Baseline Assessments and Analysis of Fish, Macroinvertebrates and Herpetofauna in the Otter Creek Coal Tracts Area of Powder River County Prepared for: Pat Farmer, Project Manager/Director Westech Environmental Services Helena Field Office Prepared by: David Stagliano, Aquatic Ecologist Montana Natural Heritage Program a cooperative program of the Montana State Library and the University of Montana January 2012 MONTANA Natural Heritage Xm> Program Executive Summary We summarize the first year of baseline surveys for the Aquatic Assessment of Fish, Macroinvertebrates, and Herpetofauna in the Otter Creek coal tracts area. Project goals were: 1) to conduct standardized surveys and collect baseline information on the aquatic and riparian communities occurring in Otter Creek and three tributaries (seasonally and spatially oriented) prior to coal development, 2) to assess aquatic community integrity and condition by interpreting key indicators recorded at sites using standardized protocols and biotic thresholds, and to compare these against reference condition standards. These data collected represent pre- development (i.e. pre-impact, BACI design) conditions at the local reach scale. Habitat assessments, herpetofauna, macroinvertebrate and fish surveys were performed seasonally at three predetermined mainstem Otter Creek reaches (control, impact and downstream) and three tributaries coinciding with established surface water quality stations during 2011. Additionally, we added fish sampling visits to Otter Creek Impact #2 (upstream of the Threemile Creek confluence) because Threemile Creek remained "dry" during all visits. In total, we performed 15 surveys for fish during the visits: 11 at four mainstem Otter Creek reaches and four surveys at two tributary streams. Fifteen macroinvertebrate samples were collected during the visits; neither taxa survey was conducted at Threemile Creek in any season due to lack of surface water present. All six a priori stream reaches were visually surveyed for amphibians or reptiles during all visits. Biological community integrity was calculated for 15 fish surveys using Fish Integrated Biotic Indices (IBI's) and Observed/Expected Models (O/E), while the 15 macroinvertebrate samples were assessed with Montana DEQ's multi-metric indices (MT MMI). Habitat Evaluations. Of the seven sampling reaches evaluated in the study area, we found three in Proper Functioning Condition (PFC) with a stable trend and four were Functional at Risk (FAR). Reasons that sites ranked FAR were likely due to anthropogenic habitat alteration by cattle (Home Creek {Otter_lA} and Threemile Creek {Otter_3m}) or stream manipulation (Otter Creek JT- Trussler and Otter Creek #16). Highest site integrity scores using both the BLM Habitat and PFC Assessment methods were recorded at the Otter Creek sites #23 (Tenmile Creek) and #22 (control- Denson reach). Sites with lower habitat scores were structurally degraded predominately by cattle use and had high associated Livestock Use Indices (LUI) (Home Creek, Threemile and Otter Creek #16-fall). Point conductivity measurements recorded at all Otter Creek mainstem sites across most seasons were above the threshold for impairment levels (>500ps, DEQ 2006), and Home Creek site 1A had visible signs of natural gas seepage from the sediments. Macroinvertebrate Communities: Overall, 104 unique macroinvertebrate taxa were reported from the 15 macroinvertebrate assessment samples. One known Montana species of concern (SOC), the mayfly, Caenis youngi was collected in fair numbers at the control site, Otter Creek #22. Average macroinvertebrate taxa richness per site was 29.5 and the highest taxa richness was 41 taxa reported at the Otter Creek JT site. Using the Montana DEQ macroinvertebrate multimetric index (MMI), four of the five sites (12 of 15 samples) were ranked non-impaired (good to excellent biological integrity), while all three samples from Tenmile Creek were ranked marginally impaired. Stream sites that maintained flowing, connected water scored higher with the MMI than sites with interrupted pool areas. Overall, mainstem sites evaluated in the Otter Creek study received significantly higher macroinvertebrate MMI scores than those in the tributaries (T-test, p <0.01). MMI's did not significantly differ on Otter Creek mainstem