Southeastern Naturalist 275 J.W. Johansen, H.T. Mattingly, and M.D. Padgett 22001166 SOUTHEASTERN NATURALIST 1V5o(2l.) :1257,5 N–2o9. 02 Population Densities of Two Rare Crayfishes, Cambarus obeyensis and Cambarus pristinus, on the Cumberland Plateau in Tennessee John W. Johansen1,2,*, Hayden T. Mattingly1,3, and Matthew D. Padgett3 Abstract - Cambarus obeyensis (Obey Crayfish) and Cambarus pristinus (Pristine Crayfish) are species of conservation concern, but basic information needed by conservation managers is lacking. To provide a quantitative measure of abundance, we conducted a mark–recapture study at six 100-m reaches per species during May–August 2013. These sites were a subset that we selected from eighty-nine 100-m reaches surveyed during 2011–2013. We built regression models to predict crayfish abundance based on single-pass capture rates for the 12 mark–recapture sites and for all occupied sites identified during the 2011–2013 surveys. We also calculated site-level density and capture efficiency for each species. Cambarus pristinus occurred at significantly lower densities across a larger range than C. obeyensis. Capture efficiency for both species varied across sites, suggesting that monitoring programs should incorporate regular, quantitative estimates of density and capture efficiency. Our results indicate that both species merit ongoing conservation attention and that C. pristinus may represent a higher conservation priority than previously recognized. Introduction North American crayfishes are a highly imperiled aquatic group with extinction rates predicted to increase in the future (Ricciardi and Rasmussen 1999, Taylor et al. 2007). Threats to freshwater crayfishes are generally the same as those affecting other freshwater organisms (e.g., climate change, overexploitation, degraded water quality, flow modification, habitat destruction or degradation, invasive species; Aldous et al. 2011, Dudgeon et al. 2006). An additional cause of imperilment is the limited geographic range of many crayfish species (Taylor et al. 2007). Agencies tasked with crayfish conservation require effective management strategies for the allocation of limited resources. Unfortunately, like many aquatic invertebrates, basic biological data to develop these strategies are lacking for many crayfish species (Cardoso et al. 2011, Welsh et al. 2010). Conservation assessments are valuable tools for identifying relative extinction risk and prioritizing species for management. However, data necessary to perform quantitative assessments are often lacking, so distributional patterns, like range size or site occupancy, are commonly used as proxies to assess imperilment (Mace et al. 2008, Masters 1991). These data have been used to assemble lists of potential species of conservation concern, but do not provide information on current population 1School of Environmental Studies, Box 5152, Tennessee Technological University, Cookeville, TN 38505. 2Current address - Department of Biology, Box 4718, Austin Peay State University, Clarksville, TN 37040. 3Department of Biology, Box 5063, Tennessee Technological University, Cookeville, TN 38505. *Corresponding author - johansenj@apsu.edu. Manuscript Editor: Lance Williams Southeastern Naturalist J.W. Johansen, H.T. Mattingly, and M.D. Padgett 2016 Vol. 15, No. 2 276 trends, threats to local populations, or options for management needed to develop effective conservation plans. For example, naturally rare species may have adaptations allowing these populations to remain viable in the absence of external disturbances (Gaston 1997), thereby requiring different strategies than a more abundant species shown to be in decline. Also, extinction processes operate at multiple scales, so finerscale data may be needed to identify local processes and threats to extinction (Hartley and Kunin 2003). The latter is particularly true of many crayfish species because their imperilment status is based primarily on limited range size. Local abundance is a finer-scale population measure that can be useful in predicting the extinction risk for a species (Mace and Kershaw 1997, O’Grady et al. 2004, Purvis et al. 2000, Rabinowitz et al. 1986). Many crayfish conservation assessments do not report measures of local abundance, or at most report qualitative or semi-quantitative abundance data (e.g., counts, catch-per-unit effort) that are collected during larger survey efforts to document regional fauna or delimit distributions of target crayfish species (Kilian et al. 2010; Rohrbach and Withers 2006; Wagner et al. 2010; Williams et al. 2004, 2006; Withers and McCoy 2005). These data lack measures of variation in sampling efficiency and detection. Such variation can lead to inaccurate assessments of abundance and an inability to effectively identify short-term population trends in the species of interest (Link and Nichols 1994, Nowicki et al. 2008). Although more expensive and time-consuming, mark–recapture methods can provide quantitative abundance estimates and provide a measure of sampling variation (Nowicki et al. 2008). Mark–recapture population estimates can be used in combination with semi-quantitative abundance estimates to efficiently and effectively assess local populations across a species’ wider geographic range (e.g., Black et al. 2013). Cambarus obeyensis Hobbs and Shoup (Obey Crayfish) and Cambarus pristinus Hobbs (Pristine Crayfish) occupy narrow ranges along the western margin of the Cumberland Plateau in Tennessee. Both species are considered to be vulnerable to extinction, and are the focus of current conservation actions (Center for Biological Diversity 2010). Recent survey efforts have delimited the extent of their range and provided valuable qualitative and semi-quantitative assessments of abundance (Rohrbach and Withers 2006, Williams et al. 2006, Withers and McCoy 2005). In response to increased conservation concern, we conducted surveys across the range of these species to (1) generate species-specific population-estimation models using a multi-technique sampling protocol, (2) assess the capture efficiency of the multi-technique sampling protocol, and (3) use the population-estimation models to assess the