Southeastern Naturalist W.K. Gibbs, J.E. Miller, S.B. Cook, and M.A. Kulp 2014 Vol. 13, No. 1 40 2014 SOUTHEASTERN NATURALIST 13(1):40–55 Habitat Use and Dispersal of a Reintroduced Etheostoma sitikuense (Citico Darter) Population W. Keith Gibbs1,*, Jason E. Miller1,2, S. Bradford Cook1, and Matt A. Kulp3 Abstract - Etheostoma sitikuense (Citico Darter), a federally protected fish endemic to the southeastern United States, was extirpated from Abrams Creek in Great Smoky Mountains National Park in 1957. The species was reintroduced from 1993–2001, but recovery efforts have thus far achieved only partial success, due in part to limited knowledge of Citico Darter habitat use. After distribution of the reintroduced population was established, we monitored Citico Darters in a 4-km section of Abrams Creek using underwater observation. We evaluated macro- and microhabitat use over four summers using principal components analysis to determine macrohabitat variables influencing Citico Darter distribution, and used classification tree methods to analyze microhabitat use. We analyzed dispersal using linear regression to compare historical stocking data with current Citico Darter distribution data. We identified percentage of pools and cobble/small boulder substrates as the most significant macrohabitat variables influencing Citico Darter presence. This species most often occupied microhabitats away from riffles under intermediate-sized cover rocks. Dispersal of reintroduced Citico Darters was limited in Abrams Creek. Results of this study can be used to identify additional reintroduction zones and assist in further conservation efforts. Introduction The southeastern United States has the greatest diversity and endemicity of freshwater fishes in North America (Warren et al. 2000); however, ratios of imperilment are also highest (Warren and Burr 1994). The southeastern physiographic region known as the Eastern Interior Highlands, particularly the Tennessee and Cumberland River drainages, comprises the most endemic, imperiled fish fauna in the United States (Starnes and Etnier 1986, Warren and Burr 1994). At least 83 indigenous taxa of the more than 290 described fish species within Tennessee merit designation under state or federal protective status due to continued threats from anthropogenic activities (Etnier and Starnes 1991, Warren et al. 2000). Characterization of habitat use of imperiled taxa is vital to understanding life-history requirements to preserve existing populations and identify potential reintroduction sites within historical ranges where anthropogenic degradation has been alleviated and suitable habitat remains. Darters, especially the subgenus Catonotus of the genus Etheostoma (Percidae), represent a large proportion of the imperiled taxa within Tennessee (Etnier and Starnes 1991). Etheostoma sitikuense Blanton (Citico Darter), a Catonotus 1Department of Biology, Tennessee Technological University, Cookeville, TN 38505. 2Current address - Burgess Falls State Natural Area, 4000 Burgess Falls Drive, Sparta, TN 38583. 3Resource Management Division, Great Smoky Mountains National Park, Gatlinburg, TN 37738. Corresponding author - kgibbs@tntech.edu. Manuscript Editor: Clifton Ruehl Southeastern Naturalist 41 W.K. Gibbs, J.E. Miller, S.B. Cook, and M.A. Kulp 2014 Vol. 13, No. 1 originally thought to represent the federally endangered Etheostoma percnurum Jenkins (Duskytail Darter), was recently re-described and considered morphologically distinct from the other 3 geographically isolated, extant populations within the Duskytail Darter complex, which also includes Etheostoma lemniscatum Blanton (Tuxedo Darter) and Etheostoma marmorpinnum Blanton (Marbled Darter) (Blanton and Jenkins 2008). The only known natural population of Citico Darters remains in Citico Creek, a tributary of the Little Tennessee River, located in Monroe County, TN. The relict distribution of this species complex within the Tennessee and Cumberland River drainages suggests a previously wide distribution (Etnier and Starnes 1993). Isolation of remaining populations is attributed to habitat degradation (i.e., siltation) and extensive impoundment within both drainages (Etnier and Starnes 1993). The 3 re-described species within the species complex, including the Citico Darter, should be considered highly endangered and included under the protective status as designated for the Duskytail Darter until individual imperilment listings and recovery plans can be established. An objective of the Duskytail Darter recovery plan is to establish additional viable populations within its historical range (USFWS 1993). An isolated population within the species complex was extirpated from Abrams Creek, Great Smoky Mountains National Park (GSMNP), Blount County, TN in 1957 as a result of rotenone application (Lennon and Parker 1959, Simbeck 1990). Comparison of specimens collected from Abrams Creek before rotenone application, plus previous free-flowing connectivity to the Little Tennessee River between Abrams and Citico Creeks (creek mouths separated by <10 km of the Little Tennessee River) indicated that they were conspecific and justified reintroductions of captively propagated Citico Darters from Citico Creek into Abrams Creek from 1993–2001 (Shute et al. 2005). Evidence of Citico Darter reproduction within Abrams Creek was first observed in 1995 (Shute et al. 2005), and observation