Pre-lmpact Control, Impact or Downstream Sites (T-test, p >0.05), despite the fish communities reflecting a decrease in biotic integrity. Fish Communities. Overall, nine fish species (five native/four introduced) were identified from 37,679 individuals collected from 15 site visits (Table 4). One potential species of concern (PSOC), the brassy minnow, was collected at five of six sites. Average fish species per Otter Creek mainstem site across all seasons was 6.5 (± 0.8 SE), while the tributary sites averaged 1.75 species (Table 4). All fish presence sites also reported at least one species of amphibian. Lake chubs had the highest site occupancy rate at 93% (14 of 15 visits) followed by fathead and brassy minnows at 80% (12 of 15 visits). Fish data collected in previous years from three sites within the study area showed similar biological integrity over time. The most diverse site in the study area was Otter Creek JT site with nine species, while the most intact sites were Otter Creek Site #22 at Denson's (four native) and the Tenmile Creek spring survey (two native spp.). Using Montana's Prairie Fish IBI, 9 of the 15 fish sites were ranked non-impaired (good biological integrity), two were slightly impaired (moderate integrity) and four were moderately impaired (poor biotic integrity). Fish IBI's decreased going downstream on Otter Creek, and the Pre-lmpact Control Site scored significantly higher than Impact or Downstream sites (T-test, p <0.05). Amphibian and Reptile Incidentals. Eight herpetofauna species were observed, collected in dipnets/seines or incidentally recorded in conjunction with the fish, habitat and macroinvertebrate surveys. We reported four amphibian species, of which, Woodhouse's Toad (Bufo woodhousii) had the highest site occupancy, occurring at six of seven sites, followed by the Northern Leopard Frog (Rana pipiens) and Tiger Salamander (Ambystoma tigrinum) recorded at five and four sites, respectively. Boreal Chorus Frog (Pseudacris maculata) was detected at three sites during the spring visits only. We also recorded four reptile species (in order of site occurrence): Western Rattlesnake (Crotalus viridis), Painted Turtle (Chrysemys picta), Terrestrial Garter Snake, (Thamnophis elegans) and Snapping Turtle (Chelydra serpentine) (SOC). Conclusions. The aquatic community sampling sites chosen for pre-impact baseline data were representative of the range of stream classes found in the Otter Coal Tracts project area: Ephemeral, Intermittent and Perennial Prairie Streams. Despite this having been an unusually high water year for the region, stream communities that we sampled encompassed the range of expected species to occur in these stream types. Biotic integrity of sites was initially higher in the upstream control reaches of Otter Creek. One extraordinary finding was the high density and large biomass of fish inhabiting the stream reach below Trusler's Ranch road crossing (~20,000 fish in the 300 m reach fall survey, 15 fish/m 2 ); this was most likely an artifact of the impassible culvert located here and the fish "stacking up" downstream of this barrier. Density dependent fish anomalies (lesions and parasites, i.e., yellow grub and anchorworm) were prevalent at this site, indicating the fish were likely experiencing stress from overcrowding. Benthic macroinvertebrate densities at this site were low, but, more diverse than other sites, indicating that the fish may have been consuming large portions of the available insect production, having a top-down community effect. Table of Contents INTRODUCTION 6 METHODS Study Area 6 Habitat Assessments 7 Macroinvertebrate Community Surveys 9 Fish and Amphibian Surveys Surveys 10 RESULTS Habitat Assessments 11 Macroinvertebrate Community Surveys 12 Fish Communities 14 Hepetofauna Communities 18 CONCLUSIONS 19 SITE PHOTOS 21 LITERATURE CITED 25 APPENDIX A. Fish data and IBI metric calculations collected from Otter Creek Project Sites APPENDIX B. Macroinvertebate taxa list, abundance and metrics for the 15 collection sites List of Figures gure 1. USGS gage data for Otter Creek at Ashland in 2011 8 gure 2. Macroinvertebrate D-net Sampling procedure 9 gure 3. Brassy minnows and lake chubs in the seine at Otter 22 10 gure 4. Photo of a structurally cattle-damaged, functional-at-risk (FAR) stream reach 11 gure 5. DEQ MMI scores by site and season 13 gure 6. Average macroinvertebrate MMI scores by site type 14 gure 7. Fish IBI seasonality recorded during 2011 surveys 16 gure 8. Comparison of the Fish IBI and O/E scores between sites and season 17 gure 9. Average fish IBI scores by BACI design site type 17 List of Tables Table 1. Otter Creek Study site GPS locations of reach 7 Table 2. Impairment determinations from the MMI and O/E (RIVPACS) models 9 Table 3. Fish metrics and classification offish species captured in the study area 11 Table 4. Macroinvertebrate sample characteristics and metrics used for the DEQ MMI 13 Table 5. Fish abundance, IBI and O/E scores for the 12 surveys reporting fish 15 Table 6. Species of vertebrates collected during the aquatic surveys in 2011 18 Table 7. Otter Creek Temporal Habitat Scores, Fish IBIs and MMIs 19 Acknowledgements We would like to thank Pat Farmer of Westech Environmental Services of Helena for funding this project under agreement # L08AC13222. Report review and editing was provided by Stephen Fernandes and Linda Vance. Field work coordination, landowner contacts and logistics were greatly expediated by Carl Lenz and Heidi Kaiser (Hydrometrics, Inc.). Fieldwork assistance provided by Stephen Fernandes (MTNHP) was invaluable. All photos in the report were taken by MTNHP personnel, unless otherwise noted Introduction The Otter Creek basin in southeast Montana is currently undergoing exploration for a large, open- pit coal mine. Baseline data on the condition of the ecosystem prior to coal mine development (pre-impact) is essential to determine what effects the coal extraction might have on the fish and wildlife in, and downstream of, the effected extraction area (post-impact). The initial mining rights transfer determined that no Federally Listed species under the ESA would be affected in the coal tracts area (BLM EA 2002). But this EA did not take into account the presence of Montana species of concern (SOC) or other ecologically sensitive native species assemblages. Despite numerous projects undertaken to document and monitor biological communities in the Powder and Tongue River watersheds in the wake of CBNG development (Confluence Consulting Inc. 2003, 2004; Stagliano 2006; Davis et al. 2009; Maxell 2009; Petersen et al. 2009; Senecal 2009, Stagliano 2011), large gaps still exist in basic baseline surveys for riparian macroinvertebrate, fish, and herpetofauna in the Otter Creek basin. Fish communities have been documented to be seasonally variable in prairie streams (Bramblett and Fausch 1991), thus sampling across all seasons is a good strategy to document baseline community differences. Many of the remaining data gaps involve small prairie streams that constitute the Northwestern Great Plains Intermittent Prairie Stream ecological system (Stagliano 2005). This stream system may have downstream connectivity early in the season for potential fish spawning and nursery areas (Smith and Hubert 1989, Bramblett 2000) or no fish colonization at all, but by summer often becomes a string of isolated pools that are important breeding and rearing areas for amphibians (Stagliano 2011). Identifying baseline communities and conditions in streams of the coal tracts area (i.e., the presence of fish, macroinvertebrate, and herpetofauna) prior to coal development is essential to understanding and potentially mitigating impacts to habitats and species during and after coal extraction. Methods Study Sites The aquatic community sampling sites chosen for pre-impact baseline data are representative of the range of streams found in the Otter Coal Tracts project area: Ephemeral, Intermittent and Perennial Prairie Streams. Three predetermined mainstem Otter Creek reaches (control, impact and downstream) and three tributaries coinciding with established surface water quality stations were visited seasonally (May, July, October) during 2011 (Table 1). Additionally, we added fish sampling visits to Otter Creek site #2 near the Threemile Creek confluence, since Threemile Creek remained "dry" during all visits (see Site Photos). Study site visits were timed with season and discharge (Figure 1); although we tried to coordinate field visits with baseflow water levels, 2011 was an Table 1. Otter Creek Coal Study Site GPS Locations at the top (T) and bottom (B) of the