Environ Biol Fish (2011) 92:141–150 DOI 10.1007/s10641-011-9823-1 Patterns of shield darter, Percina peltata, distribution in the Eastern Piedmont of Maryland, USA Patrick Ciccotto & Scott Stranko Received: February 2010 / Accepted: April 2011 / Published online: 27 April 2011 # Springer Science+Business Media B.V 2011 Abstract Detailed analyses of habitat associations with rare species are typically constrained by limited sample size and the availability of habitat data The dense spatial coverage of stream sampling by the Maryland Biological Stream Survey provides ample data to quantitatively examine correlations between habitat and rare species distributions The shield darter, Percina peltata, has a widespread distribution on the Atlantic Slope of the United States, but is uncommon throughout its range in Maryland Associations of in situ physical habitat, water chemistry, and alterations in landscape with shield darter presence in the Eastern Piedmont physiographic province in Maryland were examined Shield darter occurrence was associated with larger sized streams in concordance with the species’ known ecology Shield darter distribution was further associated with stream segments with deep riffle habitats with diverse velocities, low concentrations of chloride and sulfate, low levels of urbanization in upstream catchments, and several pollution intolerant fish species Although P Ciccotto (*) University of Maryland Baltimore County, 1000 Hilltop Circle, Baltimore, MD 21250, USA e-mail: c76@umbc.edu S Stranko Maryland Department of Natural Resources, Monitoring and Non-Tidal Assessment Division, 580 Taylor Avenue, Annapolis, MD 21401, USA the exact mechanism of the effects is not clear, results indicate that the shield darter is sensitive to urban development and habitat and water quality alteration that typically accompanies urbanization Shield darter conservation in Maryland necessitates the protection and restoration of minimally urbanized watersheds where they are known to occur The results from this study indicate that habitat information on rare species may be important in elucidating important habitat associations that are not evident via examination of community level data Keywords Shield darter Habitat quality Urbanization Introduction The shield darter, Percina peltata, is a small (up to 90 mm) freshwater species of the family Percidae This species typically inhabits the riffles of warm, streams and rivers with substrates ranging from detritus and silt to gravel, cobble, and boulders Spawning typically occurs over gravel substrates (New 1966; Jenkins and Burkhead 1994; Schmidt and Daniels 2004).The diet of shield darter consists mainly of aquatic insects Based on its ecology, it is considered intolerant to anthropogenic disturbances (Southerland et al 2005) The species has a large Atlantic Slope distribution, ranging from the Hudson and Susquehanna Rivers in New York south to the 142 Environ Biol Fish (2011) 92:141–150 James River in Virginia in the Eastern Piedmont and onto the upper Coastal Plain (Rohde et al 1994; see inset Fig 1) In Maryland, the shield darter is currently uncommon throughout the state, ranging from lower Susquehanna River tributaries to Potomac River watersheds in the Eastern Piedmont Shield darter populations are also known to occur in watersheds of the Choptank River basin on the eastern shore of the state as well as in lower portions of Antietam Creek and Catoctin Creek in the Ridge and Valley physiographic province in western Maryland The shield darter is currently on the Watch List in Maryland (MDNR 2007) Despite the importance of this species to mid-Atlantic aquatic biodiversity (particularly Maryland), research on the effects of habitat and water quality alteration has not been conducted The family Percidae is the second most diverse family of freshwater fishes in North America Despite their diversity however, almost half of the 191 described percid species are considered imperiled (Jelks et al 2008) Threats such as pollution, fragmentation, and habitat-loss as a result of human- induced landscape alteration appear to be consistent with those facing other aquatic taxa (Allan and Flecker 1993; Burkhead et al 1997) Seven of the nine native darters in Maryland are listed on the state’s list of rare, threatened, and endangered species (MDNR 2007) Habitat degradation and expanding urbanization from the Baltimore and Washington D.C metropolitan areas threaten several of these species, including the shield darter, which in Maryland is found within close proximity to these urban centers Urbanization from Washington D.C and Baltimore has caused major anthropogenic disturbances to many of the streams within Maryland’s portion of the shield darter distribution by altering physical habitat and chemical composition of lotic systems Changes in fish communities have been widely documented to result from these alterations in urbanized watersheds with the loss of endemic species and shifts from intolerant to tolerant and/or non-native species in coincidence (Klein 1979; Weaver and Garman 1994; Wang et al 1997, 2000; Paul and Meyer 2001; Tabit and Johnson 2002; Walters et al 2003; Morgan and Cushman 2005) Fig Map of site locations where shield darter were present (black circle) and absent (white circle) used in 422 sites sampled in the Easter Piedmont from 2000–2009 in Maryland Gray lines represent the boundaries of the Eastern Piedmont, while the dashed black line represents the division between