Foraminiferal Densities and Pore Water Chemistry in the Indian River, Florida Bit cia it = “MARTIN A>BUZAS.. i ai 3 i. oe % “a ee ii aie x fe ae “= ne te ae sree ill tian. mae , a i se i oa a ; lll spustitianon “KENNETH P SEVERIN. NIAN LCONTHIBETION S giving an account of the new diecovenes| in: f the. z _ in all branches of knowledge.” This. theme of basic res arch has been adhered to thr. years by thousands of titles issued in series publications under the Smithsoni commencing with Smithsonian foo to mowed? | in | 1846 and continui following active series: “Smihscnian Contributions to Anthropology _ Smithsonian Contributions to Botany Smithsonian Contributions to the Earth Sciences _ Smithsonian Contributions to the Marine Sciences _ Smithsonian Contributions to Paleobiology ___ Smithsonian Contributions to Zoology ___ Smithsonian Folklife Studies Smithsonian Studies in Air and Space ‘Siatiiconian Studies in History and Technology in these < series, the Institution publishes small papers and full-scale monographs that ) the research and collections of its various museums and bureaux or of professional colleagu _ in the world of science and scholarship. The publications are distipiled by realiog lis libraries, universities, and similar institutions throughout the world. 2 Papers or monographs submitted for series publication are received by the Smi Institution Press, subject to its own review for format and style, only through dep _ Various Smithsonian museums or bureaux, where the manuscripts are given sub PlESs = feauiomenic for manuecret ae art preparation are outlines or the inside bac co ‘Smith : nian a Tosi tio SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES Foraminiferal Densities and Pore Water Chemistry in the Indian River, Florida Martin A. Buzas and Kenneth P. Severin SMITHSONIAN INSTITUTION PRESS Washington, D.C. 1993 ¢e NUMBER 36 ABSTRACT Buzas, Martin A., and Kenneth P. Severin. Foraminiferal Densities and Pore Water Chemistry in the Indian River, Florida. Smithsonian Contributions to the Marine Sciences, number 36, 38 pages, 32 figures, 29 tables, 6 appendices, 1993.—Two stations were established about 10 m apart at a depth of about | m at Link Port, Florida. One consisted of quartz sand and the other of quartz sand with a dense stand of seagrass. At the surface of each station and at a depth of 10 cm at the grass site, four replicate samples consisting of 5 ml each were taken every fortnight from 27 March to 6 November 1978 (17 sampling times, 204 samples). The taxa Quinqueloculina, Elphidium, Ammonia, Bolivina, and Ammobaculites comprising 98% of the fauna were enumerated. In addition, pore water chemistry was measured for temperature, salinity, oxygen, pH, Eh, NH,, PO,, Si, NO,, and NO, + NO,. General linear models were used to analyze the bare surface-grass surface, and grass surface-grass 10 cm data sets. Foraminiferal densities were evaluated for differences between sites, periodicity, sites x periodicity (interaction), and environmental variables. Differences in overall density between the bare surface-grass surface sites were not significant for the three most abundant taxa (Quinqueloculina, Elphidium, and Ammonia). At the grass site the density for all taxa were significantly lower at 10 cm than at the surface (very few individuals were observed at 10 cm). Hypotheses for periodicity and interaction were significant for all taxa in all comparisons except for Bolivina in the bare surface-grass surface analysis. At the bare surface, maximum densities occurred in spring while at the grass surface in summer. Although densities were low at 10 cm, no synchronization between the grass surface and 10 cm was evident. The environmental variables were significant for all taxa in both comparisons. The environmental variables are, however, highly correlated. To alleviate this difficulty, a principal component analysis was performed on these variables. The first three components included all of the 10 variables. Subsequent multiple regression of foraminiferal densities and the principal components indicated that usually at least two components, accounting for most of the variables, were Statistically significant. Thus, no simple relationship between pore water chemistry and density is apparent. The very large difference in density between the grass surface and 10 cm depth is much more strongly related to the pore water chemistry than the smaller differences with time at the surface sites. OFFICIAL PUBLICATION DATE is handstamped in a limited number of initial copies and is recorded in the Institution’s annual report, Smithsonian Year. SERIES COVER DESIGN: Seascape along the Atlantic Coast of eastern North America. Library of Congress Cataloging-in-Publication Data Buzas, Martin A. Foraminiferal densities and pore water chemistry in the Indian River, Florida / Martin A. Buzas and Kenneth P. Severin. p. cm.—(Smithsonian contributions to the marine sciences ; no. 36) Includes bibliographical references (p. _ ). 1. Foraminifera—Florida—Indian River. 2. Protozoan populations—Florida—Indian River. 3. Population den- sity—Florida—Indian River. 4. Pore water—Florida—Indian River. I. Severin, Kenneth P. II. Title. III. Series. QL368.F6B875 1993 593.1’204526323’097592-dce20 92-46076 The paper used in this publication meets the minimum requirements of the American National Standard for Permanence of Paper for Printed Library Materials Z39.48—1984. Contents Page Introductl ont: Saree eae cu Tim eM eg hie gu 1 ACKNOWICUSINEIS mere Renna erm ret tae tn ores eee ioe YS el ex sie aisle bw te 1 IMethodSieremertr ieee iitniccrm ec eh nt oeN auc ecin aco eg eS Al eee ap ads eens 1 Ja 5 oes dy co oO HO OOF GTO GENCE OR Se om eo a 1 MEAD OLALOnY A Reet ME Mee TMi aay sien s nce slave on clita. 6. ooh 8) Kulan ee locos 2 Statistical @ieecmre reruns ne In cae Rk CGO. oly ye pe ae @ secs Beng 2 BarersuntacerandsiGrassssurtace epee asiam aces meneincriel orisiiaiteom ele ss ees eee ee 3 ETVinOnMmentals Vatiableswee wae emi net ne Tuusnt eles Mn eats ee) os ae 3 Species Densities, Station Differences, Periodicity, and Environmental Wania blest wmsnt emer aii emcee os Ge oat ina Sore eG Stale ylayite niece 10 Ouingueloculingmrest rene Tol sac ees ee ieee ten ana re ss aiden a ae kehie Yon 10 LEY ONTO: 6,5 3. 0%, Bo Gd BLOC MOR RRR nO pea are eee ri INGO DTD. 26.5 SIS) OHO SOHO AOE ENC E O ND ATT RSI ID P an arete 11 BOliV INGE are Rea OR eA gee rs, eee eo oo ken ere eee, ys ile 11 ATTN ODAGCUIILCS tear ure Maer aL alcatel is oe Wena tte a Ue, re Wen teats y 12 GrasseSurfacerandsGrassplOlCMpusyuek sees se is cel eee ee i ee eee 13 EnviToOnmentalmvaltableSw-arwrusge Gb ry cue sl ale ten ye lie eas 13 Species Densities, Station Differences, Periodicity, and Environmental Wanita blestese anni Segre rane ara rituals, © iin sa ewe: tivsh suse s tS Ouingueloculinammra Rear: a rcerares ceri sieht) ae Gh ee 6 cee ete 2) ts 15 EE Didi ara aaa ie ele eee ons) east ret een Sake eel, oe eg lela ss 16 INGOT NGG, S55 SS a BR aa rar a ee cere ea 16 BOLING eR mre EM MAINE ae gids eyoci se etn oar ein eae eT sce ta Satna fend 21 PATIMODUCHINCS aa Ae ieee is coe eects foc: Siete. oe ote ee eee 22 CompanisontolsAnalyses soe ca iu ot ee tie ce ee sake eh oe 23 Companisongwitht@OthemStudies secu 6 ees ee cee es ee ese 24 AP DENGIXw ces ales SUI ACE Mime aS, ircuc-) ile. o eeeel te os etek eso) RG, vw, Be) Sakep sy 29 ZAPDENGIXG@? -BGTASSHOUTIACE pn meme tuped lcri fis ie ee ey elise ee be Gls) s Sue eto coe se 31 FAPDENGIXES Grassi OMCs seep cera cea leery ce teriol el Aue eee a nosy) Sc is see 33 PAP DENGIXEA MES ALCL SUIPACC Pw irc tie ey thes siya cies, We Cee eh we es a ee 35 PAP PENGIXE) a GLASSROULPACE ist take Lcciiionsc tat ee eos eee site elie ee See 36 XD PENGIXE Oss GIAaSSmlORGIMN naan sales es ear) Slee Secs een ael Si sas seeds si ot wi ss ueti'es ees 37 Miteraturem Citedmar spree eur iia oie vant ie ies