The availability of high quality water is critical to both humans and ecosystems. A recent proposal was made by rapidly expanding municipalities in central Oklahoma to begin transferring groundwater from the Arbuckle-Simpson aquifer, a sensitive sole-source aquifer in south-central Oklahoma. Concerned citizens and municipalities living on and getting their drinking water from the Arbuckle-Simpson lobbied the legislature to pass a temporary moratorium on groundwater transfer to allow for a comprehensive study of the aquifer and its ecosystems. We conducted an instream flow assessment using Physical Habitat Simulation (PHABSIM) on springs and streams with four spring-dependent species: two minnows, southern redbelly dace (Phoxinus erthyrogaster) and redspot chub (Nocomis asper); and two darters, least darter (Etheostoma microperca) and orangethroat darter (Etheostoma spectabile). Spring habitats are unique compared to other river habitats because they have constant flow and temperature, small and isolated habitat patches, and a general lack of predators.

Our study sites included two spring-fed streams, one larger stream with high groundwater inputs, and a river with both groundwater and surface water inputs that is adjacent to the small spring-fed streams. These habitats meet the criteria for groundwater dependent ecosystems because they would not exist without the surface expression of groundwater. A total of 99 transects in all four sites were surveyed for channel elevation, and three sets of water surface elevation and water velocity were measured. Habitat suitability criteria were derived for the species at each site using nonparametric confidence limits based on underwater observations made by snorkelers. Simulations of flow were focused on declines in discharge, which is the expected effect of the proposed groundwater diversion.

Our results show that only a small proportion of the total available area in each habitat is considered to be preferred habitat (Weighted Usable Area [WUA]) by the four target species. In the spring habitats, a maximum of 10% of the total area is preferred habitat and that dropped to as little as 3% with decreased flows. The quantity of WUA decreased when lower discharges were simulated for all the target species. Declines in the small amount of habitat that is already available would likely degrade these populations of fishes. In the larger river habitat, the highest WUA occurred at the lowest discharge, which leads us to conclude that in the event of dewatering of the spring habitats, the river should provide some refuge habitat for spring dependent species.

Based on the findings of this study, groundwater removal from the aquifer near springs may have adverse impacts on fish habitat availability for spring dependent fish populations if seasonal trends in spring discharge are not maintained (higher in winter and lower in late summer). Quantifying the relationship of streamflow between gaged and ungaged springs in the Arbuckle-Simpson is a possible method to monitor and maintain flows in springs.