Precipitation Morphology in the Western United States: Its Relationship to Ambient Atmospheric Conditions and Future Changes

Extreme precipitation is generally projected to increase in a warmer climate as atmospheric moisture increases with temperature. However, changes in the spatial distribution of extreme precipitation within the precipitation features associated with different storm types are less known but they have important implications for changing flood characteristics in the future. Analysis of precipitation features and drivers of their changes has been limited by the lack of coherent datasets of precipitation and atmospheric condition at high spatial resolution. Recent advances in convection-permitting modeling have enabled storm-resolving simulations that produce more realistic precipitation features. In this study, computer vision technique is used to extract the spatial information of more than 60,000 daily precipitation events from a 6-km regional climate simulation over the western U.S. during 1981-2010. The physical and morphological characteristics of these events are analyzed: total precipitation volume, precipitation area, and the average as well as peak precipitation intensity. We also examine the statistical relationship between these characteristics and their ambient atmospheric conditions (air temperature profile, moisture availability, vertical air motion, and atmospheric instability) to identify the key factors controlling each precipitation characteristics and how they vary seasonally and spatially. Applying the same analysis to a five-member ensemble of pseudo-global warming simulations for 2040-2070, changes of the precipitation features under global warming will be investigated. By analyzing the top 50 extreme precipitation events from each pair of historical and future simulation, we will explore the changes in precipitation features and atmospheric conditions and how well the former may be explained by the latter based on their statistical relationships identified from the historical climate simulations. Through this analysis, we expect to gain an improved understanding of how precipitation morphology of storms in the western U.S. responds to climate change in the future, and what meteorological factors are driving these responses.