Downscaling and bias-correction contribute considerable uncertainty to local climate projections in CMIP6

Presented at the Interdisciplinary Workshop on Weather and Climate Extremes in Clemson, SC in May 2023.

Transcript

1. David Lafferty & Ryan Sriver, Department of Atmospheric Sciences, University of Illinois Urbana-Champaign MOTIVATION • Efforts to diagnose the risks of a changing climate often rely on downscaled and bias-corrected climate information. • What are the uncertainties and potential biases of this approach? Downscaling and bias-correction contribute considerable uncertainty to local climate projections in CMIP6 davidcl2@illinois.edu david0811.github.io This work was supported by the U.S. Department of Energy, Office of Science, Biological and Environmental Research Program, Earth and Environmental Systems Modeling, MultiSector Dynamics under Cooperative Agreement DE-SC0022141. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the US Department of Energy. Lafferty, D.C. & Sriver, R.L. (2023) Downscaling and bias-correction contribute considerable uncertainty to local climate projections in CMIP6. ESS Open Archive. METHODS 1. Compile all global, publicly available, downscaled and bias-corrected CMIP6 outputs, including the NEX-GDDP- CMIP6, CIL-GDPCIR, ISIMIP3b, and carbonplan ensembles. 2. Calculate metrics of climate change, including annual temperature and precipitation averages and indices of climate extremes. 3. Perform a simple variance decomposition to partition uncertainty among scenario uncertainty, model uncertainty, downscaling & bias- correction uncertainty, and interannual variability. RESULTS Downscaling and bias-correction are important sources of uncertainty in near-term projections, in projections of precipitation, and in projections of climate extremes. Fig. 1: (a) Projections of annual average temperature in Seattle for each downscaled & bias-corrected model output. Colored lines of different styles show ensemble- scenario means. (b) Resulting variance decomposition. Fig. 2: Global variance decomposition for (a) annual maximum of daily maximum temperature and (b) annual maximum 1-day precipitation. Each column shows the contribution from a different source of uncertainty and each row shows a 20-year period representing either the early, mid, or late 21st century. 2 3 563  55. 2 % 3 (5-10 1 515 . 7 3- 0(  ( 0 3-1 1) . 1804( .-0, 3- -.-59