abundance patterns across the range of these species. Study Area and Target Species The Cumberland Plateau, hereafter referred to as the Plateau, is an important area for crayfish diversity, with numerous endemic and taxonomically unique species (Bouchard 1976a, b; Crandall and Buhay 2008). In Tennessee, steep escarpments separate the Plateau from the Interior Highlands to the west and Ridge and Valley to the east; the Sequatchie Valley divides the southern half of the Plateau. Southeastern Naturalist 277 J.W. Johansen, H.T. Mattingly, and M.D. Padgett 2016 Vol. 15, No. 2 Pennsylvanian-aged rocks and acidic soils of low-to-moderate fertility characterize this region with lotic systems of low productivity (Bouchard 1976a, Smalley 1982). Although regional aquatic diversity is high, local (fine scale) diversity and abundance tend to be low compared to neighboring physiographic provinces (Bouchard 1976a). Historically, only a small percentage of the Plateau has been affected by anthropogenic activities, but increasing human population, land-use change, climate change, and establishment of invasive species represent serious threats to the unique fauna of this region (Cordell and Macie 2002, Dale et al. 2009). Cambarus obeyensis is reported to occupy headwater streams in the East Fork Obey River watershed (Williams et al. 2006). However, the species is currently believed to be restricted to the Hurricane Creek system, and collections of C. obeyensis from Dripping Springs Creek and other localities outside the Hurricane Creek sub-watershed are considered erroneous (D. Withers, Tennessee Department of Environment and Conservation [TDEC], Nashville, TN, pers. comm.). Cambarus obeyensis may be particularly susceptible to stochastic environmental events. It was reported that 50% of the population was lost following an extreme drought event in 2008 (Cordeiro and Thoma 2010a), but no data were presented to verify this putative decline. Cambarus obeyensis is currently classified as highly vulnerable to extinction in several conservation classification-schemes (Corderio and Thoma 2010a, Natureserve 2014, Taylor et al. 2007), and the US Fish and Wildlife Service has been petitioned to federally list the species (Center for Biological Diversity 2010). Although it currently occupies a larger geographic range than C. obeyensis, 2 distinct forms of C. pristinus have been identified—the Caney Fork form and the Sequatchie form (Williams et al. 2009). Future taxonomic work may ultimately yield 2 distinct species; thus, our study focused on the Caney Fork form representing the type specimen. The Caney Fork form is thought to be restricted to headwater streams in the Bee Creek and Upper Caney Fork watersheds. Cambarus pristinus is uncommon across its range and is absent from several sites that appear to have suitable habitat (Bouchard 1976b, Hobbs 1965, Rohrbach and Withers 2006). Although C. pristinus is considered threatened and was also included in the federal listing petition (Center for Biological Diversity 2010, NatureServe 2014, Taylor et al. 2007), the IUCN Redlist requires more data to accurately assess imperilment of this species (Cordeiro and Thoma 2010b). Methods We surveyed eighty-nine, 100-m reaches for the target species from May through September in 2011, 2012, and 2013 within the East Fork Obey River, West Fork Obey River, Upper Caney Fork, and Bee Creek watersheds (Fig. 1). We selected the first 80 reaches based on proximity to available access points (e.g., bridge crossings, trail crossings, proximity to road) using an equal-stratified sampling design. Reaches were equally distributed across 1st- through 4th-order streams in each watershed. We superimposed a grid over a map of the study area and used a random-number generator to select our sites. If a cell contained multiple reaches, we used a random-number generator to select the reach based on stream order. If no Southeastern Naturalist J.W. Johansen, H.T. Mattingly, and M.D. Padgett 2016 Vol. 15, No. 2 278 access point existed in a cell, we selected surrounding cells at random until we encountered an alternative reach. During site visits, if we determined that a reach was inaccessible, we replaced it with the nearest alternative site of equal stream order. To increase the proportion of occupied sites for each species for our mark–recapture experiments (explained below), we selected 9 additional reaches in 2013. At each reach, we isolated a 100-m length of stream by placing block nets at the upstream and downstream ends of the reach. A 2-person crew sampled all available aquatic habitats during daylight hours using a combination of visual searches, dip nets, and seines. We used these techniques because they were applicable to the variety of flows and habitat types encountered within each stream and across the study area. We made collections only during periods when the turbidity was low (assessed qualitatively) and we could easily see the stream bottom. The proportion of time we devoted to each collection technique varied from reach-to-reach and was contingent on site-specific characteristics. Visual searches were the primary mode of capture and consisted of disturbing available cover (e.g., cobbles, boulders, or large woody debris) and hand-capturing any exposed crayfish. Figure 1. Locations of survey reaches on the Cumberland Plateau in Tennessee with HUC 12 watersheds delineated for (A) Cambarus obeyensis in the East Fork Obey (dark grey) and West Fork Obey (light grey) rivers and (B) Cambarus pristinus in Bee Creek (dark grey) and the upper Caney Fork (light grey) during May–September of 2011–2013. For C. obeyensis, a black diamond indicates an occupied reach, and a white diamond indicates a mark-recapture reach. For C. pristinus, a black triangle indicates an occupied reach, and a white triangle indicates a mark-recapture reach. For both species, a black cross indicates a failure to detect the species, and a white cross indicates a failure to detect the species at a historical locality. Southeastern Naturalist 279 J.W. Johansen, H.T. Mattingly, and M.D. Padgett 2016 Vol. 15, No. 2 We employed visual searches in shallow areas (<0.5 m) throughout the reach, and effort