Illuminating hotspots of imperiled aquatic biodiversity in the southeastern US

Illuminating hotspots of imperiled aquatic biodiversity in the southeastern US

Global Ecology and Conservation 19 (2019) e00654 Contents lists available at ScienceDirect Global Ecology and Conservation journal homepage: http://...

3MB Sizes 0 Downloads 17 Views

Global Ecology and Conservation 19 (2019) e00654

Contents lists available at ScienceDirect

Global Ecology and Conservation journal homepage:

Illuminating hotspots of imperiled aquatic biodiversity in the southeastern US Duncan Elkins a, *, Sarah C. Sweat b, Bernard R. Kuhajda b, Anna L. George b, Katie S. Hill a, Seth J. Wenger a a b

University of Georgia River Basin Center, 203 D.W. Brooks DR, Athens, GA, 30602-5017, USA Tennessee Aquarium Conservation Institute, 175 Baylor School Road, Chattanooga, TN, 37405-2506, USA

a r t i c l e i n f o

a b s t r a c t

Article history: Received 7 November 2018 Received in revised form 6 May 2019 Accepted 6 May 2019

The southeastern United States is a global hotspot for aquatic biodiversity but has relatively little land under protection. To guide conservation investment in this region, we developed a prioritization of watersheds to highlight areas of exceptional conservation potential. Using range maps for 1043 fish, mussel, and crayfish species derived from museum and research collections as well as state natural heritage databases, we derived a conservation priority score for each watershed that incorporates biodiversity, endemism, and imperilment. The highest-scoring watersheds are primarily in the Middle Tennessee River and eastern Mobile River systems. On average, only 1.5% of the area of the topscoring watersheds is currently protected. Further, average federal spending under the Endangered Species Act is 2.1e46.6 times greater on aquatic species found outside the region than on southeastern endemics. These results provide the foundation for systematic freshwater conservation planning in the Southeast and argue for increased investment in innovative preservation strategies in this globally significant region. © 2019 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (

Keywords: Fish Mussel Crayfish Prioritization

1. Introduction The southeastern United States is a global hotspot for freshwater biodiversity (Abell, 2000; Collen et al., 2014). Approximately one-third the world's crayfish species (Richman et al., 2015) and almost 40% of the world's freshwater mussel species are found in the Southeast, including 91% of US mussel species (Neves et al., 1997; Graf and Cummings, 2007). Of the 831 freshwater fishes in the U.S. and Canada (Page and Burr, 2011), over 550 (79%) are found in the Southeast. These aquatic communities also include highly diverse assemblages of salamanders, turtles, and aquatic insects such as mayflies and caddisflies. Unfortunately, the southeastern landscape has been extensively altered by human activities over the past 150 years, and these modifications have taken a toll on aquatic species (Benz and Collins, 1997). A 2000 assessment of southeastern fishes determined that 28% of southeastern fishes were imperiled and noted that this represented “a 75% increase in jeopardized southern fishes since 1989 and a 125% increase in 20 years” (Warren et al., 2000). This trend in imperilment continues; in just eight years an additional 13 newly recognized taxa of southeastern fish species were considered imperiled and the

* Corresponding author. E-mail addresses: [email protected] (D. Elkins), [email protected] (S.C. Sweat), [email protected] (B.R. Kuhajda), [email protected] (A.L. George), [email protected] edu (K.S. Hill), [email protected] (S.J. Wenger). 2351-9894/© 2019 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license ( licenses/by-nc-nd/4.0/).