cn et Ait) tts Moh, ote ee ee ae 38 ili Foraminiferal Densities and Pore Water Chemistry in the Indian River, Florida Martin A. Buzas and Kenneth P. Severin Introduction A basic variable for ecological studies is density, the number of individuals per volume or area. Densities of foraminiferal species, like those of all organisms, vary in space and time. Geographic changes in foraminiferal species densities have been documented between all marine environments from marshes to the abyss. Large differences in space like those between a marsh and the abyss or the Arctic and the tropics are easily recognized. A general qualitative correlation between observed species densities and the environment is easily accepted as an explanation for these changes. Similarly, differences over vast amounts of geologic time are easily recognized, and explained by the interplay of evolution and the environment. As we decrease the scale of our observations in space and time, however, and, at the same time, increase our effort to achieve quantitative results, differences in species densities and their explanation become much more difficult. Nevertheless, densities do vary over a matter of meters and within a time scale measured in weeks, months, or years. The present study is an analysis of quantitative measurements of species densities and environmental variables observed at two stations about 10 m apart which were sampled every fortnight for 9 months. During 1978 the chemistry group of the Harbor Branch Oceanographic Institution, Ft. Pierce, Florida monitored 10 pore water chemistry variables on a continual basis at two sites (stations) about 10 m apart at a depth of about 1 m in the Indian River (a shallow lagoon of nearly normal marine salinity on the central east coast of Florida). One station was located on bare quartz sand, the other on quartz sand with a covering of Martin A. Buzas, Department of Paleobiology, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560. Kenneth P. Severin, Department of Geology and Geophysics, University of Alaska-Fairbanks, Fairbanks, Alaska 99775-0760. seagrass (mostly Halodule wrightii and Thalassia testidium). We viewed their study as an ideal opportunity to conduct a study of the foraminifera with an experimental design allowing us to test statistically for differences in density between stations and with time as well as for the statistical significance of 10 environmental variables. Foraminiferal densities were also enumerated at a depth of 10 cm within the sediment at the grass station. Few living foraminifera were observed at 10 cm and most of the water chemistry variables exhibited a dramatic difference compared to the measurements made at the surface. To test the efficacy of the statistical procedures used in this study, and in others, the same statistical analyses were employed in evaluating the differences between the grass surface and at 10 cm as for the two surface stations. ACKNOWLEDGMENTS.—We thank the former Chemistry group at the Harbor Branch Oceanographic Insititution, especially J. Montgomery, M. Hucks, G. Peterson, M. Price, and C. Zimmermann. In the field and laboratory K. Carle, D. Mook, and H. Sheng were most helpful. J. Jett prepared the tables, figures, and appendices. L.S. Collins, L.C. Hayek, and B.K. Sen Gupta offered helpful suggestions on the manuscript. We thank J.C. Warren for his careful copy editing. This is contribution No. 319 from the Smithsonian Marine Station at Link Port. Methods FIELD.—Two stations were established about 100 m south of the Link Port jetty at a depth of about 1 m. The stations were marked with four poles encompassing an area of about 1 m* so that the same area could be re-occupied easily. The sediment at one station consisted of bare quartz sand with a silt-clay content of about 2%. The other was on the same substrate, but had a dense stand of Halodule wrightii with some Thalassia testidium. The sediment was sampled by pushing plastic coring tubes (inner diameter 3.5 cm) into the substrate. At each station four sediment samples were taken indiscriminately (not statistically randomized) at each sampling time which con- sisted of every fortnight from 27 March until 6 November 1978 (17 sampling times). The temperature, salinity, oxygen, pH, Eh, NH, PO,, Si, NO,, NO, + NO, were measured on pore waters by the chemistry group of the Harbor Branch Oceanographic Institu- tion throughout the field experiment (Montgomery et al., 1979); LABORATORY.—Immediately upon return to the laboratory (within a half-hour), 5 ml of sediment was removed from each core top, and at a depth of 10 cm. Each 5 ml sample was washed over a 63 tum sieve and preserved in 95% ETOH. Prior to examination for foraminifera the sample was stained overnight in rose bengal, dried, floated in a mixture of tetrabromine and acetone (specific gravity 2.3), and re-wet using “photo-flo” as a wetting agent. The foraminifera were enumerated while underwater, a procedure which facilitates the recognition of vividly stained protoplasm. The taxa counted were Quinqueloc- ulina (mostly Q. impressa and Q. seminulum), Elphidium (mostly E. mexicanum and E. gunteri), Ammonia (A. beccarii), Bolivina (B. striatula), and Ammobaculites (A. exiguus). These taxa are sufficiently dissimilar so that they can easily be identified under a binocular microscope and account for 98% of the foraminiferal fauna. The second replicate sample taken at the grass surface on 22 May 1978 was destroyed in a laboratory accident. The mean number of individuals from the other three replicates was used to estimate the missing data (Appendix 2). The systematics of the taxa used here are treated by Buzas and Severin (1982). Enumeration was made by Severin. STATISTICAL.—We have, then, two stations: one bare sand and the other with a stand of seagrass. For this study, the bare surface, the grass surface, and a depth of 10 cm within the sediment at the grass station were sampled for foraminifera. These three sites were sampled 17 times with four replicates of 5 ml each taken for examination. There are, then, N = 68 replicates or observations at each of the three sites. The number of individuals counted for each of five taxa and the total (the five taxa accounted for over 98% of the total) for each replicate is tabulated for the bare surface in Appendix 1, for the grass surface in Appendix 2, and for grass 10 cm in Appendix 3. We obtained the measurements made by the chemistry group for 10 water chemistry variables at each of the three sites at each sampling time. These measurements for the bare surface are tabulated in Appendix 4, for the grass surface in Appendix 5, and for the grass 10 cm in Appendix 6. The data were divided into two sets for statistical analyses, bare: surface-grass surface, and grass surface-grass 10 cm. For each analysis we wished to test the following hypotheses; difference between sites (bare surface vs. grass surface or grass surface vs. grass 10 cm), differences with time (periodicity), SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES different periodicities at each site, and the significance of the environmental variables. To accomplish this, we constructed a general linear model (GLM) similar to the one used by Buzas et al. (1977). In matrix notation the GLM is written as @ : x = LZ’ B + e (n x 1) (nxq) (qx 1) (n x 1) where x, the “dependent” variable is a vector of observed species densities for the n = 136 observations (n, =n, = 68), Z’ is a matrix of q “independent” variables, the composition of which will be discussed below, B is a vector of q parameters “regression coefficients” explaining the observations, and e is a vector of “errors” or “residuals,” assumed to have a normal distribution. The original counts x were transformed to In(x + 1) to make the data more Normal and stabilize the variance. Restricted Q models containing s parameters are constructed by equating