ranged from 1 to 4 h depending on the amount of appropriate habitat and the number of crayfish encountered. We sampled undercut banks, root mats, leaf packs, and depositional areas using 0.5-mm-mesh triangular aquatic dip-nets (30 x 30 x 30 cm). Dip-netting consisted of 2–3 jabs per habitat patch (average = ~24 jabs per reach; range = 0–60 jabs per reach). To perform a jab, we thrust the dip net into the lower portion of the habitat patch, pulled it upward as it was removed, and examining the collected debris for crayfish. We used a 1.0 m x 1.5 m seine (3.2-mm mesh) to sample deep-flowing habitats (>0.5 m). One person held the seine while the second disturbed a 2 m x 1 m area immediately upstream, flushing crayfish into the net. We conducted ~2–3 seine sets per 10 m of appropriate habitat (average = ~10 seine sets per reach; range = 0–25 seine sets per reach). We kept all collected crayfish in aerated buckets until sampling was complete. Following collection efforts, we identified, sexed, and enumerated all crayfish. We measured carapace length (CL; tip of rostrum to posterior edge of carapace) for all collected individuals of the target species; the smallest C. obeyensis encountered during these survey efforts was 7.5 mm CL and the smallest C. pristinus encountered was 9.9 mm CL. We used this measurement to set a minimum-size limit for each species for the mark–recapture study (detailed below). We preserved voucher specimens in 70% ethanol for verification in the lab and redistributed the remaining crayfishes evenly across the reach. We measured channel wetted width perpendicular to stream flow at the upper end, middle, and downstream end of the reach to estimate wetted surface area, and thalweg depth every 10 m along the thalweg to calculate mean thalweg depth. We made Petersen mark–recapture population estimates at a subset of reaches during the summer of 2013. For each species, we selected 6 reaches covering its geographical extent and range of occupied stream sizes as determined by link magnitude (i.e., the number of 1st-order tributaries upstream of a study reach). The reaches selected for mark–recapture of C. obeyensis were 2nd- to 4thorder streams with link magnitudes of 2–47 (Fig. 1, Table 1). These reaches had mean wetted channel widths of 1.9–7.9 m and mean thalweg depths of 9–37 cm (Table 1). The reaches selected for mark–recapture of C. pristinus were 2nd- and 3rd-order streams with link magnitudes of 2–31 (Fig. 1, Table 1). These reaches had mean wetted channel widths of 2.9–8.1 m and mean thalweg depths of 9–55 cm (Table 1). We conducted sampling at mark–recapture reaches in the same manner as single-pass surveys with the following differences. We sampled each reach on 2 consecutive days, with the second pass beginning ~16–18 h following the completion of the first pass. On the first day, we counted target crayfish larger than the minimum size limit set during previous surveys, marked individuals by using scissors to remove ~25% of the outermost right uropod, and redistributed the crayfish evenly across the reach. We left block nets in place overnight and followed the same protocol to make a second collection, during which we counted, sexed, identified as marked or unmarked, and released all tar get crayfish. Southeastern Naturalist J.W. Johansen, H.T. Mattingly, and M.D. Padgett 2016 Vol. 15, No. 2 280 We made population estimates using the Chapman modification of the Petersen index with confidence intervals (95%) estimated using a binomial distribution (Krebs 1989). We calculated population densities (crayfish per 100 m2) by dividing the population estimate for each reach by the wetted surface area (m2) of the reach and multiplying that value by 100. We used t-tests to compare abundance estimates and densities of the target species. We determined single-pass capture efficiency for the mark–recapture portion of the study by dividing single-pass capture rates by population estimates. We performed Spearman rank-order correlation analysis to determine if capture efficiency was related to stream size (link magnitude, mean wetted channel width, and mean thalweg depth) or density of target species. Capture efficiencies for C. obeyensis and C. pristinus were not significantly different during this portion of the study (t = 1.311, P = 0.23); thus, we combined the data from both species to examine correlations. To predict population estimates based on single-pass capture rates, we used the mark–recapture data to construct species-specific models by regressing log10 (population estimate) onto log10 (single-pass capture rate) (e.g., Black et al. 2013). Collections made from 2011 and 2012 in our study, or collections that fall outside the range of single-pass capture rates used to construct the model (11–134 individuals for C. obeyensis; 6–56 individuals for C. pristinus), should be interpreted cautiously because they fall outside the range of parameters used to construct the regression models. However, to provide an initial assessment of variation across the range of each species, we used the species-specific regression models to estimate abundance and density at all sites where we detected C. obeyensis and C. pristinus. Finally, we employed t-tests to compare predicted abundance estimates and densities of these Table 1. Characteristics of twelve 100-m sampling reaches where we used Petersen mark–recapture methods to quantify Cambarus obeyensis (Obey Crayfish) and Cambarus pristinus (Pristine Crayfish) population densities during summer 2013 on the Cumberland Plateau of Tennessee. Reaches are ranked first by stream order and then by link magnitude for each species. Mean channel Mean Stream order wetted width thalweg Stream reach County (link magnitude) (m) depth (cm) Cambarus obeyensis sites Hurricane Creek 3 Overton 4 (47) 7.90 34 Hurricane Creek 1 Putnam 3 (17) 6.10 37 Little Hurricane Creek 2 Overton 3 (13) 6.17 35 Little Hurricane Creek 1 Overton 3 (7) 3.65 11 Piney Creek 1 Putnam 2 (6) 3.76 23 Little Piney Creek 3 Overton 2 (2) 1.93 9 Cambarus pristinus sites Caney Fork 1 Cumberland 3 (31) 8.07 55 West Fork 2 Cumberland 3 (20) 6.53 50 Pokepatch Creek 2 Cumberland 3 (9) 4.88 29 West Fork Little Cane Creek 1 