The distribution of the daily average air temperature with high spatial resolution is vital for hydro-ecological applications. The air temperature usually recorded at fixed-point stations provides little distribution information and easily suffers from the scarce amount and uneven distribution of the stations in the data sparse regions. In this study, a method based on multisource spatial data was developed to estimate the spatial distribution of daily average temperature, especially for data sparse regions. In this method, the instantaneous temperature was retrieved first using the moderate resolution imaging spectroradiometer data, which was then transformed to a daily value using transformation equations. Second, the global land data assimilation system air temperature data were spatially downscaled and used to improve the data accuracy from step 1 at low temperatures. This method was applied in the Ili River basin in Central Asia, and the results were evaluated against data from two stations’ observations and in situ data from a field test site. The results showed the correlation coefficient varies from 0.90 to 0.94 and the root mean square deviation is ∼3°C , indicating the generated temperature matched the observations well. This suggests the method is an alternative for data sparse regions.