STAR-ESDM: A Generalizable Approach to Generating High-Resolution Climate Projections Through Signal Decomposition

Abstract

High-resolution climate projections are critical to assessing climate risk and developing climate resilience strategies. However, they remain limited in quality, availability, and/or geographic coverage. The Seasonal Trends and Analysis of Residuals empirical statistical downscaling model (STAR-ESDM) is a computationally-efficient, flexible approach to generating such projections that can be applied globally using predictands and predictors sourced from weather stations, gridded data sets, satellites, reanalysis, and global or regional climate models. It uses signal processing combined with Fourier filtering and kernel density estimation techniques to decompose and smooth any quasi-Gaussian time series, gridded or point-based, into multi-decadal long-term means and/or trends; static and dynamic annual cycles; and probability distributions of daily variability. Long-term predictor trends are bias-corrected and predictor components used to map predictand components to future conditions. Components are then recombined for each station or grid cell to produce a continuous, high-resolution bias-corrected and downscaled time series at the spatial and temporal scale of the predictand time series. Comparing STAR-ESDM output driven by coarse global climate model simulations with daily temperature and precipitation projections generated by a high-resolution version of the same global model demonstrates it is capable of accurately reproducing projected changes for all but the most extreme temperature and precipitation values. For most continental areas, biases in 1-in-1000 hottest and coldest temperatures are <0.5°C and biases in the 1-in-1000 wet day precipitation amounts are <5 mm/day. As climate impacts intensify, STAR-ESDM represents a significant advance in generating consistent high-resolution projections to comprehensively assess climate risk and optimize resilience globally.