March 6, 2024
On the uncertainty of long-period return values of extreme daily precipitation.
Regional average uncertainty associated with GEV estimate of extreme precipitation return value for sample sizes of 25 to 200. Uncertainty is standard deviation of the samples of estimates (see text), normalized by the GEV estimate of precipitation for the given region and return period. Uncertainties for both MLE (dashed) and L-moments (solid) estimates of precipitation extreme values are calculated for 20-year return period (black), 50-year return period (blue), 100-year return period (red), and 200-year return period (green).
The Science
Methods for calculating return values of extreme precipitation and their uncertainty are compared using daily precipitation rates over the Western U.S. and Southwestern Canada from a large ensemble of climate model simulations. The roles of return-value estimation procedures and sample size in uncertainty are evaluated for various return periods. We compare two different generalized extreme value (GEV) parameter estimation techniques, namely L-moments and maximum likelihood (MLE), as well as empirical techniques.
The Impact
Even for very large datasets, confidence intervals calculated using GEV techniques are narrower than those calculated using empirical methods. The more efficient L-moments parameter estimation techniques result in narrower confidence intervals than MLE parameter estimation techniques at small sample sizes, but similar best estimates.
Summary
This study provides several recommendations to climate data analysts about statistical analyses of extreme precipitation and temperature. The primary contributions of this paper are four-fold: to compare uncertainty in estimated precipitation return values between various methods of estimating return values; to compare uncertainty in estimated precipitation return values across various sample sizes; to compare practical ways of quantifying uncertainty when the sample size is limited; and to evaluate whether uncertainty approaches zero at increasing sample size.
Contact
Michael F. Wehner
Lawrence Berkeley National Laboratory
Funding
Work at LBNL was supported by the U.S. Department of Energy, Office of Science, as part of research in Regional and Global Model Analysis, Earth and Environmental System Modeling Program under Contract No. DE340AC02-05CH11231 through the Calibrated and Systematic Characterization Attribution and Detection of Extremes (CASCADE) Scientific Focus Area.
Publications
Wehner MF, Duffy ML, Risser M, Paciorek CJ, Stone DA and Pall P (2024) On the uncertainty of long-period return values of extreme daily precipitation. Frontiers in Climate. 6:1343072. doi: 10.3389/fclim.2024.1343072