D. Elkins et al. / Global Ecology and Conservation 19 (2019) e00654

conservation status of 25 imperiled species declined, with only 11 species improving (Jelks et al., 2008). Other taxa may be even more imperiled; the extinction rates for gastropods and mussels exceed that of fishes (Johnson et al., 2013; Haag and Williams, 2014). However, conservation planning for these species has been complicated by a limited understanding of the natural history of many taxa (Johnson et al., 2013), ongoing revisions of systematics (e.g., Layman and Mayden, 2012), and an incomplete picture of spatial priorities. The most recent region-wide prioritization based on best-available species data was conducted by The Nature Conservancy in 2002 (Smith et al., 2002). While there are notable state-level data-driven prioritizations (e.g., Alabama's strategic habitat units; and the Southeastern Aquatic Resources Partnership maintains a region-wide fish richness map (, there has been no comprehensive analysis of the spatial pattern of aquatic biodiversity in the region in over a decade. A region-wide prioritization of diversity-rich watersheds would allow resource agencies to efficiently conserve aquatic species at an ecosystem-level rather than focusing on species-specific protection (Williams et al., 1989; Noss et al., 1995). In this study, we assembled a region-wide database of aquatic organism distributions using datasets of over 1000 fish, crayfish, and mussel species. We formed committees of experts to validate the range maps of each species using published distributions and professional experience. We assessed the level of endemism for each species and combined this with current assessments of imperilment to create a priority score for each of 290 watersheds from southern Louisiana to southern Virginia. We then compared prioritization scores with the degree of land protection in each watershed to assess the current state of aquatic conservation across the region. We also estimated the amount of government conservation spending on federally listed aquatic species and compared it to the amount spent on taxa outside of the region. Our objectives were (1) to identify the highest aquatic conservation priorities in the Southeast, and (2) assess the degree to which these areas are currently protected. 2. Methods We defined the project area using a combination of geographic and biogeographic boundaries based upon the distribution of fishes, the best understood freshwater taxon in the region (Burr and Warren, 1986; Swift et al., 1986; Jenkins and Burkhead, 1994). (Fig. 1). The area encompasses all or part of 32 USGS 4-digit hydrologic unit code (“HUC-4”) basins (Fig. 1), which include eight Atlantic-slope drainages, nine that flow directly into the Gulf of Mexico, five tributaries to the Ohio, four basins within the Tennessee River System, and five that drain into the Lower Mississippi River. These are subdivided into 290 “HUC8s” (technically termed “sub-basins,” but here we use the common-language term “watersheds”), which constitute our primary unit of analysis. Our overall strategy and analysis were developed in consultation with an advisory committee of regional experts (see Acknowledgements), with whom we met throughout the project. To identify the watersheds which, if protected and restored, would contain the highest biodiversity of native aquatic organisms in the Southeast, we compiled datasets of field observations from university researchers, museums, state agencies, and online databases derived from these sources (see fish, crayfish, and mussel sections for full list of data sources). We found that the number and distribution of observations were sufficient to build maps for fishes, crayfishes, and mussels, but not for other invertebrates such as aquatic snails. We excluded amphibians because speciation and the resultant distributional patterns in this group often do not follow traditional watershed boundaries due to their ability to use terrestrial habitat, therefore our analysis might be confounded. Most of the observations consisted of point records, reflecting one survey at a specific time, but some agencies provided us with polygon coverages, reflecting broader areas in which a particular species has been collected at least once. Polygon coverages were more typical for imperiled species. We aggregated all point and polygon collection data by watershed (HUC-8). This resulted in species range maps covering 290 planning units for the Southeast with an average size of 3500 square kilometers (1351 square miles) each. For all taxonomic groups, we only included native species. We included undescribed species if they were recognized in the literature (published papers, books, State Wildlife Action Plans) and there was information available on their distribution and imperilment status. We did not include species known to be extinct but retained historic records of species thought to be currently extirpated in the watershed, on the assumption that re-introduction from another population could be possible. Where possible, we excluded introduced ranges. Species which had their entire range within the 290 HUC-8 sub-basin area were classified as southeastern endemics. Predictably, many of the original records contained errors, either spatial or taxonomic. We organized and corrected obvious errors (e.g., spelling, west/east longitude confusion) in the raw data and produced draft maps for all species. Further corrections were made by other team members, advisory board members, and other experts, as described in the subsequent sections. 2.1. Fishes Fish data were downloaded from Multistate Aquatic Resources Information System (MARIS), FishNet2, and the Global Biodiversity Information Facility (GBIF). Aggregated fish data were vetted by species and HUC-8 sub-basins by B. Kuhajda using published “Fishes of” state books, online atlases, or primary literature for recently described species (a list of the references consulted is provided in Appendix 1). As a group, fishes are the best-studied freshwater taxon in the Southeast, both with regard to taxonomy and distribution, with numerous distributional references at the county, state, and drainage levels.

D. Elkins et al. / Global Ecology and Conservation 19 (2019) e00654


Fig. 1. River drainages of the Southeast included in the analysis.

We were therefore able to rely on published sources rather than convening a panel of experts to resolve taxonomic and distributional uncertainties (which was necessary for crayfishes and mussels). We assigned imperilment status for fish species using the ranks in Jelks et al., 2008, modified in some cases for new taxonomy or where an updated assessment was available. Fish taxonomy is current as of August 2017.

2.2. Crayfishes We contacted southeastern astacologists beginning with those who had attended the 2015 symposium “Conservation, Ecology, and Taxonomy of Southeastern Crayfish” at the annual meeting of the Southern Division of the American Fisheries Society in Savannah, Georgia, and asked if they had relevant datasets of crayfish distributions that they would be willing to have aggregated for this project. In some cases, they referred us to another researcher or a museum database. Ultimately, we received polygon or point data from 17 sources (Table 1), including one query of the GBIF online database for records from the Florida Museum of Natural History and one query covering most of Georgia from the Smithsonian Museum's database, which returned records that we manually georeferenced using road and stream intersections (see Appendix 1 for query details and URLs).


D. Elkins et al. / Global Ecology and Conservation 19 (2019) e00654

Table 1 Astacologists who contributed data or reviewed crayfish distribution maps. Name


Provided Data

In-Person Review

Email Review

Susie Adams Tyler Black Chris Skelton Arnie Eversole Bob Jones Zach Loughman Guenter Schuster Chris Taylor Roger Thoma Bronwyn Williams Carl Williams David Withers Geological Survey of AL IL Natural History Survey KY Department of Fish & Wildlife Resources Jeff Simmons Smithsonian NMNH GBIF Stuart McGregor Rebecca Bearden

USFS NC Wildlife Resources Commission HNTB Corporation Clemson Univ. MS Museum of Natural Science West Liberty Univ. Eastern KY Univ. (retired) IL Natural History Survey Midwest Biodiversity Institute NC Museum of Natural Sciences TN Wildlife Resources Agency TN Department of Environment and Conservation




