Margulies et al., 5 Hassani S., and B. Stequert. 1990. Sexual maturity, spawning and fecundity of the yellowfin tuna (Thunnus al- bacares) of the Western Indian Ocean. Coll. Vol. IPTP (Indo- Pacific Tuna Development and Management Program) Doc. Vol. 4, p. 91-107. IPTP PO. Box 2004, Colombo, Sri Lanka. 6 Sun, C., W. Wang, and S. Yeh. 2005. Reproductive biology of yellowfin tuna in the central and western Pacific Ocean. Western and Central Pacific Fisheries Commission (WCPFC). WCPFC-SC1, BI WP-1, 1-14 p. 262 Fishery Bulletin 111(3) Gonad-free weight (g) 0 10P00 20P00 30P00 40000 50P00 60P00 Figure 6 Linear regression between batch fecundity (BF) and fish gonad-free weight (Wg), and power function between BF and fork length (FL) for Yellowfin Tuna (Thun- nus albacares) sampled in the western Indian Ocean in 2009 and 2010 2007). Lauth and Olson (1996) suggested that tuna may boost batch fecundity in response to greater amounts of forage, in accordance with the results reported by Itano2 for the Pacific Yellowfin Tuna. Batch fecundity enhancement by feeding might explain the observed high interindividual variation in the BF. Apart from prey availability, other biotic and abiotic parameters, such as fish condition, genetic variation, geographic dif- ferences in sampling locations, and temperature, have been reported to be associated with variability in the produc- tivity of fish species (Schaeffer, 1998; Murua and Motos, 2006). Conclusions Advances in the management of fish stocks rely to a great extent on updated knowledge of repro- ductive dynamics and its imple- mentation in stock assessments. The results described here will contribute to the improvement of the understanding of the re- productive dynamics of Indian Ocean Yellowfin Tuna. The L50 was estimated at 75 cm FL with the maturity threshold set at the CA stage of ovarian develop- ment. Two reproductive periods were detected in which reproduc- tive activity was higher in large females than in small ones. Sim- ilarly, large females had higher contribution of fecundity than did small females; however, more research on fecundity is required to investigate the principal fac- tors that affect its variability at individual and population levels. We suggest that Yellowfin Tuna follow a capital-income breeder strategy during ovarian development, by mo- bilizing accumulated energy and incorporating energy from feeding. However, further research on the proxi- mate composition (e.g., lipids and proteins) of this spe- cies through analyses of different tissues is required to obtain more precise descriptions of energy allocation and mobilization during the reproductive cycle of Yel- lowfin Tuna. Table 6 Estimates of the batch fecundity (BF) and the relative batch fecundity (BFrel) of Yellowfin Tuna ( Thunnus albacares ) reported for previous studies in different areas of Pacific and In- dian Oceans. Values are expressed in millions for BF and in oocytes per gram of gonad-free weight for BFrel. Studies Area BF BFrel Hassani and Stequert {see fn ■ 5 m tke text) Western Indian Ocean 0.50-8.00 - Timochina and Romanov (see fn- 3 ln the text) Western Indian Ocean 3.27 - Schaefer 1996 Eastern Pacific Ocean 1.57 68.0 Schaefer 1998 Eastern Pacific Ocean 2.50 67.3 Itano(see fn' ^ the text) Hawaii area 3.45 63.5 Itano2 Equatorial western Pacific 2.16 54.7 Sun 6t al (see fn‘ 6 *n the text) Western Pacific Ocean 2.71 62.1 Zudaire et a!.: Reproductive potential of Thunnus albacares in the western Indian Ocean 263 Acknowledgments We thank X. Salaberria for his work in processing samples. We greatly appreciate the assistance of J. Murua, G. Ocio, and A. Sanchez for sample collections on board the purse seiner. This research benefited from the financial support of the Department of Agriculture, Fisheries and Food of the Basque Government, and the Pesqueria Vasco Montanesa S.A. (PEVASA) fishing company. This work was partly funded by a grant from the Fundacion Centros Tecnologicos Inaki Goenaga to Iker Zudaire for his doctoral work. We also are grateful for the contribution received from EMOTION project (ANR JSV7 007 01). This paper is contribution no. 628 from AZTI-Tecnalia (Marine Research Department). Literature cited Alonso-Fernandez, A., and F. Saborido-Rey. 2012. Relationship between energy allocation and re- productive strategy in Trisopterus luscus. J. Exp. Mar. Biol. Ecol. 416-417:8-16. Arrington, D. A., B. K. Davidson, K. O. Winemiller, and C. A. Layman. 2006. Influence of life history and seasonal hydrology on lipid storage in three neotropical fish species. J. Fish Biol. 68:1347-1361. Ashton, W. D. 1972. The logit transformation with special reference to its uses in bioassay, 87 p. Hafner Publ. Co. Inc., New York. Brown-Peterson, N. J., D. M. Wyanski, F. Saborido-Rey, B. J. Macewicz, and S. K Lowerre-Barbieri. 2011. A standardized terminology for describing re- productive development in fishes. Mar. Coast. Fish. 3:52-70. Claramunt, G., R. Serra, L. R. Castro, and L. Cubillos. 2007. Is the spawning frequency dependent on fe- male size? Empirical evidence in Sardinops sagax and Engraulis ringens off northern Chile. Fish. Res. 85:248-257. Collette, B. B., and C. E. Nauen. 1983. FAO species catalogue. Vol. 2: Scombrids of the world: an annotated and illustrated catalogue of tunas, mackerels, bonitos, and related species known to date, 137 p. FAO, Rome. Dominguez-Petit, R., and F. Saborido-Rey. 2010. New bioenergetic perspective of European hake ( Merluccius merluccius L.) reproductive ecology. Fish. Res. 104:83-88. Dominguez-Petit, R., F. Saborido-Rey, and I. Medina. 2010. Changes of proximate composition, energy storage and condition of European hake ( Merluccius merluc- cius, L. 1758) through the spawning season. Fish. Res. 104:73-82. Henderson, B.A., and G. E. Morgan. 2002. Maturation of walleye by age, size and surplus en- ergy. J. Fish Biol. 61:999-1011. Hunter, J. R., B. J. Macewicz, and C. A. Kimbrell. 1989. Fecundity and other aspects of the reproduction Sablefish, Anoplopoma fimbria, in central California water. CalCOFI Rep. Vol. 30, 61-72 p. Korta, M. 2010. New methodologies applied to quantify the dy- namics of the ovary in indeterminate fecundity species. Ph.D. diss., 208 p. Univ. Basque Country, Leioa, Spain. Lauth, R. R., and R. J. Olson. 1996. Distribution and abundance of larval Scombridae in relation to the physical environment in the north- western Panama Bight. Inter-Am. Trop. Tuna Comm. Bull. 21:127-167. Lowerre-Barbieri, S. K., N. Henderson, J. Llopiz, S. Walters, J. Bickford, and R. Muller. 2009. Defining a spawning population (spotted seatrout Cynoscion nebulosus ) over temporal, spatial, and demo- graphic scales. Mar. Ecol. Prog. Ser. 394:231-245. Lowerre-Barbieri, S. K., K. Ganias, F. Saborido-Rey, H. Murua, and J. R. Hunter. 2011. Reproductive timing in marine fishes: Variabil- ity, temporal scales, and methods. Mar. Coast. Fish. 3:71-91. Margulies, D., J. M. Suter, S. L. Hunt, R. J. Olson, V. P. Schol- ey, J. B. Wexler, and A. Nakazawa. 2007. Spawning and early development of captive yellow- fin tuna (Thunnus albacares). Fish. Bull. 105:249-265. Marquardt, D. W. 1963. Algorithm for least-squares estimation of nonlin- ear parameters. J. Soc. Ind. Appl. Math. 11:431-441. Marshall, C.T., N. A. Yaragina, Y. Lambert, and O. S. Kjesbu. 1999. Total lipid energy as a proxy for total egg produc- tion by fish stocks. Nature 402:288-290. McPherson, G. R. 1991. Reproductive biology of yellowfin tuna in the east- ern Australian fishing zone, with special reference to the North-western Coral Sea. Aust. J. Mar. Freshw. Res. 42:465-477. Morgan, M., H. Murua, G. Kraus, Y. Lambert, G. Marteinsdot- tir, C. Marshall, L. OBrien, and J. Tomkiewicz. 2009. The evaluation of reference points and stock productivity in the context of alternative indices of stock reproductive potential. Can. J. Fish. Aquat. Sci. 66:404-414. Murua, H., and L. Motos. 2006. Reproductive strategy and spawning activity of the European hake Merluccius merluccius (L.) in the Bay of Biscay. J. Fish Biol. 69:1288-1303. Murua, H., and F. Saborido-Rey. 2003. Female reproductive strategies of commercially important fish species in the North Atlantic. J. North- west Atl. Fish. Sci. 33:23-32. Murua, H., G. Kraus, F. Saborido-Rey, P. R. Witthames, A. Thorsen, and S. Junquera. 2003. Procedures to estimate fecundity of marine fish species in relation to their reproductive strategy. J. Northwest Atl. Fish. Sci. 33:33-54. Murua, H., L. Ibaibarriaga, P. Alvarez, M. Santos, M. Korta, M. Santurtun, and L. Motos. 2010. The daily egg production method: A valid tool for application to European hake in the Bay of Biscay? Fish. Res. 104:100-110. Rideout, R. M., and J. Tomkiewizs. 2011. Skipped spawning in fishes: More common than you might think. Marine and Coastal Fisheries. Mar. Coast. Fish. 3:176-189. Saborido-Rey, F., and S. Junquera. 1998. Histological assessment of variations in sexual 264 Fishery Bulletin 1 11 (3) maturity of cod ( Gadus morhua L.) at the Flemish cap (north-west Atlantic). J. Mar. Sci. 55:515-521. Schaefer, K. 1987. Reproductive biology of black skipjack, Euthynnus lineatus, an eastern Pacific tuna. Inter-Am. Trop. Tuna Comm. Bull. 19:166-260. 1996. Spawning time, frequency, and batch fecun- dity of yellowfin tuna, Thunnus albacares, near Clip- perton Atoll in the eastern Pacific Ocean. Fish. Bull. 94:98-112. 1998. Reproductive biology of yellowfin tuna (Thunnus albacares) in the eastern Pacific Ocean. Inter-Am. Trop. Tuna Comm. Bull. 21:205-221. 2001. Reproductive biology of tunas. In Tuna: Physiol- ogy, ecology, and evolution (B. A. Block and E. D. Ste- vens), p. 225-270. Academic Press, San Diego, CA. Secor, D. H. 2007. Do some some Atlantic bluefin tuna skip spawning? Col. Vol. Sci. Pap. ICC AT. 60(4): 1141-1153. Stequert, B., and F. Marsac. 1989. Tropical tuna-surface fisheries in the Indian Ocean. FAO Fish. Tech. Pap. 282, 238 p. FAO, Rome. Stequert, B., and B. Ramcharrun. 1996. La reproduction du listao ( Katsuwonus pelamis) dans le bassin ouest de l'ocean Indien. Aquat. Living Resour. 9:235-247. Stequert B., J. Nunez Rodriguez, B. Cuisset, and F. Le Menn. 2001. Gonadosomatic index and seasonal variations of plasma sex steroids in skipjack tuna ( Katsuwo- nus pelamis ) and yellowfin tuna (Thunnus albacares) from the western Indian ocean. Aquat. Living Resour. 14:313-318. Thorsen, A., P. Witthames, G. Marteinsdottir, R. Nash, and O. Kjesbu. 2010. Fecundity and growth of Atlantic cod (Gadus morhua L.) along a latitudinal gradient. Fish. Res. 104:45-55. Trippel, E. 1999. Estimation of stock reproductive potential: history and challenges for Canadian Atlantic gadoid stock as- sessments. J. Northwest Atl. Fish. Sci. 25:61-81. Wright, P. J. 2007. Understanding the maturation process for field in- vestigations of fisheries-induced evolution. Mar. Ecol. Prog. Ser. 335:279-283. Zaboukas, N., H. Miliou, P. Megalofonou, and M. Moratiou-Apostolopoulou. 2006. Biochemical composition of the Atlantic bonito Sarda sarda from the Aegean Sea (eastern Mediter- ranean Sea) in different stages of sexual maturity. J. Fish Biol. 69:347-362. Zhu, G., L. Xu, Y. Zhou, and L. Song. 2008. Reproductive biology of yellowfin tuna T. albac- ares in the west-central Indian Ocean. J. Ocean Univ. China. 7:327-332. Zudaire, I., H. Murua, M. Grande, M. Korta, H. Arrizabalaga, J. J. Areso, and A. Delgado-Molina. 2013. Yellowfin tuna (Thunnus albacares) fecundity reg- ulation strategy in the Western Indian Ocean. Fish. Res. 138:80-88. 265 Abstract— With the southern New England lobster fishery in distress, lobster fishermen have focused more effort toward harvesting channeled whelk (Busycotypus canaliculatus). However, minimal research has been conducted on the life history and growth rates of channeled whelk. Melongenid whelks generally grow slowly and mature late in life, a characteristic that can make them vulnerable to overfishing as fish- ing pressure increases. We sampled channeled whelk from Buzzards Bay, Massachusetts, in August 2010 and in July 2011, studied their gonad development by histology, and aged them by examining opercula. Males had a slower growth rate and a low- er maximum size than females. Male whelk reached 50% maturity (SM50) at 115.5 mm shell length (SL) and at the age of 6.9 years. Female whelk reached SM50 at 155.3 mm SL and at the age of 8.6 years. With a mini- mum size limit of 69.9 mm (2.75 in) in shell width, males entered the fishery at 7.5 years, a few months after SM50, but females entered the fishery at 6.3 years, approximately 2 years before SM50. Increased fishing pressure combined with slow growth rates and the inability to reproduce before being harvested can eas- ily constrain the long-term viability of the channeled whelk fishery in Massachusetts. Manuscript submitted 25 September 2012. Manuscript accepted 28 May 2013. doi 10.7755/FB.111.3.5 Fish. Bull. 111:265-278 (2013). The views and opinions expressed or implied in this article are those of the author (or authors) and do not necesarily reflect the position of the National Marine Fisheries Service, NOAA. Age, size, and sexual maturity of channeled whelk ( Busycotypus canaliculatus ) in Buzzards Bay, Massachusetts Bhae-Jin Peemoeller (contact author)1 Bradley G. Stevens2 Email address for contact author: bhaejin@gmail.com 1 Department of Natural Sciences University of Maryland Eastern Shore Carver Hall Princess Anne, Maryland 21853 Present address for contact author: 5013 Smith Farm Road Virginia Beach, Virginia 23455 2 Living Marine Resources Cooperative Science Center Department of Natural Sciences University of Maryland Eastern Shore Carver Hall Princess Anne, Maryland 21853 The channeled whelk (Busycotypus canaliculatus-. Melongenidae) sup- ports a small but growing fishery in Massachusetts. Most fishing is conducted by lobstermen during the off-season (spring and fall) or when the lobster fishery is slow; therefore, fishing of this species typically is done on a part-time basis. However, channeled whelk landings in Mas- sachusetts increased substantially after 2000, as the southern New England lobster stock declined, and reached 1400 metric tons in 2011 with a value of $6.2 million (Glenn and Wilcox1). In addition, exvessel prices have nearly doubled from 2007 to 2011, increasing the incentive to expand effort in this fishery (Glenn and Wilcox1). Fishing pressure may affect the average size of whelks be- cause many fishermen may focus on catching larger whelks (>160 mm shell length [SL]). Davis and Sisson (1988) reported declines in popula- tion density and mean shell width 1 Glenn, R., and S. Wilcox. 