Spatiotemporal characteristics of future changes in precipitation and temperature in Central Asia.

The arid and semi‐arid areas in Central Asia have scarce water resources and fragile ecosystems that are especially sensitive and vulnerable to climate change. Reliable information regarding future projections of change in climate is crucial for long‐term planning of water resources management and structural adjustment of agriculture in this region. However, the low‐density meteorological observation network brings great challenges for investigating the effects of climate variations. In this study, variations of precipitation and temperature in Central Asia were examined by a combination of gridded climate dataset of the Climate Research Unit and the latest general circulation models (GCMs) under a representative concentration pathway 4.5. Three downscaling methods, Delta, Advanced Delta, and Bayesian model averaging (BMA) methods, translate the coarse GCMs to local climatic variations for the period 2021–2060 relative to 1965–2004. Major results suggested that the Advanced Delta and BMA methods outperformed the Delta method in precipitation downscaling. Projected precipitation exhibited a general increasing trend at a rate of 4.63 mm/decade for entire Central Asia with strong spatiotemporal heterogeneity. While a declining trend was observed in the southwestern and central parts of Central Asia in summer. The projected temperature was revealed an obvious ascending at 0.37°C/decade, while the warming rate accelerated in higher latitude and mountainous areas. [Correction added on 03 December 2018, after first online publication: The preceding statement has been corrected in this version.] The surface land coverage had significant effects on the variations of precipitation and temperature, respectively. The driven factors of local climate were suggested by analysing the relationships between climate variations and large‐scale atmospheric circulation fields anomalies. The findings of this study aims to provide useful information to improve our understanding for future climate change and benefit local decision makers. The performance of three downscaling methods in terms of precipitation and temperature from general circulation model (GCM) outputs were compared in this study. The relationships between climatic variations and different land cover types were detected based on Land‐Use and Land‐Cover Change (LUCC) and NDVI datasets. The possible drivers of local climatic variations were also suggested by analysing the relationships between local climatic variations and large‐scale atmospheric circulation fields. [ABSTRACT FROM AUTHOR]