TN Valley Authority

Geological Survey of AL Geological Survey of AL








We convened a meeting in Chattanooga, TN, on June 1 and 2, 2016, that was attended by most of the researchers who had provided data. In this meeting, we reviewed HUC-8 level range maps generated by the combination and aggregation of the input datasets. This initial list included cave species and species not classified as primary burrowers, and it contained some species with unclear or disputed taxonomy. The group corrected taxonomic and geographic errors and assigned southeast endemism for most species, although approximately twenty species were flagged for further review by individuals not at the meeting or where a more extensive literature search was required. These maps were subsequently corrected via email communications. On the advice of the crayfish committee, we excluded most species classified as “primary burrowers,” which tend to be terrestrial, with the exception of a small number that the group considered stream-dependent. The excluded species constituted approximately 15% of described species, including almost a third of those with “critically imperiled” conservation status (Welch and Eversole, 2006). We assigned crayfish imperilment ranks based on consultation with Chris Taylor of the Illinois Natural History Survey, who maintains an updated list based on the most recent American Fisheries Society crayfish status assessment (Taylor et al., 2007). Crayfish taxonomy is current as of December 2016. 2.3. Mussels Museum records were the primary source of mussel point locations. We requested all mussel records for the study area or queried the online databases of the Ohio State University Museum of Biological Diversity, the North Carolina Museum of Natural Science, and the Mississippi Museum of Natural Science. We also obtained the state databases for Alabama, Kentucky, and Georgia. All contributors are listed in Table 2. These point records (HUC-12 polygons for Kentucky) were aggregated and species range maps were produced as for fishes. We employed an expert-opinion approach, emailing collections of range maps to malacologists with regional expertise (Table 2) who assigned endemism and delivered corrected maps in writing or over the phone. Most areas were assigned to more than one reviewer, and conflicts were rare. However, this process was not as thorough as the multi-party discussion that occurred within the crayfish review group, so we convened a meeting in Chattanooga, TN, on Feb 15e16, 2017, for an additional session of in-person review as described for the crayfish panel. Mussel imperilment scores used during this session were drawn from an in-press distribution and imperilment appendix for mussels from Jim Williams, developed for the Freshwater Mollusk Conservation Society. Mussel taxonomy is current as of July 2017. 2.4. Richness, endemism and imperilment scores We calculated species richness for fishes, crayfishes, and mussels for each watershed. We then calculated two other scores for each watershed: endemism and imperilment. Endemism was based only on species whose entire range fell within the project area. We derived an endemism score for each of these, calculated as the reciprocal of the number of watersheds in which it occurred. Thus, a narrow endemic which occurred in a single watershed received a score of 1/1 (1), while a more widely-distributed species occurring in 10 watersheds received a score of 1/10 (0.1). We summed endemism scores of all the fish, crayfish, or mussel species in each watershed.

D. Elkins et al. / Global Ecology and Conservation 19 (2019) e00654


Table 2 Malacologists who contributed data or reviewed mussel distribution maps. Name



In-Person Review

Jeff Garner Stuart McGregor Jason Wisniewski Bob Jones Art Bogan Jim Williams Bob Butler Wendell Haag Jess Jones Don Hubbs Brian Watson Paul Hartfield Gerry Dinkins

AL Department of Conservation & Natural Resources Geological Survey of AL GA Department of Natural Resources MS Museum of Natural Science NC Museum of Natural Sciences Florida Museum of Natural History US Fish & Wildlife Service US Forest Service VA Department of Fish and Wildlife Conservation TN Wildlife Resources Agency Virginia Dept. of Game and Inland Fish US Fish and Wildlife Service University of Tennessee KY Department of Fish and Wildlife Geological Survey of AL GA Department of Natural Resources Ohio State University Museum of Biological Diversity




Email Review



While endemism is a strong correlate of imperilment, some widely-distributed taxa are also considered imperiled (Angermeier, 1995; Warren et al., 1997). We devised a simple score for taxon imperilment by assigning 3 points for each endangered species, 2 points for each threatened species, and 1 point for each vulnerable species. These categories were assigned based on current assessments from the literature or relevant scientific society, not federal status under the Endangered Species Act. For fishes we used a southeastern U.S. and North American list of imperiled freshwater and anadromous species (Warren et al. (2000) and Jelks et al. (2008) as well as a literature survey by one of the authors (BRK); for mussels we used Williams et al. (1993) and updates from Jim Williams (Pers. Comm.); for Crayfishes we used Taylor et al. (2007) and updates from Chris Taylor (Pers. Comm.). A list of our imperilment scores for all species is included as Appendix 2. We then summed imperilment scores for all species in each watershed. We also calculated a “conservation prioritization score.” We converted each of the individual scores (richness, imperilment, and endemism) into a common scale (0e1) and then summed the scores by watershed. We acknowledge that this aggregation method is simple; many other approaches are reasonable.

2.5. Protected areas and conservation spending We derived a GIS coverage in ArcGIS v10.3 (ESRI, 2011) of protected areas in the Southeast using the Protected Areas Database of the US (PAD-US) version 1.4 (US Geological Survey 2016), which contains an attribute field for the International Union for the Conservation of Nature (IUCN) protected area status (Dudley, 2008) for most polygons. We processed this database to remove overlapping polygons and assigned the highest level of protection for any overlapping areas. We then performed a union between the protected areas polygons and the watershed outlines and tabulated the areas under each protection type within every watershed. Since some of the IUCN categories are based on cultural rather than ecological designations, we summed the area under what we judged to be strict protection, including polygons with IUCN categories Ia and Ib describing “Strict Nature Reserves” and “Wilderness Areas”, IUCN category II for “National Parks,” and category IV designating “Habitat/species management protected areas.” This strict definition also included areas designated category VI, for “protected area with sustainable use” and areas with a “Not Reported” IUCN status that “meet the definition of IUCN protection (i.e. GAP Status Code 1 or 2) but where IUCN Category has not yet been assigned and categorical assignment is not appropriate” (US Geological Survey 2016). Together, these areas covered just 4.23% of the project area. Since the total area of protected polygons in the Southeast is relatively small and we reasoned that even parcels designated for cultural reasons, e.g., landmarks or battlefields, would be subject to some sort of environmental review prior to development, such as an analysis under the National Environmental Policy Act, we performed a similar spatial analysis using comprehensive protection criteria that included all categories, including those labeled “unassigned” and “other” in PAD-US. The total area protected under this comprehensive designation was 16.05% of the project area. To calculate the amount of federal and state spending on federally listed species, we compiled expenditures available for fiscal years 2012-2016 (US Fish and Wildlife Service, 2012; 2013, 2014, 2015, 2016) for freshwater species found completely inside of the southeastern United States (as defined above) and compared to species found completely outside of this region. Species found in both areas were not included in the comparisons.