2012. Profile of the channeled whelk pot fishery, 9 p. Report to the Massachusetts Marine Ad- visory Commission. Massachusetts Di- vision of Marine Fisheries, Invertebrate Fisheries Program, 1213 Purchase St., New Bedford, MA 02740. (maximum distance across shell) for channeled whelk in Nantucket Sound between 1978 and 1981. Bruce (2006) reported a decrease in mean SL be- tween 1994 and 2004 for a related species, knobbed whelk ( Busycon carica), subject to a dredge fishery in Delaware Bay. Most whelk research has been conducted on knobbed whelk, and minimal research has been done on channeled whelk (Avise et ah, 2004; Bruce, 2006; Castagna and Kraeuter, 1994; Eversole et ah, 2008; Kraeu- ter et ah, 1989; Power et al.2; Walk- er et al., 2005; Walker et ah, 2007). Because of limited information on growth rates and size at maturity, managers do not know if the current minimum size limit of 69.9 mm (2.75 in) in shell width (SW) is appropri- ate to ensure the reproduction and longevity of channeled whelk in Mas- 2 Power, A. J., C. J. Sellers, and R. L. Walk- er. 2009. Growth and sexual maturity of the knobbed whelk, Busycon carica (Gmelin, 1791), from a commercially harvested population in coastal Georgia, 24 p. Occasional Papers of the Univer- sity of Georgia Marine Extension Ser- vice, vol. 4. Marine Extension Service, Univ. Georgia, Shellfish Research Labo- ratory, Savannah, GA. 266 Fishery Bulletin 111(3) sachusetts and to sustain the fishery for this species. The minimum size limit was established by the State of Massachusetts on the basis of the size of market ac- ceptability (Glenn and Wilcox1). With increased fishing pressure and limited biological information, channeled whelk can easily become overfished, especially if they are not able to reproduce before they enter the fishery. The channeled whelk ranges from Cape Cod, Mas- sachusetts, to Cape Canaveral, Florida (Edwards and Harasewych, 1988). Growth rate and size at maturity for channeled whelk are virtually unknown, but whelks of the family Melongenidae typically are slow growing, late maturing animals. In the seaside lagoons of Vir- ginia, knobbed whelk reach a mean size of 176.1 mm SL in 9-11 years (Kraeuter et ah, 1989). In South Caro- lina, knobbed whelk <90 mm SL grow faster than larger whelk (up to 7 times faster), although some knobbed whelk have minimal or negative growth (Eversole et ah, 2008). It has been suggested that channeled whelk have low fecundity because they lay egg strings only once a year (Edwards and Harasewych, 1988). Betzer and Pilson (1974) reported an annual change in gonad index (fresh weight of the gonad/fresh weight of whole soft tissues) of channeled whelk in Narragansett Bay, Rhode Island, with spawning most likely occurring in late summer and fall. No studies have been published that provide the spawning season of channeled whelk or the environmental factors, such as temperature and salinity, at which they spawn. In a study of this species in aquaria, channeled whelk began hatching from an egg string on 18-30 April 2010 at water temperatures of 15-18°C; the egg string was collected on 1 March 2010 near Cedar Island, Virginia (Harding, 2011). Channeled whelk may have a similar reproductive cycle to that of knobbed whelk. On intertidal flats in Virginia, knobbed whelk copulated in June and July and laid egg strings from mid-August to November; hatch- ing occurred from mid-March to early May (Castagna and Kraeuter, 1994). Knobbed whelk egg cases found in Cedar Island, Virginia, in 1977 yielded an average of 3770 whelk per string (Castagna and Kraeuter, 1994). However, information on the fecundity of channeled whelk is needed. The mode of reproduction for whelks also needs investigation because Castagna and Kraeu- ter (1994) suggested that knobbed whelk may be pro- tandrous hermaphrodites. Knobbed whelk raised in a laboratory were all males at 9 years, but, after 13 years, some males changed sex, and at the age of 14 years, pro- duced viable offspring (Castagna and Kraeuter, 1994). This outcome is contrary to the findings of Avise et al. (2004), who determined that knobbed whelk are geneti- cally dioecious and sex is determined at birth. Aging opercula gives insight on the growth and lon- gevity of channeled whelk. llano et al. (2004) reported that striae form annually on the operculum of Bucci- num isaotakii and can be used to estimate age. Heude- Bertherlin et al. (2011) counted the number of striae on the operculum of waved whelk ( Buccinum undatum ) to determine age. Kraeuter et al. (1989) aged knobbed whelk by embedding opercula in plastic resin and then sectioning them. For validation of aging, they used labo- ratory-reared knobbed whelk and embedded the opercu- la of 3 knobbed whelk of 6+ years and 3 knobbed whelk of 7+ years. Kraeuter et al. (1989) reported the average ages of these knobbed whelk at 6.0 and 7.2 years, re- spectively. Another aging technique involves bleaching knobbed whelk opercula and counting annuli (Bruce et al.3). In the opercula of older knobbed whelk, Power et al.2 found a “bubbling effect,” from growth ring overlap, due to decreased growth rates. There are no published reports on the histological staging of channeled or knobbed whelk gonads. We be- gan this study to provide useful biological information, such as size and age at sexual maturity, for managers of the channeled whelk fishery in Massachusetts. Data on the size at sexual maturity will provide managers with information needed to set minimum size limits that al- low females to spawn at least once and, therefore, to help prevent overfishing (Gordon, 1994). We sampled channeled whelk in Buzzards Bay, Massachusetts, and expected that they would be protandrous hermaphro- dites as reported for laboratory-reared knobbed whelk by Castagna and Kraeuter (1994). We hypothesized that SL would increase with age and that slower growth would occur at older ages. We provide the first informa- tion on the relationship between size, age, and stages of gonad development for channeled whelk and on the size and age at 50% maturity (SM50) of male and female channeled whelk. Materials and methods Sampling Sampling was conducted off Massachusetts in Buzzards Bay, a large, semi-enclosed estuary in the northeastern United States (Fig. 1). Buzzards Bay is uniformly shal- low, with depths mostly at 10-15 m, and is open to the sea at multiple locations. As a result, the water column in this bay is extremely well mixed, and differ- ences between surface and bottom temperatures rare- ly exceed 1°C, and salinities are almost uniformly 30 ppt (Turner et al., 2009). Commercial wooden or wire mesh conch traps, baited with the Atlantic horseshoe crab ( Limulus polyphemus), were used to collect chan- neled whelk from Buzzards Bay in August 2010 and in July 2011. Traps varied in size but were generally 50x50x30 cm, and they were set at depths of 10-15 m at 10-12 different sites in each year to maximize catch- es. Traps were allowed to soak for 1 week, and they were retrieved weekly over a 4-week period each year. Channeled whelk were identified according to Pollock 3 Bruce, D. G., R. Wong, and M. Greco. 2006. Delaware Bay whelk (conch) fishery assessment 2005, 36 p. Delaware Di- vision of Fish and Wildlife, 89 Kings Highway, Dover, DE 19901. Peemoeller and Stevens: Age, size, and sexual maturity of Busycotypus canaliculatus in Buzzards Bay, Massachusetts 267 (1998). Whelk from different sites were mixed; sites sampled in 2010 were mostly in eastern Buzzards Bay, and those sites sampled in 2011 were mostly in western Buzzards Bay. Because we mixed whelk from a large number of sites, we did not measure water temperature at individual sites; instead, seawater temperatures were exam- ined from the NOAA Data Buoy Station BZBM3, at the Northeast Fisheries Science Center dock in Woods Hole, Massachusetts (NOAA station 8447930, http://www.ndbc.noaa.gov/station_his- tory.php?station=bzbm3). We measured whole wet weight (Wt) of shell and tissue to the nearest 0.1 g, SL to the nearest 0.1 mm, and lip width (LW, maximum distance across the upper edge of the largest whorl) to the nearest 0.1 mm. Length of the male copula- tory organ (penis length), measured to the near- est 0.1 mm, was recorded for 93 males only in 2011. We also measured the SW for the first 166 whelk. Although SW is very similar to the mea- surement used by fishermen and managers for the minimum size limit, channeled whelk have an asymmetrical shell structure, and there is no standard procedure for measuring SW. We used LW instead of SW as our primary width mea- surement because it was more easily replicated and more precise than SW; the mean coefficient of variation for whelk in 5 -mm intervals of LW was 2.14 times greater for SW than for LW, with a range from 1.21 to 4.13 for SW. For this study, 292 channeled whelk (155 fe- males and 137 males) were sampled through the random selection of at least 20 whelk of various sizes in each 10-mm-SL interval. In 2010, we filled most of our SL intervals except for the in- tervals for very small (<100 mm SL) and very large (>190 mm SL) whelk. In 2011, we focused on the collection of whelk from these unfilled SL intervals to extend the sampling range for this study. After whelk were weighed and measured, we cracked their shells with a hammer to deter- mine sex. Males were defined as those whelk with a penis, and females were defined as those whelk with a nidamental gland. Scissors were used to separate the gonad from the digestive gland (gonads were not weighed). We removed the entire gonad from smaller whelk (<120 mm SL), and larger whelk had an ap- proximately 2-cm section removed from the terminal end of the gonad (closest to the tip of the shell spire) to the start of the gonad (closest to the operculum). Gonads were stored in 10% formalin for 1 week before they were transferred to 80% ethanol (EtOH) (Stoner et al., 2012). Age analysis Opercula from our dissected channeled whelk were re- moved with a scalpel, labeled, and stored in 10% for- Figure 1 Map of approximate capture sites (gray circles) of channeled whelk (Busycotypus canaliculatus ) collected in August 2010 and July 2011 in Buzzards Bay, Massachusetts. Specimens came from numerous sites within Buzzards Bay; however, capture site was not recorded for each whelk, malin for 1 week, and then transferred to 70% EtOH. Whelk opercula were taken out of EtOH, bleached for 30 min with 5% sodium hypochlorite, and then blotted dry with paper towels. Annuli were defined as growth striae that form on the interior margin and extend across the operculum to the exterior margin (Fig. 2; Bruce et al.3). The first annulus of each operculum gen- erally had a 6-mm linear distance that ranged from the interior to the exterior margin. To count annuli, opercula were placed under a light source and pho- tographed with an Olympus Stylus 4004 4-megapixel digital camera (Olympus Optical Co., Ltd., Center Val- ley, PA). Opercula did not dry and curl during this pro- 4 Mention of trade names or commercial companies is for iden- tification purposes only and does not imply endorsement by the National Marine Fisheries Service, NOAA. 268 Fishery Bulletin 111(3) Figure 2 Image of the operculum of a 9-year-old male (142.9 mm shell length) channeled whelk ( Busycotypus canaliculatus). Readers counted annuli to determine the age of whelk col- lected for our study in August 2010 and July 2011 in Buzzards Bay, Massachusetts. Dots are placed on each annulus for reference. The interior margin is at the top of this opercula, and the exterior margin is at the bottom. For the first annulus of each operculum examined in our study, there generally was a 6-mm linear distance from the interior to the exterior margin. cess, and shrinkage was not a concern because counting of annuli did not depend on accurate measurement of widths of annuli. Each operculum was read by 3 readers to estimate age (Kraeuter et al., 1989). Each reader recorded their confidence for each operculum, a technique similar to the one used by Herbst and Marsden (2011) to age oto- liths for Lake Whitefish ( Coregonus clupeaformis). Con- fidence levels for aging channeled whelk opercula were defined as 1) not confident — the initial year’s growth was at least partially missing and annuli were difficult to distinguish because of opaque or dark zones on the operculum; 2) somewhat confident — the initial year’s growth was partially missing but the shape was rea- sonably approximated; annuli were partially obscured by washed out or dark zones but were reasonably ap- proximated; 3) confident — operculum showed initial year’s growth and annuli were easily readable; or 4) very confident — operculum exhibited strongly promi- nent annuli that were easy to read. We analyzed operculum only from whelk with an av- erage confidence rating >2. If 2 of the 3 readers record- ed the same age for an operculum, that age was used. If all 3 readers recorded different ages, the mean value (rounded to the nearest year) was used if variance was <1 (equiva- lent to a confidence bound of ±2 years); whelk with greater variance were de- termined to be unaccept- able. Only 227 whelk (115 females and 112 males) of 292 total dissected whelk (155 females and 137 males) were used for oper- culum analysis (Table 1). Histology and gonad staging A histological examina- tion of gonads from 112 males and 115 females was undertaken to deter- mine gonad development stage. Not all dissected whelks (155 females and 137 males) were used for analysis because some opercula were not de- termined acceptable for operculum analysis and, therefore, could not be used for the comparison of age with gonad develop- ment. Two tissue samples were processed from each whelk gonad with a Tissue- Tek VIP-E150 (Sakura Fineteck USA, Inc., Torrence, CA) automated tissue processor that contained a de- hydration series of 70% EtOH, 95% EtOH, and various concentrations of 100% EtOH and clearing agent and melted paraffin. Tissue samples were taken from a one- third portion and a two-thirds portion (approximately 5 mm apart) of the gonad sample to address synchrony within the gonad. The processed tissues were embed- ded in paraffin and sectioned to 10 pm (male) or 14 pm (females). Each tissue sample produced 4 repli- cates, which were mounted on slides and dried on a slide warmer. The best 2 slides from each tissue sample (4 slides per gonad) were stained with a haematoxylin and eosin stain; coverslips were then mounted with a resin mounting medium. Four slides from each whelk gonad were examined to determine sex and gonad stage. The dominant stage was defined as the stage that represented at least 50% of the section (llano et al., 2003). If both tissue samples from a whelk’s gonad had different dominant stages, then the most advanced stage was recorded. The clas- sifications for each stage followed descriptions of Buc- Peemoeller and Stevens: Age, size, and sexual maturity of Busycotypus canaiicuiatus in Buzzards Bay, Massachusetts 269 cinum isaotakii (e.g., llano et al., 2003) and Rapana venosa (e.g., Mann et al., 2006). Male gonads exhibit- ed 4 stages: immature (I), early developing (ED), late developing (LD), and mature (M). Male maturity was classified on the basis of the presence of spermatozoa in the tubule and not on the basis of density. For classi- fication in a fifth stage, the recovering (R) stage, males would be expected to exhibit elongate tubules with few remaining spermatozoa. No males were considered in the R stage because all males with elongate tubules had >50% of tubules with spermatozoa and were sub- sequently reported as in the mature stage. Female go- nads exhibited 5 stages: I, ED, LD, M, and R. Female maturity was determined