Abstract:Climate change is expected to alter the magnitude and spatiotemporal patterns of hydro-climate variables such as precipitation, which has significant impacts on the ecosystem, human societies and water security. Global Climate Models are the major tools to simulate historical as well as future precipitation. However, due to imperfect model structures, parameters and boundary conditions, direct model outputs are subject to large uncertainty, which needs serious evaluation and bias correction before usage. In this study, seasonal precipitation predictions from 30 Coupled Model Inter-comparison Project Phase 6 (CMIP6) models and Climate Research Unit observations are used to evaluate historical precipitation climatology in global continents during 1901-2014. A grid based model heterogeneity oriented Convolutional Neural Network (CNN) is proposed to correct the ensemble mean precipitation bias ratio. Besides, regression based Linear Scaling (LS), distribution based Quantile Mapping (QM) and spatial correlation CNN bias correction approaches are employed for comparison. Results of model performance evaluation indicate that generally precipitation prediction is more reliable in JJA than DJF on the global scale. Most models tend to have larger bias ratio for extreme precipitation. In addition, current CMIP6 models still have certain issues in accurate simulation of precipitation in mountainous regions and the regions affected by complex climate systems. Moreover, the proposed grid based model heterogeneity oriented CNN has better performance in ensemble mean bias correction than LS, QM, and spatial correlation CNN, which could consider the relative model performance and capture the features similar to actual climate dynamics. Climate change continuously affects the magnitude and spatiotemporal distributions of precipitation around the world, which further significantly impacts the ecosystems, human society, and water resources. Global Climate Models (GCMs) are commonly used to predict historical as well as future precipitation. However, these models are not perfect and their predictions may not be accurate enough. In this study, we evaluated precipitation predictions of 30 different GCMs for different parts of the global continents based on precipitation observations from Climate Research Unit during 1901-2014. We find that GCMs tend to be more accurate in predicting precipitation in JJA than in DJF. Moreover, GCMs often have trouble in predicting extreme precipitation events. Results also indicate that GCMs have struggles in accurately predicting precipitation in mountainous regions and regions with complex climate patterns. To correct the biases of GCMs, we proposed a method called grid based model heterogeneity oriented Convolutional Neural Network (CNN) and compared it with three traditional methods. The results show that grid based model heterogeneity oriented CNN has better performance in ensemble mean precipitation bias correction. Models perform worse in mountainous regions with complex climate systemThe wet model becomes wetter and dry model becomes drier for extreme precipitation predictionsThe proposed grid-based model heterogeneity oriented Convolutional Neural Network (CNN) has better bias correction performance than Quantile Mapping, Linear Scaling, and spatial correlation CNN

On deep learning-based bias correction and downscaling of multiple climate models simulations

Statistical Downscaling of Temperature Using Global Climate Model Outputs - Effect of Bias correction

Validation of a transient dynamical downscaling simulation approach with multi-model ensemble-based GCM bias corrections

A Generative Adversarial Network–based Unified Model Integrating Bias Correction and Downscaling for Global SST

Customized deep learning for precipitation bias correction and downscaling

An Intercomparison of Deep-Learning Methods for Super-Resolution Bias-Correction (SRBC) of Indian Summer Monsoon Rainfall (ISMR) Using CORDEX-SA Simulations

A Deconfounding Approach to Climate Model Bias Correction

A Climate Downscaling Deep Learning Model Considering the Multiscale Spatial Correlations and Chaos of Meteorological Events

Downscaling Precipitation with Bias-informed Conditional Diffusion Model

Two deep learning-based bias-correction pathways improve summer precipitation prediction over China

Evaluation of bias correction techniques for generating high-resolution daily temperature projections from CMIP6 models

Deep learning for bias-correcting CMIP6-class Earth system models

Non‐stationary Bias Correction of Monthly CMIP5 Temperature Projections over China Using a Residual‐based Bagging Tree Model

DeepSD: Generating High Resolution Climate Change Projections through Single Image Super-Resolution

A High-Resolution Land Surface Temperature Downscaling Method Based on Geographically Weighted Neural Network Regression

Machine learning-based spatial downscaling and bias-correction framework for high-resolution temperature forecasting

Conditional diffusion models for downscaling & bias correction of Earth system model precipitation

Statistical downscaling of daily temperature and precipitation over China using deep learning neural models: Localization and comparison with other methods

Investigating the Performance of CMIP6 Seasonal Precipitation Predictions and a Grid Based Model Heterogeneity Oriented Deep Learning Bias Correction Framework

Multivariate bias correction of regional climate model boundary conditions

A two-step downscaling method for high-scale super-resolution of daily temperature — a case study of Wei River Basin, China