Sensitivity of Extreme Rainfall to Atmospheric Moisture Content in Arid/Semiarid Southwestern United States: Implications for Probable Maximum Precipitation Estimates

Sensitivity of extreme rainfall to atmospheric moisture content for the 19 August 2014 storm in central Arizona was investigated based on the Weather Research and Forecasting (WRF) model. Our study is motivated by providing improved understanding of the roles of complex terrain and atmospheric moisture content in dictating spatial and temporal variability of extreme rainfall in arid/semi-arid southwestern US. We also shed light on estimates of probable maximum precipitation (PMP) over complex terrain based on physical models. Extreme rainfall for the 19 August 2014 storm occurred within a 12-hour period and was characterized by two distinct storm episodes (in terms of synoptic environment, storm evolution and structure, etc.). The control WRF simulation identifies complex interactions of low-level moisture transport and orographic lift as key elements in producing extreme rainfall for the first storm episode, while rainfall for the second storm episode is related to the passage of a frontal system. Two sensitivity experiments are performed by increasing the relative humidity by 10% and 20% in both initial and boundary conditions used for the control simulation. Changes in atmospheric moisture content modify the storm dynamics, including instability of the storm environment as well as interactions of synoptic flow with complex terrain. The two storm episodes exhibit contrasting responses to the increase of moisture content, with rainfall accumulation and maximum convective available potential energy monotonically increasing for the second storm episode as opposed to a more complex relationship for the first storm episode. A moisture maximization simulation was also designed, for which moisture content was increased to almost saturation for both initial and boundary conditions, to mimic as closely as possible the framework of moisture maximization in PMP estimates. Our results highlight the nonlinear relationship between extreme rainfall and atmospheric moisture content, especially in complex environments where small-scale convection dominates the spatial and temporal variability of extreme rainfall. Physical model simulations provide valuable insights into the physical understandings of extreme rainfall and guidance for PMP estimates. Limitations of PMP estimates based physical models will also be discussed.