on the basis of the amount of vitellogenic oocytes (VOs) present in the cross section. Statistical analyses Linear regression relationships between SL, LW, and Wt were calculated and compared through the use of covariance analysis to determine if there was a rela- tionship between sex and shell size or shape (Stoner et al., 2012). A significant coefficient CP<0.05) meant that either slope or intercept differed between males and females. As a prerequisite for covariance analysis, an F-test was used to ensure that the variance between males and females was not significantly different. Data were logio transformed for the relationship between SL and Wt because Wt increased nonlinearly to SL. For the comparison of logio SL and sex to logio Wt, sex was a factor and logjo SL was a covariate. The von Bertalanffy growth model has been used to describe the growth of gastropods, such as the queen conch ( Strombus gigas) (e.g., Berg and Olsen, 1989) and Buccinum isaotakii (e.g., llano et al, 2004). Shell length and estimated age for each channeled whelk were used to fit a von Bertalanffy growth model, for each sex separately (Ricker, 1975): La - Linf (1 - e K(a V), where ^inf — K = a = to = the SL (in millimeters) at age (years); the theoretical maximum SL; the growth coefficient; the age; and the theoretical age at length 0. A von Bertalanffy growth model that compared LW (in millimeters) and estimated age also was conducted, with LW replacing SL where appropriate. The annual growth rate (in SL) for each sex was calculated by sub- tracting the size at each age from the size at age+1 from the von Bertalanffy growth model. Growth rate and age were square-root (sqrt) transformed and co- variance analysis was used to determine if sex affects growth. The sqrt transformation was used because a log10 transformation yielded nonlinear results. The re- gression of sqrt-transformed SL growth versus age and sex included sex as a factor and sqrt (age) as a covari- ate and was expressed (in terminology of R statisti- cal software [R Development Core Team, 2011]) in this manner: sqrt (SL growth) ~ sqrt (age) * sex, where ~ indicates a relationship (“modeled as”) and * indicates the combination of the factors sqrt (age) + sex and the interaction of sqrt (age) by sex. All further regression equations used this terminology and were computed with R statistical software (R Development Core Team, 2011). On the basis of results of reproductive histology, we classified as mature any whelk with gonads in stage M or R. These data were used to calculate a nonlin- ear logistic regression by using a general linear model (GLM) with a binomial link function, with SL, LW, or age as the size measurement, and the GLM regression coefficients were used to estimate the size at which channeled whelk of each sex reached SM50 with the following formula: SM50 = where Bq = the intercept; and B\ = the slope. Because the SM50 equation gives only a single value for each data set (male or female), a bootstrap routine was used to resample each data set with replacement 1000 times, and the results were used to calculate the bias and standard error (SE) of the original mean value of SM50. Bias was calculated through the subtraction of the original (full data set) value of SM50 from the resa- mpled mean value of SM50. SM50 is a widely accepted predictor of sexual maturity for various shellfishes, including the queen conch (e.g., Stoner et al., 2012), waved whelk (e.g., Heude-Bertherlin et al., 2011), and Zidona dufresnei (e.g., Gimenez and Penchaszadeh, 2003). In our study, the fitted values of the GLM, rep- resenting the proportion of mature whelk, were plotted against either SL or LW. For comparison, the propor- tion of mature whelk was calculated from the raw data within 10-mm increments of SL or 5-mm increments of LW and plotted. To compare the LW at which each sex reached SM50 with the current minimum size limit for harvest, we conducted a regression that compared the LW of 166 channeled whelk to their SW, the measure- ment similar to the one used to specify the legal size limit. All SM50 values are given as mean ±1 SE, and all statistics were computed with R (R Development Core Team, 2011). Results Staging gonads Seawater temperatures at Woods Hole in the sampling months of August 2010 and July 2011 averaged 22.2°C (SE 0.7) and 22.2°C (SE 0.6), respectively, and were not significantly different, but mean temperatures in August were 1.2°C greater in 2011 than in 2010. To our 270 Fishery Bulletin 111(3) knowledge, there is no other published infor- mation about temperatures during the spawn- ing season of channeled whelk, but our data do not allow inferences to be made about temper- ature effects on spawning because we sampled only during 1 month in each of 2 years and did not sample throughout the entire spawn- ing season. Histological examination showed that male channeled whelk classified as stage I of gonad development had small circular tubules that contained no spermatozoa and few spermato- gonia, spermatocytes, and spermatids (Fig. 3A). The cross section contained connective tis- sue and empty space. Males assigned to stage ED had <25% of tubules with spermatozoa (Fig. 3B). There were few spermatogonia, sper- matocytes, and spermatids present, and some empty space was still visible in the cross sec- tion. Males in stage LD had 25-50% of tubules with spermatozoa; more spermatogonia, sper- matocytes, and spermatids were present and tubules were more elongate than they were in males in earlier stages (Fig. 30. Males classi- fied as stage M had >50% of tubules with sper- matozoa; tubules were elongate and contained many spermatogonia, spermatocytes, and sper- matids (Fig. 3D). Female channeled whelk classified in de- velopment stage I had gonads with minimal previtellogenic oocytes (PVOs), no nuclei pres- ent, and empty space in the cross section; con- nective tissue was more prevalent than it was in later stages, as seen in males classified as stage I (Fig. 3E). Females in stage ED had more developed PVOs that contained round nuclei and visible nucleoli than did females in the earlier stage (Fig. 3F). Females identified as LD generally had <50% of the cross section with VOs (Fig. 3G). The VOs were larger and more elongate than the PVOs, although some PVOs still were present in the gonad. Females classified as M had >50% of the cross section with VOs, which were elongate and full of large yolk granules (Fig. 3H). Females in stage R had some elongate oocytes and some small oocytes that were empty or had minimal PVOs or VOs (Fig. 3D. Morphological relationships The ratio of females to males in our sample of dissected channeled whelk was near unity at the size range of 90-110 mm SL (Table 1) and ages of 4—6 years (Table 2). The smallest size intervals (70-90 mm SL) were male domi- nated, although not many individuals in that range were caught (Table 1). Male whelk had a lower maximum size and age than female Figure 3 Histological photographs of the stages of gonad development for (A-D) male and (E-I) female channeled whelk ( Busycotypus canaliculatus ) used in our study to classify whelk collected in Buzzards Bay, Massachusetts, in July 2010 and August 2011. Ob- served male gonad stages: (A) immature; (B) early developing; (C) late developing; and (D) mature, with arrow pointing to sper- matozoa. Observed female gonad stages: (E) immature; (F) early developing, with arrow pointing to a nucleus (also containing a nucleolus) from a previtellogenic oocyte; (G) late developing; (H) mature female with vitellogenic oocytes full of large yolk gran- ules; and (I) recovering. Scale bars are set at 100 pm for panels A-F and 200 pm for panels G-I. Peemoeller and Stevens: Age, size, and sexual maturity of Busycotypus cana/iculatus in Buzzards Bay, Massachusetts 271 Table 1 Number and proportion of each sex by interval of shell length (SL) for channeled whelk ( Busycotypus canaliculatus ) collected in August 2010 and July 2011 in Buzzards Bay, Massachusetts. Lower and Upper SL describe the range for each interval and is measured in millimeters. For the age and gonad analyses in our study, we used 112 males and 115 females. Proportion Proportion Lower SL Upper SL Males Females male female 70 80 3 0 1.00 0.00 80 90 7 1 0.88 0.13 90 100 9 6 0.60 0.40 100 110 7 8 0.47 0.53 110 120 13 5 0.72 0.28 120 130 16 2 0.89 0.11 130 140 20 1 0.95 0.05 140 150 20 2 0.91 0.09 150 160 16 3 0.84 0.16 160 170 1 11 0.08 0.92 170 180 0 19 0.00 1.00 180 190 0 19 0.00 1.00 190 200 0 16 0.00 1.00 200 210 0 18 0.00 1.00 210 220 0 4 0.00 1.00 whelk. Males dominated the size range of 110-160 mm SL and ages of 7-9 years, and females dominated the size intervals above 160 mm SL and ages of 10- 14 years (Tables 1 and 2). For the 137 dissected male channeled whelk, the maximum SL recorded was 175.0 mm and the maximum Wt was 490.6 g. Females typi- cally grew larger than males, reaching a maximum SL of 214.2 mm and a maximum Wt of 930.0 g (n=155). Penis length increased with SL in males (Table 3). The single exception was a 199.3-mm male that was removed from the study because histology showed that this whelk’s gonad was nonfunctional. Its gonad was empty of any gonadal precursors and contained only connective tissue. The sex of the whelk could not be determined on the basis of the gonad. It was recorded as a male only because a small penis (15 mm in length) was detected; a mature male at 165 mm SL normally would have a penis approximately 37 mm long. There was an exponential relationship between SL and Wt and a linear relationship between SL and LW. Covariance analysis of log10VW to logi0 SL and sex (Ta- ble 4) showed that the male intercept and male slope were not significantly different from the female inter- cept and female slope. The results of this covariance analysis of log1()Wt indicate that log10VF^ regression equations for males and females are similar (Table 3). Because sex was not a significant factor, a combined regression equation that included both sexes was calcu- lated and indicated that logioWt was significantly re- lated to logio SL (Tables 3 and 4). Covariance analysis of LW versus SL and sex showed that slopes differed between sexes and that male and female intercepts were similar (Tables 3 and 4). The LW covariance analysis indicates that males and females have a similar initial size, but male LW may differ from female LW as SL increases. Growth rate The von Bertalanffy growth models for the sexes were significantly differ- ent (Table 5; Fig. 4, A and B). The old- est female was 14 years old, and males reached a maximum age of 12 years. Males and females had similar SL until about the age of 4 years, after which the curve for males started to plateau (Fig. 4C). Females reached a larger maximum size and had higher annual growth rates than males (Fig. 4, C and D). The growth rates for males and females decreased as age increased (Fig. 4D). Covariance anal- ysis of sqrt (SL growth) versus sqrt (age) and sex indicated that the intercepts were not significantly different, and the male slope was significantly less than the female slope (Table 4). The sqrt (SL growth) covariance analysis indicates that males and females had a similar initial size, but male and female growth rates differed over time. Size at 50% maturity Females classified in stage I of gonad development did not exceed 130 mm SL (Fig. 5A) and 8 years of age (Fig. 5C). Females identified as ED did not exceed 160 Table 2 Number and proportion of each sex by age in years for channeled whelk (Busycotypus canaliculatus) collected in August 2010 and July 2011 in Buzzards Bay, Massa- chusetts. For the age and gonad analyses in our study, we used 112 males and 115 females. Age Male Female Proportion male Proportion female 4 1 1 0.50 0.50 5 5 5 0.50 0.50 6 12 10 0.55 0.45 7 27 5 0.84 0.16 8 26 5 0.84 0.16 9 27 8 0.77 0.23 10 9 21 0.30 0.70 11 3 34 0.08 0.92 12 2 15 0.12 0.88 13 0 5 0.00 1.00 14 0 6 0.00 1.00 272 Fishery Bulletin 111(3) Table 3 Regression equations, sample sizes, coefficients of multiple determination ( R 2), standard errors, and P-values for penis length, whole wet weight (Wt), lip width (LW), and shell width (SW) as functions of shell length (SL), LW, or SW from our study of channeled whelk ( Busycotypus cana- liculatus) collected in August 2010 and July 2011 in Buzzards Bay, Massachusetts. Standard error (SE) is ±1 SE. Regression Equation Sample size ( n ) R2 Standard error P-value Penis length -249.171 + 128.982 * logic SL 92 0.801 4.048 <2.2e-16 Female logioWt -4.437 + 3.139 * logio SL 115 0.978 0.049 <2.2e-16 Male logioWt -4.204 + 3.027 * log10 SL 112 0.956 0.053 <2.2e-16 LogioWt -4.385 + 3.114 * logio SL 227 0.979 0.051 <2.2e-16 Female LW 3.699 + 0.504 * SL 115 0.976 2.779 <2.2e-16 Male LW 6.453 + 0.473 * SL 112 0.958 2.177 <2.2e-16 SW -9.607 + 1.233 * LW 166 0.938 2.933 <2.2e-16 LW 11.954 + 0.761 * SW 166 0.938 2.304 <2.2e-16 mm SL (Fig. 5A) and 9 years of age (Fig. 5C). No LD or R females exceeded 190 mm SL (Fig. 5A), although there were females assigned as LD at the age of 12 years and females identified as stage R at the age of 14 years (Fig. 50. The largest males classified as stage I were 120 mm SL; males identified as ED did not exceed 130 mm SL, and LD males did not exceed 140 mm SL (Fig. 5B). Similar to females in stages I and ED, males in stages I and ED did not exceed the ages of 8 years and 9 years, respectively; however, the ages of LD males also were not older than 9 years (Fig. 5D). Females were assigned to stage M as small as 159 mm SL and at 8 years of age, and males were classified in stage M as small as 104 mm SL and at 6 years of age (Fig. 5). Both males and females included relatively few LD individuals, indicating that spawn- ing had not yet occurred and development was not yet complete. Table 4 Covariance analysis on regressions of channeled whelk ( Busycotypus canaliculatus) growth in weight (Wt), lip width (LW), and square-root (Sqrt) shell length (SL) from our study of this species collected in August 2010 and July 2011 in Buzzards Bay, Massachusetts. Sample size for each regression was 112 male and 115 female channeled whelk. The female intercept and slope are as stated. The actual male intercept is calculated by adding the male intercept esti- mate to the female intercept estimate. The actual male slope is calculated by adding the male slope estimate to the female slope estimate. Standard error (SE) is ±1 SE. Standard Regression Coefficient Estimate error P-value LogioWt ~ logio SL * sex Female intercept -4.437 0.103 <2e-16 Female slope 3.139 0.046 <2e-16 Male intercept 0.233 0.162 0.151 Male slope -0.112 0.075 0.138 LogioWt - logi0 SL Intercept -4.385 0.066 <2e-16 Slope 3.114 0.030 <2e-16 LW ~ SL * sex Female intercept 3.699 1.168 0.002 Female slope 0.504 0.007 <2e-16 Male intercept 2.754 1.814 0.130 Male slope -0.031 0.013 0.015 Sqrt (SL growth ) - sqrt (age) * sex Female intercept 7.919 0.052 <2e-16 Female slope -1.496 0.018 <2e-16 Male intercept -0.039 0.073 0.594 Male slope -0.209 0.026 4.63e-08 Peemoeller and Stevens: Age, size, and sexual maturity of Busycotypus canalicu/atus in Buzzards Bay, Massachusetts 273 The average proportion of mature channeled whelk in each interval (10 mm SL or 5 mm LW) were plotted along with the fitted logistic maturity curves from the GLM for each sex in Figures 6A and 6B, respectively. Estimated male SM50 was 115.5 mm SL (SE 2.2), and 61.3 mm LW (SE 0.9) (Table 6; Fig. 6, A and B). Estimated female SM50 was 155.3 mm SL (SE 3.0) and 80.6 mm LW (SE 1.9) (Table 6; Fig. 6, A and B). Es- timated male and female age at SM50 was 6.9 years (SE 0.2) and 8.6 years (SE 0.3), respectively (Table 6). Conver- sion of LW to SW through regression yielded an SM50 of 66.0 mm SW (SE 2.9) and 89.7 mm SW (SE 2.9) for males and females, respectively (Table 3). The SW regression was not sex specific. The current mini- mum size limit (69.9 mm SW) was equivalent to a size of 65.1 mm LW (SE 2.3) (Table 3). From the von Ber- talanffy growth model, age at minimum size limit was calculated as 7.5 years (7. 