The Interannual Cyclicity of Precipitation in Xinjiang During the Past 70 Years and Its Contributing Factors
Precipitation cyclicity plays a crucial role in regional water supply and climate predictions. In this study, we used observational data from 34 representative meteorological stations in the Xinjiang region, a major part of inland arid China, to characterize the interannual cyclicity of regional pre...
Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-05-01
|
| Series: | Atmosphere |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2073-4433/16/5/629 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Precipitation cyclicity plays a crucial role in regional water supply and climate predictions. In this study, we used observational data from 34 representative meteorological stations in the Xinjiang region, a major part of inland arid China, to characterize the interannual cyclicity of regional precipitation from 1951 to 2021 and analyze its contributing factors. The results indicated that the mean annual precipitation in Xinjiang (MAP_XJ) was dominated by a remarkably increasing trend over the past 70 years, which was superimposed by two bands of interannual cycles of approximately 3 years with explanatory variance of 56.57% (Band I) and 6–7 years with explanatory variance of 23.38% (Band II). This is generally consistent with previous studies on the cyclicity of precipitation in Xinjiang for both seasonal and annual precipitation. We analyzed the North Tropical Atlantic sea-surface temperature (NTA<sub>SST</sub>), El Niño-Southern Oscillation (ENSO), North Atlantic Oscillation (NAO), Arctic Oscillation (AO), and Indian Summer Monsoon (ISM) as potential forcing factors that show similar interannual cycles and may contribute to the identified precipitation variability. Two approaches, multivariate linear regression and the Random Forest model, were employed to ascertain the relative significance of each factor influencing Bands I and II, respectively. The multivariate linear regression analysis revealed that the AO index contributed the most to Band I, with a significance score of −0.656, whereas the ENSO index with a one-year lead (ENSO<sub>−1yr</sub>) played a dominant role in Band II (significance score = 0.457). The Random Forest model also suggested that the AO index exhibited the highest significance score (0.859) for Band I, whereas the AO index with a one-year lead (AO<sub>−1yr</sub>) had the highest significance score (0.876) for Band II. Overall, our findings highlight the necessity of employing different methods that consider both the linear and non-linear response of climate variability to driving factors crucial for future climate prediction. |
|---|---|
| ISSN: | 2073-4433 |