Spatiotemporal variations and its driving factors of ground surface temperature in China

Abstract

The ground surface temperature (GST) serves as a crucial indicator for understanding land-atmosphere mass and energy exchange. The shift from manual measurement to automated station for GST in China after 2002 introduced inconsistencies at certain stations, potentially distorting research findings. Here, daily automatedly observed GST from 2003 to 2017 at 615 selected meteorological stations were updated by constructing linear regression model based on manually observed air temperature (AT) and GST from 1960 to 2002. Then, the spatiotemporal variations of GST from 1960 to 2017 and its driving factors were investigated. Results indicated that: (1) the AT-GST linear regression model could effectively mitigate the inconsistency caused by the change of GST observation methods, enhancing data reliability. (2) GST in China showed little change from 1960–1980, but increased significantly across all regions from 1980 to 2000, with the increase rate slowed down except in the Qinghai–Tibet plateau (QTP) and southwest China after 2000. Notable GST increase is concentrated in colder regions, including the QTP, northeast (NEC), and northwest China (NWC). (3) Evapotranspiration (ET) and vapor pressure deficit were the primary drivers of annual GST variations at the regional scale, while their contributions to GST variations exhibited notable seasonal variability. Our findings could offer valuable scientific insights for addressing climate change, enhancing surface environmental models, and safeguarding ecological environments.

Keywords

• ground surface temperature (GST)
• data correction
• spatiotemporal variation
• driving factors
• China