Deep-Learning-Based Harmonization and Super-Resolution of Near-Surface Air Temperature from CMIP6 Models (1850–2100)

Abstract

Abstract. Future global temperature change would have significant effects on society and ecosystems. Earth system models (ESM) are the primary tools to explore the future climate change. However, ESMs still exist great uncertainty and often run at a coarse spatial resolution (The majority of ESMs at about 2 degree). Accurate temperature data at high spatial resolution are needed to improve our understanding of the temperature variation and for many applications. We innovatively apply the deep-learning(DL) method from the Super resolution (SR) in the computer vision to merge 31 ESMs data and the proposed method can perform data merge, bias-correction and spatial-downscaling simultaneously. The SR algorithms are designed to enhance image quality and outperform much better than the traditional methods. The CRU TS (Climate Research Unit gridded Time Series) is considered as reference data in the model training process. In order to find a suitable DL method for our work, we choose five SR methodologies made by different structures. Those models are compared based on multiple evaluation metrics (Mean square error(MSE), mean absolute error(MAE) and Pearson correlation coefficient(R)) and the optimal model is selected and used to merge the monthly historical data during 1850–1900 and monthly future scenarios data (SSP1-2.6, SSP2-4.5, SSP3-7.0, SSP5-8.5) during 2015–2100 at the high spatial resolution of 0.5 degree. Results showed that the merged data have considerably improved performance than any of the individual ESM data and the ensemble mean (EM) of all ESM data in terms of both spatial and temporal aspects. The MAE displays a great improvement and the spatial distribution of the MAE become larger and larger along the latitudes in north hemisphere, presenting like a ‘tertiary class echelon’ condition. The merged product also presents excellent performance when the observation data is smooth with few fluctuations in time series. Additionally, this work proves that the DL model can be transferred to deal with the data merge, bias-correction and spatial-downscaling successfully when enough training data are available. Data can be accessed at https://doi.org/10.5281/zenodo.5746632 (Wei et al., 2021).

Diffusion model-based probabilistic downscaling for 180-year East Asian climate reconstruction

“…Table 1 unequivocally highlights the superiority of DPDM in downscaling results compared to other statistical methods. This includes deterministic models such as Enhanced Deep Residual Networks for Super-Resolution with Generative Adversarial framework (EDSR-GAN), known for their superiority over traditional statistical techniques [40][41][42][43] , as well as the widely-used linear interpolation methods (Lerp) in meteorology 44,45 (PSNR), Structure Similarity Index Measure (SSIM), and Root Mean Square Error (RMSE). It is worth mentioning that this superiority is particularly evident in the case of precipitation and temperature downscaling, which are very important for social life in a warming climate.…”

Machine-learning ensembled CMIP6 projection reveals socio-economic pathways will aggravate global warming and precipitation extreme

“…Studies have demonstrated that deep learning can reproduce data in pattern coupling with excellent performance (Sun and Archibald,. 2021;Wei et al, 2021). In this study, considering temporal variation, the application of neural network and machine learning reproduce dataset with higher ability of projecting climatological rainfall and temperature under SSP1-2.6, SSP2-4.5 and SSP5-8.5.…”

“…(3) development of machine learning (ML) with nonlinear function to train selected models adjusted by bias correction (Xu et al, 2020;Wei et al, 2021).…”

“…Nowadays, ML applications in data-driven geoscience mainly focus on downscaling (Tran Anh et al, 2019;Vandal et al, 2019), land cover transmission (Condro et al, 2019;Gianinetto et al, 2020) and inversion model construction (Jiang et al, 2019a;Liu and Grana, 2019), etc. To correct climate models, ML has been proved to be an effective tool in taking advantage of excellent features from GCMs in several studies (Wei et al, 2021;Jose et al, 2022). Jose.…”