assessment reach. Aquatic Ecological System (AES) code described in text followed Stagliano (2005). Elev Site Code Site Name Type AES code Latitude Longitude (m) Comment Spring survey for fish Otter_23 Tenmile Creek (T) Control D005/E005 45.43409 -106.13403 961 above the road reach, dry during July, Oct. Otter_23 Tenmile Creek (B) Control D005/E005 45.43465 -106.13253 958 visits Top of reach began ~60m below Tenmile Otter_22 Otter Creek 22 (T) Control D005/C005 45.43035 -106.14428 951 Creek road, proceeded 300m Otter_22 Otter Creek 22 (B) Control D005/C005 45.43274 -106.14366 948 downstream Top of reach began Otter_16 Otter Creek 16 (T) Impact C005 45.48514 -106.16487 938 ~60m above the stream crossing and Otter_16 Otter Creek 16 (B) Impact C005 45.48365 -106.16725 937 went 240m below Spring survey Otter_3m Threemile Creek (T) Impact E005 45.51054 -106.16288 933 contained standing water in cattle hoof prints, dry during July, Otter_3m Threemile Creek (B) Impact E005 45.50955 -106.16960 928 Oct. visits Site added for fish Otter_2 Otter Creek 2 (T) Impact C005 45.50475 -106.17493 929 surveys during July visit (unseinable in Spring), downstream Otter_2 Otter Creek 2 (B) Impact C005 45.50561 -106.17561 928 of road crossing Top of reach began ~80m below the ranch OtterJT Otter Creek JT(T) Down C005 45.55675 -106.21798 910 road/impassable culvert, proceeded Otter JT Otter Creek JT(B) Down C005 45.55782 -106.21770 909 300m down Bottom of reach Otter_lA Home Creek (B) Down D005/E005 45.54483 -106.18717 952 began ~500m above the Otter Creek road crossing, proceeded Otter_lA Home Creek (T) Down D005/E005 45.54422 -106.18947 950 300m upstream extreme high water year (avgerage yearly discharge 14.9 vs. 3.96 cfs in 2010). Therefore, we had to take advantage of sampling conditions as they presented themselves, sometimes not during the most ideal sampling periods (e.g., May visit @ 19 cfs between two high water pulses, Figure 1). Summer and fall visits were closer to base flow at 11 and 7 cfs, respectively. Habitat assessments, herpotofauna, macroinvertebrate and fish surveys were performed during the same site visit. In total, we surveyed 15 reaches for fish during the visits: four mainstem Otter Creek reaches (11 surveys) and two tributary streams (4 surveys). Fifteen macroinvertebrate samples were collected during the visits; neither survey was conducted at Threemile Creek in any season due to lack of surface water present. Spring and fall macroinvertebrate samples were collected outside of the MTDEQ recommend sampling time frame (June l st -September 15 th ) (MTDEQ 2006), but this period was largely derived for mountain streams. All six a priori stream reaches were visually surveyed for amphibians or reptiles during all visits. Figure 1. Discharge reported at the USGS gage in Ashland, MT. Arrows indicate sampling dates. ^ 300.0 8 200. i H _l H c 19.0 Jan 2611 USGS ©6307749 Otter Creek at Ashland MT Har 2011 Hay 2011 Sep 2611 Mow 2011 Daily nean discharge • Flow at station affected by ice — Estinated daily nean discharge ^—Period of provisional data ^^ Period of approved data Riparian Assessments The assessment stream reach was divided into 10 equally spaced transects according to BLM and EMAP protocols (BLM 2008b; Lazorchak et al. 1998). The downstream transect was marked (GPS, flagging and photo point) as the bottom of the reach and all ecological assessment protocols started from this point and continued upstream for 300m (designated the assessment area or "AA") to the top of the reach which was also marked. Parameters recorded at each transect were: wetted width, three channel depth measurements, % large woody debris and riparian shading. On-site habitat assessments were conducted using the rapid assessment protocol developed for the EPA by Barbour et al. (1999) with modifications for the BLM by the National Aquatic Assessment Team (scores 0-24) (BLM 2008). The process for determining Proper Functioning Condition followed Pritchard et al. (1993). Basic water parameters (temperature, pH, conductivity) were recorded prior to biological sampling using a Horiba H-10 water monitor, calibrated for the higher conductivity range. The Livestock Use Index ("Cowpie" CPI) was assessed by walking a randomly chosen 75m reach on both sides of the stream channel in the riparian area within the