the appropriate individual or groups of Bto 0. The sum of squares of the residuals, e’e, for each model is a scalar and is estimated by a least squares solution (Buzas et al., 1977, give the equations). Restricted models can be compared to the general model by the ratio (en — een )r Gs) TG oA ae (q-s\(n-q) een / Maiq) The mglh program of “SYSTAT” was used to calculate the sum of squares of the residuals for each model, and these residuals were then used to calculate the F-ratio given above. The results of the analysis are most easily displayed in the standard ANOVA table. The composition of the matrix Z’ of the @ model is given in Table 1. The vector z, is composed of 1’s so that B, is a constant, Z, contrasts the difference between sites by assigning +1 to one site and —1 to the other. The vectors z, and z, are made up of sin (m x 7/3) and cos (m x 77/3), respectively, where m = l,...,17. The vectors z; and z, are composed of sin (m x 7/6) and cos(m xX 1/6), respectively. These vectors are components of a periodic regression (Bliss, 1958) and account for a possible overall periodicity in the observations. Figures 1 and 2 illustrate these vectors over the period of our observa- tions. The possibility exists that the two sites may exhibit periodicity, but that the periodicity differs at the two sites. The interaction vectors Z, = Z, X Z3, Zz = Z, X Z4, Zy = Z, X Zs, and 29 = Z, X Z, account for this. We have, then, 8 vectors to examine the possible periodicity in our data. Had we constructed instrumental variables to examine the differences between the 17 sampling times and their interaction, we would have required 32 vectors, and made the model much more complicated. Finally, vectors z,, through Z,, contain the water chemistry variables completing the Z’ matrix for the GLM. NUMBER 36 SIN (m x 17/3) COS (mx m/ 3) —) TABLE 1.—Composition of the Z’ matrix. a vector of units +1 for bare surface, —1 for grass surface sin (m x 7/3), m=1,...,17 cos (m x 7/3), m=1,... ,17 sin (m x 7/6), m=1,...,17 cos (m x 1/6), m=1,...,17 Z, XZ, Z, XZ, Z, XZ; Z, XL, temperature salinity oxygen pH Eh 1978 FIGURE 1.—1/3 periodicity. cS zB ~ fw Zz Nn -1 2 Lb Ie, hb 4 % 4, Le, (a) t, gj Cy © & G € C>, g to, % ‘b Ny

10 5 9 9 4, 4 4 4 7%, Sy “i, |% | % | | |, Sue Fo, a © %,, py ® &y % Gp p “, 1978 im FIGURE 3.—Temperature measurements in °C. are presented in Table 2. No significant difference was observed between stations, but, once again, differences with time were highly significant. Very low to zero values were recorded in the spring and fall with maxima at both stations in the summer. PO, values are plotted in Figure 9 and ANOVA results are presented in Table 2. No statistical difference was observed between stations or with time. The measured values were generally very low to zero with the exception of the bare station in September which appears to be an outlier. The measured Si values are plotted in Figure 10 and the ANOVA results are presented in Table 2. No statistical difference was observed between stations, but differences with time were significant. Zero values were recorded at both stations in NUMBER 36 12 GRASS SURFACE GRASS SURFACE A A Lt =28.79 Ne = 4.47 35 $ = 335 10 Oo = 1.8 > z= z 5 i 5 —_— = Z a © BARE SURFACE 12 BARE SURFACE nw 40 i =29.68 ye ‘ fi 2 G = 4.48 Ane 6 =2.89 > = = iS 4 co) - > n ~ ~ — 1978 FIGURE 4.—Salinity measurements in °/oo. FIGURE 5.—Oxygen measurements in mg-at/I. pH pH 9.0 GRASS SURFACE ft =8.01 nw 0 =0.24 1978 FIGURE 6.—pH measurements. Eh Eh SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES GRASS SURFACE fi = -4.06 G = 78.29 200 150 100 50 0 -50 -100 -150 -200 -250 q Sy (9) te, Uy e, te eS + o) % Kz %, “te, ep 1978 ‘2 300 BARE SURFACE fi = 6.09 6 = 12.38 250 9 August; however, values generally increase from March to November. NO, values are plotted in Figure 11 and the ANOVA results are presented in Table 2. Significant differences were again noted only with time. In general, values are low with maxima in summer and fall at the bare station and spring, summer, and fall at the grass station. Both stations had minima values from late August until early October. NO, + NO, values are plotted in Figure 12, and ANOVA results are presented in Table 2. No significant differences were observed between stations or with time. An unexplainably high value was recorded in April at the grass station. Table 3 shows the correlation coefficients between the environmental variables. Temperature and NH, are positively correlated with one another and negatively with oxygen, pH, and Eh which are all positively correlated with one another. Consequently, any hypothesis concerning the significance of a particular variable on species density is not independent of the others. A way to avoid this difficulty is to transform the original variables to principal components. Principal component analy- sis is a technique which produces a succinct parsimonious summarization of many correlated (non-zero covariance) variables by transforming the original variables to independent (zero covariance) variables called principal components. An additional advantage of the technique is that the first principal component will account for most of the variability in the data, the second less, and so on (Seal, 1964). Eigenvalues were calculated from the correlation matrix (Table 3) using the SYSTAT statistical package. The first three eigenvectors account for 63.43% of the variability. The factor score coefficients (standardized vectors which when multiplied by the original standardized variables produce the principal components) indicate that all the environmental variables contribute substan- tially to the first three principal components (Table 4). The coefficients (Table 4) indicate that the first principal compo- nent (PC1) accounting for 31% of the variability contrasts TABLE 3.—Correlation matrix for chemical variables for bare surface and grass surface. 0.05 level is underlined. 200 ioe) S ~ 150 + oS > 100 50 » 1s) 4, te 4 %, G 1%, %, © oy % Tp No tp ey “e, on %, 1978 FIGURE 12.—NO, + NO, measurements in j1g-at/I. Temperature Salinity Oxygen Temperature 1.00 Salinity —0.14 1.00 Oxygen —0.68 0.07 1.00 pH —0.36 0.03 0.69 Eh —Or92 —0.08 0.69 NH, 0.40 0.29 0.48 PO, 0.14 0.24 0.03 Si —0.06 0.39 —0.05 NO, —0.15 —0.04 —0.08 NO, + NO, —0.26 0.24 0.15 pH 1.00 0.44 -0.44 0.03 —0.33 0.14 0.10 Eh NH, PO, Si NO, NO,+NO, 1.00 0.52 1.00 0.07 ~0.03 1.00 0.04 0.04 0.13 1.00 0.09 0.32 ~0.05 0.21 1.00 ~0.06 0.02 -0.07 0.23 0.12 1.00 10 TABLE 4.—Factor score coeffecients for chemical variables for bare surface and grass surface. Chemical variable Factor PC2(18%) PC1(31%) PC3(14%) Temperature Salinity 0.01 ~0.45 0.04 Oxygen 0.29 0.00 0.04 pH 0.24 0.07 ~0.19 Eh 10000 GRASS SURFACE QUINQUELOCULINA ft = 254.68 Aa © = 350.65 1000 100 MEAN DENSITIES 1978 10000 BARE SURFACE QUINQUELOCULINA fi = 328.59 an 6 = 585.57 1000 100 MEAN DENSITIES %, ee, 1978 FIGURE 13.—Mean number of individuals of Quinqueloculina per 5 ml of sediment (density). SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES temperature and NH, with oxygen, pH, and Eh. The second principal component (PC2) accounting for 18% of the variability consists mainly of salinity, Si, and NO, + NO,, although PO,, NH,, and Eh also contribute and a line of demarkation is not as clear cut as for PCl. The coefficients (Table 4) indicate that the third principal component (PC3) accounting for 14% of the variability consists mainly of NO,, Si, and temperature, but again there is no dramatic demarka- tion. The most highly correlated variables (Table 3) are all concentrated on PC1. SPECIES DENSITIES, STATION DIFFERENCES, PERIODICITY, AND ENVIRONMENTAL VARIABLES.