Cumberland 3 (7) 3.37 21 Oldfield Branch 1 Cumberland 3 (4) 3.40 9 Whiteoak Creek 1 Cumberland 2 (2) 2.88 11 Southeastern Naturalist 281 J.W. Johansen, H.T. Mattingly, and M.D. Padgett 2016 Vol. 15, No. 2 species. All statistical tests were performed in JMP Version 10 (SAS Institute, Inc. 2014), and we set α = 0.05. Results We detected C. obeyensis in 9 of 46 reaches in the East and West Fork Obey River watersheds (Fig. 1, Table 2). All occupied reaches were located in the Hurricane Creek sub-watershed and included 2 new localities in the Little Piney Creek drainage. Single-pass capture rates ranged from 2 to 134 crayfish per 100 m (Table 2). We detected C. obeyensis in 1st- to 4th-order streams with a wide range of link magnitudes (1 to 47; Table 2). Occupied reaches had a mean (± SE) wetted channel width of 5.3 ± 0.8 m (range = 1.3–10.3 m) and mean thalweg depth of 22 ± 3.9 cm (range = 5–39 cm) (Table 2). We documented C. pristinus in 13 of 43 reaches surveyed in the Bee Creek and Upper Caney Fork watersheds, including new localities in Little Cane Creek, East Fork Little Cane Creek, and Pokepatch Creek (Fig. 1, Table 3). We did not detect C. pristinus at 4 historic localities: Spring Creek 1, Laurel Creek 2, Henderson Branch 1, and Caney Fork 2. Single-pass capture rates ranged from 1 to 56 crayfish per 100 m (Table 3). Reaches where we detected C. pristinus were located in 2nd- to 4th-order streams with link magnitudes ranging from 2 to 39. Occupied sites had a mean (± SE) wetted channel width of 5.1 ± 0.6 m (range = 2.2–9.1 m) and mean thalweg depth of 22 ± 4.0 cm (range = 4–55 cm) (Table 3). Table 2. Characteristics of nine 100-m sampling reaches in the Hurricane Creek watershed where Cambarus obeyensis was detected during May–September of 2011–2013. Reaches are listed from largest to smallest based on link magnitude. Reach ranks are provided for single-pass capture rates, abundance (crayfish/100 m) was predicted by the species-specific regression model, and density (crayfish/100 m2) was predicted by the species-specific regression model. Mean Stream channel Mean Abundance order wetted thalweg Single- Predicted Density (link width depth pass estimate estimate Stream reach Date surveyed magnitude) (m) (cm) (rank) (rank) (rank) Hurricane Creek 3 23 July 2013B 4 (47) 7.90 34 46 (7) 202 (6) 26 (9) Hurricane Creek 2 20 July 2011 4 (44) 10.32 39 54 (5) 247 (5) 48 (6) Little Hurricane 5 Sept 2012 4 (33) 6.75 27 2 (11) 4 (11) 1 (11) Creek 3 Hurricane Creek 1 31 May 2011 3 (15) 6.10 14 131 (2) 764 (3) 140 (2) 6 June 2013B 3 (15) 5.46 37 89 (3) 467 (2) 77 (3) Little Hurricane 19 July 2012 3 (13) 4.91 9 39 (8) 163 (8) 33 (8) Creek 2 19 June 2013B 3 (13) 6.17 35 81 (4) 414 (4) 67 (5) Little Hurricane 14 August 2013B 3 (7) 3.65 11 134 (1) 786 (1) 194 (1) Creek 1 Piney Creek 1 31 July 2013B 2 (6) 3.76 23 43 (6) 185 (7) 43 (7) Little Piney Creek 3A 26 June 2013B 2 (2) 1.93 9 11 (10) 33 (10) 17 (10) Little Piney Creek 2A 27 June 2013 1 (1) 1.30 5 25 (9) 93 (9) 71 (4) APreviously undocumented locality. BPetersen mark–recapture event. Southeastern Naturalist J.W. Johansen, H.T. Mattingly, and M.D. Padgett 2016 Vol. 15, No. 2 282 For the reaches where we conducted mark–recapture sampling, Petersen mark– recapture population estimates for C. obeyensis ranged from 35 to 608 (mean ± SE = 358 ± 107) crayfish per 100 m, with a mean density estimate of 72 ± 24 (range = 18–167) crayfish per 100 m2 (Table 4). For the reaches where we conducted mark– recapture sampling, Petersen mark–recapture population estimates for C. pristinus ranged from 12–168 (mean ± SE = 79 ± 28) crayfish per 100 m with a mean density estimate of 18 ± 8 (range = 4–48) crayfish per 100 m2 (Table 4). Mean ± SE capture efficiency during the 2013 mark–recapture study was 24 ± 3.8% (range = 14–38%) for C. obeyensis and 35 ± 8.1% (range = 23–72%) for C. pristinus. This result indicates that during a single pass, our 2-person crew collected an average of 24% C. obeyensis individuals with CL ≥ 7.5 mm and 35% of C. pristinus individuals with CL ≥ 9.9 mm present in an occupied reach. Capture efficiencies were negatively correlated with all 3 characteristics of stream size measured mean wetted channel width (rs = -0.62; P = 0.03), mean thalweg depth (rs = -0.61; P = 0.04), and link magnitude (rs = -0.63; P = 0.03), and with local density of the target species (rs = -0.62; P = 0.03). The regression model for C. obeyensis, log10 yo = 0.1813 + 1.2731 (log10 xo), where yo is C. obeyensis population estimate (crayfish per 100 m) and xo is Table 3. Characteristics of thirteen 100-m sampling reaches in the Bee Creek and upper Caney Fork watersheds where we detected Cambrus pristinus during May–September of 2011–2013. Reaches are listed from largest to smallest based on link magnitude. Reach ranks are provided for single-pass capture rates; abundance (crayfish/100 m) was predicted by the species-specific regression model, and density (crayfish/100 m2) was predicted by the species specific regression model. Mean Stream channel Mean Abundance order wetted thalweg Single- Predicted Density (link width depth pass estimate estimate Stream reach Date surveyed magnitude) (m) (cm) (rank) (rank) (rank) Meadow Creek 2 8 June 2011 4 (39) 6.88 9 12 (6) 36 (6) 5 (8) Wilkerson Creek 2 19 May 2012 4 (36) 9.13 24 1 (15) 2 (15) 0.5 (16) Caney Fork 1 18 July 2012 3 (31) 8.30 42 4 (12) 10 (12) 1 (13) 17 July 2013B 3 (31) 8.07 55 12 (6) 36 (6) 7 (7) Laurel Creek 1 13 June 2012 4 (25) 5.15 26 2 (14) 5 (14) 1 (13) Little Cane Creek 2A 30 July 2012 4 (21) 7.30 12 10 (9) 29 (9) 4 (11) West Fork Creek 2 27 July 2011 3 (20) 6.11 33 11 (8) 33 (8) 5 (8) 30 May 2013B 3 (20) 6.53 50 16 (4) 50 (4) 8 (6) Maple Creek 1 28 June 2012 3 (9) 2.20 9 1 (15) 2 (15) 1 (13) Pokepatch Creek 2A 30 July 2013B 3 (9) 4.88 29 56 (1) 205 (1) 42 (3) West Fork Little 7 June 2011 3 (7) 2.25 9 42 (2) 148 (2) 44 (2) Cane Creek 1 11 June 2013B 3 (7) 3.37 21 38 (3) 132 (3) 59 (1) Pokepatch Creek 1 25 June 2012 3 (6) 2.39 33 14 (5) 43 (5) 18 (4) East Fork Little 11 July 2012 2 (5) 2.93 4 4 (12) 10 (12) 4 (11) Cane Creek 1A Oldfield Branch 1 4 June 2013B 3 (4) 3.40 9 6 (11) 16 (11) 5 (8) Whiteoak Creek 1 30 July 2013B 2 (2) 2.88 11 9 (10) 26 (10) 