Changes of snow occurrence across the central and eastern United States under future warming for the late 21st century are investigated, by applying an empirical hyperbolic tangent function to both observed and downscaled high spatial resolution (~12.5 km) daily temperature and precipitation, to compare the historical (1981–2000) and future (2081–2100) snow occurrence.
The observed distributions of snow frequency show that snow-rain transition zones are mainly zonally distributed, since they are largely determined by temperature, with slight shifts to the south over the Appalachian Mountains. The snow-rain transition zone is located around 38–46°N for November and March, and 32–42°N for winter months (DJF). These observed patterns are reproduced well for the historical period by an ensemble average of multiple general circulation models (GCMs). The probabilistic projections show that the snow-rain transition zone will shift to the north under the background of global warming at magnitudes of 2–6 °C, indicating that large areas will experience a partial, or even a very large, loss of snow occurrence in the future. The northward shifts are about 2° latitude under the representative concentration pathways 4.5 (RCP4.5) scenario and 4° latitude under the RCP8.5 scenario. The percentages of the area losing snow occurrence are also assessed.
This project carries critical implications for agriculture, hydrology, ecosystems dependent on snow, and municipal planning.
- Ning L & Bradley RS, 2015. Snow occurrence changes over the eastern United States under future warming scenarios. Scientific Report, 5:17073; DOI: 10.1038/srep17073
- Ning L & Bradley RS. The risk of losing snow under future warming over the eastern United States. AGU Dec 2014.