2017 ESA Annual Meeting (August 6 -- 11)

COS 174-6 - Effects of snow extent on Dall sheep recruitment indicate population declines under climate change

Friday, August 11, 2017: 9:50 AM
D129-130, Oregon Convention Center
Madelon van de Kerk1, David Verbyla2, Anne W. Nolin3, Kelly J. Sivy4 and Laura R. Prugh1, (1)University of Washington, (2)University of Alaska Fairbanks, Fairbanks, AK, (3)Department of Geography, University of Nevada-Reno, Reno, NV, (4)School of Environmental and Forest Sciences, University of Washington
Background/Question/Methods

Northern alpine ecosystems are especially vulnerable to climate change because they occur at high latitudes and high elevations, but knowledge about the specific impacts of climate change on these systems is lacking. Dall sheep are an iconic species endemic to Northern alpine ecosystems, and recent population declines have been attributed to harsh spring weather. We investigated the effect of spring snow cover on Dall sheep recruitment using 15 years of range-wide remote sensing and Dall sheep survey data collected from 2000-2015. We used lamb:ewe ratios as an index of recruitment, and evaluated the effect of the following covariates on this ratio: snow-off date, snowline elevation on May 15 (peak of the lambing season), elevation, latitude, and climatic group for each mountain range. We used additive and interactive linear mixed models with a random effect of mountain range to test for the effect of the covariates, and ranked the models based on the Akaike Information Criterion (AIC). Using predictions from the highest ranked model, we then assessed the effect that expected increases in precipitation might have on Dall sheep population dynamics.

Results/Conclusions

We gathered data from 1800 surveys in 24 mountain ranges, with an average of 72 (SE 3.26) surveys spanning 8 (SE 0.22) years per mountain range. The average range-wide lamb:ewe ratio was 0.31 (SE 0.01). Lamb:ewe ratios increased with higher snowline elevations on May 15 (slope: 0.066, 95% CI: 0.026 – 0.106) and decreased with later snow-off dates (slope: -0.003, 95% CI: -0.004 – 0.001). Out of all the covariates, the snow-off date was the strongest predictor for lamb:ewe ratios (AIC parameter weight: 0.95), followed by latitude (AIC parameter weight: 0.81). The best-supported model had an AIC model weight of 0.41 and included an interactive effect of snow-off date and latitude. This model predicted that the negative effect of snow-off date on recruitment was strongest at high latitudes, and increasingly weaker at lower latitudes. Predictions from this model indicated that, compared to mean weather conditions, population growth would decrease by 2% in years with a late snow-off date at intermediate northern latitudes, and by 4% at high latitudes. These results demonstrate that the timing of snowmelt affects Dall sheep throughout their range, and indicate that climate change could adversely affect the population dynamics of this iconic northern species.