0-7. 9 years) and 6.3 years (6. 1-6.6 years) for males and females, respectively (Table 5). Discussion Table 5 Estimates for the parameters of shell length (SL) or lip width (LW) for male and female channeled whelk (Busycotypus canaliculatus) from the von Bertalanffy growth model used in our study of this species in Buz- zards Bay, Massachusetts, in August 2010 and July 2011. Each estimate is given as value ±1 standard error (SE). Llnf=theoretical maximum SL; /f=growth coefficient; ^theoretical age at length 0. Sample Model size (n) T-inf K to Female SL 115 247.15 (21.63) 0.15 (0.04) 1.78 (0.59) Female LW 115 126.40 (10.10) 0.16 (0.04) 1.73 (0.59) Male SL 112 177.80 (20.01) 0.20 (0.08) 1.63 (0.88) Male LW 112 89.11 (8.79) 0.22 (0.08) 1.40 (0.95) 2 years before they reach SM50. In contrast, males en- ter the fishery at an age of 7.5 years, a few months af- ter they reach SM50. Although the annual growth rate was higher for female channeled whelk than for male whelk, females took longer to reach SM50. Male waved This article describes the first published study on reproductive maturity and growth in channeled whelk, and it provides the first estimates of size at matu- rity for this species. Males reached SM50 2 years before females did. In addition, on the basis of SL, male SM50 was 40 mm less than fe- male SM50. Male channeled whelk reached SM50 at 66.0 mm SW (SE 2.9), which is below the minimum legal size limit in Massachu- setts. However, females reached SM50 at 89.7 mm SW (SE 2.9), 20 mm great- er than the minimum size limit. This result indicates that males have a greater chance of copulation before they are harvested than do females, many of which are captured before they reach sexual maturity. Females enter the fishery at an age of 6.3 years, approximately D Age (years) Figure 4 Results from the von Bertalanffy (VB) growth model used in our study of channeled whelk (Busycotypus canaliculatus ) collected in Buzzards Bay, Massachusetts, in Au- gust 2010 and July 2011; (A) VB growth model of females (n=115); (B) VB growth model of males (n = 112); (C) a comparison of male and female VB growth models; and (D) the calculated growth per year for males and females from the VB growth model. 274 Fishery Bulletin 111(3) Maturity 70 80 90 100 110 120 130 140 150 160 170 180 190 200 210 220 Female shell length (mm) 70 80 90 100 110 120 130 140 150 160 170 180 190 200 210 220 Male shell length (mm) 6 7 8 9 10 11 12 Female age (years) 13 14 7 8 9 10 11 Male age (years) Figure 5 Proportions of channeled whelk (Busycotypus canaliculatus ) classified at different stages of gonad development in our study of this species in Buzzards Bay, Massachusetts, in August 2010 and July 2011, shown by varying shell lengths (for (A] females and [B] males) and by varying ages (for [C] females and [D] males). The maturity stages used in our study and presented here are immature (I), early developing (ED), late developing (LD), mature (M), and recovering (R). For gonad analysis, we used 115 females and 112 males. whelk also reach 50% maturity before females do, at an age of 3 years versus 4 years (Heude-Bertherlin et al., 2011). Sustainability of the channeled whelk popu- lation is of concern for this fishery because effort and incentives are biased toward catch of larger whelk, which are almost entirely female. The highest growth rate for males and fe- males occurred in the first few years of life. Growth rate progressively decreased as chan- neled whelk aged (Fig. 4D). There was a wide range of SL at each age for this species (Fig. 4, A and B). This variation also was evident in knobbed whelk in Delaware Bay (Bruce et al.3) and Buccinum isaotakii (llano et al., 2004), both of which were aged by examining opercula. Harding (2011) reported an average size of 3.8 mm SL at hatching for channeled whelk cultured from hatch in the laboratory. At 171 days after hatching, the average SL was 48.4 mm; a linear growth model for age-at-lengfh resulted in a growth rate of 0.21 mm/day (Harding, 2011). In our study of wild channeled whelk, the von Bertalanffy growth model pre- dicted that an average SL of 48.4 mm would not be achieved until the age of 3 years, indi- cating a much slower growth rate for whelks in our study than the one found by Harding (2011). Although we did not capture channeled whelk younger than 4 years old, the von Berta- lanffy growth model predicted that 3-year-old males and females would be only 41.3 mm SL and 43.5 mm SL, respectively (Fig. 4C; Table 5). The discrepancy in growth rates between channeled whelk in our study and the whelk in the Harding (2011) study possibly reflects individual variation, although food availability, predator abundance, and habitat may affect shell growth as well. Channeled whelk held in laboratory tanks may be more protected from shell damage than channeled whelk in their natural environment. Channeled whelk have thin shells and can eas- ily chip their shells (or siphons) while they feed or move around on the ocean floor. In ad- dition, channeled whelk in a laboratory may be exposed to more food than they would be if they were in the ocean. Bourdeau (2010) re- ported that the frilled dogwinkle ( Nucella lam- ellosa), a marine snail, had thicker shells and reduced shell growth when in the presence of red rock crab ( Cancer productus). Food-limited snails did not significantly differ from snails exposed to crab, indicating food consumption, instead of a physiological response from pre- dation, ultimately affected growth (Bourdeau, 2010). These factors that affect SL growth may partially explain the quicker growth rates in Harding (2011) compared with the rates found in our study. Although knobbed whelk are in a different genus and have thicker shells than channeled whelk (Mag- alhaes, 1948), both whelks share similar growth pat- Peemoeller and Stevens: Age, size, and sexual maturity of Busycotypus canaliculatus in Buzzards Bay, Massachusetts 275 Table 6 Estimates of size at 50% maturity (SM50) by shell length (SL), lip width (LW), and age for male and female channeled whelk ( Busyco- typus canaliculatus ) collected for our study in Buzzards Bay, Massa- chusetts, in August 2010 and July 2011. We resampled each data set with replacement 1000 times with a bootstrap routine to calculate the resampled SM50 estimate and the bias and standard error (SE) of the original SM50 estimate. Bias is the difference of the original SM50 estimate from the resampled SM50 estimate. Standard error (SE) is ±1 SE around the original SM50 estimate. Sex Sample size (n) Subject SM50 estimate Bias SE Female 115 Age 8.6 years 0.0014 0.3 SL 155.3 mm 0.24 3.0 LW 80.6 mm 0.11 1.9 Male 112 Age 6.9 years -0.0027 0.2 SL 115.5 mm 0.12 2.2 LW 61.3 mm 0.04 0.9 terns. From the von Bertalanffy growth model, estimates of SL for 10-year-old male and female channeled whelk were 145.7 mm and 174.8 mm, respectively (Fig. 4C; Table 5). Kraeuter et al. (1989) reported knobbed whelk in the seaside lagoons of Virginia with SL of 176 mm after 10 years (average for ages 9-11). Bruce et al.3 re- ported 10-year-old knobbed whelk in Del- aware Bay with average SL of 112.8 mm and 127.6 mm for males and females, re- spectively. The Delaware Bay population of knobbed whelk appears stunted (possibly because of heavy fishing pressure from the whelk dredge fishery) when compared with the Virginia population. Error in aging opercula also could attribute to shorter SL at age in Delaware Bay. Channeled whelk and knobbed whelk share sexual dimor- phism in maximum size. The largest male knobbed whelk in Delaware Bay reached an SL of 161 mm; whereas the largest female reached an SL of 197 mm (Bruce et al.3). In Buzzards Bay, Massachusetts, the second-largest male had an SL of 175 mm, but the largest female reached an SL of 214.2 mm. The largest male found in Buz- zards Bay was deemed reproductively unviable and subsequently discarded from our analysis. Females had a slightly larger LW at larger SL when compared with males. This finding could be due to the presence of the nidamental gland, which takes up a large portion of shell volume in mature females. Fe- male waved whelk invest more energy in reproduction than do males of that species: 3.84 kj-m-2-yr_1 versus 0.26 kj-in~2-yr_1, respectively (Kideys et ah, 1993). The nidamental gland is essential for forming egg cases and may require channeled whelk females to spend more energy on reproduction than do males. Channeled whelk males only need to form a penis and testis, which may explain why males have a lower maximum size. Males and females had a similar relationship of Wt to SL, although females grew to larger sizes. We expected that the weights of males and females would differ be- cause females form ova- ries and a nidamental gland, which would alter the relationship between Wt and SL at larger SL. Castagna and Kraeu- ter (1994) reported the seasonal gonad index of nidamental gland per meat weight (N/MW) at a range of 9.5-18.3% for female knobbed whelk. N/MW was lowest in the spring and in October; there has been no field observations report- ed for knobbed whelk spawning in the spring, although observed egg laying peaked in October (Castagna and Kraeuter, 1994). Despite the in- A B Figure 6 Size at 50% maturity (SM50), on the basis of (A) shell length (SL) or (B) lip width (LW), for male (M) and female (F) channeled whelk (Busycotypus canaliculatus) collected in August 2010 and July 2011 in Buzzards Bay, Massachusetts. Lines indicate predictions from a generalized linear model (GLM); points represent the average proportion within each size interval of 10 mm SL or 5 mm LW. Vertical lines indicate SM50. For age and gonad analysis, we used 115 females and 112 males. 276 Fishery Bulletin 111(3) crease in LW and the subsequent increase in volume of female shells, the lack of any observed difference in the relationship of Wt and SL could have occurred be- cause the nidamental gland has a lower density than the density of muscle tissue. During dissections of channeled whelk in our study, there were no signs of hermaphroditism; all whelk had either a penis or a nidamental gland. No whelk were found to contain both male and female gonads in his- tological sections. Male penis length increased with SL, a finding that also indicates channeled whelk may not be protandric hermaphrodites. In knobbed whelk, the penis of males that changed sex to females shrank to a round protuberance (Castagna and Kraeuter, 1994). The results of Castagna and Kraeuter ( 1994) indicate knobbed whelk can be protandric hermaphrodites, al- though sex is determined at birth and there is a 1:1 sex ratio at hatching (Avise et ah, 2004). We found an equal sex ratio for male and female channeled whelk at early ages, but later, males domi- nated at ages of 7-9 years and females dominated at ages of 10-14 years (Table 2). Males did not live as long as females (Table 2), and there was a greater pro- portion of males at the size range of 120-160 mm SL (Table 1). Both factors could explain why there were more males than females at the ages of 7-9 years. It also is possible that the fishery has inflated the pro- portion of males at smaller sizes. Males have a lower maximum size and mature at smaller sizes than fe- males. We define large whelk as those individuals >160 mm SL because this size is slightly above the female SM,5o and is equivalent to the 97.5 percentile of male SL (on the basis of a sample of 9460 whelk; B. Stevens, unpubl. data). Because the fishery is directed at catch of large whelk, which are mostly females, fishermen do not actively target males and males may accumulate at smaller sizes. However, it is possible that knobbed whelk may ex- hibit protandric behavior in extreme conditions; Cast- agna and Kraeuter (1994) held only male knobbed whelk in the laboratory. Whelks generally grow slowly and move slowly, and, at low population levels, inter- action between sexes may be minimal. Protandry in knobbed whelk may be opportunistic, and not every individual may be capable of it. In addition, unknown pollutants potentially could cause the sex change seen in knobbed whelk by Castagna and Kraeuter (1994). Whether or not protandry occurs in channeled whelk, the fishery will remain in peril if mature females are not protected. There is a lack of evidence to support the hypothesis that protandry may occur in channeled whelk. Further, no channeled whelk with evidence of imposex were found in our study. In gastropods, tributyltin (TBT) causes higher testosterone, which in turn can cause the penis in male bruised nassa ( Nassarius vibex) to devel- op earlier and males to mature earlier (Demaintenon, 2001). Female bruised nassa with imposex developed a penis, and that development caused sterility in some cases because the penis blocked the oviduct (Demain- tenon, 2001). Females with imposex can be confused with sequential hermaphrodites in sex transition be- cause individuals may have both male and female sex organs. Castagna and Kraeuter (1994) indicated that knobbed whelk could be sequential hermaphrodites be- cause some male knobbed whelk switched sex to female in the laboratory, and some of the newly formed female knobbed whelk laid viable egg strings. However, it is possible that their knobbed whelk were originally fe- males with imposex, or that sex change was the result of social and physiological effects of confinement in a laboratory setting. Gonad histology of the “male” channeled whelk that was abnormally large (more than 20 mm longer than the second-largest male) revealed that its gonad was nonfunctional. The gonad was empty of any male or fe- male gonadal precursors and contained only connective tissue. The sizes of the penis and gonads were much smaller in this abnormally large male than in fully ma- ture male channeled whelk in the size range of 150- 160 mm SL. Because the testis and penis were very small, this whelk may never have been reproductively viable or was too old to reproduce. However, it is still possible that this large male was exhibiting protandric or even imposex conditions. The gonads of only 3 of 115 female channeled whelk contained different dominant stages in the replicate slides; males did not have different dominant stages (n=112). This finding