Abstract: Soil heat flux (G) is one of key factors in affecting the surface energy balance over the Qinghai-Tibetan Plateau. The estimation and spatiotemporal distribution analysis can provide a reference for the surface energy balance in the region. In this paper, the applicability and accuracy of the model was evaluated by combining SEBAL model inversion data with observation data over the Qinghai-Tibetan Plateau from 2003 to 2018 based on the MODIS, the China regional surface meteorological element driven data set, and the 1 km all-weather surface temperature data set in Western China. We found that SEBAL model had high accuracy for inverting soil heat flux (G) over the Qinghai-Tibetan Plateau. On this basis, soil heat flux (G) value were inverted by remote-sensing data, and analyzed the spatiotemporal distribution characteristics of G in the region from 2003 to 2018. The results showed that: (1)The mean value of G showed a fluctuating downward trend in multiyear, with the maximum valley value appearing in 2011, and the maximum peak value in 2016. The mean value of G showed a fluctuating downward trend in every season, except in winter. The mean value ranges of G in different seasons showed a trend of summer > spring > autumn > winter , and the order of magnitude of fluctuation changes was consistent with former. (2)The mean value of G showed spatial distribution characteristics that the highest area in the northern Qaidam Basin and its surrounding areas, the southwestern region such as Ngari region and circumjacent areas followed, the rest of the area was mostly low in general, and with obvious spatial heterogeneity. The mean value of G in different seasons showed spatial distribution characteristics were basically consistent with former. (3)The central and southeastern areas showed an increasing trend, while the northern, western and southwestern areas showed a decreasing trend. The proportion of areas with an increasing trend was highest in winter and lowest in summer, the proportion of areas with a decreasing trend was highest in summer and lowest in winter. The results of this study demonstrated the applicability of the SEBAL model for inverting G, and enriched the studies of surface energy balance over the Qinghai-Tibetan Plateau.