—Quinqueloc- ulina: Quinqueloculina was the most abundant taxon making up about 75% of the total living foraminifera at the surface stations. A plot of the mean densities observed at the bare and grass surface at the two stations is shown in Figure 13. The ANOVA table for six hypotheses is shown in Table 5. We recall that each hypothesis is formulated by equating the desired B to zero. For example, the model used to test for station differences deletes B,, for m/3 periodicity and inter- action 6, = 6, = B= B, = 0, and so on. Table 5 indicates that all hypotheses except for station differences are significant. The overall mean density at the bare station is higher than at the grass station, but is not statistically significant at the chosen 0.05 level. Figure 13 shows that the bare station had high densities in spring while the grass station had high densities in summer. The environmental variables are significant as a group. Because they are not independent, testing for the significance of each individually is risky. Nevertheless, using the standard errors of the ty s from the @ model for calculating confidence limits indicates oxygen, Eh, PO,, Si, and NO, + NO, are significant. Again, emphasizing that the variables are not independent, the ease of calculation using “SYSTAT” made the calculation of simple regressions for each of the variables vs. density irresistible. The results shown in Table 6 indicate the regressions for oxygen, pH, PO,, Si, and NO, are significant. A way of avoiding the correlations between variables is to calculate a multiple regression using principal components instead of the original variables. Because the TABLE 5.—Statistical analysis of GLM for Quinqueloculina for bare surface and grass surface. Sum of squares Variability on account of Stations 1.86 1 1.86 3.64 0.06 1/3 periodicity 22.53 4 5.63 11.04 0.00 and interaction 1/6 periodicity 27.50 4 6.88 13.38 0.00 and interaction 1/3 interaction 9.88 2 4.94 9.68 0.00 7/6 interaction 20.45 2 10.23 20.05 0.00 Environmental 37.34 10 3.73 7.32 0.00 variables Residual 0.51 NUMBER 36 TABLE 6.—Values of F-ratio’s for simple regressions on species densities and environmental variables at bare surface and grass surface. (+ indicates signficant (.05 level) positive value of B; — significant negative value of B.) Environmental variables Temperature principal components are orthogonal, each hypothesis is independent and the analysis is similar to a one-way ANOVA. The results of an analysis on the log densities of Quinqueloc- ulina and the first three PC’s of the environmental variables are shown in Table 7. The overall F-ratio is significant and the test for the significance of each PC, which are independent, indicates that PC1 and PC3 are significant at the 0.05 level while PC2 is nearly so. We recall from Table 4 that temperature, oxygen, pH, Eh, and NH, all contribute nearly equally to the first PC, and the third PC is weighed heavily on temperature, Si, and NO,. Thus, the analysis using principal components requires seven of the 10 environmental variables for two PC’s and 10 for three to explain the results. All of the above analyses indicate that the identification of one or two variables as solely significant for the observed densities is impossible. Elphidium: Elphidium constitutes about 14% of the total living population and mean densities at the bare and grass stations are plotted in Figure 14 and the results of the GLM TABLE 7.—Regression of Quinqueloculina and principal components for bare surface and grass surface. Coefficient Standard t error Variable P(2 tail) R2=0.11 Constant PCI PC2 PC3 0.20 0.09 2.28 0.02 —0.17 0.09 Regression Residual 132 11 analysis is presented in Table 8. Although the mean density at the bare station is again higher than at the grass station, it is not Statistically significant. The 2/3 periodicity and interaction hypotheses are significant, but the 7/6’s are not. Figure 14 indicates a spring high at the bare station, but no pronounced summer high at the grass station as was observed with Quinqueloculina. The group of environmental variables are significant, and the ty s for Eh, PO,, NO,, and NO, + NO, were significant. Regressions of density vs. each individual environ- mental variable yielded a significant F-ratio for temperature, salinity, oxygen, pH, and Si (Table 6). The results of multiple-regression of Elphidium densities and the first three PC’s of the environmental variables are shown in Table 9. The first two PC’s are significant and these account for all of the environmental variables except NO,. Here, we note that individual tests of the ty s from the @ ri10deI and individual ty s from simple regressions do not agree testifying to the difficulty encountered when variables are highly correlated. Ammonia: Ammonia makes up about 8% of the total living population and mean densities at the bare and grass stations are plotted in Figure 15. The results of the GLM analysis are presented in Table 10. The overall mean density is slightly higher at the bare station, but not significantly so. The 7/3 periodicity and interaction hypotheses are significant and Figure 15 indicates an early spring maximum at the bare station while both stations have summer and fall maxima. Environ- mental variables are significant as a group and the B’s for Eh, NO,, and NO, + NO, were significant. Individual simple regressions on density vs. environmental variables yielded significant F-ratios for salinity and NO, (Table 6). The results of the multiple regression on densities of Ammonia and the PC’s of the environmental variables are shown in Table 11. The results present us with a small dilemma because the F-ratio for the overall analysis has a probability of 0.06 which is slightly above our chosen level, and, therefore, is not significant. On the other hand, PC3 is significant (Table 11). If we choose to regard the third principal component as significant, then temperature, Si, and NO, are the major contributors, especially NO, (Table 4). The F-ratio for environmental variables in the @ model, while significant, is the smallest encountered for any of the five taxa analyzed. Ammonia appears, then, to be the least influenced by the 10 variables measured in this study. Bolivina: Bolivina constitutes about 1% of the total living population and mean densities at the bare and grass stations are plotted in Figure 16. The results of the GLM analysis are shown in Table 12. Although the differences in densities between stations are small, they are, nevertheless, statistically signifi- cant, and the highest density is at the grass station. None of the hypotheses for periodicity are significant. There does appear to be a decrease in density over the course of sampling, but the fluctuations in density shown in Figure 16 are small compared to those considered previously (note the difference in the scale of the ordinate). The environmental variables are significant as 12 1000 GRASS SURFACE MEAN DENSITIES 1000 BARE SURFACE ELPHIDIUM (i = 80.96 G = 145.58 MEAN DENSITIES xy) 1978 2 FIGURE 14.—Mean number of individuals of Elphidium per 5 ml of sediment (density). a group, and the B’s for salinity and Si were significant. F-ratios for simple regressions are significant for salinity, PO,, Si, and NO, + NO, (Table 6). Multiple regression on densities of Bolivina and the first three PC’s of the environmental variables are shown in Table 13. Only PC2 consisting mostly of salinity, Si, and NO, + NO, (Table 4) is significant. The relationship with salinity and to a lesser extent with Si are notable for this species. Ammobaculites: Ammobaculites also constitutes about 1% of the total living population and mean densities for the bare and grass stations are plotted in Figure 17. The results of the GLM analysis are shown in Table 14. The hypothesis for station differences is significant with the highest densities occurring at the grass surface. The 1/3 periodicity and interaction hypotheses are significant, and Figure 17 indicates the now familiar spring maximum was observed at the bare station after which time the densities remained very low (Appendix 1). At the grass surface the densities increased overall during the sampling duration (Appendix 2). The hypothesis for the environmental variables is significant. The SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES TABLE 8.