Confidence Interval Estimation for Quantitative Precipitation Forecasts (QPF) Using Short-Range Ensemble Forecasts (SREF)

Abstract

The Hydrometeorological Prediction Center (HPC) at the NCEP has produced a suite of deterministic quantitative precipitation forecasts (QPFs) for over 40 yr. While the operational forecasts have proven to be useful in their present form, they offer no information concerning the uncertainties of individual forecasts. The purpose of this study is to develop a methodology to quantify the uncertainty in manually produced 6-h HPC QPFs (HQPFs) using NCEP short-range ensemble forecasts (SREFs). Results presented herein show the SREFs can predict the uncertainty of HQPFs. The correlation between HQPF absolute error (AE) and ensemble QPF spread (SP) is greater than 0.5 at 90.5% of grid points in the continental United States, exceeding 0.8 at 10% of these, for the 6-h forecast in winter. On the basis of the high correlation, the linear regression equations of AE on SP are derived at each point on a grid covering the United States. In addition, the regression equations for data categorized according to the observed and forecasted precipitation amounts are obtained and evaluated. Using the regression model equation parameters for 15 categorized ranges of HQPF at each horizontal grid point for each season and individual forecast lead time, an AE associated with an individual SP is predicted, as is the 95% confidence interval (CI) of the AE. Based on the AE CI forecast and the HQPF itself, the 95% CI of the HQPF is predicted as well. This study introduces an efficient and advanced method, providing an estimate of the uncertainty in the deterministic HQPF. Verification demonstrates the usefulness of the CI forecasts for a variety of classifications, such as season, CI range, HQPF, and forecast lead time.

Testing the Tests: What Are the Impacts of Incorrect Assumptions When Applying Confidence Intervals or Hypothesis Tests to Compare Competing Forecasts?

“…To more accurately make inferences about the true performance of an NWP model, the uncertainty in the estimate should be included. More recently, some model comparisons have included estimates of sampling uncertainty through confidence intervals on verification statistics (e.g., Im et al 2006;Wolff et al 2014). Thus, users can more accurately assess the statistical significance of any differences they see in standard verification measures.…”

Improving Probabilistic Ensemble Forecasts of Convection through the Application of QPF–POP Relationships

“…The HPC provides POP forecasts for thresholds of precipitation through their Excessive Rainfall and Winter Weather forecasts. The HPC provides deterministic QPFs, as well, though these forecasts do not provide a measure of uncertainty (Im et al 2006). …”

“…In particular, Olson et al (1995) noted how QPFs were noticeably, consistently worse in the warm season compared to the cool season. Im et al (2006) found that the skill of HPC deterministic QPFs deteriorated as the amount of precipitation increased. The use of ensemble forecasts, however, has advantages over the use of deterministic QPF forecasts.…”