indicates that gonadal develop- ment was mostly synchronous. Mann et al. (2006) re- ported asynchrony for 2 of 3 male specimens of Rapana venosa but not for female specimens collected in June from the Chesapeake Bay. The gonad developmental stages determined for channeled whelk in this study are point-in-time estimates and do not provide a sea- sonal evaluation. Gonad samples from our study were collected in July and August, most likely at the be- ginning of the spawning season for channeled whelk (Betzer and Pilson, 1974). The I and ED stages were differentiated to determine when spermatozoa (males) and nuclei (females) were first produced. Male maturity was determined mostly by the pres- ence and amount of spermatozoa. We believed that a male was capable of spawning effectively if it contained spermatozoa in at least 50% of its tubules, and we la- beled such whelk in our study as mature. A LD-stage male that contained spermatozoa in 25-50% of its tu- bules could potentially spawn, but it may not contrib- ute enough spermatozoa to be considered effective. There is no paternity information on egg strings of channeled whelk, but this species could follow a similar reproductive strategy to that of knobbed whelk. Walker et al. (2005) reported sex-linked markers in knobbed whelk that can be used to determine the biological par- ents of embryos. A knobbed whelk egg string contained 7 different fathers, indicating that the female knobbed whelk most likely used a “well-blended sperm pool” for fertilization (Walker et al., 2007). Female knobbed Peemoeller and Stevens: Age, size, and sexual maturity of Busycotypus canaliculatus in Buzzards Bay, Massachusetts 277 whelk may copulate with multiple males to increase fitness or because males do not excrete enough sperma- tozoa to fertilize an entire female brood. If male chan- neled whelk copulate with multiple females, then it is possible that males do not use all their spermatozoa on one female; this “sperm conservation” strategy of sperm competition is common among snow crab ( Cliionoece - tes opilio) when the ratio of females to males is high (Rondeau and Sainte-Marie, 2001; Sainte-Marie et ah, 2002). Female maturity was determined mainly by the presence of VOs. We assumed a gonad with VOs that covered more than half the cross section could be capa- ble of reproduction, although fecundity could be lower in a female with such a gonad than in a fully7 matured female. A LD-stage female contained VOs in less than half the cross section and most likely would not repro- duce until 1-2 years later. Staging of gonad development for male channeled whelk was more consistent than staging for females because there were a greater proportion of males at smaller SLs. Males accumulated at smaller size class- es (110-160 mm SL) because of a slower growth rate and a lower maximum size than the growth and size of females. In contrast, females grew rapidly past the shorter SL of males. Male and female channeled whelk classified as stage LD may not spawn until the follow- ing year because channeled whelk typically grow slow- ly and would have to overcome a large deficit in gonad- al development in just a few months to be considered mature. This large deficit in gonadal development is especially evident in females that probably invest more energy into reproduction than males. Females in stage R were not present at sizes >190 mm SL (Fig. 5A). It is not known if larger females (>190 mm SL) spawn every year, every other year, or multiple times per year. Bruce et al.3 suggested that female knobbed whelk with gravid ovaries (comparable to a mature female) may stay in this state year-round and conse- quently, not spawn annually. In our study, smaller fe- male channeled whelk (<190 mm SL) classified in the R stage could have spawned earlier in the year or in the previous year. Large females also may not com- pletely deplete their ovaries when they spawn, as we observed in some smaller females, and, therefore, may have a shortened generation time. Large females also may have been ready to spawn after we sampled them in July and August. This circumstance may have been the reason why few females were classified as stage R. Reduced fecundity at older age or larger size, al- though unlikely, also could be a reason that no females in stage R were found at sizes >190 mm SL. Conclusions On the basis of the morphological and histological evidence collected in our study, we suggest that most channeled whelk do not change sex, although chan- neled whelk under some conditions may exhibit pro- tandric-like symptoms (as did the unusually large male in our study). We did not find any channeled whelk with both a penis and nidamental gland or both ovary and testes. All female whelk contained a nidamental gland and ovary. All male whelk contained a penis and testes, except for the large male that contained a pe- nis and an inactive gonad. Penis length increased with SL in males, indicating that channeled whelk are not protandric. Females reached a larger maximum size and age and had a quicker growth rate than males. With the current minimum size limit, this fishery cap- tures small females before they reach SM50 and males just after SM50. To prevent the occurrence of overfish- ing, fishery managers need to consider the sex-specific growth rates, SM50, and fecundity of channeled whelk. More information on population estimates are needed to understand if the channeled whelk population in Massachusetts is being overfished. The results from this study provide information necessary for managers to work with lawmakers to enact appropriate legisla- tion on size limits to secure the longevity of the Mas- sachusetts whelk fishery and to allow potential mature females to spawn. Acknowledgments This project was supported by the Saltonstall-Kennedy (S-K) Grant Program (grant no. NA10NMF4270007) and the Living Marine Resources Cooperative Science Center for salary of B. Stevens. We thank R. Bemis, S. Lawrentz, and K. Amagada for their help capturing and measuring channeled whelk and C. Conroy and S. Lawrentz for their contributions in aging opercula. S. Lawrentz also assisted greatly with histology. Special thanks go to the University of Massachusetts Dart- mouth for allowing us to use their Sea Water Labora- tory and to fisherman J. Drake for his assistance and experience in capturing channeled whelk. Literature cited Avise, J. C., A. J. Power, and D. Walker. 2004. Genetic sex determination, gender identification and pseudohermaphroditism in the knobbed whelk. Busycon carica (Mollusca: Melongenidae). Proc. R. Soc. Lond., Ser. B: Biol. Sci. 271:641-646. Berg, C. J., Jr., and D. A. Olsen. 1989. Conservation and management of queen conch (Strombus gigas) fisheries in the Caribbean. In Marine invertebrate fisheries (J. F. Caddy, ed.), p. 421-442. John Wiley and Sons, New York. Betzer, S. B., and M. E. Q. Pilson. 1974. The seasonal cycle of copper concentration in Busycon canaliculatum L. Biol. Bull. Mar. Biol. Lab. Woods Hole 146:165-175. Bourdeau, P. E. 2010. An inducible morphological defence is a passive 278 Fishery Bulletin 111(3) by-product of behavior in a marine snail. Proc. R. Soc. Lond., Ser. B: Biol. Sci. 277:455-462. Bruce, D. G. 2006. The whelk dredge fishery of Delaware. J. Shell- fish Res. 25:1-13. Castagna, M., and J. N. Kraeuter. 1994. Age, growth rate, sexual dimorphism and fecun- dity of knobbed whelk Busycon carica (Gmelin, 1791) in a western mid-Atlantic lagoon system, Virginia. J. Shellfish Res. 13:581-585. Davis, J. P., and R. T. Sisson. 1988. Aspects of the biology relating to the fisher- ies management of New England populations of the whelks, Busycotypus canaliculatus and Busycon cari- ca. J. Shellfish Res. 9:453-460. Demaintenon, M. J. 2001. Ontogeny of the pseudohermaphroditic reproduc- tive system in Nassarius vibex (Gastropoda: Buccinidae: Nassariinae). J. Molluscan Stud. 67:51-57. Edwards, A. L., and M. G. Harasewych. 1988. Biology of recent species of the subfamily Busyco- ninae. J. Shellfish Res. 7:467-472. Eversole, A. G., W. D. Anderson, and J. J. Isely. 2008. Age and growth of the knobbed whelk Busycon carica (Gmelin 1791) in South Carolina subtidal waters. J. Shellfish Res. 27:423-426. Gimenez, J., and P. E. Penchaszadeh. 2003. Size at first sexual maturity in Zidona dufresnei (Caenogastropoda: Volutidae) of the south-western At- lantic Ocean (Mar del Plata, Argentina). J. Mar. Biol. Assoc. U.K. 83:293-296. Gordon, D. A. 1994. Lingcod fishery and fishery monitoring in south- east Alaska. Alaska Fish. Res. Bull. 1(2):140-152. Harding, J. M. 2011. Observations on the early life history and growth rates of juvenile channel whelks Busycotypus canalicu- latus (Linnaeus, 1758). J. Shellfish Res. 30(3):901-903. Herbst, S. J., and J. E. Marsden. 2011. Comparison of precision and bias of scale, fin ray, and otolith age estimates for lake whitefish (Coregonus clupeaformis ) in Lake Champlain. J. Great Lakes Res. 37:386-389. Heude-Berthelin, C., L. Hegron-Mace, V. Legrand, A. Jouaux, B. Adeline, M. Mathieu, and K. Kellner. 2011. Growth and reproduction of the common whelk Buccinum undatum in west Cotentin (Channel), France. Aquat. Living Resour. 24:317-327. llano, A. S., K. Fuginaga, and S. Nakao. 2003. Reproductive cycle and size at sexual maturity of the commercial whelk Buccinum isaotakii in Fun- ka Bay, Hokkaido, Japan. J. Mar. Biol. Assoc. U.K. 83:1287-1294. llano, A. S., A. Ito, K. Fujinaga, and S. Nakao. 2004. Age determination of Buccinum isaotakii (Gas- tropoda: Buccinidae) from the growth striae on opercu- lum and growth under laboratory conditions. Aquacul- ture 242:181-195. Kideys, A. E., R. D. M. Nash, and R. G. Hartnoll. 1993. Reproductive cycle and energetic cost of repro- duction of the neogastropod Buccinum undatum in the Irish Sea. J. Mar. Biol. Assoc. U.K. 73:391-403. Kraeuter, J. N., M. Castagna, and R. Bisker. 1989. Growth rate estimates for Busycon carica (Gmelin, 1791) in Virginia. J. Shellfish Res. 8:219-225. Magalhaes, H. 1948. An ecological study of the genus Busycon at Beau- fort, North Carolina. Ecol. Monogr. 18(31:377-409. Mann, R., J. M. Harding, and E. Westcott. 2006. Occurrence of imposex and seasonal patterns of gametogenesis in the invading veined rapa whelk Rapa- na venosa from Chesapeake Bay, USA. Mar. Ecol. Prog. Ser. 310:129-138. Pollock, L. W. 1998. A practical guide to the marine animals of north- eastern North America, 367 p. Rutgers LTniv. Press, New Brunswick, NJ. Ricker, W. 1975. Computation and interpretation of biological sta- tistics of fish populations, 382 p. Bull. Fish. Res. Board Can. No. 191. R Development Core Team. 2011. R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. [Available from http://www.R-project. org.l Rondeau, A., and B. Sainte-Marie. 2001. Variable mate-guarding time and sperm allocation by male snow crabs ( Chionoecetes opilio ) in response to sexual competition, and their impact on the mating suc- cess of females. Biol. Bull. Mar. Biol. Lab. Woods Hole 201:204-217. Sainte-Marie, B., J.-M. Sevigny, and M. Carpentier. 2002. Interannual variability of sperm reserves and fe- cundity of primiparous females of the snow crab ( Chi- onoecetes opilio) in relation to sex ratio. Can. J. Fish. Aquat. Sci. 59:1932-1940. Stoner, A. W., K. W. Mueller, N. J. Brown-Peterson, M. H. Da- vis, and C. J. Booker. 2012. Maturation and age in queen conch ( Strombus gi- gas ): urgent need for changes in harvest criteria. Fish. Res. 131-133:76-84. Turner, J. T., D. G. Borkman, J. A. Lincoln, D. A. Gauthier, and C. M. Petitpas. 2009. Plankton studies in Buzzards Bay, Massachusetts, USA. VI. Phytoplankton and water quality, 1987 to 1998. Mar. Ecol. Prog. Ser. 376:103-122. Walker, D., A. J. Power, and J. C. Avise. 2005. Sex-linked markers facilitate genetic parent- age analyses in knobbed whelk broods. J. Hered. 96:108-113. Walker, D., A. J. Power, M. Sweeney-Reeves, and J. C. Avise. 2007. Multiple paternity and female sperm usage along egg-case strings of the knobbed whelk, Busycon carica (Mollusca; Melongenidae). Mar. Biol. 151: 53-61. 279 Abstract-From 2003 to 2006, 44,882 Yellowtail Flounder (Limanda ferru- ginea) were captured and released with conventional disc tags in the western North Atlantic as part of a cooperative Yellowtail Flounder tagging study. From these releases, 3767 of the tags were recovered. The primary objectives of this tagging program were to evaluate the mor- tality and large-scale movement of Yellowtail Flounder among 3 stock areas in New England. To explore mortality, survival and recovery rate were estimated from traditional Brownie tag-recovery models fitted to the data with Program MARK. Models were examined with time- and sex-dependent parameters over several temporal scales. The models with a monthly scale for both sur- vival and recovery rate had the best overall fit and returned parameter estimates that were biologically rea- sonable. Estimates of survival from the tag-recovery models confirm the general magnitude of total mortality derived from age-based stock assess- ments but indicate that survival was greater for females than for males. In addition to calculating mortality estimates, we examined the pattern of release and recapture locations and revealed frequent movements within stock areas and less frequent movement among stock areas. The collaboration of fishermen and sci- entists for this study successfully re- sulted in independent confirmation of previously documented patterns of movement and mortality rates from conventional age-based analyses. Manuscript submitted 9 April 2012. Manuscript accepted 31 May 2013. doi 10.7755/FB. 11 1.3.6 Fish. Bull. 111:279-287 (2013). The views and opinions expressed or implied in this article are those of the author (or authors) and do not necesarily reflect the position of the National Marine Fisheries Service, NOAA. Mortality and movement of Yellowtail Flounder {Limanda ferrugined) tagged off New England Anthony D. Wood (contact author)' Steven X. Cadrin2 Email address for contact author: anthony.wood@noaa.gov 1 Northeast Fisheries Science Center National Marine Fisheries Service, NOAA 166 Water Street Woods Hole, Massachusetts 02543 2 School for Marine Science and Technology University of Massachusetts Dartmouth 200 Mill Road, Suite 325 Fairhaven, Massachusetts 02719 The Yellowtail Flounder ( Limanda ferruginea ) is one of the principal re- sources of the groundfish complex in the northeastern United States, with major fishing grounds on Georges Bank, off southern New England, and off Cape Cod. The fishery for Yellowtail Flounder is among the most productive and valuable in New England, yielding 1832.5 metric tons and $4.78 million to U.S. fishermen in 2011 (NMFS, 2012). Flowever, the potential yield of Yellowtail Floun- der is much greater than the current yield. The estimated maximum sus- tainable yield from the 3 stocks in New England is 29,483.5 metric tons (NEFSC1). The 3 stock areas exam- ined in this study were Cape Cod- Gulf of Maine, Georges Bank, and southern New England-Mid-Atlantic. Despite substantial investments in sampling Yellowtail Flounder fisheries and resources, uncertainty persists in the age-based stock as- sessments from the 2008 Ground- fish Assessment Review Meeting of 1 NEFSC (Northeast Fisheries Science Cen- ter). 