13 (5) APreviously undocumented locality. BPetersen mark–recapture event. Southeastern Naturalist 283 J.W. Johansen, H.T. Mattingly, and M.D. Padgett 2016 Vol. 15, No. 2 C. obeyensis single-pass capture rate, was significant (P = 0.002) and explained 93% of the variation in C. obeyensis population estimates (Fig. 2). The regression model for C. pristinus, log10 yp = 0.3258 + 1.1371 (log10 xp), where yp is C. pristinus population estimate (crayfish per 100 m) and xp is C. pristinus single-pass capture rate, was also significant (P = 0.009) and explained 85% of the variation in C. pristinus population estimates (Fig. 2). Across all sites, predicted abundances (t = -3.07; P = 0.01) and densities (t = -2.92; P = 0.01) for C. obeyensis were significantly higher than for C. pristinus. The mean population estimate predicted for C. obeyensis at occupied reaches during the 2011–2013 surveys was 305 ± 272 crayfish per 100 m (range = 4–786 crayfish per 100 m) and mean density was 65 ± 17 crayfish per 100 m2 (range = 1–194 crayfish per 100 m2) (Table 2). The mean population estimate predicted for C. pristinus at occupied reaches during the 2011–2013 surveys was 49 ± 59 crayfish per 100 m (range = 2–205 crayfish per 100 m) and mean density was 14 ± 18 crayfish per 100 m2 (range = 0.5–59 crayfish per 100 m2) (Table 3). Discussion A better understanding of local abundance patterns, life-history strategies, and habitat affinities is needed to assess the current status of most crayfish species and help inform conservation efforts (Welsh et al. 2010). We found that C. obeyensis occupied a very limited range and occurred only in the Hurricane Creek sub-watershed (Fig. 1). Within that range, the species occupied a high proportion of sites across a variety of stream sizes. These populations share a high degree of physical Table 4. Summary of Petersen mark–recapture population estimates conducted at twelve 100-m reaches within 10 streams during summer 2013, where M = number of crayfish captured and marked during the first day, C = total number of crayfish captured during the second day, R = number of marked crayfish recaptured on the second day, N = population estimate (number of crayfish per 100 m), and 95 % CI = confidence interval for N (lower–upper) . We obtained the density (crayfish per 100 m2) by dividing N by the wetted surface area of each reach and multiplying by 100. Stream reach M C R N 95% CI Density Cambarus obeyensis sites Hurricane Creek 3 46 57 11 226 139–511 28.6 Hurricane Creek 1 89 100 14 605 387–1271 99.2 Little Hurricane Creek 2 81 95 13 561 352–1013 91.0 Little Hurricane Creek 1 134 175 38 608 419–957 166.6 Piney Creek 1 43 43 16 113 90–159 30.0 Little Piney Creek 3 11 14 4 35 23–79 18.1 Cambarus pristinus sites Caney Fork 1 12 20 5 45 24–109 5.5 West Fork 2 16 25 6 62 38–200 9.5 Pokepatch Creek 2 56 91 30 168 122–255 34.5 West Fork Little Cane Creek 1 38 41 9 163 95–422 48.3 Oldefield Branch 1 6 9 2 22 10–300 6.6 Whiteoak Creek 1 9 11 8 12 9–26 4.3 Southeastern Naturalist J.W. Johansen, H.T. Mattingly, and M.D. Padgett 2016 Vol. 15, No. 2 284 connectivity, increasing the risk that a substantial portion of the population could be impacted in the event of a large-scale upstream disturbance (Hitt and Chambers 2014). However, the relatively high rate of occupancy and density throughout the watershed creates potential refugia from disturbances that could offer a natural source for recolonization, or potential stock for captive propagation and reintroduction (Sedell et al. 1990, Townsend 1989). While the fragmented distributional pattern of C. pristinus makes it unlikely that a single event would impact the entire species, this distributional pattern might increase the potential for localized extirpations (Fig. 1; Fagan et al. 2002). Density patterns observed in our study for C. obeyensis and C. pristinus were similar to those reported for other Cambarus species considered to be imperiled (i.e., threatened or endangered; Taylor et al. 2007) because of limited geographic ranges. For example, Cambarus scotti Hobbs (Chattooga River Crayfish; threatened), Cambarus unestami Hobbs and Hall (Blackbarred Crayfish; threatened), and Cambarus cracens Bouchard and Hobbs (Slenderclaw Crayfish; endangered) had mean density estimates of 16, 40, and 3 crayfish per 100 m2, respectively (Kilburn et al. 2014). Comparatively widespread, non-imperiled species (i.e., currently Figure 2. Regression models for Cambarus obeyensis (triangles; dashed line) and Cambarus pristinus (circles; solid line) constructed by regressing log10 (Petersen population estimate) onto log10 (single-pass capture rates). Each model was constructed using mark–recapture data collected from six 100-m reaches for each species during May–August 2013. Southeastern Naturalist 285 J.W. Johansen, H.T. Mattingly, and M.D. Padgett 2016 Vol. 15, No. 2 stable; Taylor et al. 2007) like Cambarus bartonii (Fabricius) (Common Crayfish) or Cambarus hubbsi Creaser (Hubbs’ Crayfish) have been reported to occur at densities exceeding 500 crayfish per 100 m2 (Flinders and Magoulick 2003, Griffith et al. 1996, Mitchell and Smock 1991). The lower densities observed for imperiled crayfish may reflect the natural rarity of these species and may represent a particular demographic pattern that, in the absence of external pressures, allows them to persist despite an overall increased risk of extinction (Kunin and Gaston 1993). Alternatively, C. cracens, C. obeyensis, C. pristinus, and C. unestami are endemic to the Plateau, and it is possible that this density pattern may be representative of the lower productivity characteristic of lotic systems on the Plateau (Bouchard 1976a). Low local abundance also may be attributed to occupancy of poor-quality habitats (Fagen 1988). Cambarus obeyensis generally occurred in densities greater than 20 crayfish per 100 m2 (Table 2). Two surveyed reaches in which crayfish densities fell below this level had obvious anthropogenic impacts (a large impoundment or active livestock and agriculture) with associated in-stream habitat degradation. We do not feel that these sites represent naturally low densities because comparably sized streams and nearby sites supported