—Statistical analysis of GLM for Elphidium for bare surface and grass surface. Variability on Sum of df Mean F account of squares square ve) Stations 0.88 1 0.88 1.78 0.18 1/3 periodicity 8.28 4 2.07 4.21 0.00 and interaction 1/6 periodicity 2.94 4 0.74 1.49 0.21 and interaction 1/3 interaction 3.73 2 1.87 3.79 0.03 1/6 interaction 2.91 2 1.45 2.95 0.06 Environmental 28.43 10 2.84 5.78 0.00 variables Residual 57.04 116 0.49 TABLE 9.—Regression of Elphidium and principal components for bare surface and grass surface. Coefficient Standard t error Variable P(2 tail) R?=0.18 Constant 0.07 50.73 PCl 0.21 0.07 3.05 0.00 PC2 —0.30 0.07 —4,32 0.00 PC3 0.07 —0.40 Analysis of variance Source Sum of df Mean F-ratio P squares square Regression 18.25 3 6.08 9.39 0.00 Residual 85.54 132 0.65 B's of the @ model for the variables pH, PO,, Si, NO,, and NO, + NO, were significant. The F-ratios for temperature, salinity, oxygen, and NH, were significant for simple regres- sions on density and environmental variables (Table 6). The results of a multiple regression on the densities of Ammobacu- lites and the first three PC’s of the environmental variables are shown in Table 15. The first two PC’s are significant indicating that all the variables except for NO, (Table 4) are important contributors. Again, we note the inconsistencies obtained by testing the environmental variables individually. In summary, we recognize no station differences for the three most abundant taxa. For the rare species Bolivina and Ammobaculites which together constitute only about 2% of the total living population, densities are higher at the grass station. All of the taxa except for Bolivina exhibited periodicity. Quinqueloculina, Elphidium, and Ammonia all showed high densities in spring at the bare surface station and in summer at the grass surface. Bolivina exhibited an overall decreasing density from spring onward at both stations while Ammobacu- lites increased at the grass surface station and decreased at the NUMBER 36 1000 GRASS SURFACE AMMONIA Ht =27.54 Aa © = 29.74 wn = E 100 7) r4 = a & 410 = 1 4 Y, ‘p % Oe Lb, {9 ©, ©; xe <2 eI SS KS Sy ep 1978 1000 BARE SURFACE AMMONIA aA = iu = 33.94 6 = 29.78 3 MEAN DENSITIES S 1978 FIGURE 15.—Mean number of individuals of Ammonia per 5 ml of sediment (density). bare surface. The environmental variables were significant as a group for all taxa, however, individual variables cannot be evaluated with confidence. Grass Surface and Grass 10 cm ENVIRONMENTAL VARIABLES.—The recorded temperature values are plotted in Figure 18, and the results of a two-way ANOVA testing for differences between surface and 10 cm and with time are shown for temperature and all the other environmental variables in Table 16. The hypothesis testing for differences in mean temperature between the surface and 10 cm depth is not significant, however, the hypothesis testing for time is. As everyone knows, the water is warmer in the summer and cooler in spring and fall. A very high temperature was recorded in June and a very low one in August at 10 cm. The measured salinity values are shown in Figure 19. Only the hypothesis for time is significant. Salinities were highest in 13 TABLE 10.—Statistical analysis of GLM for Ammonia for bare surface and grass surface. Variability on Sum of df Mean F p(F) account of squares square Stations 0.06 1 0.06 0.13 0.72 1/3 periodicity 11.29 4 2.82 5.98 0.00 and interaction 1/6 periodicity 4.24 4 1.06 2.24 0.07 and interaction 1/3 interaction 6.38 2 3.19 6.76 0.00 1/6 interaction 1.94 2 0.97 2.05 0.13 Environmental 16.60 10 1.66 3.52 0.00 variables Residual 54.70 116 0.47 TABLE 1 1.—Regression of Ammonia and principal components for bare surface and grass surface. Standard Variable Coefficient P(2 tail) R?2=0.06 error Constant 3.14 0.07 45.92 0.00 PC1 -0.01 0.07 -0.16 0.87 PC2 -0.12 0.07 -1.72 0.09 PC3 0.15 0.07 2.18 0.03 Sum of squares 4.92 3 1.64 2.58 0.06 Regression Residual spring except for the first observation in March at 10 cm. Both stations experienced a dip in salinity in August. The recorded oxygen values are shown in Figure 20. The mean square for differences between surface and 10 cm is large and highly significant, and the mean square for time, although relatively much smaller, is nearly significant. Figure 20 illustrates and Appendix 6 tabulates zero recordings for oxygen during spring and summer at 10 cm making the difference between the surface and 10 cm (depth hypothesis) so dramatic. The measured pH values are plotted in Figure 21. The depth hypothesis is significant, and the pH is always lower at 10 cm (Figure 21, Appendix 5, 6). Even at 10 cm, however, only one reading (November) recorded a value below 7. Eh values are plotted in Figure 22. The depth hypothesis is significant. Values of Eh at 10 cm are always negative, and usually highly so, while the values at the surface fluctuate from positive to negative with a slightly negative mean (Figure 22, Appendix 5, 6). 14 100 GRASS SURFACE BOLIVINA ft = 7.26 =7.98 i S MEAN DENSITIES 100 BARE SURFACE BOLIVINA ft = 3.38 G = 3.82 MEAN DENSITIES 1978 FIGURE 16.—Mean number of individuals of Bolivina per 5 ml of sediment (density). NH, values are plotted in Figure 23. The depth hypothesis is significant. The values at 10 cm are usually two orders of magnitude greater than at the surface (Appendix 5, 6). Two zero values were recorded at 10 cm, one in April and one in October. We suspect these anomalies are due to problems with instrumentation. PO, values are plotted in Figure 24. The hypothesis for depth is significant. The values for PO, were very low at the surface, and one or two orders of magnitude higher at 10 cm (Appendix 5, 6). A very high value was recorded at 10 cm in June. Si values are plotted in Figure 25 and the hypothesis for depth is significant. The values for Si at the surface are an order of magnitude smaller than at 10 cm (Appendix 5, 6). At 10 cm zero values were recorded in April, August, and October. Except for the zero value in August, the pattern is very similar to that observed for NH, at 10 cm. NO, values are plotted in Figure 26. The hypothesis for depth is significant, and values for NO, are low. At 10 cm very low values were recorded in all months except October. SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES TABLE 12.—Statistical analysis of GLM for Bolivina for bare surface and grass surface. Variability on Sum of df Mean p(F) account of squares square Stations 4.03 1 4.03 7.45 0.01 1/3 periodicity 4.38 4 1.09 2.02 0.10 and interaction 1/6 periodicity 3:33 4 0.83 1.54 0.20 and interaction 7/3 interaction 1.25 2 0.62 eS 0.32 1/6 interaction 0.36 2 0.18 0.33 0.72 Environmental 42.21 10 4.22 7.80 0.00 variables Residual 62.77 116 0.54 TABLE 13.—Regression of Bolivina and principal components for bare surface and grass surface. Standard Variable Coefficient t P(2 tail) R2=0.28 error Constant 1.38 0.07 19.10 0.00 PCl 0.04 0.07 0.47 0.64 PC2 -0.52 0.07 -7.14 0.00 PC3 0.01 0.07 0.19 0.85 Analysis of variance Source Sum of df Mean F-ratio P squares square Regression 36.34 3 12.11 17.07 0.00 Residual 93.66 132 0.71 NO, + NO, values are plotted in Figure 27. There is no significant difference with depth or with time. Except for one measurement at the surface in April (Figure 27, Appendix 5) all the measurements were low. The two-way ANOVA’s for environmental variables at the bare surface vs. grass surface showed most significant differences are with time (Table 2). In contrast, the analysis for grass surface vs. grass 10 cm shows most of the significant differences are with depth (Table 16). Moreover, the F-ratios for the latter are usually much higher. We have, then, a situation where, except for temperature, salinity, and NO, + NO,, there are much larger differences than at the surface stations. The environmental variables are highly correlated as shown in Table 17. High positive values occur between oxygen and Eh, oxygen and NO,, pH and Eh, Eh and NO,, NH, and POQ,, NH, and Si, and PO, and Si. High negative values occur between oxygen and NH,, oxygen and PO,, oxygen and Si, pH and NH,, pH and Si, Eh and NH, Eh and PO,, and Eh and Si. The correlations differ from those at the bare and grass surface NUMBER 36 100 GRASS SURFACE AMMOBACULITES MEAN DENSITIES 100 BARE SURFACE AMMOBACULITES MEAN DENSITIES 1 q, “p, % %, % Np tp 1978 FIGURE 17.