2008. Assessment of 19 north- east groundfish stocks through 2007: re- port of the 3rd Groundfish Assessment Review Meeting (GARM III), Northeast Fisheries Science Center, Woods Hole, MA, 4-8 August 2008. NMFS, Northeast Fish. Sci. Cent. Ref. Doc. 08-15, 884 p. + xvii. [Available from National Marine Fisheries Service, 166 Water Street, Woods Hole, MA 02543-1026.] Northeast Fisheries Science Center (NEFSC1). The sources of this un- certainty are not well understood but may result from the movement of flounder among stock areas, lack of information on the effect of closed areas on population dynamics, insuf- ficient surveying of areas closed to fishing, inaccurate age determina- tions, misrepresentative sampling of distributional patterns, underreport- ed catch, or inaccurate assumptions about natural mortality (NEFSC1). In 2002, a cooperative tagging study was designed to provide independent information on mortality and move- ment, to complement the current stock assessment methods for Yellow- tail Flounder, and to improve the re- liability of scientific advice for effec- tive fishery management (NEFSC2). Movement of Yellowtail Flounder off New England has been addressed 2 NEFSC (Northeast Fisheries Science Center), Workshop Organizing Commit- tee (S. Tallack, ed., R Rago, chairperson, T. Brawn, workshop coordinator, and (al- phabetically) S. Cadrin, J. Hoey, and L. Taylor Singer). 2005. Proceedings of a workshop to review and evaluate the design and utility of fish mark-recap- ture projects in the northeastern United States; Nonantum Resort, Kennebunk- port, Maine, 19-21 October 2004. U.S. Dep. Commer., Northeast Fish. Sci. Cent. Ref. Doc. 05-02; 141 p. [Available from National Marine Fisheries Service, 166 Water St., Woods Hole, MA 02543- 1026.] 280 Fishery Bulletin 111(3) by several historical and more recent tagging studies. A study that examined fish tagged and released off the northeastern coast of the United States from 1942 to 1949 concluded that groups were relatively localized and exhibited short seasonal migrations. Most tagged fish were recovered within 80 km of the release site and little mixing was observed between fishing grounds, except for frequent movement from the Mid-Atlantic Bight to the waters of southern New England (Royce et al., 1959). Lux (1963) confirmed these observations and concluded that groups ofYellowtail Flounder move sea- sonally within fishing grounds, and a small amount of seasonal mixing occurs among groups. More recent tag- ging studies ofYellowtail Flounder in Canadian waters have confirmed the sedentary nature of this species. Fish tagged from 3 studies on the Grand Bank, Cana- da, traveled an average of 59 km (Walsh and Morgan, 2004). A recent summary of all previously published studies that detailed Yellowtail Flounder movements off the northeastern United States indicated 95% resi- dence in current stock areas (Cadrin, 2010). The cooperative Yellowtail Flounder tagging study described in this article was designed to address sourc- es of uncertainty in Yellowtail Flounder stock assess- ments. Two objectives of this study were to provide 1) estimates of mortality independent of mortality esti- mates from current Yellowtail Flounder stock assess- ments and 2) extensive release and recovery informa- tion and documentation of movement of fish between stock areas. Materials and methods The general approach to tagging in this study involved a sampling design with geographic coverage that repre- sented the entire Yellowtail Flounder resource off New England. All phases of the proposed research, from the field protocol to public outreach, were developed co- operatively with New England groundfish fishermen. We contracted commercial fishermen and their vessels to work with scientists to tag and release fish in all stock areas. The geographic design was developed on the basis of fishing areas, with releases in each area proportional to values of relative abundance of Yellow- tail Flounder from groundfish surveys conducted by the Northeast Fisheries Science Center. The field protocol and analytical design were considered during peer re- view to be valid approaches to meet the objectives of this tagging study (NEFSC2). Fish were captured with commercial otter trawls or gillnets with large mesh (16.5 cm) and relatively short trawl tows (30 min) or gillnet sets (<6 h). Yel- lowtail Flounder were identified by using Collette and Klein-MacPhee (2002). All legal-size fish (>33 cm in fork length) in viable condition, and some sublegal-size fish from tows in low-density areas, in the southern New England-Mid-Atlantic stock area were tagged with Petersen disc tags (22-mm in diameter). Viability was classified as excellent, good, or poor. Excellent fish exhibited body flexion and operculum or mouth move- ment, and no apparent damage from capture. Good fish also exhibited activity but had minor damage (e.g., scale loss, minor abrasion, and net marks). Poor fish, which exhibited no activity and had major abrasions or bleeding, were not tagged. Fish were released during the spawning season (May-August), with the exception of 1% of the releases, which occurred in the autumn of 2003. Tags were collected from fish recaptured from a year-round commercial fishery with some seasonal geographic closures. The reward system for reporting recaptures included a $1000 lottery for all returned tags and 280 high-value ($100) rewards. The outreach system involved reward posters, brochures, a website (http://www.cooperative-tagging.org), annual letters to Yellowtail Flounder fishermen, press releases, and a toll-free phone number. Patterns of tag-recovery rates were analyzed statistically with contingency tables (G- test; Sokal and Rohlf, 1995) of frequencies of releases and reported recaptures by sex, size, condition code, and damage code. To analyze the Yellowtail Flounder tagging data, Program MARK was used to fit multiperiod tagging models (White and Burnham, 1999). Program MARK facilitates the application of various types of mark- recapture models and estimates model parameters through the use of maximum likelihood. Brownie-type dead-recovery models (Brownie et al., 1985) were used to estimate the probability ofYellowtail Flounder sur- vival ( S ) and a recovery rate probability if). Recovery rate in this study was a compound parameter that rep- resented the probabilities that a tagged fish was cap- tured, the tag was retrieved, and the tag was reported. The model assumes that the probabilities of survival and recovery are the same for all marked animals and that tagging is instantaneous during sampling occasions. A preliminary analysis, in which alternative tag- ging models with time-steps over different temporal scales (weekly, monthly, biannual and annual) were examined, indicated that the data were best suited for models with a monthly time step. Recovery data were entered into Program MARK with a classic recovery matrix format (Brownie et ah, 1985; White and Burn- ham, 1999). The recovery matrix was examined with a suite of models that exhibited both time-dependent and constant survival, as well as time-dependent and constant recovery rate. Sex was included in the model as a group effect on survival and recovery rate. In addi- tion, commercial catch was used as a proxy for fishing effort and was explored as a covariate on recovery rate in multiple models. Matrices of expected values were developed for each model, and recoveries were modeled as multinomial random variables. Parameters were es- timated with maximum likelihood estimation. Akaike’s information criterion (AIC) was used to rank and select the model that achieved an optimal Wood and Cadrin: Mortality and movement of Limanda ferruginea tagged off New England 281 balance between the parsimony of the model and the goodness-of-fit, where parsimony decreases as the number of parameters in the model increases. Model fit was judged with the model likelihood (L): AIC = 2K - 2 ln(L), where K - the number of parameters. To determine whether the general model (fully pa- rameterized) was a reasonable fit to the tag-recovery data, goodness-of-fit was tested. A simulation proce- dure was used to calculate an estimate of overdisper- sion (c). Data were simulated at varying levels of over- dispersion (simulated c), and the deviance of each data set was divided by the degrees of freedom to obtain a range of values. A logistic regression was used to esti- mate the level of c where 50% of the simulated values were above and 50% were below the observed deviance divided by degrees of freedom of the general model. A model solution that perfectly conforms to the assumed error distribution would produce an expected vari- ance equal to the observed variance and would have a c value of 1.0. Deviations of c above or below 1.0 indicate over- or underdispersion, respectively. Gener- ally, a c value >3.0 indicates poor model fit because the model deviance is greater than the expected deviance (Lebrenton et al., 1992; Burnham and Andersen, 2002). To account for c and for differences in effective sam- ple size (N), a quasi-likelihood adjusted AIC (QAICc) was used to adjust fit of the top selected models (An- derson et al., 1994; Burnham and Anderson, 2002): QAICc=2K + c N-K-l The adjusted results from the top ranked models de- termined through the model selection criterion were then examined for biologically realistic parameter es- timates. Models that estimated multiple parameters at their boundaries (e.g., S=1.0) were rejected in favor of the next ranked model. Results Researchers worked with commercial fishermen to tag 44,882 Yellowtail Flounder with conventional disc tags, and 3767 of these tags were recovered from the com- mercial fishery. Of all the lottery tags and $100 high- reward tags, 8% and 14% were returned, respectively. The relative return rate of lottery tags to high-value tags indicates a 59% reporting rate, assuming that 100% of the high-reward tags were reported (Table 1). The results from the analysis of observed recovery rates by sex, size, condition code, and damage code in- dicate that females had a greater recapture rate than males (particularly small males). Fish categorized as good had the same recovery rates as fish that were excellent. All damage codes had similar recovery rates, except for the slightly lower recovery rates for those Table 1 Total releases and recaptures of tagged Yellowtail Flounder ( Limanda ferruginea) by tag type from 2003 to 2006 in a cooperative tagging study off New England. The ratio of recovery rate from lottery tags and high- value tags indicates a 59% reporting rate, assuming 100% reporting of high-value tags. Percentage Tag type Releases Recovered recovered Lottery tags 44,501 3713 8.3 $100 tags 381 54 14.2 Total 44,882 3767 8.4 fish with net marks (5% recovered) and those showing evidence of lymphocystis (3% recovered). Releases occurred in monthly batches over a 39-month period from June 2003 to August 2006, most- ly in summer (Fig. 1). The full recovery matrix (with known month of capture) included a total recapture rate of 7.9% (Table 2). There was a higher rate of re- capture for females (8.4%) than for males (6.5%). Mortality The release and recovery data were audited, and only tags with both location and fish sex information were included in the modeling. The final recovery matrix used to estimate survival included 43,907 releases and 3457 recaptures. Several model variations with both time-dependent and time-independent parameter es- timates and with sex-dependant parameter estimates were successfully fitted to the data. All of the top-ranked models determined through model selection had a time-dependent recovery rate parameter iff) with varying levels of sex dependence on both recovery rate and survival (Sg). Results from procedures for the simulation of goodness of fit indi- cate that the general model fitted the tag-recovery data well, returning a c estimate of 2.12 (Fig. 2). After the goodness-of-fit adjustment to the models, full weight was given to the model with a time-dependent survival and time-dependent sex-specific recovery rate. The 2 best models had a time-dependent survival estimate. However, many of the estimates were at the upper boundary of survival (S=1.0) because of sparseness in the data, and therefore the 2 models with time-depen- dent survival were not considered. The optimum mod- el was a model with constant survival and time- and sex-dependent recovery rate (Table 3). On the basis of goodness-of-fit and model validation diagnostics, the optimum model appears to be a reliable representation of the data. A constant rate of survival of 0.89/month with a standard error of 0.016 was estimated from the best model. The annual rate of survival was calculated at 282 Fishery Bulletin 111(3) 0.25, a value equivalent to a total annual mortality (Z) of 1.4. Abundance-weighted F-values for age 6+ Yel- lowtail Flounder from an age-based stock assessment were 0.49, 1.49, 1.01, and 0.85, for the years from 2003 to 2006, respectively. The tag-based estimate of Z is within the range of these stock-assessment F-values, assuming a natural mortality of 0.2 for these years (Fig. 3). Estimates indicate that recovery rates ranged from 0.001 to 0.078 for males and 0.001 to 0.037 for females. Noticeable peaks in the recovery rate for fe- males were observed in the period of May-August of each year (Fig. 4). These peaks were present but less evident in the trend for male recovery rate. Confidence intervals of recovery rate estimates indicate that re- covery rates were greater in summer for females, but not for males, presumably because the estimates for females were more precise. Movement Results indicate frequent movements within the Cape Cod-Gulf of Maine and Georges Bank stock areas with less frequent movement among stock areas. Recapture data with known recapture locations indicate 96% resi- dence in the Cape Cod-Gulf of Maine stock area (with 3% movement to the Georges Bank stock area and 1% Table 2 Annual data of release and recapture of tagged Yellowtail Flounder (Limanda ferruginea) used to estimate survival in this study of the mortality and movement of this species off New England. Year marked Number marked 2003 Number recaptured/year 2004 2005 2006 2007 Total recaptured Percentage recaptured 2003 10,122 526 403 72 18 3 1022 10.1 2004 17,783 628 492 96 18 1234 6.9 2005 6084 440 96 23 559 9.2 2006 9918 490 152 642 6.5 Totals 43,907 526 1049 1004 700 196 3457 7.9 Wood and Cadrin: Mortality and movement of Limanda ferruginea tagged off New England 283 Observed deviance/df=6.1 1 £ 1 o 1 ° § ° o 1 | Median c=2.12 1.0 1.5 2.0 2.5 3.0 Simulated c Figure 2 Results from a goodness-of-fit test used to estimate overdispersion (c) for the general model fitted to tag- recovery data (time- and sex-dependent parameter- ization) for Yellowtail Flounder (Limanda ferruginea) tagged off New England from June 2003 to August 2006. The estimates of simulated deviance c are shown for a range of simulated c values. The