D. Elkins et al. / Global Ecology and Conservation 19 (2019) e00654

3. Results Our final database included 589 fish species, 234 mussel species, and 221 crayfish species. A ranked table of the standardized richness, endemism, and imperilment scores for all 290 watersheds appears as an online research data supplement. The overall distribution of standardized scores for species richness differed somewhat from those for imperilment and endemism (Fig. 2). Here, we present the top-scoring watersheds for each element of our priority index.

3.1. Species richness We found that overall species richness is highest in the Lower Tennessee River system (Lower Tennessee, Middle Tennessee-Elk HUC-4s), but is also high in the Green River, the Alabama River, portions of the Cumberland River, and the Pearl River in Mississippi and Louisiana (Fig. 3A). Fish richness tends to be high in the same areas, particularly the Middle/Lower Tennessee and Alabama systems (Fig. 3B). Mussel richness is high in the Middle and Lower Tennessee systems along with the Green and Cumberland systems, and relatively low across the Atlantic drainages (Fig. 3C). Crayfish species richness is patchier (Fig. 3D), though the Middle Tennessee, Tombigbee, Barren, and Pascagoula systems score high.

3.2. Endemism The areas of highest endemism are the Middle Tennessee River system, followed by the Alabama River, the Upper Tennessee, and Cumberland River systems (Fig. 4A). For fishes, the areas of highest endemism include the Alabama, Middle Tennessee-Elk, and Green River systems (Fig. 4B). The Alabama River system is also the area of highest mussel endemism, followed by the Cumberland and the coastal plain systems of the Apalachicola and Choctawhatchee-Escambia river systems

Fig. 2. Histogram of standardized values for species richness, endemism, and imperilment for the 290 project watersheds.

D. Elkins et al. / Global Ecology and Conservation 19 (2019) e00654

Fig. 3. Normalized Species Richness scores for all taxa (A) and group richness scores for fishes (B), mussels (C), and crayfishes (D).

Fig. 4. Normalized Southeast endemism score for all taxa (A) and group richness scores for fishes (B), mussels (C), and crayfishes (D).



D. Elkins et al. / Global Ecology and Conservation 19 (2019) e00654

(Fig. 4C). Crayfish endemism is highest in the Middle Tennessee-Elk, Choctawhatchee-Escambia, and Cumberland river systems (Fig. 4D). 3.3. Imperilment Imperilment is strongly concentrated in the greater Tennessee River System, with nine of the top ten watersheds belonging to the Middle Tennessee-Elk or Upper Tennessee HUC-4s and the remaining watershed in the Cumberland River System (Fig. 5A). For fishes, the watersheds highest in imperilment were concentrated in the Alabama, Upper Tennessee, and Middle Tennessee-Elk river systems (Fig. 5B). The Middle Tennessee-Elk was also the watershed with the highest imperilment scores for mussels and crayfishes, followed by the Upper Tennessee and Cumberland river systems for mussels (Fig. 5C), and the Pascagoula, Choctawhatchee-Escambia, and Mobile-Tombigbee systems for crayfishes (Fig. 5D). 3.4. Overall priority The two watersheds with the highest overall priority scores are Wheeler Lake and Pickwick Lake, both in the Middle Tennessee-Elk system, with additional high-ranking areas in the Upper and Lower Tennessee, Alabama, and Cumberland river systems (Fig. 6A). Twenty-seven of the top 30 watersheds fall within the drainages of the Tennessee (14), Alabama (10), or Cumberland rivers (3), with the remainder coming from the Green (2) and Mobile-Tombigbee (1). Fishes (6B), mussels (6C) and crayfish (6D) all had their highest priority watersheds in the Middle-Tennessee-Elk, although the groups showed different patterns of secondary priorities. The watershed scores for overall and taxa-specific priorities are available as an online data supplement, summarized for the region or by HUC4 basin. 3.5. Protected areas in priority watersheds We found no correlation between the percentage of a watershed under protection (Top 30 watersheds shown in Table 3. Data for all watersheds is available at (, 1), calculated using either strict (Pearson's r ¼ 0.06) or comprehensive criteria (r ¼ 0.06), and its conservation priority. Three watersheds have more than 20% of their area protected under the strict criteria: the Tuckaseegee and Pigeon watersheds, both in the Upper Tennessee River System, and Bulls Bay watershed, in the Edisto-Santee River System. The overall

Fig. 5. Normalized imperilment scores for all taxa (A) and group richness scores for fishes (B), mussels (C), and crayfishes (D).

D. Elkins et al. / Global Ecology and Conservation 19 (2019) e00654


Fig. 6. Overall priority score by watershed for all taxa (A), fishes (B), mussels (C) and crayfishes (D). The maximum possible score for any watershed is 3.

priority rankings for these three watersheds are 184, 178, and 286, respectively, out of 290. The top 10% of watersheds based on overall priority score average 1.5% of their area under protection, with 22 having less than 1% of their area protected. The most protected of this top 10% is the Lower Little Tennessee watershed (ranked #25, overall), in the Upper Tennessee River System, which is 18.1% protected. We found similar trends using the comprehensive protection criteria. The mean percentage under protection in the top 10% of watersheds is 10.0%, with 10 having less than 5% of their area under protection using the comprehensive protection definition. Again, the Lower Little Tennessee River watershed scores highest with 47.7% of its area under protection (mostly within National Forests).

