Characterizing species co‐occurrence patterns of imperfectly detected stream fishes to inform species reintroduction efforts

Abstract

Species reintroduction efforts can improve the recovery of imperiled species, but successful implementation of this conservation strategy requires a thorough understanding of the abiotic and biotic factors influencing species viability. Species interactions are especially understudied, in particular by omitting the effect of imperfect detection on negative, neutral, or positive associations within a community. Using repeat surveys from 5 southern Ontario, Canada, Great Lakes tributaries, we quantified species co‐occurrence patterns with the eastern sand darter (ESD) (Ammocrypta pellucida), listed as federally threatened, and characterized how imperfect detection during sampling can influence inference regarding these relationships. We used a probabilistic framework that included 3 approaches of increasing complexity: probabilistic co‐occurrence analysis ignoring imperfect detection; single‐species occupancy models with subsequent co‐occurrence analysis; and 2‐species occupancy models. We then used our occupancy models to predict suitable sites for potential future reintroduction efforts while considering the influence of negative species interactions. Based on the observed data, ESD showed several positive associations with co‐occurring species; however, species associations differed when imperfect detection was considered. Specifically, a negative association between ESD and rosyface shiner (Notropis rubellus) was observed only after accounting for imperfect detection in the Grand River. Alternatively, positive associations in the Grand River between ESD and northern hogsucker (Hypentelium nigricans) and silver shiner (Notropis photogenis) were observed regardless of whether imperfect detection was accounted for. Our models predicted several potential reintroduction sites for ESD in formerly occupied watersheds with high levels of certainty. Overall, our results demonstrate the importance of investigating imperfect detection and species co‐occurrence when planning reintroduction efforts.

Habitat associations of the Threatened pugnose minnow (Opsopoeodus emiliae) at the northern edge of the species range

“…Imperfect detection describes the situation when a species is undetected during sampling despite being present (MacKenzie et al, , ) and is known to affect analyses of fish species–habitat associations (e.g. Dextrase, Mandrak, & Schaeffer, ; Lamothe, Dextrase, & Drake, in press). Given the associations of pugnose minnow to vegetated, clear‐water systems across its global range (Robertson et al, ; Schneider, ; Trautman, ), we hypothesised positive relationships between pugnose minnow occupancy probability and water clarity and the extent of submergent aquatic vegetation at a site.…”

Modelling patterns of coexistence of three congeneric headwater fishes

“…Our results provide further support that biotic interactions may not be meaningful in describing patterns of co-occurrence of stream fishes; the influence of these interactions on species occupancy may only be detectable at specific spatial extents. There is much support for the regulation of co-occurrence within a stream-reach as a consequence of habitat preferences at the stream-reach and landscape scales (Giam & Olden, 2016;Grossman et al, 1998;Jackson, Pedro, & Olden, 2001;Peres-Neto, 2004); however, there is also limited support that stream fishes may be structured by biotic interactions at the stream-reach scale as well (Lamothe, Dextrase, & Drake, 2019;Peoples & Frimpong, 2016). In our study, we sought to examine the influence of biotic and abiotic factors on non-random patterns of co-occurrence of three headwater congeners at the stream-reach scale using a method that accounts for imperfect detection.…”

Abiotic and biotic associations between the round gobyNeogobius melanostomusand tubenose gobyProterorhinus marmoratuswith the endangered northern madtomNoturus stigmosusin Canada

“…Occupancy models have been shown to be useful tools for estimating abiotic and biotic habitat associations of imperilled species while accounting for measurement error and other sampling biases (e.g. Dextrase, Mandrak, & Schaefer, ; Falke, Fausch, Bestgen, & Bailey, ; Lamothe, Dextrase, & Drake, ; MacKenzie et al, , ). Occupancy models were developed to account for the challenges of imperfect detection in species monitoring data (MacKenzie et al, ).…”

“…Imperfect detection occurs when a species is not detected despite its presence at a site and is accounted for in occupancy models by modelling detection probability explicitly using a repeated sampling design (MacKenzie et al, , ). For example, Dextrase, Mandrak, and Schaefer () used a repeated sampling design to develop single‐season, single‐species occupancy models that identified important habitat associations for the Threatened eastern sand darter ( Ammocrypta pellucida ) in south‐western Ontario rivers, and subsequently to identify potential sites for the reintroduction of the species (Lamothe et al, ). Their models had good predictive performance and accounted for imperfect detection despite the challenges of capturing imperilled freshwater fishes in the wild.…”