COS 147-3 - Losing small wetlands in agricultural landscapes worst case for wetland biodiversity

Thursday, August 10, 2017: 2:10 PM
B118-119, Oregon Convention Center
David C. Deane1, Damien A. Fordham2, Fangliang He1 and Corey JA Bradshaw3, (1)Renewable Resources, University of Alberta, Edmonton, AB, Canada, (2)University of Adelaide, Adelaide, Australia, (3)Biological Sciences, Flinders University, Adelaide, Australia
Background/Question/Methods

Wetlands are generally obstacles to agricultural development. As landscapes are progressively cleared for productive use, water-logged areas must either be drained, or avoided. If drainage is not practical or economical, a pattern of land clearance emerges where any native vegetation in non-viable areas remains in a semi-intact state. The resulting network of remnant native-habitat patches can persist even in heavily cleared contemporary landscapes. While local biodiversity of individual patches might be valued, any landscape-scale contribution of wetland networks might not be recognised, let alone quantified. In a drying climate, widespread reduction in wetland habitat is plausible, with transition to productive land use a likely consequence — particularly for smaller, privately-owned wetlands. Negative local impacts on native biodiversity would be inevitable under such a scenario, but it is generally unknown what the widespread loss of multiple wetland patches could mean for regional biodiversity.

We analysed the impact of progressive loss of wetland patches from an agricultural landscape on native plant species richness across all patches. We simulated the destruction of individual wetland patches starting with the smallest, most vulnerable sites, and compared the reduction in total species number across all wetlands against the generalised area-loss prediction of species- and endemics-area curves.

Results/Conclusions

Removing entire wetland patches always resulted in more species extinctions than the equivalent loss arising from all wetlands in the network according to both species- and endemics-area models. Wetlands loss in order of smallest to largest size resulted in the highest rate of species loss. For a 10% loss in total wetland area, the small-to-large patch-loss prediction was > 350% higher than the worst-case area-loss model (7.9% of species versus 2.2%). This reflects the distribution of regional endemic species, which were present in even the smallest wetlands. Wetlands < 1.5 ha contained 16% of endemic species in only 5% of total wetland area.

Most wetlands in this region of Mediterranean-type climate depend on reliable winter-spring rainfall to maintain water balance, making them particularly vulnerable to projected drying. A rapid transition from water-logged to productive land risks the loss of many species from wetland habitats. Importantly, species extinctions would not be restricted to wetland plants, because diversity in these seasonal wetlands is largely a function of the distribution of rare terrestrial species. Managers need to factor in this possibility and encourage plans to establish native terrestrial vegetation wherever practical.