: Impoundments, diversion dams, and stream dewatering have created a mosaic of large river fragments throughout the Great Plains of central North America. Coincident with these habitat changes are massive declines in the distribution and abundance of Great Plains fishes belonging to the “pelagic‐spawning” reproductive guild. We analyzed longitudinal fragment lengths (measured in river kilometers, rkm) and literature accounts of population status for eight species from this guild across 60 fragments to derive thresholds in stream length associated with extirpations. Fragment length predicted population status (F2,21 = 30.14, P < 0.01), with lengths averaging 136 ± 21 rkm for extirpated, 226 ± 69 rkm for declining, and 458 ± 137 for stable populations. Fragment length explained 71% of reported extirpations and estimated thresholds in fragment length explained 67% of variation in population persistence. Our findings provide insight into appropriate spatial scales for conducting riverscape conservation approaches that address the hierarchical effects of fragmentation on stream‐dwelling fishes.

Abstract. Biodiversity in stream networks is threatened globally by interactions between habitat fragmentation and altered hydrologic regimes. In the Great Plains of North America, stream networks are fragmented by .19 000 anthropogenic barriers, and flow regimes are altered by surface water retention and groundwater extraction. We documented the distribution of anthropogenic barriers and dry stream segments in five basins covering the central Great Plains to assess effects of broad-scale environmental change on stream fish community structure and distribution of reproductive guilds. We used an informationtheoretic approach to rank competing models in which fragmentation, discharge magnitude, and percentage of time streams had zero flow (a measure of desiccation) were included to predict effects of environmental alterations on the distribution of fishes belonging to different reproductive guilds. Fragmentation caused by anthropogenic barriers was most common in the eastern Great Plains, but stream desiccation became more common to the west, where rivers are underlain by the depleted (i.e., extraction . recharge) High Plains Aquifer. Longitudinal gradients in fragmentation and desiccation contributed to spatial shifts in community structure from taxonomically and functionally diverse communities dominated by pelagic reproductive guilds where fragmentation and desiccation were least, to homogenized communities dominated by benthic guilds where fragmentation and desiccation were common. Modeling results revealed these shifts were primarily associated with decline of pelagic reproductive guilds, notably small-bodied pelagophilic and lithopelagophilic fishes that declined in association with decreased fragment length and increased number of days with zero flow. Graph theory combined with a barrier prioritization approach revealed specific fragments that could be reconnected to allow fishes within these guilds to colonize currently unoccupied fragments with the mitigation or removal of small dams (,10 m height). These findings are useful for natural resource managers charged with halting or reversing the prevailing pattern of declining fish diversity in the Great Plains. Our study represents one of the most comprehensive assessments of fish diversity responses to broad-scale environmental change in the Great Plains and provides a conservation strategy for addressing the simultaneous contributions of fragmentation and flow alteration to the global freshwater biodiversity crisis.

Stream fragmentation alters the structure of aquatic communities on a global scale, generally through loss of native species. Among riverscapes in the Great Plains of North America, stream fragmentation and hydrologic alteration (flow regulation and dewatering) are implicated in the decline of native fish diversity. This study documents the spatio–temporal distribution of fish reproductive guilds in the fragmented Arkansas and Ninnescah rivers of south‐central Kansas using retrospective analyses involving 63 years of fish community data. Pelagic‐spawning fishes declined throughout the study area during 1950–2013, including Arkansas River shiner (Notropis girardi) last reported in 1983, plains minnow (Hybognathus placitus) in 2006, and peppered chub (Macrhybopsis tetranema) in 2012. Longitudinal patterns in fish community structure in both rivers consisted of strong breaks associated with dams, and pelagic‐spawning fishes were missing from shorter fragments upstream of those barriers. Among downstream and longer fragments, probability of occurrence for pelagic‐spawning fishes declined or fell to zero during periods of drought. Based on these data, interactions between fragmentation and drying are hypothesized as operating as an ecological ratchet mechanism in which forward movement toward pelagic‐spawning fish extirpation occurs during desiccation, and reciprocated reverse movement toward recolonization following return of flows is blocked by fragmentation. The ratchet mechanism is capable of explaining the long‐term 'ratcheting down' of fish diversity in Great Plains rivers and has implications for managing biodiversity in fragmented riverscapes where water is scarce or might become so in the future. Copyright © 2014 John Wiley & Sons, Ltd.

Effects of fragmentation on the ecology of organisms occupying dendritic ecological networks (DENs) have recently been described through both conceptual and mathematical models, but few hypotheses have been tested in complex, real-world ecosystems. Stream fishes provide a model system for assessing effects of fragmentation on the structure of communities occurring within DENs, including how fragmentation alters metacommunity dynamics and biodiversity. A recently developed habitat-availability measure, the "dendritic connectivity index" (DCI), allows for assigning quantitative measures of connectivity in DENs regardless of network extent or complexity, and might be used to predict fish community response to fragmentation. We characterized stream fish community structure in 12 DENs in the Great Plains, USA, during periods of dynamic (summer) and muted (fall) discharge regimes to test the DCI as a predictive model of fish community response to fragmentation imposed by road crossings. Results indicated that fish communities in stream segments isolated by road crossings had reduced species richness (alpha diversity) relative to communities that maintained connectivity with the surrounding DEN during summer and fall. Furthermore, isolated communities had greater dissimilarity (beta diversity) to downstream sites notisolated by road crossings during summer and fall. Finally, dissimilarity among communities within DENs decreased as a function of increased habitat connectivity (measured using the DCI) for summer and fall, suggesting that communities within highly connected DENs tend to be more homogeneous. Our results indicate that the DCI is sensitive to community effects of fragmentation in riverscapes and might be used by managers to predict ecological responses to changes in habitat connectivity. Moreover, our findings illustrate that relating structural connectivity of riverscapes to functional connectivity among communities might aid in maintaining metacommunity dynamics and biodiversity in complex dendritic ecosystems.

Human environmental change influences freshwaters as well as the regulating, provisioning, and cultural services that ecosystems provide worldwide. Here, we assess the global human impact on the potential value of six freshwater ecosystem services (ES) and estimate the proportion of each used globally (the mean value across all countries is in parentheses): biodiversity (0.37), disturbance regulation (0.24), commodities (0.39), greenhouse gases (0.09), water availability (0.10), and water quality (0.33). We also created a composite index of the impact. Using different valuation schemes, we found that humans have used potential global freshwater ES scaled by a relative value of roughly 4-20%, with a median of 16%. All countries use a considerable amount of the potential ES value, invalidating the idea that wealthier countries have less impact on their ES once they have developed. The data suggest that humans have diminished the potential ES provided by freshwaters across the globe and that factors associated with high population growth rates are related to the overall degradation.