higher densities. For example, Little Piney Creek 3 had a density estimate 4 times lower than Little Piney Creek 2. We sampled both reaches during the same period and the sites were separated by less than 50 m of stream. Little Piney Creek 3 was directly downstream of a large impoundment, while Little Piney Creek 2 was a small tributary that avoided direct impacts from the impoundment. Another potential example can be inferred from collections at Hurricane Creek 1. We collected at this site in 2011 and then again in 2013. The single-pass capture rate, predicted abundance, and estimated density were lower in the 2013 survey, during which we noted an increase in sandy depositional material. Future studies should address the potential impact of in-stream habitat degradation on this species. Cambarus pristinus was generally less abundant than C. obeyensis, with ~85% (11 of 13 sites) of occupied reaches having single-pass counts of less than 20 C. pristinus per 100 m. Lower reach-scale abundance could be related to habitat specificity at the mesohabitat scale. Previous surveys suggested C. pristinus preferentially occupied pools with large, slab-shaped rocks (Rohrbach and Withers 2006, Williams et al 2004, Withers and McCoy 2005); we also observed a patchy distributional pattern, often finding every captured individual in a 10–20-m section of the 100-m reach. However, pool-riffle-scale habitat specificity does not entirely explain the consistently low densities observed across the range of this species. Alternative explanations for this pattern include source–sink population dynamics, strong ecological interactions with heterospecifics, or large-scale habitat degradation. Further study is needed to clarify the underlying causes of this distributional pattern and to better guide conservation efforts. Second-pass capture rates (C) were significantly higher than first-pass capture rates (M) (t = 3.09, P = 0.01; Table 4) in the mark–recapture portion of our study. Given this significant pattern, we considered the following potential ramifications to our study results. First, if unmarked individuals were migrating into the reach, then we overestimated the population size (Krebs 1989). The ability of crayfish Southeastern Naturalist J.W. Johansen, H.T. Mattingly, and M.D. Padgett 2016 Vol. 15, No. 2 286 to circumvent barriers was a concern of Larson et al. (2008), but this ability may be limited (Kerby et al. 2005), and there was no logical reason that immigration rates would consistently exceed emigration rates. Second, if uropod clips or handling crayfish affected the catchability of marked individuals, making them easier to capture on the second day, then we underestimated the population size (Krebs 1989). However, uropod clipping has been shown to be a suitable marking technique in other studies of crayfish abundance (Guan 1997, Larson et al. 2008, Rabeni et al. 1997). Finally, increased second-pass capture rates may have resulted from improved second-day capture efficiency by the survey crew, possibly related to (1) within-reach crayfish movement or exposure resulting from the disturbance of habitat during the first pass, or (2) an improved search image for the target species by sampling crew members. In both of these latter scenarios, there would likely be a proportional increase in the probability of capture of marked and unmarked individuals, resulting in no strong directional bias in our population estimates. Researchers can assume that not all individuals will be detected during monitoring surveys (MacKenzie et al. 2002, 2003). Our single-pass capture efficiencies were comparable to those reported in other aquatic surveys (e.g., Black et al. 2013, Davis et al. 2011) and indicate that population estimates of these species based solely on single-pass capture rates likely underestimate population size for these species. If this detection bias was equal across sampling sites and sampling periods, relative population trends could still be assessed. However, capture efficiency has been shown to vary even when standardized sampling protocols are used, limiting the usefulness of these data for monitoring programs (Link and Nichols 1994, Yoccoz et al. 2001). We observed site-specific variation in capture efficiency related to stream size and target-species density. We would also expect capture efficiency to vary between seasons, making single-pass capture rate ineffective for detecting population trends. Quantitative monitoring can account for this variation (Bailey et al. 2004; Link and Nichols 1994; MacKenzie et al. 2002, 2003) and as demonstrated here and elsewhere (e.g., Black et al. 2013), programs that mix qualitative surveys with quantitative abundance assessments can balance cost and effort of data collection while still providing the data necessary for making conservation decisions. Due to their limited geographic range, C. obeyensis and C. pristinus should continue to receive attention from regional conservation managers. Previous survey work identified C. obeyensis as a higher conservation priority because of its extremely limited range and a qualitative assessment of population decline (Corderio and Thoma 2010a, 2010b; Natureserve 2014; Taylor et al. 2007). However, conservation efforts on the Plateau in Tennessee should consider prioritizing C. pristinus based on its significantly lower densities, fragmented population, and apparent extirpation at 4 historic localities. For C. obeyensis, conservation efforts could focus on habitat protection and restoration, and the development of an adaptive management plan. For C. pristinus, identifying the underlying causes of low densities and local extirpation events should be addressed to help determine the direction of future conservation efforts. We recommend continued monitoring of population size and habitat conditions for both species. We also suggest that monitoring protocols Southeastern Naturalist 287 J.W. Johansen, H.T. Mattingly, and M.D. Padgett 2016 Vol. 15, No. 2 be designed to maximize the usefulness of the data for identifying population trends by mixing quantitative and semi-qualitative sampling designs. Finally, additional studies are needed for both species