Community Ecology and Restoration of Desert Springs

Freshwater ecosystems are among the most important and most imperiled of ecological resources, especially in arid landscapes where aquifer-fed surface waters are biodiversity hotspots that can harbor high levels of endemic and often endangered biota. These desert springs are threatened by water mining, land-use change, biological invasions, and other global change phenomena. My dissertation describes the community ecology of a desert spring system (Ash Meadows, Nevada), including the environmental parameters that drive community composition, the effects of ecological restoration, and spatiotemporal dynamics in stable and restored springs. I used a combined environmental DNA - metagenetic survey method to assess the composition of whole eukaryotic communities from environmental samples of algal mats, the water column, and benthic sediments. Spring size, water temperature, and invasive species (red swamp crayfish or largemouth bass) were all correlated with community composition. This has important implications for conservation management of desert springs, which could decrease in size and increase in temperature with aquifer pumping and climate change. Next, I conducted two chronosequence studies to assess the effects of ecological restoration and habitat creation on community composition and temporal variation. The restoration of a low-flow, high-temperature spring showed remarkable success: after prolonged desiccation and structural modification, the spring community exhibited a successional trajectory towards a historic, natural composition, suggesting environmental filtering during community assembly. The second chronosequence study compared a natural habitat to its constructed analog. The natural habitat, Devils Hole, is a small opening into a deep aquifer, the surface of which comprises the entire range of the endangered Devils Hole pupfish (Cyprinodon diabolis). Federal management agencies constructed an artificial refugium for $4.5 million to harbor a backup population of the pupfish; the 380,000 L refuge tank was designed to exactly replicate the Devils Hole environment. Despite seeding protocols intended to recreate the biotic community found in Devils Hole, and controls over water temperature, insolation, and other abiotic parameters, the refuge tank community differed from the natural habitat in composition and seasonal variability. The community outcomes revealed by these chronosequence studies highlight the importance of monitoring to gauge progress towards ecological goals in managed systems