3.6. Conservation spending Federal and state expenditures on federally listed aquatic species in the United States over five fiscal years (USFWS, 2012, 2013, 2014, 2015, 2016) showed lower spending on species found solely within the Southeast versus those found solely outside of the region. We found that federally listed southeastern freshwater mussels represented 84.4% of total mussel diversity but only received 72.5% of the funding, corresponding to 2.1 times more funding for mussel species outside of the Southeast. Freshwater crustaceans in the Southeast represented 19.3% of overall diversity but only received 4.2% of the funds, being outspent 7.0 times by funding for crustaceans outside of the southeastern United States. Finally, southeastern freshwater dependent fishes represented on average 29.1% of overall diversity but only received 0.9% of funding. This means that fishes outside of the Southeast received 46.6 times more funding per species than those within the Southeast.

4. Discussion We found that the rich freshwater biodiversity of the Southeast is highly concentrated in portions of the Tennessee, Alabama and Cumberland River systems. These systems also rise to the top of our priority lists for endemism and imperilment, although there are at least some local endemics in most river systems. Fishes, crayfishes, and mussels all reflected this overall pattern, despite some substantial differences in secondary priorities. For aquatic conservation purposes, the clear priorities are within the Tennessee, Alabama, and Cumberland systems, led by the Pickwick and Wheeler Lake watersheds that straddle the Tennessee/Alabama state line.


D. Elkins et al. / Global Ecology and Conservation 19 (2019) e00654

Table 3 The thirty highest-scoring watersheds in the project area, with ranks for individual taxa. See methods for protected area category definitions. Overall Rank Priority Sum HUC8 Name

HUC4 Name

Fish Crayfish Mussel Total Protected Strictly Protected Rank Rank Rank Area (%) Area (%)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

Mid. Tennessee-Elk Mid. Tennessee-Elk Up. Tennessee Low. Tennessee Alabama Cumberland Cumberland Alabama Alabama Cumberland Green Low. Tennessee Up. Tennessee Alabama Green Low. Tennessee Mid. Tennessee-Hiwassee Up. Tennessee Mid. Tennessee-Elk Alabama Alabama Up. Tennessee Alabama Mid. Tennessee-Elk Up. Tennessee Alabama Up. Tennessee Mobile-Tombigbee Alabama Alabama

9 1 2 6 3 25 19 11 4 16 10 15 13 8 18 40 29 33 139 17 5 56 22 35 14 12 51 27 36 20

2.87 2.83 2.15 2 1.96 1.86 1.77 1.76 1.74 1.74 1.72 1.69 1.69 1.69 1.68 1.67 1.66 1.64 1.62 1.62 1.59 1.58 1.51 1.49 1.47 1.44 1.44 1.43 1.4 1.4

Wheeler Lk. Pickwick Lk. Up. Clinch, Tennessee, Virginia Low. Duck Cahaba Caney Up. Cumberland-Lk. Cumberland Mid. Coosa Etowah South Fork Cumberland Barren Up. Duck Watts Bar Lk. Low. Coosa Up. Green Low. Tennessee-Beech Mid. Tennessee-Chickamauga Powell Guntersville Lk. Low. Alabama Conasauga Holston Coosawattee Up. Elk Low. Little Tennessee Low. Tallapoosa Low. Clinch Mid. Tombigbee-Lubbub Up. Alabama Mid. Alabama

1 2 168 43 22 16 65 50 73 35 24 80 161 118 37 11 17 58 9 15 109 156 28 74 180 27 132 12 26 69

2 1 6 13 17 5 9 3 40 12 32 14 11 10 20 23 18 8 7 22 26 4 38 16 28 77 15 34 25 41

6.2 3.4 10.9 2.1 11.8 6.2 7.6 13.2 9.9 30.2 0.2 4.4 13 6.6 3.2 3 6 7.4 7.2 1.6 24.2 4.4 11.2 5.4 46.5 1.4 0.6 1.3 1.8 2

2.4 0.9 3.1 0.5 0.3 3 1.5 2.9 5.5 9 0 1.9 10.1 0 2.7 1.6 1.6 3.3 2.8 0.4 7.7 0.4 2.8 0.1 19.9 0.1 0 0.6 0.2 0

We were surprised that the Pickwick Lake and Wheeler Lake HUCs topped the prioritization rankings, as these are not often cited as biodiversity hotspots. The eponymous “lakes,” are, in fact, reservoirs, but both watersheds encompass diverse habitats from caves and springs to larger tributaries. Though the reservoirs have eliminated nearly all of the mainstem habitat, a remarkable number of species persist in the large tributaries, such as Shoal Creek, Cypress Creek, and the Paint Rock River. In reviewing the literature on habitat associations for each of the fish species in these basins, we found that only 6.7% of the Wheeler fishes and 6.3% of the Pickwick fishes were migratory or primarily mainstem species, whereas 55% and 62% of the species were restricted to tributaries, with the remainder (38.3% and 31.7%) found in both habitats. The overall high rankings of these two watersheds are due to a combination of high species richness (across all taxa) combined with high endemism. Fourteen species are endemic to one or both of these watersheds. Seven are found only in caves and two are endemic to springs, both habitats a product of the underlying karst geology of the region. Two more species are restricted to headwater habitats. Further, we examined the fish fauna of the top six watersheds to assess the effects of our decision to include extirpated species in our watershed scores. Documented extirpations ranged from 0 in the Lower Duck River and Caney Fork to 11 in Pickwick Lake, or 0e7.7% of fish species, but this drops to 0e5.6% once reintroductions of Acipenser fulvescens, Erimoxax monacha, and Etheostoma wapiti are included. Mussel extirpations have undoubtedly been more extensive, as are current propagation and reintroduction efforts (Haag and Williams, 2014; Bouska et al., 2018), particularly in some of the areas of our highest priority (Smith et al., 2017). While we argue that conservation is almost always a sounder investment than restoration, the utility of propagation, augmentation, and reintroduction as a short-term strategy (McMurray and Roe, 2017) justifies the inclusion of extirpated species over the majority of our project area. We chose to develop a combined prioritization for fishes, mussels, and crayfishes under the assumption that management activities that would protect or restore aquatic habitat would benefit all three, but it should be noted that there is substantial variability in the areas that appear to be most important for the conservation of each group. While Pickwick and Wheeler Lakes are top priorities for all three taxonomic groups, many of the other top-priority watersheds are relatively low priorities for either crayfishes, mussels or fishes. The highest priority areas for fishes and mussels are more similar to each other than either is to the priority areas for crayfishes, perhaps due to the coevolutionary relationships between the groups (i.e., all native North American mussels rely on fish hosts in early life stages). However, these differences mean that conservation targeted at the top priority for one group might not be sufficient to arrest the decline of other groups in the Southeast. This may be particularly true for gastropods, which were excluded from our analysis.