on other aspects of their biology such as life history, ecology, habitat affinities, and genetic variation. Acknowledgments Funding for this project was provided by the US Fish and Wildlife Service, Tennessee Wildlife Resources Agency, and The Nature Conservancy (Tennessee Chapter) as part of the Cumberland Habitat Conservation Plan. Additional support was provided by the Center for the Management, Utilization, and Protection of Water Resources and the Department of Biology at Tennessee Technological University, and the Center of Excellence for Field Biology at Austin Peay State University. We thank Johnathan Davis, J. Rufus Darden, Rebecca Johansen, Brianna Zuber, Christine Peterson, Meiko Camp, and Nika Cantrell for assistance in the field, and Bledsoe State Forest and private landowners who allowed us access to their property to conduct crayfish surveys. C.A. Brown, S.B. Cook, E.A. Hart, G.L. Norris, and 2 anonymous reviewers improved the manuscript through their editorial reviews and comments. Literature Cited Aldous, A., J. Fitzsimmons, B. Richter, and L. Bach. 2011. Droughts, floods, and freshwater ecosystems: Evaluating climate-change impacts and developing adaptation strategies. Marine and Freshwater Research 62:223–231. Bailey, L.L., T.R. Simons, and K.H. Pollack. 2004. Estimating site occupancy and speciesdetection probability parameters for terrestrial salamanders. Ecological Applications 14:692–702. Black, T.R., J.E. Detar, and H.T. Mattingly. 2013. Population densities of the threatened Blackside Dace, Chrosomus cumberlandensis, in Kentucky and Tennessee. Southeastern Naturalist 12 (Special Issue 4):6–26. Bouchard, R.W. 1976a. Geography and ecology of crayfishes of the Cumberland Plateau and Cumberland Mountains, Kentucky, Virginia, Tennessee, Georgia, and Alabama. Part I: The genera Procambarus and Orconectes. Pp. 563–584, In J.W. Avault Jr. (Ed.). Freshwater Crayfish. Louisiana State University Division of Continuing Education, Baton Rouge, LA. 676 pp. Bouchard, R.W. 1976b. Geography and ecology of crayfishes of the Cumberland Plateau and Cumberland Mountains, Kentucky, Virginia, Tennessee, Georgia, and Alabama. Part II: The genera Fallicambarus and Cambarus. Pp. 585–605, In J.W. Avault Jr. (Ed.). Freshwater Crayfish. Louisiana State University Division of Continuing Education, Baton Rouge, LA. 676 pp. Cardoso, P., T.L. Erwin, P.A. Borges, and T.R. New. 2011. The seven impediments in invertebrate conservation and how to overcome them. Biological Conservation 144:2647–2655. Center for Biological Diversity. 2010. Petition to list 404 aquatic, riparian, and wetland species from the southeastern United States as threatened or endangered under the Endangered Species Act. Available online at http://www.biologicaldiversity.org/programs/ biodiversity/1000_species/the_southeast_freshwater_extinction_crisis/pdfs/SE_Petition. pdf. Accessed 11 July 2014. Southeastern Naturalist J.W. Johansen, H.T. Mattingly, and M.D. Padgett 2016 Vol. 15, No. 2 288 Cordeiro, J., and R.F. Thoma. 2010a. Cambarus obeyensis. The IUCN Red List of Threatened Species. Version 2014.1. Available online at www.iucnredlist.org. Accessed on 11 July 2014. Cordeiro, J., and R.F. Thoma. 2010b. Cambarus pristinus. The IUCN Red List of Threatened Species. Version 2014.1. Available online at www.iucnredlist.org. Accessed on 11 July 2014. Cordell, H.K., and E.A. Macie. 2002. Population and demographic trends. Pp. 11–35, In L.A. Hermansen and E.A. Macie (Eds.). The Southern Wildland-Urban Interface Assessment. US Forest Service, Southern Research Station, Asheville, NC. 160 pp. Crandall, K.A., and J.E. Buhay. 2008. Global diversity of crayfish (Astacidae, Cambaridae, and Parasticidae-Decapoda) in freshwater. Hydrobiologia 595:295–301. Dale, V.H., K.O. Lannom, M.L. Tharp, D.G. Hodges, and J. Fogel. 2009. Effects of climate change, land-use change, and invasive species on the ecology of the Cumberland forests. Canadian Journal of Forest Research 39:467–480. Davis, J.G., J.E. Miller, M.S. Billings, W.K. Gibbs, and S.B. Cook. 2011. Capture efficiency of underwater observation protocols for three imperiled fishes. Southeastern Naturalist 10:155–166. Dudgeon, D., A.H. Arthington, M.O. Gessner, Z.I. Kawabata, D.J. Knowler, C. Lévêque, R.J. Naiman, A. Prieur-Richard, D. Soto, M.L.J. Stiassny, and C.A. Sullivan. 2006. Freshwater biodiversity: Importance, threats, status, and conservation challenges. Biological Reviews 81:163–182. Fagan, W.F., P.J. Unmack, C. Burgess, and W.L. Minckley. 2002. Rarity, fragmentation, and extinction risk in desert fishes. Ecology 83:3250–3256. Fagen, R. 1988. Population effects of habitat change: A quantitative assessment. Journal of Wildlife Management 52:41–46. Flinders, C.A., and D.D. Magoulick. 2003. Effects of stream permanence on crayfishcommunity structure. American Midland Naturalist 149:134–147. Gaston, K.J. 1997. What is rarity? Pp. 30–47, In W.E. Kunin and K.J. Gaston (Eds.) The Biology of Rarity: Causes and Consequences of Rare–Common Differences. Chapman and Hall, London, UK. 280 pp. Griffith, M.B., L.T. Wolcott, and S.A. Perry. 1996. Production of the crayfish Cambarus bartonii (Fabricius, 1798) (Decapoda, Cambaridae) in an acidic Appalachian stream (USA). Crustaceana 69:974–984. Guan, R.Z. 1997. An improved method for marking crayfish. Crustaceana 70:641–652. Hartley, S., and W.E. Kunin. 2003. Scale dependency of rarity, extinction risk, and conservation priority. Conservation Biology 17:1559–1570. Hitt, N.P., and D.B. Chambers. 2014. Temporal changes in taxonomic and functional diversity of fish assemblages downstream from mountaintop mining. Freshwater Science 33:915–926. Hobbs, H.H. 1965. A new crayfish from the genus Cambarus from Tennessee with an amended definition of the genus (Decapoda, Astacidae). Proceedings of the Biological Society of Washington 78:265–273. Kerby, J.L., S.P.D. Riley, L.B. Kats, and P. Wilson. 2005. Barriers and flow as limiting factors in the spread of an invasive crayfish (Procambarus clarkii) in southern California streams. Biological Conservation 126:402–409. Kilburn, S.L., C.A. Taylor, and G.A. Schuster. 