Reconstructing high-resolution temperature for the past 40 years in the Tianshan Mountains, China based on the Earth system data products

Abstract

The impacts of climate change on runoff and water resources have been widely concerned. Meteorological stations and observation data are very limited in the mountainous areas of Northwest China, thus challenging accurate hydrological forecasting and water resource evaluation. To address this issue, we developed a method to reconstruct high-resolution temperatures in mountainous areas. Based on the ERA-Interim temperature and high-resolution DEM, nonlinear regression models were fitted and corrected via bilinear interpolation. Using the developed method, monthly temperatures from January 1979 to August 2019 with a spatial resolution of 90 m × 90 m were reconstructed and then verified against observations from 30 meteorological stations. The verification results indicate the high accuracy of the reconstructed data. Based on the reconstructions, we found that over the past 40 years, annual mean temperatures in the Tianshan Mountains had increased by 0.3 °C per decade. The highest annual mean temperature of 8.32 °C was found in East Tianshan (87°47′E-96°E), followed by 5.21 °C in West Tianshan (73°E-80°E), and 3.62 °C in Middle Tianshan (80°E-87°47′E). Altitude is the main factor affecting the spatial distribution of temperature in the study region. We also simulated runoff in the Aksu River and the Kaidu River, both of which originate from the Tianshan Mountains, using reconstructed temperature data. Comparing the accuracy of the simulated runoff with that of simulated using the observed temperature and the raw ERA-Interim temperature data revealed that the simulated runoff using our reconstructions showed the highest accuracy.