Loading...
Extreme events over the contiguous United States portrayed in a CESM-WRF dynamical downscaling framework
Cai, Lei
Cai, Lei
Citations
Altmetric:
Abstract
A dynamical downscaling framework is adopted to explore historical (1950-1999) and projected (2050-2099) behavior of extreme precipitation (PR), maximum temperature (TMAX) and minimum temperature (TMIN) events within the contiguous United States. Compared to reanalysis data, simulations represent temperature better than precipitation, and the model performs better east of the Rocky Mountains than over the mountainous west. Extreme events are defined according to exceedances of percentiles of the distribution of precipitation and temperature variables (typically the 90th, 95th, and 99th percentiles), as well as the actual magnitudes corresponding to the percentiles. After applying a bias-correction to all three variables, extreme percentile thresholds show broadly higher values for all three variables in the projected simulation compared to the historical simulation. Precipitation extremes show no systematic trends of frequency or intensity in either the historical or the projected simulations. Trends of TMAX and TMIN have frequency and intensity that are consistently positive in the historical simulation, but the positive trend patterns are somewhat different in the projected simulation. In the projected simulation, all climate zones exhibit consistent increases in PR and TMAX extremes, and a decrease in TMIN extremes, both in the frequency and intensity. Northern zones such as Dfa and Dfb exhibit more changes of extremes in the projected simulation compared to other zones. On the other hand, the patterns of extreme frequency and intensity in all zones suggest their dependence on regional climatologies (e.g., B class zones have more frequent TMAX and less frequent PR than other zones, while D class zones have lower TMAX and TMIN intensity than other zones). In the projected simulation, PR intensity increases more significantly than frequency, while frequency increases more than intensity for TMAX and TMIN. The projected heat waves (defined as high temperature events lasting multiple days) are more severe in both number and duration, which results predominately from the increasing of the mean TMAX.
Description
Date
2014-12-31
Journal Title
Journal ISSN
Volume Title
Publisher
University of Kansas
Research Projects
Organizational Units
Journal Issue
Keywords
Atmospheric sciences, Climate change, bias correction, climate classification, dynamical downscaling, extreme events, heat waves