—Mean number of individuals of Ammobaculites per 5 ml of sediment (density). (Table 3) in that PO,, Si, and NO, now are highly correlated so that the matrix of correlation coefficients has many more high values. To succinctly summarize the environmental variables, and remove the covariance between them, a principal compo- nent analysis was calculated on the correlation matrix. The first three eigenvalues account for-about 70% of the variability. The factor score coefficients are given in Table 18. The first vector accounting for 45% of the variability has high values for oxygen, pH, Eh, NH, PO,, Si, and NO,. The second vector accounting for 15% of the variability has the highest values for salinity and NO, + NO,, and the third accounting for 10% of the variability has a high value for temperature. Thus, all of the water chemistry variables are accounted for by using the first three principal components, and PC1 accounts for all of them except temperature, salinity, and NO, + NO, which, interest- ingly, are the variables without significant differences with depth (Table 16). SPECIES DENSITIES, DEPTH DIFFERENCES, PERIODICITY, AND 15 TABLE 14.—Statistical analysis of GLM for Ammobaculites for bare surface and grass surface. Variability on Sum of df Mean F p(F) account of squares square Stations 21.59 1 21.59 44.79 0.00 m/3 periodicity 8.56 4 2.14 4.44 0.00 and interaction 7/6 periodicity 3.13 4 0.78 1.62 0.17 and interaction 1/3 interaction 4.90 2 2.45 5.08 0.01 1/6 interaction 0.08 2 0.04 0.08 0.92 Environmental 18.26 10 1.83 3.79 0.00 variables Residual 55.87 116 0.48 TABLE 15.—Regression of Ammobaculites and principal components for bare surface and grass surface. Coefficient Standard t error Variable P(2 tail) R?=0.06 Constant i PC1 0.18 0.08 Zale 0.03 PC2 0.17 0.08 2.07 0.04 PC3 Sum of df Source F-ratio P squares square Regression 7.97 3 2.66 3.04 0.03 Residual 115.46 132 0.88 ENVIRONMENTAL VARIABLES.—Quinqueloculina: Quin- queloculina mean densities are plotted in Figure 28, and analysis by the GLM is shown in Table 19. As noted earlier, the densities for Quinqueloculina at the grass surface exhibited a summer maximum in July and August (Appendix 2). At 10 cm maxima were observed in March and November. The large increase in density observed in the summer at the surface is not reflected at 10 cm. All of the hypotheses tested by the GLM were significant (Table 19). The mean square for differences with depth is very large reflecting the two orders of difference in magnitude of the mean density between the surface and 10 cm. The set of environmental variables are significant and in the @ model B’s for oxygen, Eh, and NH, were significant. Simple regressions on density vs. environmental variables indicate the F-ratios for all variables except temperature are significant (Table 20). The results of a multiple regression on the densities of Quinqueloc- ulina and the first three PC’s of the environmental variables are 16 TABLE 16.—Analysis of variance for chemical variables on grass surface and 10 cm. Variable | Source PENS ap rea p(F) squares Temperature | depth 0.54 1 0.54 0.12 0.74 time 273.64 16 17.10 3512 0.01 residual 73.63 16 4.60 Salinity depth By PA ana 522 2.83 0.11 time 311.96 16 19.50 9.64 0.00 residual 32°35); 16 2.02 Oxygen depth 87.68 1 87.68 30.77 0.00 time 96.21 16 6.01 21 0.07 residual 45.59 16 2.85 pH depth 2.950 2.95 45.09 0.00 time 1.03 16 0.06 0.98 0.52 residual 1.05 16 0.07 Eh depth 469412.50 1 469412.50 109.31 0.00 time 107511.88 16 6719.49 1.57 0.19 residual 68712.00 16 4294.50 NH, depth 1148940.80 1 1148940.80 32.13 0.00 time 601797.43 16 37612.34 1.05 0.46 residual 572067.08 16 35754.19 PO, depth 12189.21 1. 12189.21 8.61 0.01 time 22394.25 16 1399.64 0.99 0.51 residual 22652.76 16 1415.80 Si depth 1477313.29. = -1-—-«:1477313.29 41.34 0.00 time 635755.12 16 39734.70 1.11 0.42 residual 571797.45 16 35737.34 NO, depth 0.11 1 0.11 7.82 0.01 time 0.19 16 0.01 0.85 0.62 residual 0:23— 16 0.01 NO, + NO, depth 293323) atl 2933.23 1.33 0.27 time 35026.37 16 2189.15 0.99 0.51 residual 35275.14 16 2204.70 shown in Table 21. The first and third PC’s are significant indicating that only NO, + NO, did not contribute substantially to the significance of the analysis. The F-ratios for the simple regressions are much larger than for the bare surface and grass surface (compare Tables 6 and 20). Similarly, the F-ratio and R? for the multiple regression using the PC’s is much larger for the grass surface and grass 10 cm than for the bare surface and grass surface (compare Tables 7 and 21). This is, of course, a reflection of the very large differences between densities and environmental variables between the surface and 10 cm. The signs of the factor score coefficients for PCI (Table 18), the correlation matrix (Table 17), and the B’s of the simple regressions (Table 20) show that the analysis contrasts oxygen, pH, Eh, NO, with NH,, PO,, Si. Elphidium: Elphidium mean densities at the grass surface and grass 10 cm are plotted in Figure 29. The summer maximum observed at the surface for Quinqueloculina was not observed for Elphidium (Appendix 2). Minima at the surface were observed in June and November which is similar to the SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES GRASS SURFACE 40 As ft =26.61 S = 2.96 =) i= << o je) i*I = ea} = 40 GRASS 10 CENTIMETERS nw LU = 26.86 nw So = 3.60 j=) = < $2 & (5 = ica) » ty ch ‘a Sy 1978 4 ? FIGURE 18.—Temperature measurements in °C. pattern of Quinqueloculina (Appendix 2). At 10 cm a maximum occurred in March due to high number of individuals in two of the four replicates (Appendix 3). Once again, however, the densities at 10 cm were very low, averaging less than two individuals. Table 22 indicates the largest mean square is for the hypothesis contrasting depth. The 7/3 periodicity with interac- tion is significant while the 1/6 is not. The set of environmental variables are significant, and individual B’s of the @ model for Eh and NH, were significant. F-ratios of all variables except temperature and salinity are significant for simple regressions (Table 20). The results of a multiple regression on the densities of Elphidium and the first three PC’s of the environmental variables indicate that all three PC’s, and, consequently, all the environmental variables contribute substantially (Table 23). Ammonia: Ammonia densities for the grass surface and grass 10 cm are plotted in Figure 30. A maximum density was NUMBER 36 GRASS SURFACE [it = 28.79 35 G = 335 > (> _— Zz S| = nn GRASS 10 CENTIMETERS ft =29.62 35 6 = 321 30 > ‘= a Z = = nN 25 20 ty, ) ie % %, % q # Tp Ay, a Co, SO Xe ¢ Pe. 1978 FIGURE 19.—Salinity measurements in °/oo. OXYGEN OXYGEN GRASS SURFACE A UL =4.47 cas O =1.83 Ty 4 %, y q, Lp q B Cp_ (“A Ch, % e Ve tp %, “2p, >| “ 1978 GRASS 10 CENTIMETERS ft =1.26 G =2.35 1978 iw) FIGURE 20.—Oxygen measurements in mg-at/1. 18 pH pH GRASS SURFACE ft =8.01 6 = 0.24 GRASS 10 CENTIMETERS “a UW =7.42 6 =0.27 Se Pe Nene al see eel 1978 FIGURE 21.—pH measurements. Eh Eh SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES GRASS SURFACE fl = -4.06 6 = 78.29 200 100 0 -100 -200 -300 -400 dp a7 %, %, 4, Ye Gg, 4, | |) | el See ee % > % 4 %, ane. % 1978 200 GRASS 10 CENTIMETERS ft = -239.06 100 S= 69.89 0 -100 -200 -300 -400 4 4 4b %, Y oe S (a) 4 cK Sl MM Ie RG IIS. 2, |S & o Sp y 7) “py, @, 77) ©, bp, ep 1978 FIGURE 22.—Eh measurements in mV. NUMBER 36 19 300 GRASS SURFACE GRASS SURFACE 1000 ff = 4.28 ft =0.54 6 = 8.05 250 G =0.57 100 200 oj ¢ = 150 10 100 50 1 q o Y 4, S, ca) 4, “4, % I. Go % es, 0 op > “Gp p %, 1978 GRASS 10 CENTIMETERS a “aw ft =371.93 GRASS 10 CENTIMETERS 6 =270.74 ft =38.41 nw a ah) 6 = 53.06 100 S [<0 foe) = Zs 10 1 wy | a ty, % % 1%, |% Q % 7 27 7 C . ©. 7) y g S, fa) 4 lip %, ip Np ep G as % Le ty Ip tp *, %, 4, %, On, Ve % ‘ “ “Ny e, ey, Go| % ME Ne

