Swimming through sand: connectivity of aquatic fauna in deserts

Abstract

Freshwater ecosystems in arid regions range from highly fragmented to highly connected, and connectivity has been assumed to be a major factor in the persistence of aquatic biota in arid environments. This review sought to synthesize existing research on genetic estimation of population connectivity in desert freshwaters, identify knowledge gaps, and set priorities for future studies of connectivity in these environments. From an extensive literature search, we synthesized the approaches applied, systems studied, and conclusions about connectivity reached in population genetic research concerning desert freshwater connectivity globally. We restrict our scope to obligate aquatic fauna that disperse largely via freshwaters and exclude those with active aerial dispersal abilities. We examined 92 papers, comprising 133 studies, published from 1987 to 2014. Most described studies of fishes and invertebrates in the deserts of Australia and North America. Connectivity declined with increasing scale, but did not differ significantly among arid regions or taxonomic classes. There were significant differences in connectivity patterns between species with different dispersal abilities, and between spring and riverine habitats at local scales. Population connectivity in desert freshwaters is typically most influenced by the ecology of the species concerned and hydrological connectivity. Most studies did not assess predefined models of connectivity, but described gene flow and/or genetic structure. Climate change and anthropogenic impacts worldwide are likely to increase the incidence and impact of habitat fragmentation in already threatened desert freshwaters. To reduce this risk, biodiversity conservation and environmental management must address connectivity, but often the required information does not exist. Researchers can provide this by explicitly considering the effects of hydrology and species’ ecology on connectivity, and incorporating these into connectivity models, which are vital for understanding connectivity in desert freshwaters.

Variation among macroinvertebrate communities suggests the importance of conserving desert springs

“…A study by Bogan et al () found that macrohabitat type was more important than geographic location in structuring aquatic invertebrate communities in the Sonoran Desert. Alternatively, previous studies have shown that local environmental conditions in desert springs may not be useful predictors of macroinvertebrate community composition (Rader et al, ; Sada, Fleishman, & Murphy, ), and that connectivity among spring systems may be a better predictor of species turnover than abiotic parameters (Murphy, Pavlova, Thompson, Davis, & Sunnucks, ). Despite differences in physicochemical conditions among spring sites, results from this study suggest that geographic distance is the most influential factor in community composition among springs in the Chihuahuan Desert.…”

Going, going, gone? Loss of genetic diversity in two critically endangered Australian freshwater fishes, Scaturiginichthys vermeilipinnis and Chlamydogobius squamigenus, from Great Artesian Basin springs at Edgbaston, Queensland, Australia

“…In the past, the species' reproductive biology may have facilitated expansion into new habitats and gene flow among populations, but more recently, and in far more arid conditions, populations have very little gene flow among them. Once individual spring populations become extirpated, higher levels of fragmentation and genetic isolation/structure result, a common pattern in desert freshwater taxa (Murphy et al , ).…”

The role of dispersal in river network metacommunities: Patterns, processes, and pathways

“…Extreme events like large‐scale floods, which are set to play an increasingly strong role in river ecosystems (Woodward et al., ), may override many of these dispersal constraints that network‐constrained dispersers face. Extreme flood events may increase connectivity, leading to movement of organisms between locations that would not occur under baseflow conditions, particularly in arid systems (Murphy, Pavlova, Thompson, Davis, & Sunnucks, ). Mossop et al.…”