watersheds draining directly into the Chesapeake Bay and those draining into the Potomac River Environ Biol Fish (2011) 92:141–150 Rare and imperiled species have often been excluded in examinations of broad landscape-scale alterations on stream quality because of the paucity of records that can muddle statistical results However, understanding the influence of human alterations to streams on these rare species is of particular importance for the conservation of the species themselves and for efforts to maintain regional biodiversity The Maryland Department of Natural Resources, Maryland Biological Stream Survey (MBSS), has a unique opportunity by providing relatively extensive records of some imperiled fish species concomitantly with fairly comprehensive stream habitat, chemical, and landscape data The purpose of this paper is to determine correlations of shield darter presence and absences in the Eastern Piedmont physiographic province with fish assemblage, landscape alterations, water chemistry, and physical habitat data to investigate the potential influence of human-related alterations to stream conditions within the Baltimore and Washington D.C metropolitan areas on shield darter distribution in Maryland Methods We screened sites from the MBSS database based on inherent physiographic and stream size related variables to target areas where shield darters are likely to occur The purpose of screening sites by physiographic province and stream size variables was to identify potential influences on shield darter distribution that may be addressed from a management standpoint Natural phenomena that cannot be practically addressed in conservation practices (e.g., biogeography or upstream watershed areas) were not included Water chemistry, habitat quality, and landscape alterations are potential influences on shield darter distribution that are addressable and were thus the focus of our second group of analyses After the potentially confounding effects of physiographic province (Utz et al 2009) and stream size were removed, water chemistry, fish habitat, and fish assemblage data were used to investigate the potential influences on shield darter distribution We limited our analyses to the Eastern Piedmont because 1) it is the “center” of shield darter distribution within the state (i.e., contains the largest proportion of sites where shield darter have been 143 observed) and 2) this area is undergoing an increase in urban development, which is hypothesized to be detrimental to shield darter populations Fishes were sampled from 422 randomly-selected stream sites 75 m in length from 2000–2009 in the Eastern Piedmont to provide for the most statistically rigorous methodology of assessing the patterns in shield distribution Due to the random selection of stream sites, several stream reaches or watersheds were subject to multiple sampling events We treat these as independent samples however, based on the generally limited dispersals observed in stream fishes, particularly in some adult Percina species (Freeman 1995; Warren and Pardew 1998; Skyfield and Grossman 2008) Fish collections at these sites consisted of double-pass electrofishing at low flow conditions during June–September To minimize fish movement into and out of the sites, block nets were placed at the upstream and downstream ends of the segments Fish species were identified in the field before being released We focused on larger sites (i.e., streams with wider, deeper habitats and higher stream flows) where shield darters are typically observed to best explain patterns of shield darter occurrence with anthropogenic effects without the potentially confounding influence of stream size Empirical observations of shield darters in larger streams in Maryland are consistent with observations of this species throughout its range (Jenkins and Burkhead 1994; Rohde et al 1994) To quantify stream size, five measurements were taken in the 75 m site Average thalweg depth (the deepest part of each transect), average thalweg velocity (water velocity at the deepest part of each transect), and average wetted width were averages of four wetted measurements taken at 25 m intervals on the 75 m sites Maximum depth was the deepest area found anywhere in the 75 m site Stream discharge followed U.S Geological Survey (USGS) methods (Rantz 1982) The area of the upstream site catchment was also calculated to quantify stream size for each individual site Site catchments were hand digitized based on USGS quarter quad topographic lines and their areas were calculated in ArcMap 9.1(ESRI 2005) Sites were categorized as shield darter present or absent and we conducted a principal component analysis (PCA) on the stream-size variables For significant axes (where the eigenvalue was greater than the broken-stick eigenvalue, see McCune and 144 Environ Biol Fish (2011) 92:141–150 Grace 2002) we calculated 95% confidence intervals around the mean axis scores of sites where shield darter was absent in ordination space These confidence intervals were used to eliminate sites where shield darter are expected to be absent by removing all sites below the confidence interval limit The remaining larger-sized sites were then analyzed to identify patterns of shield darter occurrences with fish community, water chemistry, physical habitat, and landscape variables Water chemistry data were based on laboratory analyses of grab samples, and included measurements of pH, conductivity, dissolved organic carbon, chloride, sulfate, total