The characterization and prediction of sub-diurnal extreme precipitation in the midwestern United States
Abstract
Heavy and extreme rainfall pose a significant forecasting and warning challenge, made worse by consequences on life, property, and agriculture. This study seeks to characterize sub-diurnal extreme precipitation and the systems that produce these occurrences using objective, quantitative measures, and to test the skill of the WRF model in simulating occurrences of extreme precipitation with varying scales of initial and boundary conditions. In chapter 2 rain gauge data is used to identify sub-diurnal extreme precipitation events using the 10-yr return level for a 6-h period at each gauge. Extreme events are characterized using an object-oriented approach with Stage II and Stage IV precipitation data, a multi-sensor product that augments radarderived precipitation estimates with rain gauge data. In chapter 3 occurrences of convective rainfall in the Midwestern United States are identified and characterized using an object-oriented approach, and extreme precipitation ‘objects’ are identified based on within-object maximum precipitation. The systems that produced these extreme occurrences are characterized using an object-oriented approach with radar reflectivity data. In chapter 4 the WRF model is used to simulate two cases of extreme precipitation occurrences with progressively coarsened initial and boundary conditions. An object-oriented approach is used to identify convective precipitation objects from the WRF simulations, and a Euclidean distance approach is used to evaluate the model’s performance by matching forecast objects with the extreme precipitation occurrences observed in Stage II data. It is found that the systems that produce occurrences of sub-diurnal extreme precipitation vary greatly in areal size from relatively small, isolated cells to large mesoscale convective systems. The larger systems are better predicted by mesoscale models, owing in part to greater synoptic-scale forcing, while smaller systems are not as well predicted, likely due to greater forcing at the mesoscale that is not well-resolved by the model.
Degree
Ph.D.
Advisors
Trapp, Purdue University.
Subject Area
Meteorology|Atmospheric sciences
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.