Date of Award
Master of Science
Department of Engineering Physics
Michael K. Walters, PhD
Probabilistic quantitative precipitation forecasts (PQPF) based on the medium range forecast (MRF) ensemble are currently in operational use below their full potential quality (i.e., accuracy and reliability). This unfulfilled potential is due to the MRF ensemble being adversely affected by systematic errors which arise from an imperfect model and less than ideal ensemble initial perturbations. This thesis sought to construct a calibration to account for these systematic errors and thus produce higher quality PQPF. Systematic errors were explored with the use of the verification rank histogram, which tracks the performance of the ensemble. The information in these histograms was then used in interpreting MRF ensemble forecasts to produce calibrated PQPF. While the calibration technique did noticeably improve the quality of PQPF, its usefulness was bounded by the natural predictability limits of cumulative precipitation. It was discovered that higher levels of cumulative precipitation cannot be reliably predicted in the medium range. Due to this limit of predictability, for significant levels of precipitation (high threshold), the calibration designed in this thesis was found to be useful only for short range PQPF. For low precipitation thresholds, the calibrated PQPF did prove to be of value in the medium range.
DTIC Accession Number
Eckel, Frederick Anthony, "Calibrated Probabilistic Quantitative Precipitation Forecasts Based on the MRF Ensemble" (1998). Theses and Dissertations. 5623.