SI Ry Date g RY RY ‘ Y 9 (1978) Ni RS g £ s eS ¥ € $ € Ny ® y ~ % «9 28 Aug 163 19 14 0 5 208 196 17 10 2 2 248 196 3 14 1 2 221 185 33 4 Z 6 247 11 Sep 117 31 11 1 12 178 77 24 11 1 11 128 76 18 20 0 4 127 95 41 19 1 6 164 25 Sep 208 24 18 0 16 270 162 29 3 0 17 217 131 18 22 0 9 187 199 20 15 2 a 255 9 Oct 152 33 27 3 16 239 159 19 25 g) 16 227 145 17 19 0 10 207 67 10 4 0 8 108 23 Oct 224 25 30 il 22 329 304 81 56 5 18 480 197 34 54 3 19 321 238 25 37 7 16 336 6 Nov 115 20 33 2 14 208 19 17 19 0 1 57 17 6 6 0 3 39 17 9 5 2 9 55 Appendix 3 Grass 10 cm Number of individuals per 5 ml of sediment. mS | td As ~ 54 100 11 20 10 11 16 16 29 15 15 19 53 17 14 15 32 33 18 24 27 22 14 < 27 Mar 10 Apr 24 Apr 8 May 22 May 5 Jun 19 Jun 3 Jul 17 Jul 31 Jul 14 Aug SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES 34 Appendix 3.—Continued. Date (1978) 27 Mar 10 Apr 24 Apr 8 May 22 May 5 Jun 19 Jun 3 Jul 17 Jul 31 Jul 14 Aug 28 Aug 11 Sep 25 Sep 9 Oct 23 Oct 6 Nov Temperature 24.5 26.5 23.5 27.0 28.3 32.0 27.0 30.0 31.0 30.0 31.0 30.5 30.5 29.5 24.0 24.0 23.0 Salinity 32.0 32.5 32.5 3S 39.0 35.0 33.0 32.0 30.5 28.0 20.0 26.0 28.0 26.0 25.5 26.0 27.0 Oxygen 7.4 9.1 10.2 0.4 1.8 0.6 Sal 0.4 3.3 4.4 23 2.4 3.4 5.4 6.0 3:3 4.9 pH 8.63 8.70 8.75 7.50 8.30 7.50 8.10 MSS 7.50 8.30 8.08 8.21 8.06 8.12 TOM 8.00 8.35 Appendix 4 Bare Surface Pore water chemistry. Eh 35 NH, 0.1425 0.0920 0.1149 0.1204 21.2810 95.9300 2.1058 65.8357 1.4197 0.1717 21.3100 0.1721 1.8610 0.1638 0.1790 1.2926 3.1530 PO, 0.7354 0.2202 0.0320 0.5373 0.0412 0.8373 0.0560 1.1206 0.0524 0.1915 1.6034 0.5327 2.2904 16.4580 0.0742 0.4990 1.8657 Si 13.3037 26.7167 7.1574 11.5527 49.3007 70.7123 29.4110 32.1760 68.8000 55.7310 51.5230 0.0000 104.0400 66.4087 133.5466 105.5033 69.4803 NO, 0.0721 0.0609 0.0113 0.0218 0.0229 0.3944 0.3152 0.3722 0.3134 0.2811 0.2359 0.0224 0.0215 0.0225 0.0218 0.3724 0.6236 NO, + NO, 4.3109 3.0221 4.9155 27.1770 3.8981 4.5691 3.3008 47.3260 0.1385 1.3133 0.5315 0.0622 0.0668 0.1841 0.0844 0.6684 1.8702 Date (1978) 27 Mar 10 Apr 24 Apr 8 May 22 May 5 Jun 19 Jun 3 Jul 17 Jul 31 Jul 14 Aug 28 Aug 11 Sep 25 Sep 9 Oct 23 Oct 6 Nov Temperature 22.0 24.5 22.8 27.0 26.5 29.0 25.2 29.5 28.0 29.0 30.0 31.0 29.5 28.5 23:5 23.3 23.0 Salinity Oxygen 32.5 31.0 33.0 31.0 33.0 31.0 32.0 29.0 28.0 31.0 22.0 26.0 25.5 25.5 24.5 28.0 26.5 pH 8.03 8.25 8.35 8.00 8.15 7.60 8.00 7.50 7.74 8.08 8.17 8.21 8.03 8.16 7.72 8.00 8.19 Appendix 5 Grass Surface Pore water chemistry. Eh =A 105 = -56 -67 =1128 185 S115 38 -50 -47 = 12 =33 -37 52 43 50 36 NH, 0.1425 0.0920 0.9383 0.1204 7.8668 33.7027 0.1486 5.5559 0.0000 4.6810 6.9164 0.1721 2.9416 0.3979 0.1790 4.6589 4.2475 PO, 0.8665 0.1329 0.1563 0.0391 0.0412 0.3404 0.0560 0.4846 0.0000 0.4182 0.9911 0.3936 1.1962 0.4561 0.5758 0.7639 2.2476 Si 13.7943 22.6467 12.0900 14.7607 32.5660 50.9217 1.3497 69.2380 70.8490 43.1990 16.9515 0.0000 98.0553 72.2520 138.5800 106.3733 68.6020 NO, 0.0549 0.0535 0.2704 0.0218 ~ 0.0229 0.1105 0.1303 0.2146 0.0313 0.1782 0.1416 0.0224 0.0215 0.0225 0.0218 0.5176 0.5030 NO, + NO, 20.8870 3.2316 275.4266 5.3229 10.0146 Date (1978) 27 Mar 10 Apr 24 Apr 8 May 22 May 5 Jun 19 Jun 3 Jul 17 Jul 31 Jul 14 Aug 28 Aug 11 Sep 25 Sep 9 Oct 23 Oct 6 Nov Temperature 24.2 23.8 23°) 27.2 27.2 28.5 25.5 36.2 27.7 27.8 21.0 30.0 31.0 29.0 24.0 26.5 23.5 Salinity Oxygen 27.0 30.7 33.0 33.5 34.5 3335 32.5 32.0 30.5 30.5 24.0 26.5 27.5 27.5 26.0 2IES 27.0 pH 7.71 7.67 7.60 7.30 7.45 7.50 7.60 7.52 Use) 7.44 7.62 7.46 7.49 7.49 7.35 7.04 6.57 Appendix 6 Grass 10 cm Pore water chemistry. Eh —196 —235 Se —286 —289 ~307 —331 —312 —282 —288 — 27) —223 HH —225 —199 —33 —176 Si NH, 96.2270 0.0000 285.6600 312.6400 463.4599 836.8533 929.6766 745.4833 518.7266 387.2833 198.0666 233.1100 417.0299 470.0866 240.5966 0.0000 187.9799 PO, 12.2930 0.0000 28.0770 26.0440 29.3460 51.8203 32.1800 70.4513 46.8573 36.5710 19.7187 13.1470 25.0397 26.2023 12.2607 0.0000 22.9157 583.6932 515.6699 688.2266 659.5033 616.1066 572.8933 559.7299 785.4966 461.1333 0.0000 324.8800 Literature Cited Bliss, C.I. 1958. Periodic Regression in Biology and Climatology. Connecticut Experiment Station Bulletin, 615:3-5S. Boltovskoy, E. 1964. Seasonal Occurrences of Some Living Foraminifera in Puerto Deseado (Patagonia, Argentina). Journal du Conseil International pour I’ Exploration de la Mer, 29(2):136-145. Boltovskoy, E., and H. Lena 1969. Seasonal Occurrences, Standing Crop, and Production in Benthic Foraminifera of Puerto Deseado. Contributions from the Cushman Foundation for Foraminiferal Research, 20(3):87-95. Bradshaw, J.S. 1961. Laboratory Experiments on the Ecology of Foraminifera. Contribu- tions from the Cushman Foundation for Foraminiferal Research, 12(3):86-106. Brooks, A.L. 1967. Standing Crop, Vertical Distribution, and Morphometrics of Ammo- nia beccarii (Linne). Limnology and Oceanography, 12(4):667- 684. Buzas, M.A. 1965. The Distribution and Abundance of Foraminifera in Long Island Sound. Smithsonian Miscellaneous Collections, 149(1): 89 pages. 1969. Foraminiferal Species Densities and Environmental Variables in an Estuary. Limnology and Oceanography, 14(3):411-422. 1977. Vertical Distribution of Foraminifera in the Indian River, Florida. Journal of Foraminiferal Research, 7(3):234-237. 1978. Foraminifera As Prey for Benthic Deposit Feeders: Results of Predator Exclusion Experiments. Journal of Marine Research, 36(4):617-625. Buzas, M.A., and K.P. Severin 1982. Distribution and Systematics of Foraminifera in the Indian River, Florida. Smithsonian Contributions to the Marine Sciences, 16: 73 pages. Buzas, M.A., R.K. Smith, and K.A. Beem 1977. Ecology and Systematics of Foraminifera in Two Thalassia Habitats, Jamaica, West Indies. Smithsonian Contributions to Paleobiology, 31: 139 pages. Erskian, M.G., and J.H. Lipps 1987. Population Dynamics of the Foraminiferan Glabratella ornatissima (Cushman) in Northern California. Journal of Foraminiferal Research, 17(3):240-256. Haake, F.-W. 1967. Zum Jahresgang von populationen einer Foraminiferen-Art in der westlichen Ostsee. Meyniana, 17:13-27. Haman, D. 1969. Seasonal Occurrence of Elphidium excavatum (Terquem) in Llan- danwg Lagoon (North Wales, U.K.). Contributions from the Cushman Foundation for Foraminiferal Research, 20(4):139-142. Jones, G.D., and C.A. Ross 1979. Seasonal Distribution of Foraminifera in Samish Bay, Washington. Journal of Paleontology, 53(2):245-257. 38 Montgomery, J.R., C. Zimmermann, G. Peterson, and M. Price 1983. Diel Variations of Dissolved Ammonia and Phosphate in Estuarine Sediment Pore Water. Florida Scientist, 46(3/4):407-414. Montgomery, J.R., C. Zimmermann, and M. Price 1979. The Collection, Analysis, and Variation of Nutrients in Estuarine Pore Water. Estuarine and Coastal Marine Science, 9:203-214. Myers, E.H. 1942. A Quantitative Study of the Productivity of the Foraminifera in the Sea. Proceedings of the American Philosophical Society, 85(4): 325-342. Nyholm, G.-G., and I. Olsson 1973. Seasonal Fluctuations of the Meiobenthos in an Estuary on the Swedish West Coast. Zoon, 1:69-76. Phleger, F.B, and R.R. Lankford 1957. Seasonal Occurrences of Living Benthonic Foraminifera in Some Texas Bays. Contributions from the Cushman Foundation for Foraminiferal Research, 8(3):93-105. Schnitker, D. 1974. Ecotypic Variation in Ammonia beccarii (Linne). Journal of Foraminiferal Research, 4(4):216-223. Seal, H.L. 1964. Multivariate Statistical Analysis for Biologists. 