of eggs and young-of-year continued through the duration of this study, yet long-term viability of the Abrams Creek population is not assured, and continued monitoring and habitat protection may be necessary. The Citico Darter and other members in the species complex are relatively small (28–64 mm total length [TL]) benthic species, generally tan to brown in appearance, with 10–15 dark vertical bars along the sides, and 6–7 dark dorsal saddles (Etnier and Starnes 1993, Jenkins and Burkhead 1994). As outlined by an additional recovery plan objective, life-history characteristics of natural populations within the species complex have been documented (e.g., Eisenhour and Burr 2000, Jenkins and Burkhead 1994, Layman 1991, Rakes et al. 1992). As with other Catonotus, members of the species complex, including the Citico Darter, are eggclusterers (Layman 1984) as exhibited by males clearing and maintaining nesting cavities beneath cover rocks where females invert and lay clusters of adhesive eggs (Etnier and Starnes 1993). Members of the species complex have been confirmed to spawn from April through June (Etnier and Starnes 1993, Layman 1991, Rakes et al. 1992). As nocturnal insectivores, species within the complex feed primarily on Southeastern Naturalist W.K. Gibbs, J.E. Miller, S.B. Cook, and M.A. Kulp 2014 Vol. 13, No. 1 42 mayfly nymphs, chironomid larvae, and microcrustaceans while spending daylight hours beneath cobble and small-boulder substrate (Etnier and Starnes 1993, Layman 1991, Jenkins and Burkhead 1994). Although other members of the species complex occur in relatively large systems in low-velocity habitats with large substrates (Etnier and Starnes 1993), habitat use of the introduced population of Citico Darters within Abrams Creek is unknown. The distribution of introduced Citico Darters was recently documented in Abrams Creek (Gibbs 2009), and habitat use should be studied at multiple spatial scales to fulfill recovery plan objectives—specifically, to determine the biological and ecological requirements of Citico Darters in Abrams Creek (USFWS 1993). Macrohabitat (i.e., reach-level) features affect the distribution of fishes at a large scale, while microhabitat (i.e., habitat within 1 m2) studies are important to identify specific stream locations fish utilize based on multiple abiotic and biotic factors (Muhlfeld et al. 2001). Macrohabitat and microhabitat data are necessary to identify potential reintroduction sites with suitable habitat outside current Citico Darter distributions and to build statistical models useful for management and conservation efforts within current distributions. The overall goal of this study was to determine habitat use and dispersal of introduced Citico Darters in Abrams Creek. Specifically, our objectives were to (1) characterize macro- and microhabitat utilized by Citico Darters and (2) determine Citico Darter dispersal from reintroduction sites. Results of this study will assist National Park Service (NPS) personnel, tasked with restoring and conserving the Citico Darter within Abrams Creek, by identifying precise locations within stream reaches occupied by the species for preservation of specific stream habitats, as well as aid in potential future reintroduction efforts. Field-Site Description Abrams Creek, a moderate-size 5th-order stream, is located in the southwestern portion of GSMNP. The Abrams Creek watershed (225 km2) is comprised of ≈348.0 km of streams, and the majority of it lies within park boundaries (Parker and Pipes 1990). Limestone bedrock within the watershed increases pH, conductivity, and alkalinity (Shaffer 2004). Abrams Falls, a 6.0-m waterfall, bisects Abrams Creek into two distinct systems and functions as a barrier against upstream fish passage (NPS 2010). The lower, 23.5-km section of Abrams Creek extends from Abrams Falls to an embayment of Chilhowee Reservoir, an impoundment of the Little Tennessee River. This lower section is a coolwater stream, with temperatures averaging <23 °C in summer. Pools, deep runs, and short cascades are dominant habitat types, with cobble and bedrock as the principal substrate types (Shaffer 2004). Stream width and stream discharge average 18 m and 27.4 m3/s, respectively, and the average gradient is 3.5% (Lennon and Parker 1959, Shaffer 2004). Reintroduction sites for Citico Darters were located only in the lower section of Abrams Creek (Rakes and Shute 2007). Southeastern Naturalist 43 W.K. Gibbs, J.E. Miller, S.B. Cook, and M.A. Kulp 2014 Vol. 13, No. 1 Methods Historically, studies of stream-dwelling species have relied on backpack electrofishing equipment or seines for collections (Hayes et al. 1996, Reynolds 1996). However, due to the imperiled status of Citico Darters, and the inability to effectively seine Abrams Creek, we used underwater observation (i.e., snorkeling; Dolloff et al. 1996). This method avoids the unnecessary mortality associated with electrofishing. We conducted snorkeling surveys during daylight hours when underwater visibility was ≥2 m to maximize observations, but only surveyed after mid-July of each year to ensure spawning fish were not disturbed. We established lanes of equidistant width that were parallel to shore, and a minimum of 4 people snorkeled each lane, moving upstream to reduce startling fish and to optimize observations (Dolloff et al. 1996). Movable rocks larger than gravel were gently lifted