Two sets of bias-corrected regional UK Climate Projections 2018 (UKCP18) of temperature, precipitation and potential evapotranspiration for Great Britain

Earth system science data, Journal Year: 2025, Volume and Issue: 17(5), P. 2113 - 2133

Published: May 20, 2025

Abstract. The United Kingdom Climate Projections 2018 (UKCP18) regional climate model (RCM) 12 km perturbed physics ensemble (UKCP18-RCM-PPE) is one of the three strands latest set UK national projections produced by Met Office. It has been widely adopted in impact assessment. In this study, we report biases raw UKCP18-RCM simulations that are significant and likely to deteriorate assessments if they not adjusted. Two methods were used bias-correct UKCP18-RCM: non-parametric quantile mapping using empirical quantiles a variant developed for third phase Inter-Sectoral Impact Model Intercomparison Project (ISIMIP) designed preserve change signal. Specifically, daily temperature precipitation 1981 2080 adjusted members. Potential evapotranspiration was also estimated over same period Penman–Monteith formulation then bias-corrected latter method. Both successfully corrected range temperature, precipitation, potential metrics reduced multi-day lesser degree. An exploratory analysis projected future changes confirms expectation wetter, warmer winters hotter, drier summers shows uneven different parts distributions both precipitation. bias-correction preserved signal almost equally well, as well spread among changes. factor method benchmark show it fails capture variables, making inadequate most assessments. By comparing differences between two within members, uncertainty stemming from parameterization far outweighs introduced selecting these methods. We conclude providing guidance on use datasets. datasets bias-adjusted with ISIMIP3BA publicly available following repositories: https://doi.org/10.5281/zenodo.6337381 (Reyniers et al., 2022a) https://doi.org/10.5281/zenodo.6320707 2022b). datasets, method, at https://doi.org/10.5281/zenodo.8223024 (Zha 2023).

Language: Английский

Correcting biases in regional climate model boundary variables for improved simulation of high-impact compound events

iScience, Journal Year: 2023, Volume and Issue: 26(9), P. 107696 - 107696

Published: Aug. 21, 2023

Although climate models have been used to assess compound events, the combination of multiple hazards or drivers poses uncertainties because systemic biases present. Here, we investigate multivariate bias correction for correcting in boundaries that form inputs regional (RCMs). This improves representation physical relationships among variables, essential accurate characterization events. We address four types events result from eight different hazards. The results show while RCM simulations presented here exhibit similar performance some event types, broadly compared no univariate correction, particularly coincident high temperature and precipitation. with uncorrected tends produce a negative return period these suggesting tendency over-simulate respect observed

Language: Английский

Correcting Multivariate Biases in Regional Climate Model Boundaries: How Are Synoptic Systems Impacted Over the Australian Region?

Geophysical Research Letters, Journal Year: 2024, Volume and Issue: 51(21)

Published: Nov. 8, 2024

Abstract Synoptic climatology, which connects atmospheric circulation with regional environmental conditions, is pivotal to understanding climate dynamics. While models (RCMs) can reproduce key mesoscale precipitation patterns, biases related synoptic from the driving model, typically global (GCMs), often remain unaddressed. This study examines influence of correcting systematic bias in RCM boundaries on representation Australian systems. We utilize a structural self‐organizing map evaluate frequency, persistence, and transitions daily Our findings reveal that an multivariate bias‐corrected improves systems compared GCM, or uncorrected simply boundaries, particularly reference frequency identified. demonstrates appropriately boundary conditions helps correct many errors inherited GCM but not all.

Language: Английский