assessment area and counting all the old and new cowpies (high CPI equals high cow usage). The goal of these evaluations is to characterize local reach geomorphology, riparian and in-stream habitat, and characteristics that influence aquatic community integrity. The sites ranking higher using these protocols are determined to have higher quality local-scale habitat. Habitat assessments were performed during the same visit as the biological sampling. Macroinvertebrate Communities Macroinvertebrate communities were sampled qualitatively from the 10 transects within the 300m assessment reach using the EMAP Reach-Wide protocol (Lazorchak et al. 1998). Sampling started at the downstream transect (A) or #10 in the BLM protocol, and proceeded upstream alternating sampling with the 500-micron D-frame net to the left, right or center of the stream channel, so a random sampling of all habitats is achieved (Figure 2). Ten multi-habitat kicks were composited from the net into a 20 liter bucket. All organisms and organic matter in the bucket were elutriated from the inorganic portion Figure 2. Macroinvertebrate EMAP Sampling procedure . and washed onto a 500-micron sieve. The inorganic portion was washed and examined until no further organics or organisms were present and discarded. The organic portion on the sieve was transferred to one or two 1 liter Nalgene bottles (unless field sub- sampling was needed), labeled and preserved in 95% ethanol and brought to the MTNHP lab in Helena for processing (sorting, identification and data analysis) following protocols outlined by the BLM (2008a) and MTDEQ (2006). Macroinvertebrates were identified to the lowest taxonomic level (MTDEQ 2006), counted, imported into EDAS (Jessup 2006), and biological metrics were calculated from the data using the Montana Department of Environmental Quality's newest multimetric macroinvertebrate (MMI) protocols (Jessup et al. 2005, Feldman 2006). Metric results were scored using the MTDEQ bioassessment criteria and each sample categorized as nonimpaired or impaired according to threshold values (Table 2). The macroinvertebrate MMI score is based upon a series of metrics that measure attributes of benthic macroinvertebrate communities that are sensitive to condition changes in the stream (in the form of pollution or pollutants). The index score represents the condition of the macroinvertebrate community at the time the sample was collected within that past year. If the index score is below the impairment threshold, the individual metrics can be used to provide insight as to why the communities are different from the reference condition (Barbour et. al 1999, Jessup et. al. 2005). Table 2. Impairment determinations from the MTDEQ MMI anri n/F (RI\/PZir<;\ mnriolc Ecoregion RIVPACS MMI Impairment Determination Mountain > 0.8 or < 1.2 >63 Not impaired < 0.8 or > 1.2 <63 Impaired Low Valley > 0.8 or < 1.2 >48 Not impaired < 0.8 or > 1.2 <48 Impaired Eastern Plains > 0.8 or < 1.2 >37 Not impaired < 0.8 or > 1.2 <37 Impaired The impairment threshold set by MTDEQ is 37 for the Eastern Plains Stream Index, thus any scores above this threshold are considered unimpaired. Fish and Amphibian Surveys Fish sampling within the 300m stream channel assessment reach was conducted with 6 and 9 meter straight seines in 25-30m increments seining in a downstream direction toward the block seine (protocols outlined in Bramblett et al. 2005). Fish captured in a blocked section were transferred to holding buckets until the reach was completed, unless the reach was broken up by riffles, impassable or dry sections; in this case, fish were worked up and released within the section of capture. Fish holding in the buckets were identified to species (Holton and Johnson 2003), enumerated, examined for external anomalies (e.g. deformities, eroded fins, lesions, and tumors), and then released. At least 10% of the individuals of a species were measured for total length in millimeters (TL mm) to I obtain size structure data. Young-of-the- year fish less than 20 mm (TL) were noted on the field sheet (not included in the totals) and released. Baited minnow traps were deployed and fished overnight (12 hour minimum) during the spring visits when seining conditions were toughest. Voucher specimens were only taken in the case of uncertain field identifications, and