D. Elkins et al. / Global Ecology and Conservation 19 (2019) e00654


Although management decisions are often made at finer scales, we judged the HUC-8 watershed to be an appropriate scale for aggregation to minimize discontinuous distributions resulting from uneven sampling. It should be noted that biogeographic patterns for other taxa may not align exactly with our representation of a southeastern fauna for fishes. Species characterized as “southeastern crayfishes,” in particular, might reasonably extend into portions of Louisiana and Arkansas. We did not anticipate being able to develop a crayfish layer for the entire region when we set the project boundaries and acknowledge that this may impose a downward bias on the crayfish endemism scores for sub-basins in western Mississippi and western Tennessee. Our prioritization is a multi-criterion optimization of species counts and scores. Such optimizations are inherently sensitive to bias in the underlying scores and weighting criteria used to combine them (Hermoso et al., 2016). We acknowledge that other approaches would be defensible, particularly with respect to the weighting of imperilment scores. However, our results seem to be robust to small changes in the weighting criteria, at least with respect to the top 10% of watersheds: a ranking that prioritizes imperiled species conservation by doubling the weight of the imperilment score would capture 28 of top 29 presented here, albeit in a different relative order, while a ranking that gives double-weight to the endemism score would capture 26 of the 29. Our analysis also excluded consideration of threats, due in part to the difficulty in assembling a uniform threat coverage over such a large and diverse area, and the potential uncertainty in appropriately weighting these stressors. Adding a threat layer to the analysis might lead to alternative conservation priorities. We also found that existing protected areas do not reflect these patterns of biodiversity and conservation need. While others have concluded that existing protected areas are insufficient for comprehensive freshwater conservation (Lawrence et al., 2011; Long et al., 2012; but see Britton et al., 2017), it is striking how little correspondence we see between the priority areas in this analysis and the amount of land protected, particularly in the top priority watersheds such as Wheeler Lake (2.4% protected under the strict criteria) and Pickwick Lake (<1% protected.) Furthermore, existing protected lands in the region are often negatively affected by adjacent land uses (Long et al., 2012) and do not encompass the full range of habitats within basins (Thieme et al., 2016). We do note that a few high-value areas are relatively well-protected: The Upper Tennessee River encompasses not only much of the Great Smoky Mountain National Park but also the Little Tennessee River Native Fish Conservation Area (Williams et al., 2011). While expansion of the existing formally protected areas should not be abandoned (Lindenmayer et al., 2018), the cost and difficulty of augmenting this network is substantial in an area as settled and developed as the Southeast. Acquisition will need to be supplemented by innovative approaches to conservation that involve private landowners, such as the Native Fish Conservation Area approach. In basins where urbanization is a threat, local and state government regulations such as stormywater ordinances and riparian buffer ordinances will also be essential for effective management (Wenger et al., 2010). Lasting conservation will likely require the combination of regulations to minimize new stressors, incentives for reducing existing stressors, and acquisition to preserve the most critical habitats. 4.1. The need for increased investment The Southeast has by far the greatest concentration of freshwater biodiversity in the U.S. but receives little in the way of funding for federally threatened and endangered aquatic species. This area also has few protected lands compared to other areas in the U.S. While there have been notable conservation successes, and status reviews for a 2010 petition to list 404 aquatic species (Center for Biological Diversity, 2010) have narrowed many information gaps, this study reminds us that biodiversity may yet occur in surprising places. To slow the severe decline in the globally significant southeastern freshwater ecosystems, more investment must be made rapidly in the region, both in funding and in conservation capacity. Funding This work was supported by the National Fish and Wildlife Foundation [Grant 0102.15047037].

Acknowledgements We thank the many researchers who shared their data for our maps and the experts who generously served on this project’s advisory and technical committees and the, including: Susie Adams, US Forest Service; Paul Angermeier, Virginia Tech University; Katherine Baer, River Network; Art Bogan, NC Museum of Natural Sciences; Bob Butler, US Fish & Wildlife Service; Stephanie Chance, US Fish & Wildlife Service; Tanya Darden, SC Department of Natural Resources; Jessica Graham, Southeastern Aquatic Resources Partnership; Mike Harris, US Fish & Wildlife Service; Michael LaVoie, Eastern Band Cherokee Indians; Pat O'Neil, Geological Survey of Alabama; Peggy Shute, US Fish & Wildlife Service; Todd Slack, US Army Corps of Engineers- Engineer Research and Development Center; Matt Thomas, KY Department of Fish & Wildlife Resources. Appendix A. Supplementary data Supplementary data to this article can be found online at