to allow embedded sediment to wash downstream and determine Citico Darter presence under cover rocks within each reach. No species within lower Abrams Creek are similar in appearance to Citico Darters; however, all snorkelers were required to observe several individuals with a trained surveyor prior to making independent observations to ensure accurate identification and correct methodology. Macrohabitat We collected macrohabitat data between May and September of 2007 and 2008. Beginning at the embayment, lower Abrams Creek was divided into 2-km sections that were subdivided into consecutive 200-m reaches. A 1-m cascade was identified as a potential barrier to upstream darter passage ≈17.2 river kilometers (rkm) from the embayment. We randomly selected three 200-m reaches within each 2-km section downstream of the cascade. We delineated a 200-m reach with representative habitat above the cascade to confirm effectiveness of the barrier. This survey design resulted in 27 total reaches (≈30% of the 18 rkm upstream of the embayment) being sampled. We calculated distance from embayment from the downstream start point of each randomly selected reach using latitude-longitude coordinates obtained with a Garmin GPSmap 76Cx. We measured macrohabitat parameters in each randomly selected reach prior to snorkeling. We classified habitat types as riffle, run, or pool as described by Overton et al. (1997). We measured lengths of all units of each habitat type within each reach to quantify percentages. We delineated transects every 20 m and recorded wetted widths (m) to determine average width for each reach. We measured depth (cm) at 3 equidistant locations along each transect to calculate average reach depth (McMahon et al. 1996) and visually classified substrate using a modified Wentworth (1922) scale (silt [fines], sand [0.1–2 mm], gravel [2–60 mm], cobble [60–300 mm], small boulder [300–500 mm], large boulder >500 mm], and bedrock). We visually estimated percentages of dominant substrate types within 0.25-m radius at each location that depth was measured to calculate percentage of substrate composition within each reach. Southeastern Naturalist W.K. Gibbs, J.E. Miller, S.B. Cook, and M.A. Kulp 2014 Vol. 13, No. 1 44 Snorkeling surveys were conducted as previously described. Upon positive identification of at least 1 Citico Darter, we considered a reach to be a present location, whereas we considered sites in which no Citico Darters were encountered as absent locations. We calculated catch per unit effort (CPUE) as number of Citico Darters per person hour snorkeled within each reach. Absent locations upstream and downstream of current Citico Darter distribution were confirmed during microhabitat study. Microhabitat We collected microhabitat data from June 2009 to September 2010. Eight 200-m reaches within the Citico Darter’s distribution, identified during macrohabitat sampling, were used for microhabitat analysis. Snorkel surveys were conducted in each reach, following previously described methods. When a Citico Darter was encountered, the cover rock was carefully replaced and a lead weight (85 g) marked with highly visible forestry flagging was placed on the cover ro ck. Upon survey completion, we recorded microhabitat variables (i.e., habitat within a 0.25-m radius) at each occupied location. We measured depth (cm) with a top-setting wading rod and velocities (m/s) at substrate and at 60% total depth with a Marsh-McBirney Model 2000 Flow Meter. We measured dissolved oxygen (DO [mg/l]), temperature (°C), and conductivity (mS) at substrate with a Yellow Springs Instrument (YSI) Model 85 meter. We visually assessed habitat type, substrate beneath cover rock, and dominant substrate within a 0.25-m radius using the previously described categories. We measured dimensions (cm) of cover rocks along the two longest axes to calculate approximate surface area (cm2). We also measured the distance to nearest bank (m) and distance to nearest riffle (m). Presence of multiple specimens within a 0.25-m radius of an occupied location was recorded if observed. All microhabitat variables measured at occupied locations were measured at unoccupied locations to determine overall habitat availability within each reach and for analytical comparisons. We designated unoccupied locations as 3 randomly selected points along the 7 furthest transects from the greatest concentration of observed occupied locations, resulting in 21 unoccupied locations within each reach (Schmidt 2007). We did not measure unoccupied locations if they were within 5 m of an occupied location. If Citico Darters were widely distributed within a reach, fewer unoccupied locations were measured due to proximity of potential unoccupied points to occupied locations. Before measurements were taken, potential unoccupied locations were observed underwater to ensure no Citico Darters were present. Dispersal We obtained stocking records (i.e., locations and # of individuals stocked) from 1993–2001 from Conservation Fisheries, Inc. (CFI) to compare with CPUE of randomly selected reaches within current Citico Darter distribution. We calculated distances of stocking locations from surveyed reaches using ArcView 9.2 GIS (ESRI 2007). We tabulated stocking densities within 1 km upstream and downstream of each surveyed reach at 200-m intervals resulting in comparisons Southeastern Naturalist 45 W.K. Gibbs, J.E. Miller, S.B. Cook, and