were preserved in 10% buffered formalin. These will be deposited with the Montana State University Collections. Adult amphibians or reptiles seen while seining or walking the designated stream reach were counted and recorded even if they were not captured in the seine. Analysis of the sampled fish communities used Integrated Biotic Indices (IBI) designed for wadable prairie streams (Bramblett et. al 2005) and derived Observed/Expected (O/E) Fish Models (Stagliano 2011) to detect impairment in the biological integrity of the sites. The expected number of native fish species for a D005 classified reference stream is 2.5-3.75, while the number of fish for a C005 stream in 5.5-8.5 depending on watershed area, dividing the observed number of native fish species at a site by the expected number derives a percentage compared to reference 10 condition (>0.8 or 80% = unimpaired) (Table 1). The IBI originally proposed by Karr (1981) involved the calculation of a series of 12 metrics evaluating different attributes of the fish community (i.e. species richness, tolerance to pollutants, trophic status) (Table 3). The 10 metrics used for the prairie streams were adjusted for watershed area to calculate an overall score between and 100. Bramblett et al. (2005) did not propose threshold criteria for good, fair, and poor biological integrity for these scores. Therefore, we applied percentiles above the null criteria (no fish present score) at >30% indicates good to excellent biological integrity, 10-30% fair/good biological integrity, 0-10% indicated poor to fair biological integrity and scores below the null are indicative of poor biological integrity or severely impaired. Table 3. Fish metrics and classification of fishes captured during the Otter Creek Study (2011). Species Scientific Name Trophic Feeding Repro General Origin Total * Habitatt Guild* Tolerance ** tt Length 3 years Catostomidae White sucker Catostomus commersoni OM BE LO TOL N 229 Cyprinidae Common Carp Cyprinus carpio OM BE TOL 1 381 Brassy minnow Hybognathus hankinsoni HB BE MOD N 81 Fathead Minnow Pimephales promelas OM GE TOL§ TOL N 76 Lake Chub Couesius plumbeus OM GE MOD N 140 Sand Shiner Notropis stramineus OM GE LO MOD N 61 Centrarchidae Green Sunfish Lepomis cyanellus IC GE TOL§ TOL 1 102 Pumpkinseed Lepomis gibbosus IC GE TOL§ MOD 1 89 Ictaluridae Black Bullhead Ameiurus melas IC BE TOL§ TOL 1 152 *HB = herbivore (> 90% plants or detritus); IC = invertivore/carnivore (>25% both invertebrates and vertebrates); IN = invertivore; OM = omnivore(25-90% plants or detritus) f BE = benthic; GE = generalist; WC = water column: Brown (1971); Scott and Crossman (1973); Becker (1983) % LO=Litho-obligate Reproductive Guild; Scott and Crossman (1973); Pflieger (1997); Barbour et al. (1999) § Tolerant reproductive strategists are not litho-obligates, use parental care at spawning site: Scott and Crossman (1973); Pflieger (1997) ** INT = intolerant; MOD = moderately tolerant: TOL = tolerant; Barbour et al. (1999); ft N = native; I - introduced; Brown (1971); Holton and Johnson (2003) Results We evaluated seven stream reaches in the study area: four Otter Creek mainstem sites were classified as Great Plains Perennial Prairie Stream types (C005), and the three tributaries were the Great Plains Intermittent Prairie Stream systems (D005, E005) (Stagliano 2005). Proper 11 classification is important when determining biological integrity (Hawkins and Norris 2000) and expected species richness. The Intermittent Prairie Stream (E005) in Montana is naturally fishless about 80% of the time; therefore, absence of fish, in itself, should not be viewed as a biological impairment (Threemile and Tenmile Creek). Likewise, stream reaches of Otter Creek have become dry in previous years (Stagliano, personal observation 2005-2008), thus placing this stream section within the D005 classification. We have identified and characterized reference condition indicator assemblages for these ecosystem types previously (Stagliano 2005), and used these here to compare to our site-specific observed species. Habitat Evaluations. Of the seven sampling reaches evaluated within the study area, we found three in Proper Functioning Condition (PFC) with a stable trend and four were Functional at Risk (FAR) (Table 7). Reasons that sites ranked FAR were due