value of c, deter- mined through logistic regression, is the point (6.11) where 50% of the simulated values fall above and 50% fall below the observed deviance c for the general mod- el. df=degrees of freedom. Dev=deviance. movement to the southern New England-Mid-Atlan- tic stock area), 98% residence on the Georges Bank stock area (with 1% movement to the Cape Cod-Gulf of Maine stock area and <1% movement to the south- ern New England-Mid-Atlantic), and 26% residence in the southern New England-Mid-Atlantic stock area (with 63% movement to the Georges Bank stock area and 10% movement to the southern Cape Cod-Gulf of Maine stock area; Table 4). However, most movement from southern New England was observed for Yellow- tail Flounder released on Nantucket Shoals, near the boundary with adjacent stock areas (Fig. 5). Discussion The numbers of releases of tagged fish and tag recover- ies in this study were 10 times greater than the num- bers from any previous study of Yellowtail Flounder movement or mortality. The results from this study provide updated inferences of movement patterns and an independent estimate of mortality. Tag-recovery modeling and its application in fisheries research has increased in popularity over the past decade and has become an important tool to fisheries management (Pine et al., 2003). Large-scale tag-recapture studies provide insights into fish movement and population dynamics that are separate (except for their reliance on fishery recaptures) from the fishery-dependent and research survey data and methods used in conventional stock assessments. The results from tagging analyses should be of particular interest when there are sus- pected inconsistencies with the stock assessment data and analyses, as is the case for Yellowtail Flounder stocks off New England. The results from tag-recovery modeling in this study are consistent with the perception that the Yellowtail Flounder resource in New England is experiencing an intense rate of mortality. The total annual mortality estimate of 1.4 derived from patterns of tag recovery is similar to stock-specific, age-based mortality estimates from the Yellowtail Flounder assessments for 2003 to 2006 (NEFSC1; Fig. 3). The results from this study demonstrate that models typically used in quantita- tive ecology (e.g., the Brownie tag-recovery model) can complement conventional methods for fisheries stock assessment. The advantage of the Brownie model is that survival estimates are not conditional on an as- sumed natural mortality rate, and recovery rates are a composite of exploitation rate (i.e., natural and fishing Table 3 The top 5 most highly ranked models adjusted for a c=2.12 that were fitted to tag-recovery data in this study of the mortality and movement of Yellowtail Flounder (Limanda ferruginea) tagged off New England from 2003 to 2006. Models were ranked by quasi-likelihood adjusted AIC (QAICc). Survival (S) and recovery rate (f) were estimated by month (t), for the entire time series (.), and by sex (g). The optimal model (in bold type) was chosen on the basis of rank and parameter estimates that were biologically reasonable. Model QAICc Delta QAICc QAICc weight Model likelihood Number of Parameters Qdeviance Sit) fi g*t) 19765.56 Multiple boundary estimates for survival Sit) fit) 19813.70 Multiple boundary estimates for survival Si.) figH) 19825.54 0.00 0.73 1.00 111 1181.05 Sig) fi g*t) 19827.51 1.97 0.27 0.37 112 1181.01 S(g*t) fit) 19847.22 21.68 0.00 0.00 163 1098.07 284 Fishery Bulletin 111(3) 2.0 1.8 -\ 1.6 _ 1.4 N >- 1.2 1.0 - •c o E 76 0.8 - o H 0.6 j 0.4 0.2 j 0.0 Z from tag-recovery modeling 95% Cl for tag estimate o Z estimates from 2008 GARM 2003 2004 2005 2006 Year Figure 3 Estimates of total mortality ( Z ), with 95% confidence intervals (CIs), of Yellowtail Flounder (Limanda ferruginea) from 2003 to 2006 from tag-recovery modeling in this study and the stock as- sessments from the 2008 Groundfish Assessment Review Meeting (GARM) of the Northeast Fisheries Science Center. mortality), survival from the tagging process, tag re- tention, and tag reporting (Brownie et al., 1985). The mortality estimates derived from the tagging data corroborate stock assessment estimates; however, tag estimates could be inflated if a model assumption was violated. For the tagging models, all marked fish were assumed to have the same probability of sur- vival and recapture by the fishery. This assumption is often violated when newly released fish do not fully mix with the population and have a different recap- ture probability (Hoenig et ah, 1998). A close examina- tion of the full residual matrix for Yellowtail Flounder did not reveal any patterns consistent with nonmixing, which is typically represented with a consistent resid- ual along the recovery diagonal (Latour et al., 2001). Tagging-induced effects, both directly induced mortal- ity and short-term tag loss, also could affect survival estimates. Experiments with tanks and cages designed to test the tag retention and tag-induced mortality of Yellowtail Flounder have indicated that these effects were not a concern (Alade, 2008). Any possible influ- ence from these 2 effects likely would remain constant throughout the study and would not influence the sur- vival estimates significantly. Finally, the tag-recovery models used in this study did not estimate tag loss and tags were assumed not lost or missed. A fish that loses its tag is equivalent to a dead fish when it comes to the estimation of survival (Brownie et al., 1985). The Peterson disc tags used in this study are secure tags that pass through the body of a flounder and are anchored on both the dorsal and ventral surfaces. These tags have been widely applied in fish tagging studies and are known for a very high retention rate (Thorsteinsson, 2002). In addition, a long-term holding study with some fish held for more than 1 year showed 100% tag re- tention, and therefore tag loss was expected to be minimal (Alade, 2008). Although the tag-recovery estimate of survival is consistent with results from age- based stock assessments, several aspects of this tagging study should be considered in future assessments. The difference in recov- ery rate between the sexes indicates that population dynamics may differ between males and females. The lower recovery rate of males could be a result of greater natural mortality — a finding that is consistent with sexually dimorphic growth rates and maxi- mum sizes of Yellowtail Flounder (Lux and Nichy, 1969). However, sex-based recovery rates also could result from differences in catchability between sexes. The possibility of substantial movement between stock areas (e.g., from southern New England to Georges Bank) may also influence population dynamics of Yellowtail Flounder (Hart and Cadrin, 2004). Analyses of simulated release and recapture data, with a data structure consistent to the data used in this study, indicate that movement and mortality cannot be si- multaneously estimated because of highly correlated movement parameters (Alade, 2008). Therefore, simul- taneous estimation of both movement and mortality may require an integrated analysis of data from tag- ging surveys, fisheries surveys, and resource surveys (Maunder, 2001; Goethel et al., 2010). A summary of all published yellowtail flounder movements off the northeast United States, including those reported here, has revealed that juveniles and adults do not frequently move among fishing grounds (Cadrin, 2010). The movement from the southern New England-Mid-Atlantic stock area to the Georges Bank stock area observed in the study described in this ar- ticle is greater than the movement reported in previous studies. However, most releases in the southern New England-Mid-Atlantic stock area were near the bound- ary because that location was the only one sampled in the stock area that had high densities of Yellowtail Flounder. The pattern of tag recoveries (Fig. 5) and previous analyses of movement trends from surveys (Cadrin, 2010) indicate that the southwestern portion of Georges Bank and Nantucket Shoals is an area of stock mixing, and that relative stock size may influence stock composition in that area. Although the study described here was not designed to estimate seasonal movement, the results of this research are consistent with those of previous tagging studies of Yellowtail Flounder off New Wood and Cadrin: Mortality and movement of Limanda ferruginea tagged off New England 285 England that have indicated short, seasonal movement patterns (Royce et ah, 1959; Lux, 1963). Conclusions Current stock assessments of Yellowtail Flounder pro- vide valuable information for fishery management, al- though several major sources of uncertainty are pres- ent (NEFSC1)- The results of this study address 2 im- portant sources of uncertainty in the Yellowtail Floun- der assessments: estimates of mortality and large-scale movement patterns among stock areas. The results from the tag-recovery modeling in this study confirm that the Yellowtail Flounder population is experiencing a high level of mortality. These results were derived from data independent of the stock as- sessment data, and although we were able to confirm high levels of mortality, the direct cause remains un- known. In addition to modeling mortality, the pattern of tag-recovery locations from this study provides an updated look at stock mixing. Movement from the southern New England-Mid-Atlantic stock area to the Georges Bank stock area was greater than previously observed. However, some of this perceived movement was attributed to the study design and the locations 286 Fishery Bulletin 111(3) Table 4 Residence and movement of tagged Yellowtail Flounder ( Limanda ferruginea ) from 2003 to 2006 among managed stock areas off New England. Three stock areas were used in this study: Cape Cod- Gulf of Maine (CC-GOM), Georges Bank (GB), and southern New England-Mid-Atlantic (SNE-MA). Recapture area Recapture percentage Release area CC-GOM GB SNE-MA Total CC-GOM GB SNE-MA CC-GOM 1049 38 10 1097 0.96 0.03 0.01 GB 32 2307 11 2350 0.01 0.98 0.01 SNE-MA 14 86 36 136 0.10 0.63 0.26 Total 1095 2431 57 3583 where tagged fish were released. Tag-recovery patterns indicate that Yellowtail Flounder carried out frequent movements within stock areas and less frequent move- ment between areas. It is unlikely that uncertainties in the Yellowtail Flounder assessments are a result of substantial movement between stock areas. This study provides insight into important uncer- tainties associated with the population dynamics of Yellowtail Flounder stocks off New England. The re- sults from this study can be used to inform future as- sessments and provide additional information to aid in the management of this species. 42°N 40°N- 74°W 72°W 70°W 68°W I 66°W Release and recapture locations (color coded by stock area of release) O Cape Cod-Gulf of Maine O Georges Bank O Southern New England-Mid-Atlantic X Releases Atlantic Ocean 100 200 i Kilometers Figure 5 Map of the locations of release and recovery of tagged Yellowtail Flounder ( Limanda ferruginea) in 3 stock areas off New England from 2003 to 2006. Recaptures are color-coded on the basis of the stock area where they were tagged and re- leased. The 3 stock areas examined in this study were Cape Cod— Gulf of Maine (gold block), Georges Bank (green block), and southern New England-Mid-Atlantic (yellow block). Wood and Cadrin: Mortality and movement of Limanda ferrugmea tagged off New England 287 Acknowledgments Fishermen and researchers cooperated to develop the general approach and technical details of this tagging study. Many fishermen have contributed to this study, including 3 who were involved in all aspects of planning and decision-making: D. Goethel, R. Avila, and F. Mat- tera. Many scientists collaborated on this study, helping with tagging and other at-sea work, data modeling and analyses, and database development and maintenance. Specific thanks go to A. Westwood, J. Moser, L. Alade, D. Martins, G, DeCelles, D. Goethel, T. Miller, S. Kubis, and C. Sumi. This study was funded by the Northeast Consortium, the Northeast Cooperative Research Pro- gram of the Northeast Fisheries Science Center, the NMFS Stock Assessment Improvement Program, and the Massachusetts Marine Fisheries Institute. Literature cited Alade, L. 2008. A simulation-based approach for evaluating the performance of a yellowtail flounder (Limanda ferru- ginea) movement-mortality model. Ph.D. diss., 316 p. Univ. Maryland Eastern Shore, Princess Anne, MD. Anderson, D. R., K. P. Burnham, and G. C. White. 1994. AIC model selection in overdispersed capture-re- capture data. Ecology 75:1760-1793. Brownie, C., D. R. Anderson, K. P. Burnham, and D. S. Robson. 1985. Statistical inference from band recovery data: a handbook, 2nd ed., 305 p. U.S. Fish Wildl. Serv. Resour. Publ. 156. Burnham, K. P., and D. R. Anderson. 2002. Model selection and multimodel inference: a prac- tical information-theoretic approach, 2nd ed, 488 p. Springer- Verlag, New York. Cadrin, S. X. 2010. Interdisciplinary analysis of yellowtail floun- der stock structure off New England. Rev. Fish. Sci. 18:281-299. Collette, B. B., and G. Klein-MacPhee. 2002. Bigelow and Schroeder’s fishes of the Gulf of Maine, 3rd ed, 748 p. Smithsonian Inst. Press, Wash- ington, D.C. Goethel, D., T. J. Quinn II, and S. X. Cadrin. 2010. Incorporating spatial structure in stock assess- ment: movement modeling in marine fish population dynamics. Rev. Fish. Sci. 19: 119-136. Hart, D., and S. X. Cadrin. 2004. Yellowtail flounder ( Limanda ferruginea ) off the northeastern United States, implications of movement among stocks. In Species conservation and manage- ment: case studies (H. R. Akgakaya, M. A. Burgman, O. Kindvall, C. C. Wood, P. Sjogren-Gulve, J. S. Hatfield, and M. A. McCarthy, eds.), p. 230-244. Oxford Univ. Press, New York. Hoenig, J. M., N. J. Barrowman, K. H. Pollock, E. N. Brooks, W. S. Hearn, and T. Polacheck. 1998. Models for tagging data that allow for incomplete mixing of newly tagged animals. Can. J. Fish. Aquat. Sci. 55:1477-1483. Latour, R. J., J. M. Hoenig, J. E. Olney, and K. H. Pollock. 2001. Diagnostics for multiyear tagging models with application to Atlantic striped bass ( Morone saxatilis). Can. J. Fish. Aquat. Sci. 58:1716-1726. Lebrenton, J. D., K. P. Burnham, J. Clobert, and D. R. Andersen. 1992. Modeling survival and testing biological hypothe- ses using marked animals: a unified approach with case studies. Ecol. Monogr. 62:67-118. Lux, F. E. 1963. Identification of New England yellowtail flounder groups. Fish. Bull. 63:1-10. Lux, F. E., and F. E. Nichy. 1969. Growth of Yellowtail Flounder, Limanda ferru- ginea (Storer), on three New England fishing grounds. ICNAF Res. Bull. 6: 5-25. Maunder, M. N. 2001. Integrated tagging and catch-at-age analysis ( IT- CAAN). In Spatial processes and management of fish populations (G. H. Kruse, N. Bez, A. Booth, M.W. Dorn, S. Hills, R. N. Lipcius, D. Pelletier, C. Roy, S. J. Smith, and D.Witherell, eds.), p. 123-146. Alaska Sea Grant College Program Report AK-SG-0102, Univ. Alaska, Fairbanks, AK. NMFS (National Marine Fisheries Service). 