Simulating the Future Temperature and Precipitation Changes in the Yellow River Delta Region

Here the Statistical Downscaling Model (SDSM), the Random Forest Model (RF) and a combination of the two are used to simulate and predict the future seasonal average temperature and precipitation in the Yellow River Delta for the periods of 2030S, 2050S, 2070S. The results found that at the local scale the combination of the two had the smallest root-mean-square errors for annual, seasonal, and daily temperature forecasts; SDSM had the smallest prediction errors for annual precipitation forecasts; and RF had a higher accuracy for seasonal precipitation forecasts except for winter. Climate change trends in the study area are consistent over the next four years, with a gradual increase in average temperatures in all seasons, a gradual expansion of the high-temperature zone to the northeast in spring and summer, and an expansion to the southwest in fall and winter, and a decreasing trend in precipitation until 2030, followed by an increasing trend.