2014. Conservation assessment and habitat notes for three rare Alabama crayfishes: Cambarus cracens, Cambarus scotti, and Cambarus unestami. Southeastern Naturalist 13:108–118. Southeastern Naturalist 289 J.W. Johansen, H.T. Mattingly, and M.D. Padgett 2016 Vol. 15, No. 2 Kilian, J.V., A.J. Becker, S.A. Stranko, M. Ashton, R.J. Klauda, J. Gerber, and M. Hurd. 2010. Status and distribution of Maryland crayfishes. Southeastern Naturalist 9 (Special Issue 3):11–32. Krebs, C.J. 1989. Ecological Methodology. Harper and Row, New York, NY. 654 pp. Kunin W.E., and K.J. Gaston. 1993. The biology of rarity: Patterns, causes, and consequences. Trends in Ecology and Evolution 8:298–301. Larson, E.R., R.J. DiStefano, D.D. Magoulick, and J.T. Westhoff. 2008. Efficiency of a quadrat-sampling technique for estimating riffle-dwelling crayfish. North American Journal of Fisheries Management 28:1036–1043. Link, W.A., and J.D. Nichols. 1994. On the importance of sampling variance to investigations of temporal variation in animal population size. Oikos 69:539–544. Mace, G.M., and M. Kershaw. 1997. Extinction risk and rarity on an ecological timescale. Pp. 130–149, In W.E. Kunin and K.J. Gaston (Eds.). The Biology of Rarity: Causes and Consequences of Rare–Common Differences. Chapman and Hall, London, UK. 280 pp. Mace, G.M., N.J. Collar, K.J. Gaston, C. Hilton-Taylor, H.R. Akçakaya, N. Leader-Williams, E.J. Milner-Gulland, and S.N. Stuart. 2008. Quantification of extinction risk: IUCN’s system for classifying threatened species. Conservation Biology 22:1424–1442. MacKenzie, D.I., J.D. Nichols, G.B. Lachman, S. Droege, J.A. Royle, and C.A. Langtimm. 2002. Estimating site-occupancy rates when detection probabilities are less than one. Ecology 83:2248–2255. Mackenzie, D., J.D. Nichols, J.E. Hines, M.G. Knutson, and A.B. Franklin. 2003. Estimating site occupancy, colonization, and local extinction when a species is detected imperfectly. Ecology 84:2200–2207. Masters, L.L. 1991. Assessing threats and setting priorities for conservation. Conservation Biology 5:559–563. Mitchell, D.J., and L.A. Smock. 1991. Distribution, life history, and production of crayfish in the James River, Virginia. American Midland Naturalist 126:353–363. NatureServe. 2014. NatureServe explorer: An online encyclopedia of life [web application]. Version 7.1. NatureServe, Arlington, VA. Available online at http://www.explorer. natureserve.org. Accessed 11 July 2014. Nowicki, P., T. Tirelli, R.M. Sartor, F. Bona, and D. Pessani. 2008. Monitoring crayfish using a mark–recapture method: Potentials, recommendations, and limitations. Biodiversity and Conservation 17:3513–3530. O’Grady, J.J., D.H. Reed, B.W. Brook, and R. Frankham. 2004. What are the best correlates of predicted extinction risk? Biological Conservation 111:513–520. Purvis, A., J.L. Guttleman, G. Cowlishaw, and G.M. Mace. 2000. Predicting extinction risk in declining species. Proceedings of the Royal Society of London. Series B: Biological Sciences 267:1947–1952. Rabeni, C.F., K.J. Collier, S.M. Parkyn, and B.J. Hicks. 1997. Evaluating techniques for sampling stream crayfish (Paranephrops planifrons). New Zealand Journal of Marine and Freshwater Research 31:693–700. Rabinowitz, D., S. Cairns, and T. Dillon. 1986. Seven forms of rarity and their frequency in the flora of the British Isles. Pp. 182–204, In M.E. Soulé (Ed.). Conservation Biology: The Science of Scarcity and Diversity. Sinauer Associates, Sunderland, MA. 584 pp. Ricciardi, A., and J.B. Rasmussen. 1999. Extinction rates of North American freshwater fauna. Conservation Biology 13:1220–1222. Southeastern Naturalist J.W. Johansen, H.T. Mattingly, and M.D. Padgett 2016 Vol. 15, No. 2 290 Rohrbach, G.M., and D.I. Withers. 2006. A status survey of the Caney Fork Crayfish (Cambarus pristinus) and Hardin County Crayfish (Orconectes wrighti) with notes on the Brawley’s Fork Crayfish (Cambarus williami). Tennessee Wildlife Resources Agency, Nashville, TN. 81 pp. SAS Institute Inc. 2014. JMP® Version 10, Cary, NC. Sedell, J.R., G.H. Reeves, F.R. Hauer, J.A. Stanford, and C.P. Hawkins. 1990. Role of refugia in recovery from disturbance: Modern fragmented and disconnected river systems. Environmental Management 14:711–724. Smalley, G.W. 1982. Classification and evaluation for forest sites on the Mid-Cumberland Plateau. US Department of Agriculture, New Orleans, LA. 58 pp. Taylor, C.A., G.A. Schuster, J.E. Cooper, R.J. DiStefano, A.G. Eversole, P. Hamr, H.H. Hobbs III, H.W. Robison, C.E. Skelton, and R.F. Thoma. 2007. A reassessment of the conservation status of crayfishes of the United States and Canada after 10+ years of increased awareness. Fisheries 32:372–389. Townsend, C.R. 1989. The patch dynamics concept of stream community ecology. Journal of the North American Benthological Society 8:36–50. Wagner, B.K., C.A. Taylor, and M.D. Kottmyer. 2010. Status and distribution of Orconectes williamsi (Williams Crayfish) in Arkansas with new records from the Arkansas River drainage. Southeastern Naturalist 9(Special Issue 3):175–184. Welsh, S.A., Z.J. Loughman, and T.P. Simon. 2010. Concluding remarks: A symposium on the conservation, biology, and natural history of crayfishes from the southeastern United States. Southeastern Naturalist 9(Special Issue 3):267–269. Williams, C.E., R.D. Bivens, and B.D. Carter. 2004. A crayfish survey of Wilkerson Creek and two of its tributaries Cumberland County, Tennessee. Tennessee Wildlife Resources Agency, Nashville, TN. 9 pp. Williams, C.E., R.D. Bivens, and B.D. Carter. 2006. A status survey of the Obey Crayfish (Cambarus obeyensis). Tennessee Wildlife Resources Agency, Nashville, TN. 19 pp. Williams, C.E., R.D. Bivens, and B.D. Carter. 2009. Key to the crayfishes of Tennessee, abstracted from H.H. Hobbs Jr. 1972, H.H. Hobbs Jr. 1981, and Bouchard 1978, and an annotated list of the crayfishes of Tennessee. Tennessee Wildlife Resources Agency, Nashville, TN. 78 pp. Withers, D.I., and R.A. McCoy. 2005. Distributional surveys for Cambarus pristinus and Cambarus williami, two endangered crayfish in Tennessee. Tennessee Wildlife Resources Agency, Nashville, TN. 57 pp. Yoccoz, N.G., J.D. Nichols, and T. Boulinier. 2001. Monitoring of biological diversity in space and time. Trends in Ecology and Evolution 16:446–453.