Precipitation and vegetation dynamics predict multi-decadal occupancy of Lesser Prairie-chicken leks

Background/Question/Methods

As of February 2014, the Lesser Prairie-chicken (Tympanuchus pallidicinctus; LPC) was a candidate for federal listing as a Threatened species after decades of oscillating but declining population levels. Interannual variation in LPC populations (often indexed by lek counts) has been most strongly linked to reproductive success, which is suspected to be driven largely by spatiotemporal variation in protective cover from live and residual vegetation. While these habitat conditions are impractical to quantify in situ over large areas, gridded weather or vegetation datasets may serve as suitable proxies in investigating multi-decadal LPC dynamics. To this end, we employed 26 years (1984–2009) of lek monitoring data from 163 leks in southeastern New Mexico in multi-season models allowing for the hierarchical estimation of occupancy, colonization, extinction, and detection. Seasonal weather and vegetation covariates predicting occupancy dynamics were derived from PRISM and the Global Inventory Modeling and Mapping Studies (GIMMS) normalized difference vegetation index (NDVI) datasets.

Results/Conclusions

We found that higher levels of individual-year and multi-year precipitation reduced lek extinction probabilities. These effects appeared to be driven primarily by summer precipitation. We found weaker relationships between lek colonization and precipitation or NDVI. While NDVI had weaker relationships with dynamic occupancy than precipitation, higher NDVI values appeared to moderate the effects of dry conditions when NDVI-precipitation interactions were included in models. Our results add to evidence from previous shorter-term and local demographic studies that lagged, multi-year precipitation influences LPC population dynamics. While predictions of future North American Monsoon patterns of the region are at present uncertain, reduced winter snowfall, more rapid snowmelt, and higher rates of evapotranspiration are considered very likely in coming decades. Additional studies applying models accounting for imperfect detection of animals over large areas and multiple decades may contribute to improved forecasting of LPC population and occupancy dynamics under future climate change.