to structural habitat alteration by cattle with associated high CPI values (Home Creek {Otter_lA}, Threemile Creek {Otter_3m} and Otter #16-fall) (Figure 4) or anthropogenic stream manipulation (Otter Creek JT-Trussler and Otter Creek #16). Highest site integrity scores using both the BLM Habitat and PFC Assessment methods were recorded at the Tenmile Creek (Otter_23) and Otter Creek #22 (control-Denson reach) (Table 7). Point conductivity measurements recorded at all Otter Creek mainstem sites and tributaries across all seasons were above the threshold for impairment levels (>500ps, DEQ 2006), and Home Creek site had visible signs of natural gas seepage from the sediments. Macroinvertebrate Communities: Overall, 104 unique macroinvertebrate taxa were reported from the 15 macroinvertebrate assessment samples (Appendix B). One known Montana species of concern (SOC), the mayfly, Caenis youngi was collected in fair numbers at the control site, Otter Creek #22. Stoneflies (P) were not present at any sites, so the EPT taxa per site usually consisted of two species of tolerant mayflies (E), and one or two species of caddisfly (T); the highest EPT richness at any site was six species at the Otter JT site spring (Table 4, Appendix B). Average 12 macroinvertebrate taxa richness per site was 29.5 and the highest taxa richness was 41 taxa reported at the Otter Creek JT site. Using the MTDEQ multimetric index (MMI), four of the five sites (12 of 15 samples) were ranked non-impaired (good to excellent biological integrity), while all three samples from Tenmile Creek were ranked marginally impaired (Figure 5). Table 4. Macroinvertebrate sample characteristics and various metrics used for the DEQ MMI. Plains Total Ind. Total Total MMI EPT % Non- Site ID CollDate ID'ed Sample Taxa Index Taxa %EPT HBI Insect Otter_23 5/16/2011 605 4,840 15 36.8 1 0.3 7.7 6.9 Otter_23s 7/17/2011 515 2,060 12 35.2 0.0 7.3 15.2 Otter_23f 10/15/2011 617 2,468 13 35.6 0.0 7.1 15.9 OTTER_22 5/17/2011 628 5,024 24 46.3 3 2.7 7.4 56.2 OTTER_22s 7/14/2011 879 10,548 35 54.3 3 36.2 7.9 40.7 OTTER_22f 10/14/2011 725 5,800 35 65.9 3 22.3 7.7 43.5 OTTER_16 5/17/2011 564 4,512 34 44.4 3 3.2 7.5 55.3 OTTER_16s 7/16/2011 686 8,232 35 51.2 2 25.4 7.7 43.0 OTTER_16f 10/15/2011 603 4,824 35 46.2 4 22.9 7.2 50.3 OTTERJT 5/17/2011 610 2,440 34 64.6 6 17.4 7.4 30.2 OTTERJTs 7/18/2011 605 3,630 28 65.8 5 18.2 7.5 37.7 OTTERJTf 10/15/2011 606 2,424 41 43.4 4 7.6 6.6 10.9 0TTER_1A 5/16/2011 639 5,112 35 57.4 2 2.5 6.8 17.8 OTTER_lAs 7/17/2011 626 5,008 32 43.2 1 0.2 7.2 11.8 OTTER_lAf 10/14/2011 617 4,936 35 37.9 1 0.3 7.0 21.7 Figure 5. DEQ MMI scores across sites and seasons. Line is the impairment threshold at 37. 70 SO SS S ^ ■ Spring □ Summer □ Fall Tenmile Otter_22 Otter_16 OtterJT HomeCr. Downstream > 13 Stream sites that maintained flowing, connected water scored higher with the MMI than sites with interrupted pool areas. Overall, mainstem sites evaluated in the Otter Creek study received significantly higher macroinvertebrate MMI scores than those in the tributaries (T-test, p <0.01). MMI's did not significantly differ on Otter Creek mainstem Pre-lmpact Control, Impact or Downstream Sites (T-test, p >0.05), despite the fish communities reflecting a decrease in biotic integrity. No site had consistently high macroinvertebrate scores across all seasons. Figure 6. Average MMI scores by site type, (a) = no baseline differences between treatments. Otter Creek mainstem sites 2011 70 c O 30 i_ u a S 20 10 (a) W (a) 57.9 55.5 T 47.3 i i Control Impact Downstream Fish Communities. Overall, nine fish species (five native/four introduced) were identified from 37,679 individuals collected from 15 site visits (Table 5). One potential species of concern (PSOC), the brassy minnow, was collected at five of six sites and in very high numbers at the control site, Otter 22. Dominance by brassy minnows and lake chubs at the upstream sites shifts to a tolerant fathead minnow and sand shiner dominated community at the downstream site, Otter JT (Table 5). Otter Creek site 16 was heavily infested by introduced species during the summer and fall visits representing 41% and 93% of the fish sampled, respectively. Average number of fish species collected at Otter Creek mainstem sites across all seasons was 6.5 (± 0.8 SE), while the tributary sites