M.A. Kulp 2014 Vol. 13, No. 1 between CPUE and average number of individuals stocked within 200 m, 400 m, 600 m, 800 m, and 1000 m. Statistical analysis Due to small sample size (n = 27) of reaches, we used principal components analysis (PCA) to determine the influence of macrohabitat variables on Citico Darter distribution. PCA is a multivariate ordination technique based on eigenvectors of principal components accounting for the greatest variation among samples (Kwak and Peterson 2007). PCA is ideal for reducing dimensionality of highly correlated multivariate data, as is typical of in-stream habitat data, to ≤5 interpretable ordination axes (Clarke and Warwick 2001). Kwak and Peterson (2007) recommended retaining all components with eigenvalues >1. Macrohabitat variables calculated as percentages (i.e., habitat type and substrate composition) were approximately normalized using an arcsine transformation (Zar 1999), whereas average width and depth were kept continuous. We performed PCA using Primer v6 software (PRIMER-E 2006). We used classification tree methodology, developed by Breiman et al. (1984), to describe microhabitat use within reaches identified as being occupied by Citico Darters during macrohabitat surveys. Classification and regression trees (CART) are ideal for modeling complex ecological data with categorical and numeric explanatory variables (Breiman et al. 1984, De’ath and Fabricius 2000). Classification trees split data into the most parsimonious dichotomous groupin gs of the categorical response variable (i.e., occupied or unoccupied) by hierarchically partitioning the most influential explanatory variables (De’ath and Fabricius 2000). Continuous variables (i.e., temperature, conductivity, depth, DO at substrate, DO at 60% depth, velocity at substrate, velocity at 60% depth, cover rock area, distance to bank, and distance to riffle) were kept in original form, whereas categorical data (i.e., habitat type, cover rock type, basal substrate, and dominant substrate) were assigned numerical values (e.g., riffle = 1, run = 2, pool = 3) to retain their categorical nature. We conducted classification tree analyses using Salford Systems CART 6.0 software (Steinberg and Colla 1995). We set no stopping rules, which allowed for full tree growth with each terminal node containing individuals from only 1 response category. We used the default setting of 10-fold cross-validation to select optimal sub-trees based on relative error of misclassification rates. We selected the smallest tree within 1 standard error of the tree with the lowest relative error as the best model (Steinberg and Colla 1995). We used linear regression to associate stocking abundance to Citico Darter abundance, as represented by CPUE. We compared the number of Citico Darters stocked within 200 m, 400 m, 600 m, 800 m, and 1000 m upstream and downstream of sampled reaches with CPUE to determine dispersal ability of Citico Darters away from stocking locations. We performed regression analyses using Statistical Analysis Software (SAS) 9.2 (SAS Institute, Inc. 2008). These analyses are intended to guide future conservation efforts within Abrams Creek by allowing identification of additional habitat for monitoring natural expansion of current Citico Darter distribution and habitat suitable for reintroductions Southeastern Naturalist W.K. Gibbs, J.E. Miller, S.B. Cook, and M.A. Kulp 2014 Vol. 13, No. 1 46 outside their current range if necessary. These models should be validated with additional data collected during long-term monitoring of this reintroduced population. Results Macrohabitat Citico Darters were observed in 5 reaches spanning 4.0 rkm (≈22%) of the 18 rkm of lower Abrams Creek sampled during 2007–2008. No individuals were observed within 3.6 rkm downstream of the 1-m cascade barrier. The first 3 principal components had eigenvalues >1 and accounted for 81.0% of variation among sites (Table 1). Based on variable loadings, Citico Darters most often occurred in reaches with more pools and intermediate-sized substrates (i.e., cobble and small boulder; Fig. 1, Table 1). Most reaches occupied by Citico Darters were not typical of sampled reaches throughout lower Abrams Creek (Fig. 1). This finding may represent limited reach-level habitat suitability for Citico Darters within lower Abrams Creek and habitat limitations to distribution extensions. Microhabitat In total during 2009–2010, we collected microhabitat data from 220 locations occupied by Citico Darter from 7 sites on lower Abrams Creek within the distribution range established from macrohabitat surveys. Congruent with macrohabitat results, the majority (≈88%) of Citico Darters occupied locations away (>6.1 m) from riffles under intermediate-sized (>100 cm2) cover rocks (Fig. 2). Only 14 of 220 (6.4%) individuals occupied locations ≤6.1 m away from riffles. Of the 206 individuals located >6.1 m from a riffle, only 13 (6.3%) occupied locations under cover rocks ≤100 cm2. Most individuals (≈85%) occupying locations away from riffles under intermediate-sized cover rocks were in local habitats (i.e., ≤1 m2) dominated by cobble and small-boulder substrates (Fig. 2). Based on these results, sampling within and outside of current Citico Darter distribution and additional reintroductions should occur in long pools comprised predominately of intermediate-sized substrates. Table 1. Variable