Arctic daily temperature and precipitation extremes: Observed and simulated physical behavior
ARCTIC DAILY TEMPERATURE AND PRECIPITATION EXTREMES: OBSERVED AND SIMULATED PHYSICAL BEHAVIOR
Justin M. Glisan
Department of Geological and Atmospheric Sciences, Iowa State University, Ames, Iowa
Simulations using a six-member ensemble of Pan-Arctic WRF (PAW) were produced on two Arctic domains with 50-km resolution to analyze precipitation and temperature extremes for various periods. The first study used a domain developed for the Regional Arctic Climate Model (RACM). Initial simulations revealed deep atmospheric circulation biases over the northern Pacific Ocean, manifested in pressure, geopotential height, and temperature fields. Possible remedies to correct these large biases, such as modifying the physical domain or using different initial/boundary conditions, were unsuccessful.
Spectral (interior) nudging was introduced as a way of constraining the model to be more consistent with observed behavior. However, such control over numerical model behavior raises concerns over how much nudging may affect unforced variability and extremes. Strong nudging may reduce or filter out extreme events, since the nudging pushes the model toward a relatively smooth, large-scale state. The question then becomes - what is the minimum spectral nudging needed to correct biases while not limiting the simulation of extreme events? To determine this, we use varying degrees of spectral nudging, using WRF's standard nudging as a reference point during January and July 2007. Results suggest that there is a marked lack of sensitivity to varying degrees of nudging. Moreover, given that nudging is an artificial forcing applied in the model, an important outcome of this work is that nudging strength apparently can be considerably smaller than WRF's standard strength and still produce reliable simulations.
In the remaining studies, we used the same PAW setup to analyze daily precipitation extremes simulated over a 19-year period on the CORDEX Arctic domain for winter and summer. We defined these seasons as the three-month period leading up to and including the climatological sea ice maximum and minimum, respectively. Analysis focused on four North American regions defined using climatological records, regional weather patterns, and geographical/topographical features. We compared simulated extremes with those occurring at corresponding observing stations in the U.S. National Climate Data Center's (NCDC's) Global Summary of the Day. Our analysis focused on variations in features of the extremes such as magnitudes, spatial scales, and temporal regimes. Using composites of extreme events, we also analyzed the processes producing these extremes, comparing circulation, pressure, temperature and humidity fields from the ERA-Interim reanalysis and the model output. The analysis revealed the importance of atmospheric convection in the Arctic for some extreme precipitation events and the overall importance of topographic precipitation. The analysis established the physical credibility of the simulations for extreme behavior, laying a foundation for examining projected changes in extreme precipitation. It also highlighted the utility of the model for extracting behavior that one cannot discern directly from the observations, such as summer convective precipitation.