D. Elkins et al. / Global Ecology and Conservation 19 (2019) e00654

References Abell, R.A., 2000. Freshwater Ecoregions of North America: a Conservation Assessment. Island Press. Angermeier, P.L., 1995. Ecological attributes of extinction-prone species: loss of freshwater fishes of Virginia. Conserv. Biol. 9, 143e158. Benz, G.W., Collins, D.E., 1997. Aquatic Fauna in Peril: the Southeastern Perspective. Southeast Aquatic Research Institute, Lenz Design & Communications, Decatur, Georgia. Bouska, K.L., Rosenberger, A., McMurray, S.E., Lindner, G.A., Key, K.N., 2018. State-level freshwater mussel programs: current status and a research framework to aid in mussel management and conservation. Fisheries 43, 345e360. Britton, A.W., Day, J.J., Doble, C.J., Ngatunga, B.P., Kemp, K.M., Carbone, C., Murrell, D.J., 2017. Terrestrial-focused protected areas are effective for conservation of freshwater fish diversity in Lake Tanganyika. Biol. Conserv. 212, 120e129. Burr, B.M., Warren, M.L., 1986. A Distributional Atlas of Kentucky Fishes. Kentucky Nature Preserves Commission. 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 (Tucson, AZ). Collen, B., Whitton, F., Dyer, E.E., Baillie, J.E., Cumberlidge, N., Darwall, W.R., Pollock, C., Richman, N.I., Soulsby, A.M., Bohm, M., 2014. Global patterns of freshwater species diversity, threat and endemism. Glob. Ecol. Biogeogr. 23, 40e51. Dudley, N.E., 2008. Guidelines for Applying Protected Area Management Categories. IUCN, Gland, Switzerland. ESRI, 2011. ArcGIS Desktop: Release 10. Environmental Systems Research Institute, Redlands, CA. Graf, D.L., Cummings, K.S., 2007. Review of the systematics and global diversity of freshwater mussel species (Bivalvia: Unionoida). J. Molluscan Stud. 73, 291-291. Haag, W.R., Williams, J.D., 2014. Biodiversity on the brink: an assessment of conservation strategies for North American freshwater mussels. Hydrobiologia 735, 45e60. Hermoso, V., Linke, S., Januchowski-Hartley, S.R., Kennard, M.J., Closs, G.P., Krkosek, M., Olden, J.D., 2016. Freshwater conservation planning. In: Closs, G.P., Krosek, M., Olden, J.D. (Eds.), Conservation of Freshwater Fishes. Cambridge University Press, pp. 437e466. Jelks, H.L., Walsh, S.J., Burkhead, N.M., Contreras-Balderas, S., Diaz-Pardo, E., Hendrickson, D.A., Lyons, J., Mandrak, N.E., McCormick, F., Nelson, J.S., Platania, S.P., Porter, B.A., Renaud, C.B., Schmitter-Soto, J.J., Taylor, E.B., Warren, M.L., 2008. Conservation status of imperiled North American freshwater and diadromous fishes. Fisheries 33, 372e407. Jenkins, R.E., Burkhead, N.M., 1994. Freshwater Fishes of Virginia. American Fisheries Society, Bethesda, Maryland. Johnson, P.D., Bogan, A.E., Brown, K.M., Burkhead, N.M., Cordeiro, J.R., Garner, J.T., Hartfield, P.D., Lepitzki, D.A.W., Mackie, G.L., Pip, E., Tarpley, T.A., Tiemann, J.S., Whelan, N.V., Strong, E.E., 2013. Conservation status of freshwater gastropods of Canada and the United States. Fisheries 38, 247e282. Lawrence, D.J., Larson, E.R., Liermann, C.A.R., Mims, M.C., Pool, T.K., Olden, J.D., 2011. National parks as protected areas for U.S. freshwater fish diversity. Conserv. Lett. 4, 364e371. Layman, S.R., Mayden, R.L., 2012. Morphological diversity and phylogenetics of the darter subgenus doration (percidae: Etheostoma), with descriptions of five new species. Bull. Ala. Mus. Nat. Hist. 1e83. Lindenmayer, D., Thorn, S., Noss, R., 2018. Countering resistance to protected-area extension. Conserv. Biol. 32, 315e321. Long, J.M., Nibbelink, N.P., McAbee, K.T., Stahli, J.W., 2012. Assessment of freshwater fish assemblages and their habitats in the National Park Service system of the southeastern United States. Fisheries 37, 212e225. McMurray, S.E., Roe, K.J., 2017. Perspectives on the Controlled Propagation, Augmentation, and Reintroduction of Freshwater Mussels (Mollusca: Bivalvia: Unionoida). BIOONE. Neves, R.J., Bogan, A.E., Williams, J.D., Ahlstedt, S.A., Hartfield, P.W., 1997. Status of aquatic mollusks in the southeastern United States: a downward spiral of diversity. In: Benz, G.W., Collins, D.E. (Eds.), Aquatic Fauna in Peril: the Southeastern Perspective. Southeast Aquatics Research Institute, Lenz Design and Communications, Decatur, Georgia, pp. 44e86. Noss, R.F., LaRoe, E.T., Scott, J.M., 1995. Endangered Ecosystems of the United States: a Preliminary Assessment of Loss and Degradation. US Department of the Interior, National Biological Service, Washington, DC, USA. Page, L.M., Burr, B.M., 2011. Peterson Field Guide to Freshwater Fishes, second ed. Houghton Mifflin, New