loadings for principal components 1–3 for reach-level macrohabitat within lower Abrams Creek. Percentage of variation explained by each princip al component in parentheses. Axis Variable PC1 (36.1%) PC2 (27.0%) PC3 (17.8%) % cobble/small boulder 0.551 -0.116 -0.102 % pool 0.393 0.430 -0.006 % riffle 0.153 -0.488 0.441 % gravel 0.079 -0.414 -0.434 Avg. depth -0.004 0.604 -0.165 Stream width -0.029 0.130 0.582 % run -0.477 -0.085 -0.402 % large boulder/bedrock -0.532 0.060 0.282 Southeastern Naturalist 47 W.K. Gibbs, J.E. Miller, S.B. Cook, and M.A. Kulp 2014 Vol. 13, No. 1 Figure 1. Ordination plots of principal components analysis (PC1 [36.1%] v. PC2 [27.0%] and PC1 v. PC3 [17.8%]) for reach-level macrohabitat within lower Abrams Creek. Values above present points are catch per unit ef fort. Southeastern Naturalist W.K. Gibbs, J.E. Miller, S.B. Cook, and M.A. Kulp 2014 Vol. 13, No. 1 48 Dispersal Citico Darters were stocked in several locations throughout their current distribution in lower Abrams Creek over 8 years (Fig. 3). Average stocking densities ranged from 5.6–57.4 fish/year (Fig. 3). We tabulated average stocking densities at Figure 2. Classification tree analysis for microhabitat data of occupied (present) and unoccupied (absent) locations of Citico Darters within lower Abrams Creek. Child nodes represent the most parsimonious split of the preceding node based on the predictor variable in italics. Southeastern Naturalist 49 W.K. Gibbs, J.E. Miller, S.B. Cook, and M.A. Kulp 2014 Vol. 13, No. 1 200-m intervals upstream and downstream of each site surveyed for macrohabitat analyses. CPUE was dependent on average stocking density at all distances within 1 rkm of each sampled reach (Table 2), yet the most significant relationship and bestfitting model was between CPUE and average number of individuals stocked within 200 m (Fig. 4, Table 2). The positive relationship between CPUE and average stocking density indicates limited dispersal of Citico Darters away from stocking locations. Therefore, successful reintroductions of Citico Darters in lower Abrams Creek were dependent on >5 years of consistent yearly stocking of ≥25 individuals per year in close proximity (≤200 m) to long pool habitats with an abundance of intermediate-sized cover rocks. Figure 3. Average number of individuals stocked within current Citico Darter distribution (≈10 rkm from embayment) in relation to catch per unit effort of reaches sampled during macrohabitat surveys. Table 2. Results of linear regression analyses comparing CPUE to average stocking densities at 200-m intervals upstream and downstream from surveyed reaches within current Citico Darter distribution. Distance from site F P R2 200 m 52.0 0.0020 0.9286 400 m 18.1 0.0131 0.8190 600 m 16.8 0.0149 0.8076 800 m 19.8 0.0112 0.8319 1000 m 15.8 0.0166 0.7974 Southeastern Naturalist W.K. Gibbs, J.E. Miller, S.B. Cook, and M.A. Kulp 2014 Vol. 13, No. 1 50 Discussion Species within the genus Etheostoma inhabit a variety of habitats ranging from swift headwater streams to non-flowing swamps and ponds (Page 1983); however, relatively few lotic darter species occur away from highly oxygenated, turbulent riffle and run habitats. Typically, members of the subgenus Catonotus are riffle and run specialists (Etnier and Starnes 1993); yet, members of the Duskytail Darter species complex have generally been associated with slow-moving waters of moderate depth (0.3–1.2 m), in medium- to large-sized drainages (Etnier and Starnes 1993, Jenkins and Burkhead 1994). Etheostoma lemniscatum Blanton (Tuxedo Darter) and Etheostoma marmorpinnum Blanton (Marbled Darter), 2 other members within the species complex (Blanton and Jenkins 2008), were found to use similar habitat. Tuxedo Darters were found to use pool habitats at intermediate depths in the Big South Fork of the Cumberland River, TN and KY (Davis 2010). Marbled Darters were associated with pools and utilized cobble substrate in Little River, Blount County, TN (Layman 1991). Results from this study confirmed that Citico Darters exhibited similar habitat utilization as other members within the Duskytail Darter species complex. At the macrohabitat level, presence of Citico Darters was strongly associated with availability of intermediate-sized (i.e., cobble and small boulder) substrates and pool habitats. Reaches with Citico Darters present were dominated by cobble, and most individuals were found beneath cobble or small-boulder cover rocks. Figure 4. Relationship between catch per unit effort and average number of individuals stocked within 200 m over 8 years in lower Abrams Creek. Southeastern Naturalist 51 W.K. Gibbs, J.E. Miller, S.B. Cook, and M.A. Kulp 2014 Vol. 13, No. 1 Selectivity of Citico Darters to cobble substrates is congruent with the specialized reproductive behavior exhibited by the species complex (e.g., Etnier and Starnes 1993, Jenkins and Burkhead 1994). Although populations within the species complex are highly localized, habitat selection is consistent, particularly in the Tennessee River drainage. Throughout the life-history literature, Citico Darters and their close relatives have been associated with “slabrock” or cobble substrates during breeding seasons, depositing eggs on the undersides of these rocks (e.g., Etnier and Starnes 1993, Jenkins and Burkhead 1994, Layman 1991, Rakes et al. 1992). Percentage of pool habitat was the variable most highly