averaged 1.75 species (Table 5). All fish presence sites also reported at least one species of amphibian (Table 6). Lake chubs had the highest site occupancy rate at 93% (14 of 15 visits) followed by fathead and brassy minnows at 80% (12 of 15 visits), but fathead minnows contributed to the highest proportion of individuals (56%, see Table 5), followed by lake 14 chubs at 17%. The most diverse site in the study was Otter Creek JT site with nine species, while the most intact sites were Otter Creek Site #22 at Denson's (four native spp.) and the Tenmile Creek spring survey (two native spp.). Table 5. Fish abundance, IBI"s and O/E results for the 6 sites reporting fish, ns =not seined during visit (dry) or us =unseinable conditions. Underlined values are fish communities ranked biologically unimpaired. OTTER_ 23 OTTER_ 22 OTTER_ .16 OTTER. .2 OTTER_ JT OTTER_ 1A May Jul Oct May Jul Oct May Jul Oct May Jul Oct May Jul Oct May Jul Oct Total Black Bullhead* ns ns 5 3 186 us 6 7 13 3 5 228 Brassy Minnow 25 ns ns 81 2400 588 104 549 51 us 154 364 143 30 65 4554 Common Carp* ns ns 4 432 1314 us 14 49 20 213 138 2184 Fathead Minnow ns ns 3 40 3 12 us 56 35 2008 2995 17810 10 5 1 22978 Lake Chub 12 ns ns 153 2785 1059 16 84 9 us 105 84 5 8 103 22 12 4457 Green Sunfish* ns ns 3 6 us 3 3 3 18 Pumpkinseed* ns ns 3 3 30 us 7 10 53 Sand Shiner ns ns us 22 38 25 2755 2840 White Sucker ns ns 6 140 15 20 1 45 us 8 7 28 108 10 388 Total # species 2 3 4 7 5 6 8 8 6 8 8 9 2 2 1 9 Native Species 2 3 3 4 4 4 4 5 4 5 5 5 2 2 1 5 Total Individuals 37 240 5330 1674 184 1072 1653 372 546 2257 3384 20899 32 17 1 37,698 IBI 75.6 66.1 67.3 69.2 62.3 50.7 44.4 62.7 63.1 44.9 43.8 48.9 70.8 70.8 46.5 O/E 0.82 0.80 0.80 1.07 0.73 0.73 0.73 0.91 0.73 0.59 0.59 0.59 0.82 0.82 0.41 O/E % 81.6 80.0 80.0 106.7 72.7 72.7 72.7 90.9 72.7 58.8 58.8 58.8 81.6 81.6 40.8 Using the Prairie Fish IBI, 9 of the 15 fish sites were ranked non-impaired (good biological integrity), two were slightly impaired (moderate integrity), three were moderately impaired (poor biotic integrity) and the Home Creek fall survey was ranked severely impaired (Figure 7). The O/E models tracked the IBI scores quite well at most sites (13 of 15), except two where the O/E ranked the site as slightly impaired and the IBI ranked the sites as non-impaired (Figure 8). Figure 7. Fish IBI scores across sites and seasons. Line is the null IBI threshold (fish absent). so 70 60 . Fish IBI's decreased going downstream on Otter Creek, and the Pre-lmpact Control Site scored significantly higher than either the Impact or Downstream sites (T-test, p =0.05) (Figure 9). Within the Great Plains Intermittent Prairie Streams, the species expected to occur most often at reference condition sites are the fathead minnow and brook stickleback, but the upper Tongue and Powder River Basins are slightly out of the range of this latter species; therefore, we did not collect them or expect them to occur at these sites and revised the Observed /Expected (O/E) Fish Model predictors accordingly. The O/E scores agreed with the IBI scores in most cases, except for the Otter Creek stream reaches that could be classified as Perennial vs. Intermittent Prairie Streams (C005) (Table 5). Figure 8. Fish IBI compared to O/E model (%) scores across sites and seasons. 100 80 O oS 60 4,000 >4,000 3660 3790 3590 >4,000 3920 3790 170 na na na 3890 3700 3860 3608 3540 1950 >4,000 3209 pH 8.2 8.4 8.4 8.2 8.2 8.1 7.6 7.8 8 7.8 na na na 8.2 8.1 8.1 8.1 8.1 8.2 8.2 8.2 PFC FAR FAR FAR PFC PFC PFC PFC PFC FAR FAR FAR FAR na FAR FAR PFC PFC PFC FAR FAR FAR BLM HBI 16 17 17 22 22 22 19 19 18 17 18 18 na 17 17 19 19 19 11 12 13 Avg wetted width (m) 6.2 0.6 0.6 8.7 8.2 8.6 11.9 12.1 11.2 1.8 na na na 4.2 2.8 5.0 4.8 4.6 2.4 2.1 2.0 Avg Left CHD (cm) 25 2 2 71 67 71 103 104 93 2 na na na 30 22 53 43 36 22 28 31 Avg Center CHD (cm) 32 3 3 76 97 76 136 128 106 2 na na na 42 30 54 44 40 30 33 42 Avg Right CHD (cm) 27 2 2 70 66 70 95 92 86 2 na na na 25 20 55 41 36 18 25 29 % Fines in Reach 98 100 100 92 84 92 97 97 89 100 100 100 na 50 60 70 67 63 100 100 98 % Gravel Reach 2 4 8 4 3 3 8 na 35 30 20 26 34 1 1 3 % Cobble Reach 5 9 5 1 1 5 na 15 10 11 8 3 Livestock Use (CPI) 30 11 9 55 36 13 18 7 46 35 44 24 na 12 5 18 22 19 55 36 28 Avg. Riparian Shade 5 10 5 10 10 10 40 20 na 10 10 10 20 10 10 30 20