Search

Future global temperature change will have significant effects on society and ecosystems. Earth system models (ESM) are the primary tools to explore future climate change. However, ESMs have great uncertainty and often run at a coarse spatial resolution (usually about 2°). Accurate high‐spatial‐resolution temperature dataset are needed to improve our understanding of temperature variations and for many other applications. We apply Super resolution (SR) in computer vision using deep learning (DL) to merge 31 ESMs data. The proposed method performs data merging, bias‐correction, and spatial downscaling simultaneously. The CRU TS (Climate Research Unit gridded Time Series) data is used as reference data in the model training process. To find a suitable DL method, we select five SR methodologies with different structures. We compare the performances of the methods based on mean square error (MSE), mean absolute error (MAE) and Pearson correlation coefficient (R). The best method is used to merge the projected monthly data (1850–1900), and monthly future scenarios data (2015–2100), at the high spatial resolution of 0.5°. Results show that the merged data have considerably improved performance compared with individual ESM data and their ensemble mean (EM), both spatially and temporally. The MAE shows significant improvement; the spatial distribution of the MAE widens along the latitudes in the Northern Hemisphere. The MAE of merged data is ranging from 0.60 to 1.50, the South American (SA) has the lowest error and the Europe has the highest error. The merged product has excellent performance when the observation data is smooth with few fluctuations in the time series. This work demonstrates the applicability and effectiveness of the DL methods in data merging, bias‐correction and spatial downscaling when enough training data are available. Data can be accessed at https://doi.org/10.5281/zenodo.5746632. [ABSTRACT FROM AUTHOR]

ABSTRACT In this article, the analysis of simulated and projected climatic changes in the Indus River Basin ( IRB) is presented. The performance of the regional climate model COSMO-CLM ( CCLM) driven by the global circulation model MPI-ESM-LR is evaluated on the ability of reproducing temperature, precipitation, and wind pattern for the period 1961-2005. There exist quantitative biases, especially in precipitation pattern, which are associated to the diverse reproduction of the atmospheric circulation by the global circulation model. The overall results show that CCLM is able to satisfyingly capture the dominant spatial characteristics in annual temperature and precipitation time series, including the warming trend, and the seasonal variations. The changes in precipitation and temperature over the IRB are projected for the mid- (2046-2065) and late-21st century (2081-2100). Relative to the baseline period (1986-2005), the average annual temperature will increase during the mid- and late-21st century over the whole basin. Extreme temperature events, i.e. heat waves, are more likely to occur. An increase in summer temperature is also projected especially in the upper IRB, where the persistent increase is likely to cause further melting of glaciers. The annual precipitation will decrease in both the mid- and late-21st century with significant spatial changes. A decrease in the monsoon precipitation is also projected, particularly in the central and southern plains of lower altitudes, which might be highly related to the reduction of wet air from the Indian Ocean and the increment in outflow to the east of the IRB. For winter and spring precipitation, except for the late-21st century under the representative climate pathway (RCP) 2.6, a decreasing trend is projected, which might be caused by the reduction of water vapour from the west. This will further weaken the availability of water resources in these seasons, especially in the upper IRB of higher altitudes. [ABSTRACT FROM AUTHOR]