associated with Citico Darter presence and it should be noted that the highest observation totals (i.e., abundance) were within sites with the greatest amounts of pool habitat. Ultimately, abundance of pool habitat and intermediate-sized substrates were associated with abundance of Citico Darters in lower Abrams Creek. These results provide quantitative evidence corroborating observational data of previous studies regarding habitat use of other species within the Duskytail Darter complex. At the microhabitat level, this study determined that Citco Darters most utilized intermediate-sized (>100 cm2) cover rocks located away from riffles in areas dominated by cobble and small boulders. These results are similar to other studies’ findings that intermediate-sized substrates generally provide the best habitat for both benthic fishes and macroinvertebrates in lotic environments (Wood and Armitage 1997). However, differences in microhabitat use were most accurately illustrated by distance from riffle habitat in determining Citic o Darter presence. Most Citico Darters (>90%) observed in Abrams Creek were found in pools, and individuals were rarely found <4 m from the nearest bank. Citico Darters were generally found in microhabitats with current velocities averaging <0.1 m/s and frequently concentrated away from shallow littoral habitats at significant distances from riffles. Many habitat variables were correlated; however, CART analysis reflected the Citico Darter’s predilection for microhabitats away from riffles in moderately deep, slow-moving water, typical of pools. Presence of pool habitat, as denoted by distance from riffles during microhabitat analyses, was the most consistent habitat variable influencing Citico Darter presence in Abrams Creek at both the macro- and microhabitat spatial scales. The section of lower Abrams Creek that supports the greatest numbers of Citico Darters is not only the area that has been the most heavily stocked, but, of reaches sampled in this study, it is also the only contiguous section comprised mostly of pool habitat. Other areas within lower Abrams Creek were also extensively stocked, yet lack of suitable habitat (i.e., pools with intermediate-sized substrates) within those areas precluded establishment of viable Citico Darter populations. The definitive objective of the Duskytail Darter recovery plan is “to restore viable populations” … “to a significant portion of its historic range” and ultimately remove the species from the endangered species list (USFWS 1993). For long-term viability, Soule (1980) suggests a population size of 1500–2000 individuals. Any species within the Duskytail Darter complex can be reclassified as threatened if it is determined that “three distinct viable populations exist”, either naturally occurring Southeastern Naturalist W.K. Gibbs, J.E. Miller, S.B. Cook, and M.A. Kulp 2014 Vol. 13, No. 1 52 or reintroduced, and no threats to their survival are anticipated. Excluding reintroduced populations, the new species designations within the species complex (Blanton and Jenkins 2008) have resulted in only 1 natural population of each of the 4 species. Determination of biological and ecological requirements of each population must also be completed, and management strategies must be implemented and deemed successful prior to reclassification (USFWS 1993). This study addressed some of the biological and ecological requirements necessary for viability of this reintroduced population of Citico Darters. Although Citico Darters currently occupy <25% of the study area, reproductive success and relative abundance of the species, particularly in areas within close proximity to stocking sites, were comparable to the Citico Creek source population (Shute et al. 2005). Personal observations of nest-guarding males, as well as juveniles, provided direct evidence that the Abrams Creek population is reproducing. Considering longevity of individuals within the species complex is between 2–3 years (Layman 1991) and Abrams Creek stocking efforts ceased in 2001 (Shute et al. 2005), individuals observed during this study were, at minimum, the 2nd naturalized generation. Distributions appeared static over the course of this study. However, as this study demonstrates, short-term Citico Darter dispersal ability is limited and may require several additional years for natural extension of distribution range, provided suitable habitat is available within close proximity to current distribution. If the Abrams Creek population is viable, allowing natural dispersion is preferable due to costs associated with collecting and propagating fish in captivity. However, the species’ long-term ability to naturally disperse and colonize new areas is unknown. Population augmentation with transplanted or captively propagated individuals should be avoided until the current population density within Abrams Creek can be estimated with greater certainty (Williams et al. 1988) and long-term dispersal ability assessed through continued monitoring beyond the Citico Darter’s current range. American Fisheries Society Guidelines for reintroductions (Williams et al. 1988) suggest monitoring to determine viability of reintroduced populations and then additional stocking only if warranted. It is unknown whether current conditions warrant continued stocking, or if carrying capacity has been achieved in lower Abrams Creek. Implementation