nitrogen, total phosphorous, orthophosphate, ammonia, nitrate, and nitrogen dioxide Dissolved oxygen concentrations were measured in situ with a Hydrolab Quanta Physical habitat variables were measured using MBSS protocols (Kazyak 2001) Guidelines for physical habitat scores are reported in Table Instream habitat, pool/glide/eddy quality, and riffle/run quality were each rated on a 0– 20 scale using well-defined, quality controlled visual assessments Instream habitat scores were assigned based on perceived value of habitat to supporting the fish community Sites with a variety of habitat types and substrate particle sizes received higher instream habitat scores Pool/glide/eddy quality scores were assigned based on the variety and complexity of slowor still-water habitat at a site Higher scores were assigned to segments with undercut banks, woody debris, and other potential types of cover for fishes Riffle/run quality scores were assigned based on riffle/run depth, complexity, and importance as habitat structure for aquatic fauna Segments with deeper riffle/run areas, stable substrates, and a variety of water current velocities received the highest scores Land cover (percent urban, percent agriculture, and percent forest) was calculated for all hand digitized site catchments from the USGS 2001 National Land Cover Dataset (Homer et al 2007), such that this measurement represents all land area draining to each site from upstream In addition to the above habitat variables, we wanted to compare the fish assemblage compositions of sites where shield darters were present or absent For each site, we compiled total species richness and abundances from the electrofishing surveys Fish species that were found in less than 5% of these sites were omitted from the analysis to limit the effects of rare species on the ordination An outlier analysis using the Sorenson distance among the sites was also used and identified one site with an average distance 3.8 standard deviations away from the grand mean distance based on fish data This site was removed from further analysis We removed rare fish and outliers in order to reduce “noise” in the statistical analyses (Reynoldson and Rosenberg 1996; Rodriguez and Lewis 1997) More specifically, we did not want rare fishes or sites with outlying fish communities to skew the ordinations that would skew otherwise important patterns useful to the management of our focal species, the shield darter Water chemistry and physical habitat data were log10(x+1) transformed, while land cover data underwent an arsine square root transformation Sites were again grouped as present or absent for shield darter to assess Table Physical habitat assessment guidelines used at sampling sites (see Kazyak 2001) Habitat Optimal parameter 16–20 Instream Habitat Pool/ Glide/ Eddy Quality Sub-optimal 11–15 Greater than 50% of a variety of cobble, 30–50% of stable boulder, submerged logs, undercut banks, habitat; adequate snags, rootwads, aquatic plants, or other habitat stable habitat Complex cover and/or depth >1.5 m; both Deep (>0.5 m) areas deep (>0.5 m) and shallow (10 cm, with Run maximum depth greater than 50 cm; Quality substrate stable and variety of current velocities Riffle/run depth generally 5–10 cm, variety of current velocities Marginal 6–10 Poor 0–5 10–30% mix of stable habitat; habitat availability less than desirable Shallows ([...]... df =2, p1% of the total average gut content across the sampled populations The sum of the remaining food types are represented as ‘sum of other’ See Table 1 for description of the names... western) of each of the two bays (12 total sites): near the mouth of the bay, approximately mid-way into the bay, and near the head of the bay (inner bay) We also surveyed the channel in front of the town of Papetoai (west of Opunohu Bay) and on the forereef slope between the two bays At all sites surveys were conducted near the surface by snorkel Divers on SCUBA swam transects at a depth of 5 m (all... between the size of nocturnal and diurnal smolts In all three periods nocturnal smolts were significantly smaller than diurnal smolts (Fig 2) A significant interaction term between time of day and period (F2,10396 =3 .2, p=0.042) was due to the slightly larger mean size of diurnal smolts in P1 as opposed to P2, and the increasing mean size of nocturnal smolts with period (Fig 2) The mean sizes of nocturnal... their short duration (average of 5–7.5 h) In particular, several species described as monogamous were also described as non-territorial Roberts and Ormond (1992) predicted that longer focal-samples of pairs would yield observations of territorial behavior in some of these species Environ Biol Fish (2011) 92:167–179 There have been few recent studies of monogamy in butterflyfishes, particularly in non-corallivores,... Environ Biol Fish (2011) 92:151–157 Alerstam T, Hedenstrom A (1998) The development of bird migration theory J Avian Biol 29:343–369 Baker R (1978) The evolutionary ecology of animal migration Holmes and Meier Publishers Inc, Teaneck, 1024 pp Bakshtansky EL, Nesterov VD (1976) Hunting activities of pike and their possible effect on the diurnal pattern of downstream run of fingerlings of Atlantic salmon... July 2006 A pair of divers located either an individual or pair of C trichrous The divers waited 5 min to allow the fish to habituate to their presence Both divers then focused on the same fish Fig 1 Distribution of C trichrous by size within the bays of Moorea (p