York, NY. Richman, N.I., Bohm, M., Adams, S.B., Alvarez, F., Bergey, E.A., Bunn, J.J., Burnham, Q., Cordeiro, J., Coughran, J., Crandall, K.A., Dawkins, K.L., DiStefano, R.J., Doran, N.E., Edsman, L., Eversole, A.G., Fureder, L., Furse, J.M., Gherardi, F., Hamr, P., Holdich, D.M., Horwitz, P., Johnston, K., Jones, C.M., Jones, J.P., Jones, R.L., Jones, T.G., Kawai, T., Lawler, S., Lopez-Mejia, M., Miller, R.M., Pedraza-Lara, C., Reynolds, J.D., Richardson, A.M., Schultz, M.B., Schuster, G.A., Sibley, P.J., Souty-Grosset, C., Taylor, C.A., Thoma, R.F., Walls, J., Walsh, T.S., Collen, B., 2015. Multiple drivers of decline in the global status of freshwater crayfish (Decapoda: astacidea). Philos. Trans. R. Soc. Lond. B Biol. Sci. 370, 20140060. Smith, D.R., Butler, R.S., Jones, J.W., Gatenby, C.M., Hylton, R.E., Parkin, M.J., Schulz, C.A., 2017. Developing a landscape-scale, multi-species, and cost-efficient conservation strategy for imperilled aquatic species in the Upper Tennessee River Basin, USA. Aquat. Conserv. Mar. Freshw. Ecosyst. 27, 1224e1239. Smith, R., Freeman, P., Higgins, J., Wheaton, K., FitzHugh, T., Ernstrom, K., Das, A., 2002. Priority Areas for Freshwater Conservation Action: a Biodiversity Assessment of the Southeastern United States. The Nature Conservancy, Washington, D.C. Swift, C.C., Gilbert, C.R., Bortone, S.A., Burgess, G.H., Yerger, R.W., 1986. Zoogeography of the freshwater fishes of the southeastern United States: Savannah river to lake pontchartrain. In: Hocutt, C.H., Wiley, E.O. (Eds.), The Zoogeography of North American Freshwater Fishes. John Wiley and Sons. Taylor, C.A., Schuster, G.A., Cooper, J.E., DiStefano, R.J., Eversole, A.G., Hamr, P., Hobbs, H.H., Robison, H.W., Skelton, C.E., Thoma, R.F., 2007. A reassessment of the conservation status of crayfishes of the United States and Canada after 10þ years of increased awareness. Fisheries 32, 372e389. Thieme, M.L., Sindorf, N., Higgins, J., Abell, R., Takats, J.A., Naidoo, R., Barnett, A., 2016. Freshwater conservation potential of protected areas in the Tennessee and Cumberland River Basins, USA. Aquat. Conserv. Mar. Freshw. Ecosyst. 26, 60e77. US Fish and Wildlife Service, 2012. Federal and State Endangered and Threatened Species Expenditures: Fiscal Year 2012. US Department of the Interior, Washington, DC. US Fish and Wildlife Service, 2013. Federal and State Endangered and Threatened Species Expenditures: Fiscal Year 2013. US Department of the Interior, Washington, DC. US Fish and Wildlife Service, 2014. Federal and State Endangered and Threatened Species Expenditures: Fiscal Year 2014. US Department of the Interior, Washington, DC. US Fish and Wildlife Service, 2015. Federal and State Endangered and Threatened Species Expenditures: Fiscal Year 2015. US Department of the Interior, Washington, DC. US Fish and Wildlife Service, 2016. Federal and State Endangered and Threatened Species Expenditures: Fiscal Year 2016. US Department of the Interior, Washington, DC. US Geological Survey, National Gap Analysis Program, 2016. Protected Areas Database of the United States (PAD-US) V 1.4. USGS National Gap Analysis Program. Warren, M.L., Angermeier, P.L., Burr, B.M., Haag, W.R., 1997. Decline of a Diverse Fish Fauna: Patterns of Imperilment and Protection in the Southeastern United States. Pages 105-164 Aquatic Fanua in Peril: the Southeastern Perspective. Lenz Design Communications, Decatur, Georgia. Warren, M.L., Burr, B.M., Walsh, S.J., Bart, H.L., Cashner, R.C., Etnier, D.A., Freeman, B.J., Kuhajda, B.R., Mayden, R.L., Robison, H.W., Ross, S.T., Starnes, W.C., 2000. Diversity, distribution, and conservation status of the native freshwater fishes of the southern United States. Fisheries 25, 7e31. Welch, S.M., Eversole, A.G., 2006. The occurrence of primary burrowing crayfish in terrestrial habitat. Biol. Conserv. 130, 458e464. Wenger, S.J., Freeman, M.C., Fowler, L.A., Freeman, B.J., Peterson, J.T., 2010. Conservation planning for imperiled aquatic species in an urbanizing environment. Landsc. Urban Plan. 97, 11e21.

D. Elkins et al. / Global Ecology and Conservation 19 (2019) e00654


Williams, J.D., Warren Jr., M.L., Cummings, K.S., Harris, J.L., Neves, R.J., 1993. Conservation status of freshwater mussels of the United States and Canada. Fisheries 18, 6e22. Williams, J.E., Johnson, J.E., Hendrickson, D.A., Contreras-Balderas, S., Williams, J.D., Navarro-Mendoza, M., McAllister, D.E., Deacon, J.E., 1989. Fishes of North America endangered, threatened, or of special concern: 1989. Fisheries 14, 2e20. Williams, J.E., Williams, R.N., Thurow, R.F., Elwell, L., Philipp, D.P., Harris, F.A., Kershner, J.L., Martinez, P.J., Miller, D., Reeves, G.H., Frissell, C.A., Sedell, J.R., 2011. Native fish conservation areas: a vision for large-scale conservation of native fish communities. Fisheries 36, 267e277.