of long-term (10+ yrs.) monitoring programs can determine current population sizes, detect population trends, and infer long-term population viability (Davis et al. 2011). Monitoring protocols, if implemented, should be based on those completed for similarly imperiled species with similar habitat use, such as protocols established for the Tuxedo Darter (Davis 2010, Davis et al. 2011). Research concerning genetic viability of the population is also recommended, as results might influence decisions regarding the need for further augmentation to enhance genetic diversity. Finally, young-of-year individuals were observed during all 4 years of this study, throughout the species’ distribution within Abrams Creek. Past surveys (Rakes and Shute 2007) have also documented reproduction, providing additional evidence that the Abrams Creek population is currently self-sustaining. However, preserving Southeastern Naturalist 53 W.K. Gibbs, J.E. Miller, S.B. Cook, and M.A. Kulp 2014 Vol. 13, No. 1 this population remains a priority, and future conservation efforts will likely play a large role in determining the fate of this species. Such efforts should include long-term monitoring of population trends within lower Abrams Creek, investigation of genetic viability, and continued reintroductions if warranted by results of population and genetic analyses. Knowledge of reach-level and microhabitat use, specifically in areas with higher percentages of pools and intermediate-sized substrates in locations away from riffles under cover rocks >100 cm2, recognition of the limited dispersal ability of stocked individuals, and other suggestions presented here should be utilized to further enhance Citico Darter recove ry efforts. Acknowledgments We would especially like to acknowledge Steve Moore of the Great Smoky Mountains National Park for his guidance and help during planning and implementing this project. We are appreciative of Justin Murdock for his editorial comments. We are grateful to the anonymous reviewers whose comments improved this manuscript. Funding for this project was provided by the Tallassee Fund (Alcoa Power Co.), the National Park Service, and the Tennessee Technological University Center for the Management, Utilization, and Protection of Water Resources. The authors are especially grateful to the numerous individuals that assisted with this project, including J.R. Shute and Patrick Rakes of Conservation Fisheries, Inc., whose tireless efforts conserving rare and endangered native fishes resulted in the reintroduction efforts of Citico Darters into Abrams Creek. Literature Cited Blanton, R.E., and R.E. Jenkins. 2008. Three new species of the Etheostoma percnurum species complex (Percidae, subgenus Catonotus) from the Tennessee and Cumberland River drainages. Zootaxa 1963:1–24. Breiman, L., J.H. Friedman, R.A. Olshen, and C.G. Stone. 1984. Classification and Regression Trees. Wadsworth International Group, Belmont, CA. Clarke, K.R., and R.M. Warwick. 2001. Change in Marine Communities: An Approach to Statistical Analysis and Interpretation, 2nd Edition. PRIMER-E, Plymouth, UK. Davis, J.G. 2010. Development of long-term monitoring protocols for two rare fishes, the Tuxedo Darter and the Spotfin Chub, in Tennessee. Ph.D Dissertation. Tennessee Technological University, Cookeville, TN. 262 pp. Davis, J.G., S.B. Cook, and D.D. Smith. 2011. Testing the utility of an adaptive cluster sampling method for monitoring a rare and imperiled darter. North American Journal of Fisheries Management 31:1123–1132. De’ath, G., and K.E. Fabricius. 2000. Classification and regression trees: A powerful yet simple technique for ecological data analysis. Ecology 81(1 1):3178–3192. Dolloff, C.A., J. Kershner, and R. Thurow. 1996. Underwater observation. Pp. 533–554, In B.R. Murphy and D.W. Willis (Eds.). Fisheries Techniques, 2nd Edition. American Fisheries Society, Bethesda, MD. 732 pp. Eisenhour, D.J., and B.M. Burr. 2000. Conservation status and nesting biology of the endangered Duskytail Darter, Etheostoma percnurum, in the Big South Fork of the Cumberland River, Kentucky. Journal of the Kentucky Academy of Science 61:67–76. Environmental Systems Research Institute (ESRI). 2007. ArcView 9.2 and Extensions. Environmental Systems Research Institute, Redlands, CA. Southeastern Naturalist W.K. Gibbs, J.E. Miller, S.B. Cook, and M.A. Kulp 2014 Vol. 13, No. 1 54 Etnier, D.A., and W.C. Starnes. 1991. An analysis of Tennessee’s jeopardized fish taxa. Journal of the Tennessee Academy of Science 66(4):129–133. Etnier, D.A., and W.C. Starnes. 1993. Fishes of Tennessee. University of Tennessee Press, Knoxville, TN. 689 pp. Gibbs, W.K. 2009. Evaluation of reintroduction success of the endangered Duskytail Darter, Etheostoma percnurum, and the threatened Spotfin Chub, Erimonax monachus, in Abrams Creek, Great Smoky Mountains National Park. M.Sc. Thesis. Tennessee Technological University, Cookeville, TN. 70 pp. Hayes, D.B., C.P. Ferreri, and W.W. Taylor. 1996. Active fish capture methods. Pp. 193– 220, In B.R. Murphy and D.W. Willis (Eds.). Fisheries Techniques, 2nd Edition. American Fisheries Society, Bethesda, MD. 732 pp. Jenkins, R.E., and N.M. Burkhead. 1994. Freshwater fishes of Virginia. American Fisheries Society, Bethesda, MD. 1079 pp. Kwak, T.J. and J.T. Peterson. 