Date of Award
Master of Science
Geological and Atmospheric Sciences
William J. Gutowski
Extreme precipitation is a key topic in climate research. As the environment continues its steady increase in temperatures through increased greenhouse-gas emissions, conditions become more favorable for storm development that produce extreme events. This can have devastating effects in the upper Mississippi region, where agriculture dominates the regional economy.
To understand trends in extreme precipitation events of future climates, using climate models becomes essential. Because extreme precipitation events predominantly occur through relatively small-scale dynamics, the coarser resolution of global climate models (GCMs) is unable to capture fully such events. Therefore, regional climate models (RCMs) with higher resolution are used to simulate and investigate these events. In order to gain confidence in the models' ability to simulate future extreme events, we compare model simulations of contemporary climate with observations to determine if simulated extremes are occurring in conditions similar to the observed.
The overall objective of this work is to evaluate model performances from contemporary simulations by RCMs participating in the North American Regional Climate Change Assessment Program (NARCCAP). For comparison, we also examine output from a time-slice GCM of comparable resolution and a coarser atmosphere-ocean GCM. This work focuses on the upper Mississippi region for the winter season (December-January-February), when it is assumed that resolved synoptic circulation governs precipitation, comparing 18 years of simulations with observations.
For most of the models, average precipitation rates and percentage of days with precipitation is higher in models than observations. The high-resolution models generally reproduce well the precipitation-vs.-intensity spectrum seen in observations, with a small tendency toward producing overly strong precipitation at high intensity thresholds. Further analysis focuses on precipitation events exceeding the 99.5th percentile that occur simultaneously at 15 or more grid points in the region, yielding so-called "widespread events". Collectively, the high-resolution models also tend to produce somewhat more frequent widespread events than the observations. Models capture the inter-seasonal variability well, as December in the observations and nearly all models are the month with the highest frequency of widespread extremes. Models produce fewer consecutive widespread-extreme days compared to observations. The atmosphere-ocean GCM has lower values of extreme precipitation frequency and intensity, likely because of its coarser resolution.
Further analysis focuses on 500 hPa flow, 10-m winds, 2-m temperatures, and 2-m specific humidity. The main feature yielding widespread extreme precipitation is an environment that is more favorable for the transport of moisture from the Gulf of Mexico into the analysis domain, and positive temperature and specific humidity anomalies, allowing the atmosphere to provide more moisture for precipitation.
Kawazoe, Sho, "Regional, extreme daily precipitation in NARCCAP simulations" (2011). Graduate Theses and Dissertations. 10178.