2012. Fisheries of the United States 2011. Current Fish- ery Statistics No. 2011, 125 p. Fisheries Statistics Div., Office of Science and Technology, NMFS, Silver Spring, MD. Pine, W. E., K .H. Pollock, J. E. Hightower, T. J. Kwak, and J. A. Rice. 2003. A review of tagging methods for estimating fish population size and components of mortality. Fisheries 28C10 ): 10—23. Royce, W. F., R. J. Buller, and E. D. Premetz. 1959. Decline of the yellowtail flounder (Limanda ferru- ginea) off New England. Fish. Bull. 59:169-267. Sokal, R. R., and F. J. Rohlf. 1995. Biometry: the principles and practice of statistics in biological research, 3rd ed., 880 p. W.H. Freeman, New York. Thorsteinsson, V. 2002. Tagging methods for stock assessment and re- search in fisheries. Report of Concerted Action FAIR CT.96. 1394 (CATAG), 179 p. Mar. Res. Inst. Tech. Report 79. Marine Research Institute, Reykjavik, Iceland. Walsh, S. J., and M. J. Morgan. 2004. Observations of natural behavior of yellowtail flounder derived from data storage tags. ICES J. Mar. Sci. 61:1151-1156. White, G. C., and Burnham, K. P. 1999. Program MARK: survival estimation from pop- ulations of marked animals. Bird Study 46:120- 138. 288 Fishery Bulletin 111(3) National Marine Fisheries Service Best Paper Awards for 2012 The award for best publication of the year is given to authors who are employees of the National Marine Fisheries Service and whose article is judged to be the most noteworthy of those published in Fishery Bulletin and Marine Fisheries Review. Authors from the Na- tional Marine Fisheries Service are noted in bold font below. The winners for Fishery Bulletin Jones, Darin T., Christopher D. Wilson, Alex De Rober- ts, Christopher N. Rooper, Thomas C. Weber, and John L. Butler. Evaluation of rockfish abundance in untrawlable habitat; com- bining acoustic and complementary sampling tools. Fish. Bull. 110:332-343. The winners for Marine Fisheries Review Scott-Denton, Elizabeth, Pat F Cryer, Matt R. Duffy, Ju- dith P. Gocke, Mike R. Harrelson, Donna L. Kinsella, James M. Nance, Jeff R. Pulver, Rebecca C. Smith, and Jo Anne Williams Characterization of the U.S. Gulf of Mexico and South At- lantic penaeid and rock shrimp fisheries based on observer data. Mar. Fish. Rev. 74(4): 1 —2 7. Guidelines for authors 289 Fishery Bulletin Guidelines for authors Manuscript preparation Contributions published in Fishery Bulletin describe original research in marine fishery science, fishery en- gineering and economics, as well as the areas of ma- rine environmental and ecological sciences (including modeling). Preference will be given to manuscripts that examine processes and underlying patterns. Descriptive reports, surveys, and observational papers may occa- sionally be published but should appeal to an audience outside the locale in which the study was conducted. Although all contributions are subject to peer review, responsibility for the contents of papers rests upon the authors and not on the editor or publisher. Submission of an article implies that the article is original and is not being considered for publication elsewhere. Articles may range from relatively short contributions (10-15 typed, double-spaced pages [tables and figures not in- cluded]) to extensive contributions (20-30 typed pages). Manuscripts must be written in English; authors whose native language is not English are strongly advised to have their manuscripts checked by English-speaking colleagues before submission. Title page should include authors’ full names and mailing addresses and the senior author’s telephone, fax number, and e-mail address. Abstract should be limited to 250 words (one-half typed page), state the main scope of the research, and emphasize the authors conclusions and relevant findings. Do not review the methods of the study or list the contents of the paper. Because abstracts are circulated by abstracting agen- cies, it is important that they represent the research clearly and concisely. General text must be typed in 12-point Times New Roman font throughout. A brief introduction should convey the broad significance of the paper; the remain- der of the paper should be divided into the following sections: Materials and methods, Results, Discus- sion, Conclusions, and Acknowledgments. Headings within each section must be short, reflect a logical se- quence, and follow the rules of subdivision (i.e., there can be no subdivision without at least two subhead- ings). The entire text should be intelligible to interdisci- plinary readers; therefore, all acronyms, abbreviations, and technical terms should be written out in full the first time they are mentioned. For general style, follow the U.S. Government Print- ing Office Style Manual (2008) [available at http://www. gpoaccess.gov/stylemanual/index.html] and Scientific Style and Format: the CSE Manual for Authors, Edi- tors, and Publishers (2006, 7th ed.) published by the Council of Science Editors. For scientific nomenclature, use the current edition of the American Fisheries So- ciety’s Common and Scientific Names of Fishes from the United States, Canada, and Mexico and its compan- ion volumes ( Decapod Crustaceans, Mollusks, Cnidaria and Ctenophora, and World Fishes Important to North Americans). For species not found in the above men- tioned AFS publications and for more recent changes in nomenclature, use the Integrated Taxonomic Informa- tion System (ITIS) (available at http://itis.gov/), or, sec- ondarily, the California Academy of Sciences Catalog of Fishes (available at http://researcharchive.calacademy. org/research/ichthyology/catalog/fishcatmain.asp) for species names not included in ITIS. Citations must be given of taxonomic references used for the identification of specimens. For example, “Fishes were identified by using Collette and Klein-MacPhee (2002); sponges were identified by using Stone et al. (2011).” Dates should be written as follows: 11 November 2000. Measurements should be expressed in metric units, e.g., 58 metric tons (t); if other units of measure- ment are used, please make this fact explicit to the reader. Use numerals, not words, to express whole and decimal numbers in the general text, tables, and fig- ure captions (except at the beginning of a sentence). For example: We considered 3 hypotheses. We collected 7 samples in this location. Use American spelling. Re- frain from using the shorthand slash ( / ), an ambiguous symbol, in the general text. Word usage and grammar that may be useful are the following: Aging For our journal the word aging is used to mean both age determination and the aging process (se- nescence). The author should make clear which meaning is intended where ambiguity may arise. Fish and fishes For papers on taxonomy and biodiver- sity, the plural of fish is fishes, by convention. In all other instances, the plural is fish. Examples: The fishes of Puget Sound [biodiversity is indicated]; The number of fish caught that season [no emphasis on biodiversity]; The fish were caught in trawl nets [no emphasis on biodiversity]. The same logic applies to the use of the words crab and crabs, squid and squids, etc. Sex For the meaning of male and female, use the word sex, not gender. Participles As adjectives, participles must modify a specific noun or pronoun and make sense with that noun or pronoun. Incorrect: Using the recruitment model, estimates of age-1 recruitment were determined. [Estimates did not use the recruitment model.] Correct: Using the recruitment model, we deter- mined age-1 estimates of recruitment. [The participle now modifies the word we, those who were using the model.] 290 Fishery Bulletin 111(3) Incorrect: Based on the collected data, we concluded that the mortality rate for these fish had increased. [We were not based on the col- lected data.] Correct: We concluded on the basis of the collected data that the mortality rate for these fish had increased. [Eliminate the participle and replace it with an adverbial phrase.] Equations and mathematical symbols should be set from a standard mathematical program (MathType) or tool (Equation Editor in MS Word). LaTex is accept- able for more advanced computations. For mathemati- cal symbols in the general text (a, y2, re, ±, etc.), use the symbols provided by the MS Word program and itali- cize all variables. Do not use photo mode when creating these symbols in the general text. Number equations (if there are more than 1) for fu- ture reference by scientists; place the number within parentheses at the end of the first line of the equation. Literature cited section comprises published works and those accepted for publication in peer-reviewed journals (in press). Follow the name and year system for citation format in the “Literature cited” section (that is to say, citations should be listed alphabetically by the authors’ last names, and then by year if there is more than one citation with the same authorship. Abbrevia- tions of serials should conform to abbreviations given in Cambridge Scientific Abstracts (http://www.csa.com/ ids70/serials_source_list.php?db=aquclust-set-c). Authors are responsible for the accuracy and com- pleteness of all citations. Literature citation format: Author (last name, followed by first-name initials). Year. Title of article. Abbreviated title of the journal in which it was published. Always include number of pages. If there is a sequence of citations in the text, list chrono- logically: (Smith, 1932: Green. 1947; Smith and Jones, 1985). Digital object identifier (doi) code ensures that a publication has a permanent location online. Doi code should be included at the end of citations of published literature. Cite all software and special equipment or chemical solutions used in the study within parenthe- ses in the text (e.g., SAS, vers. 6.03, SAS Inst., Inc., Cary, NC). Footnotes are used for all documents that have not been formally peer reviewed and for observations and communications. These types of references should he cited sparingly in manuscripts submitted to the journal. All reference documents, administrative reports, inter- nal reports, progress reports, project reports, contract reports, personal observations, personal communica- tions, unpublished data, manuscripts in review, and council meeting notes are footnoted in 9 pt font and placed at the bottom of the page on which they are first cited. Footnote format is the same as that for formal literature citations. A link to the online source (e.g., [http://www/ , accessed July 2007.]), or the mail- ing address of the agency or department holding the document, should be provided so that readers may ob- tain a copy of the document. Tables are often overused in scientific papers; it is seldom necessary or even desirable to present all the data associated with a study. Tables should not be ex- cessive in size and must be cited in numerical order in the text. Headings should be short but ample enough to allow the table to be intelligible on its own. All un- usual symbols must be explained in the table legend. Other incidental comments may be footnoted with italic numeral footnote markers. Use asterisks only to indi- cate significance in statistical data. Do not type table legends on a separate page; place them above the table data. Do not submit tables in photo mode. Figures must be cited in numerical order in the text. Graphics should aid in the comprehension of the text, but they should be limited to presenting patterns rather than raw data. Figures should not exceed one figure for every four pages of text. Figures must be la- beled with the number of the figure. Place labels A, B, C, etc. within the upper left area of graphs and photos. Avoid placing labels vertically (except for the y axis). Figure legends should explain all symbols and abbre- viations seen in the figure and should be double-spaced on a separate page at the end of the manuscript. Color is allowed in figures to show morphological differences among species (for species identification), to show stain reactions, and to show gradations in temperature con- tours within maps. Color is discouraged in graphs, and for the few instances where color may be allowed, the use of color will be determined by the Managing Editor. • Notate probability with a capital, italic P. • Provide a zero before all decimal points for values less than one (e.g., 0.07). • Capitalize the first letter of the first word in all la- bels within figures. • Do not use overly large font sizes in maps and for units of measurements along axes in figures. • Do not use bold fonts or bold lines in figures. • Do not place outline rules around graphs. • Use a comma in numbers of five digits or more (e.g., 13,000 but 3000). • Place a North arrow and label degrees latitude and longitude (e.g., 170°E) in maps. • Use symbols, shadings, or patterns (not clip art) in maps and graphs. Failure to follow these guidelines and failure to correspond with editors in a timely manner will delay publication of a manuscript. Copyright law does not apply to Fishery Bulletin, which falls within the public domain. However, if an author reproduces any part of an article from Fishery Bulletin in his or her work, reference to source is con- sidered correct form (e.g., Source: Fish. Bull. 97:105). Guidelines for authors 291 Submission Submit manuscript online at http://mc.manuscriptcentral. coin/fisherybulletin. Commerce Department authors should submit papers under a completed NOAA Form 25-700. For further details on electronic submission, please contact the Associate Editor, Kathryn Dennis, at kathryn.dennis@noaa.gov When requested, the text and tables should be submit- ted in Word format. Figures should be sent as PDF files (preferred), Windows metafiles, TIFF files, or EPS files. Send a copy of figures in the original software if con- version to any of these formats yields a degraded ver- sion of the figure Questions? If you have questions regarding these guidelines, please contact the Managing Editor, Sharyn Matriotti, at sharyn.matriotti@noaa.gov Questions regarding manuscripts under review should be addressed to Kathryn Dennis, Associate Editor. Fishery Bulletin Subscription form Superintendent of Documents Publications Order Form *5178 □ YES, please send me the following publications: Subscriptions to Fishery Bulletin for $32.00 per year ($44.80 foreign) The total cost of my order is $ . Prices include regular domestic postage and handling and are subject to change. (Company or Personal Name) (Please type or print) (Additional address/attention line) (Street address) (City, State, ZIP Code) (Daytime phone including area code) (Purchase Order No.) Charge your order. IT’S EASY! Please Choose Method of Payment: I | Check Payable to the Superintendent of Documents ] GPO Deposit Account [ill1; ] VISA or MasterCard Account To fax your orders (202) 512-2104 (Credit card expiration date) (Authorizing Signature) Mail To: U.S. Government Printing Office P.O. Box 979050, St. Louis, MO 63197-9000 Thank you for your order! Also available online at http://bookstore.gpo.gov/actions/GetPublication.do7stocknumber =703-023-00000-2 SMITHSONIAN INSTITUTION LIBRARIES