Seventy years of stream‐fish collections reveal invasions and native range contractions in an Appalachian (USA) watershed

Abstract

Aim Knowledge of expanding and contracting ranges is critical for monitoring invasions and assessing conservation status, yet reliable data on distributional trends are lacking for most freshwater species. We developed a quantitative technique to detect the sign (expansion or contraction) and functional form of range‐size changes for freshwater species based on collections data, while accounting for possible biases due to variable collection effort. We applied this technique to quantify stream‐fish range expansions and contractions in a highly invaded river system. Location Upper and middle New River (UMNR) basin, Appalachian Mountains, USA. Methods We compiled a 77‐year stream‐fish collections dataset partitioned into ten time periods. To account for variable collection effort among time periods, we aggregated the collections into 100 watersheds and expressed a species’ range size as detections per watershed (HUC) sampled (DPHS). We regressed DPHS against time by species and used an information‐theoretic approach to compare linear and nonlinear functional forms fitted to the data points and to classify each species as spreader, stable or decliner. Results We analysed changes in range size for 74 UMNR fishes, including 35 native and 39 established introduced species. We classified the majority (51%) of introduced species as spreaders, compared to 31% of natives. An exponential functional form fits best for 84% of spreaders. Three natives were among the most rapid spreaders. All four decliners were New River natives. Main conclusions Our DPHS‐based approach facilitated quantitative analyses of distributional trends for stream fishes based on collections data. Partitioning the dataset into multiple time periods allowed us to distinguish long‐term trends from population fluctuations and to examine nonlinear forms of spread. Our framework sets the stage for further study of drivers of stream‐fish invasions and declines in the UMNR and is widely transferable to other freshwater taxa and geographic regions.

Remaining populations of an upland stream fish persist in refugia defined by habitat features at multiple scales

“…Intensive catchment‐wide land uses often improve habitat suitability for thermally and fine‐sediment‐tolerant species (Scott, ). Indeed, retrospective analyses by Hitt and Roberts () and Buckwalter, Frimpong, Angermeier, and Barney () in the NRD documented replacements of specialized upland natives with more generalized fishes; both studies implicated land use in these dynamics. Future research could extend our approach to examine whether multi‐scale physical features and community dynamics jointly exacerbate or reduce effects of catchment‐wide, top‐down disturbances.…”

“…Intensive catchment-wide land uses often improve habitat suitability for thermally and finesediment-tolerant species (Scott, 2006). Indeed, retrospective analyses by Hitt and Roberts (2012) and Buckwalter, Frimpong, Angermeier, and Barney (2018) in the NRD documented replacements of specialized upland natives with more generalized fishes;…”

Species traits predict stream‐fish invaders in an Appalachian (U.S.A.) river basin

“…Due to its montane aspect, south‐to‐north orientation, and large natural falls at the outlet, the New River basin has the lowest native fish species richness (44) of all 26 major drainages in the eastern U.S.A., including nine endemics (Jenkins & Burkhead, ). State‐sanctioned stockings of non‐native game and prey species, especially from the mid‐1800s to the 1970s, supplemented by continuing unauthorised introductions, have contributed at least 55 established introduced fish species, such that the basin has the greatest ratio of introduced‐to‐native species (55:44, Buckwalter, Frimpong, Angermeier, & Barney, ) among eastern U.S.A. drainages (Jenkins & Burkhead ; Peoples, Midway, DeWeber, & Wagner, ).…”

“…Introduced species reported from <3 catchments were classified as weak colonisers; species reported from three catchments were strong colonisers. We summarised species occurrence by catchment based on Buckwalter et al () (Frimpong, ).…”

“…Spreader represented invasion success during the spread stage, and applied to the complete pool of 74 introduced and native UMNR species. We adopted Buckwalter et al’s () classification of spreaders and non‐spreaders, which was based on temporal trends in detections per HU sampled (DPHS) in the UMNR since 1938, where DPHS was defined as the number of HUs a species was detected in per defined sampling period, divided by the total number of HUs sampled during the period. Herein, species that Buckwalter et al () classified as spreaders were assigned Spreader = 1; other species were assigned Spreader = 0.…”