High-resolution CMIP6 climate projections for Ethiopia using the gridded statistical downscaling method

Area of Science:

• Climate Science
• Environmental Modeling
• Data Science

Background:

• Global Circulation Models (GCMs) provide essential climate change projections but suffer from coarse resolution, biases, and uncertainty.
• These limitations hinder their direct application in regional and local adaptation strategy development.
• High-resolution climate data is vital for accurate impact assessments.

Purpose of the Study:

• To develop a high-resolution (10 km) climate change projections database.
• To improve the utility of GCM outputs for regional and local impact modeling.
• To assess the effectiveness of a statistical downscaling technique in bias correction and resolution enhancement.

Main Methods:

• Employed the bias correction constructed analogues with quantile mapping reordering (BCCAQ) statistical downscaling technique.
• Utilized data from 16 Coupled Model Intercomparison Project Phase 6 (CMIP6) GCMs.
• Generated projections under three Shared Socioeconomic Pathways (SSPs): SSP2-4.5, SSP3-7.0, and SSP5-8.5.
• Evaluated the downscaling strategy using a perfect sibling approach.

Main Results:

• The BCCAQ downscaling method substantially reduced biases present in the original GCM outputs.
• Generated daily climate data at a significantly higher spatial resolution (10 km) compared to GCMs.
• Evaluation using contrasting GCMs demonstrated the robustness of the downscaling approach.

Conclusions:

• Statistical downscaling effectively enhances the resolution and corrects biases of GCM projections.
• The resulting high-resolution database is suitable for regional and local climate change impact studies and adaptation planning.
• This work provides a valuable resource for climate change research and policy development.