A recent study led by Prof. CHEN Yaning from the Xinjiang Institute of Ecology and Geography (XIEG) of the Chinese Academy of Sciences has released the Tianshan Watershed Streamflow (TSWS) dataset (1901-2019). The dataset compiles daily streamflow data for 56 watersheds and monthly data for 89 watersheds in the Tianshan Mountains. The study was published in Scientific Data.

The researchers reconstructed streamflow observations by integrating data from both domestic and international monitoring stations. They used Hydrologiska Byråns Vattenavdelning (HBV) model and Long Short-Term Memory (LSTM) model to achieve this, and validated their results with S-tests (Nash-Sutcliffe efficiency ≥0.5, percent bias ≤25%, and ratio of the root-mean-square error to the standard deviation of the measured data ≤0.7).

The study demonstrated that the LSTM model was particularly effective at simulating long-term streamflow patterns in the Tianshan region. Streamflow simulations for 40 watersheds at the daily scale and 70 at the monthly scale passed the S-tests.

In addition, the study revealed an overall increasing trend in Tianshan streamflow, which was particularly prominent between 1990 and 2019.

"To the best of our knowledge, this is the first comprehensive and long-term streamflow modeling and data reconstruction effort at the watershed scale in the Tianshan Mountains," said LI Shuai, first author of this study.

The TSWS dataset fills the gap in the availability of comprehensive streamflow data for smaller basins in the Tianshan Mountains, and provides essential data for managing water resources and assessing the impact of climate change in the region.

 Hydrological stations, river systems, and watersheds in Tianshan Mountains. (Image by XIEG)