2007. Community indices, parameters, and comparisons. Pp. 677–763, In C.S. Guy and M.L. Brown (Eds.). Analysis and Interpretation of Freshwater Fishereis Data. American Fisheries Society, Bethesda, MD. 961 pp. Layman, S.R. 1984. The Duskytail Darter, Etheostoma (Catonotus) sp., confirmed as an egg- clusterer. Copeia 1984:992–994. Layman, S.R. 1991. Life history of the relict, Duskytail Darter, Etheostoma (Catonotus) sp., in Little River, Tennessee. Copeia 1991:471–485. Lennon, R.E., and P.S. Parker. 1959. The reclamation of Indian and Abrams creeks, Great Smoky Mountains National Park. Special Scientific Report – Fisheries No. 306. United States Department of the Interior, Fish and Wildlife Service, Washington, DC. 22 pp. McMahon, T.E., A.V. Zale, and D.J. Orth. 1996. Aquatic habitat measurements. Pp. 83–120, In B.R. Murphy and D.W. Willis (Eds.). Fisheries Techniques, 2nd Edition. American Fisheries Society, Bethesda, MD. 732 pp. Muhlfeld, C.C., D.H. Bennett and B. Martoz. 2001. Summer habitat use by Columbia River Redband Trout in the Kootenai River drainage, Montana. North American Journal of Fisheries Management 21:223–235. National Park Service (NPS). 2010. Great Smoky Mountains waterfall data. Available online at http://www.nps.gov/grsm/planyourvisit/waterfalls.htm. Accessed 9 March 2012. Overton, C.K., S.P. Wollrab, B.C. Roberts, and M.A. Radko. 1997. R1/R4 (Northern/ Intermountain Regions) fish and fish habitat standard inventory procedures handbook. General Technical Report INT-GTR-346. Intermountain Research Station, United States Department of Agriculture, Forest Service, Ogden, UT. 73 pp. Page, L.M. 1983. Handbook of Darters. T.F.H. Publications Incorporated. Neptune City, NJ. 271 pp. Parker, C.R., and D.W. Pipes. 1990. Watersheds of Great Smoky Mountains National Park: A geographical information system analysis. Research/Resources Management Report SER-91/01. Southeast United States Department of the Interior, National Park Service, Regional Office, Atlanta, GA. 126 pp. PRIMER-E. 2006. PRIMER-E Ltd, Plymouth Marine Laboratory, Plymouth, UK. Rakes, P.L., and J.R. Shute. 2007. Captive propagation and monitoring of rare southeastern fishes in Tennessee: 2006. Final Report for Tennessee Wildlife Resources Agency, [PROVIDE LOCATION], TN. 29 pp. Rakes, P.L., J.R. Shute, and P.W. Shute. 1992. Quarterly report for captive propagation of the Duskytail Darter (Etheostoma [Catonotus] sp.). Report to Tennessee Wildlife Resource Agency, Cherokee National Forest, National Park Service, and US Fish and Wildlife Service, Asheville Field Office, NC. 5 pp. Southeastern Naturalist 55 W.K. Gibbs, J.E. Miller, S.B. Cook, and M.A. Kulp 2014 Vol. 13, No. 1 Reynolds, J.B. 1996. Electrofishing. Pp. 221–253, In B.R. Murphy and D.W. Willis (Eds.). Fisheries Techniques, 2nd Edition. American Fisheries Society, Bethesda, MD. 732 pp. Rosenfeld, J. 2003. Assessing the habitat requirements of stream fishes: An overview and evaluation of different approaches. Transactions of the American Fisheries Society 132:953–968. SAS Institute, Inc. 2008. SAS/STAT® 9.2 User’s Guide. Cary, NC. Schmidt, C.U. 2007. Seasonal microhabitat use of the threatened Spotfin Chub, Erimonax monachus, in the Emory River watershed. M.Sc. Thesis. Tennessee Technological University, Cookeville, TN. 93 pp. Shaffer, G.P. 2004. Evaluation of Smallmouth Bass in Abrams Creek and Little River within Great Smoky Mountains National Park. M.Sc. Thesis. Tennessee Technological University, Cookeville, TN. 56 pp. Shute, J.R., P.L. Rakes, and P.W. Shute. 2005. Reintroduction of four imperiled fishes in Abrams Creek, Tennessee. Southeastern Naturalist 4:93–110. Simbeck, D.J. 1990. Distribution of the fishes of the Great Smoky Mountains National Park. M.Sc. Thesis. The University of Tennessee, Knoxville, TN. 128 pp. Soule, M.E. 1980. Thresholds for survival: Maintaining fitness and evolutionary potential. Pp. 151–169, In M.E. Soule and B.A. Wilcox (Eds.). Conservation Biology: An Evolutionary- Ecological Perspective. Sinauer Association, Inc., Sunderland, MA. 395 pp. Starnes, W.C., and D.A. Etnier. 1986. Drainage evolution and fish biogeography of the Tennessee and Cumberland rivers drainage realm. Pp. 325–361, In C.H. Hocutt and E.O. Wiley (Eds.). The Zoogeography of North American Freshwater Fishes. John Wiley and Sons, Inc., New York, NY. 866 pp. Steinberg, D., and P. Colla. 1995. CART: Tree-structured non-parametric data analysis. Salford Systems, San Diego, CA. US Fish and Wildlife Service (USFWS). 1993. Duskytail Darter recovery plan. Atlanta, GA. 25 pp. Warren, M.L., and B.M. Burr. 1994. Status of freshwater fishes of the United States: Overview of an imperiled fauna. Fisheries 19(1):6–18. Warren, M.L., B.M. Burr, S.J. Walsh, H.L. Bart, Jr., R.C. Cashner, D.A. Etnier, B.J. Freeman, B.R. Kuhajda, R.L. Mayden, H.W. Robison, S.T. Ross, and W.C. Starnes. 2000. Diversity, distribution, and conservation status of the native freshwater fishes of the southern United States. Fisheries 25(10):7–29. Wentworth, C.K. 1922. A scale of grade and class terms for clastic sediment. Journal of Geology 30:377–392. Williams, J.E., D.W. Sada, and C.D Williams. 1988. American Fisheries Society guidelines for introductions of threatened and endangered fishes. Fisheries 13:5–11. Wood, P.J., and P.D. Armitage, 1997. Biological effects of fine sediment in the lotic environment. Environmental Management 21:203–217. Zar, J.H. 1999. Biostatistical Analysis, 4th Edition. Prentice Hall, Inc., Upper Saddle River, NJ. 663 pp.