207 pages. New York: Wiley. Severin, K.P. 1987. Laboratory Observations of the Rate of Subsurface Movement of a Small Miliolid Foraminifer. Journal of Foraminiferal Research, 17(2):110-116. Severin, K.P., S.J. Culver, and C. Blanpied 1982. Burrows and Trails Produced by Quinqueloculina impressa Reuss, a Benthic Foraminifer, in Fine-grained Sediment. Sedimentology, 29:897-901. Walton, W.R., and B.J. Sloan 1990. The Genus Ammonia Brunnich, 1772: Its Geographic Distribution and Morphologic Variability. Journal of Foraminiferal Research, 20(2):128-156. Wefer, G. 1976. Environmental Effects on Growth Rates of Benthic Foraminifera (Shallow Water, Baltic Sea). Jn C.T. Shaffer and B.R. Pelletier, editors, First International Symposium on Benthonic Foraminifera of Continental Margins, Part A: Ecology and Biology, pages 39-50. Halifax, Nova Scotia: Editoral Board of Maritime Sediments under the auspices of the Coordinating Committee of Benthonics “75.” Wetmore, K.L. 1988. Burrowing and Sediment Movement by Benthic Foraminifera, as Shown by Time-lapse Cinematography. Revue de Paleobiologie Special Volume, 2:921-927. Young, D.K., M.A. Buzas, and M.W. Young 1976. Species Densities of Macrobenthos Associated with Seagrass: A Field Experimental Study of Predation. Journal of Marine Research, 34(4):577-592. apy vy (conducted by their originating Smithsonian museums or s) and are submitted to the Smithsonian Institution Press with I-36, which must show the approval of the appropriate y designated by the sponsoring organizational unit. Requests ial treatment—use of color, foldouts, case-bound covers, uire, on the same form, the added approval of the ring authority. of manuscripts and art by the Press for requirements of format and style, completeness and clarity of copy, and ment of all material, as outlined below, will govern, within the ent of the Press, acceptance or rejection of manuscripts and = must be prepared on typewriter or word processor, -spaced, on one side of standard white bond paper (not with 11/4” margins, submitted as ribbon copy (not carbon in loose sheets (not stapled or bound), and accompanied | art. Minimum acceptable length is 30 pages. matter (preceding the text) should include: title page with and author and no other information; abstract page with series, etc., following the established format; table of ts with indents reflecting the hierarchy of heads in the paper; word and/or preface, if appropriate. age of text should carry the title and author at the top of the second page should have only the author's name and ‘mailing address, to be used as an unnumbered footnote irst page of printed text. ar heads of whatever level should be typed with initial caps of rds, with extra space above and below the head, but no ation (such as all caps or underline, except for the ecessary for generic and specific epithets). Run-in 1 heads should use period/dashes or colons as necessary. lations within text (lists of data, often in parallel columns) can id on the text page where they occur, but they should not rules or numbered table captions. bles (numbered, with captions, boxheads, stubs, rules) be submitted as carefully typed, double-spaced copy from the text; they will be typeset unless otherwise . If camera-copy use is anticipated, do not draw rules on script copy. lic keys in natural history papers should use the -couplet form for zoology and may use the multi-level indent for botany. If cross referencing is required between key and text, nclude page references within the key, but number the taxa, using the same BUnIBETS with their corresponding 1 the text. onymy in zoology must use the short form (taxon, author, ge), with full reference at the end of the paper under ure Cited.” For botany, the long form (taxon, author, fed journal or book title, volume, page, year, with no “Literature Cited”) is optional. ext-reference system (author, year:page used within the text, full citation in “Literature Cited” at the end of the text) must be | place of bibliographic footnotes in all Contributions Series Strongly recommended in the Studies Series: “(Jones, 22)" or “... Jones (1910:122).” If bibliographic footnotes are fanuscripts intended for series publication receive substantive - REQUIREMENTS FOR SMITHSONIAN SERIES PUBLICATION required, use the short form (author, brief title, page) with the full citation in the bibliography. Footnotes, when few in number, whether annotative or biblio- graphic, should be typed on separate sheets and inserted immedi- ately after the text pages on which the references occur. Extensive notes must be gathered together and placed at the end of the text in a notes section. Bibliography, depending upon use, is termed “Literature Cited,” “References,” or “Bibliography.” Spell out titles of books, articles, journals, and monographic series. For book and article titles use sentence-style capitalization according to the rules of the language employed (exception: capitalize all major words in English). For journal and series titles, capitalize the initial word and all subsequent words except articles, conjunctions, and prepositions. Transliterate languages that use a non-Roman alphabet according to the Library of Congress system. Underline (for italics) titles of journals and series and titles of books that are not part of a series. Use the parentheses/colon system for volume (number):pagination: “4 0(2):5-9.” For alignment and arrangement of elemenis, follow the format of recent publications in the series for which the manuscript is intendéd. Guidelines for preparing bibliography may be secured from Series Section, SI Press. Legends for illustrations must be submitted at the end of the manuscript, with as many legends typed, double-spaced, to a page as convenient. ; Illustrations must be submitted as original art (not copies) accompanying, but separate from, the manuscript. Guidelines for preparing art may be secured from the Series Section, SI Press. All types of illustrations (photographs, line drawings, maps, etc.) may be intermixed throughout the printed text. They should be termed Figures and should be numbered consecutively as they will appear in the monograph. If several illustrations are treated as components of a single composite figure, they should be designated by lowercase italic letters on the illustration; also, in the legend and in text references the italic letters (underlined in copy) should be used: “Figure 9b.” Illustrations that are intended to follow the printed text may be termed Plates, and any componenis should be similarly lettered and referenced: “Plate 9b.” Keys to any symbols within an illustation should appear on the art rather than in the legend. Some points of style: Do not use periods after such abbrevia- tions as “mm, ft, USNM, NNE.” Spell out numbers “one” through “nine” in expository text, but use digits in all other cases if possible. Use of the metric system of measurement is preferable; where use of the English system is unavoidable, supply metric equivalents in parentheses. Use the decimal system for precise measurements and relationships, common fractions for approximations. Use day/month/ year sequence for dates: “9 April 1976.” For months in tabular listings or data sections, use three-letter abbreviations with no periods: “Jan, Mar, Jun,” etc. Omit space between initials of a personal name: “J.B. Jones.” : Arrange and paginate sequentially every sheet of manuscript in the following order: (1) title page, (2) abstract, (3) contents, (4) foreword and/or preface, (5) text, (6) appendices, (7) notes section, (8) glossary, (9) bibliography, (10) legends, (11) tables. Index copy may be submitted at page proof stage, but plans for an index should be indicated when the manuscript is submitted. “AME —_————