全球变化知识资源中心

The hemlock woolly adelgid (Adelges tsugae Annand - HWA) is invasive in eastern North America where it causes extensive mortality to hemlock communities. The future of these communities under projected climate change is an issue of landscape ecological interest and speculation. We employed the MaxEnt algorithm with the random subset feature selection algorithm (RSFSA) in creating HWA niche models. Final models were ensembles of 12 statistically best models with six predictors each. Out of 119 climatic, topographic, and soil variables, 42 were used in at least one final model. Soil features, followed by climate and topographic features, were most common in selected models. The three most important variables among all models were November potential evapotranspiration, slope, and percent Ochrepts soil. The potential distributions of HWA within eastern North America were projected under historical and four future climate scenarios for 2050 and 2070 under low and high CO2 emissions. The mean of the minimum values for the minimum temperature of the coldest month from the 12 MaxEnt model projections in eastern North America was -15.8 degrees C. This value was close to -15 degrees C, the extreme minimum temperature found for both HWA occurrence points and previously reported HWA cold temperature limits. These results indicate that HWA may be close to equilibrium distribution in eastern North America under current climate. We also reverse-casted the eastern North American MaxEnt model back onto the HWA native ranges in eastern Asia and western North America. The projections match best with native ranges in Asian islands, such as Japan, and the Cascade Mountains in western North America. Statistically significant HWA range shifts of 221-468 km northwards and 110-164 km eastwards were projected by the 12 models for 2050-2070. The 2070 high CO2 emission scenario models projects HWA suitability throughout most of the northern range of eastern hemlock.