Coupling effects of moisture sources and meteorological factors on stable isotopes in precipitation over the Hunshandake Sandy Land, northern China

Abstract

Effective management of water resources and the development of climate adaptation strategies in arid and semi-arid regions increasingly rely on a comprehensive understanding of water cycle processes in the context of global climate change. This study focuses on the Hunshandake Sandy Land, a key ecological barrier in northern China. It analyzes the influence of climate conditions and moisture sources on precipitation isotope composition using 394 precipitation samples collected between 2014 and 2021. Precipitation δ¹⁸O and d-excess show significant variability and distinct seasonal patterns, with δ¹⁸O values ranging from 1.78 ‰ to −24.51 ‰ and d-excess values from −26.28 ‰ to 39.08 ‰. Temperature is identified as the primary factor influencing precipitation δ¹⁸O and d-excess, accounting for 42 % of the monthly δ¹⁸O variation and 16 % of the monthly d-excess variation. The Local Meteoric Water Line, δ²H = 6.66δ¹⁸O – 3.48, highlights the dry, low-rainfall, and high-evaporation characteristics of the Hunshandake Sandy Land. The backward trajectory calculations reveal that northern China, central China, and Mongolia are the primary sources of moisture absorption, contributing 38.25 %, 26.20 %, and 14.21 % of the moisture to Hunshandake Sandy Land precipitation, respectively. The influence of monsoon climates varies over time, leading to significant variability in the contribution of different moisture source regions to Hunshandake Sandy Land precipitation. Specifically, northern China contributes 36.09 % of the moisture in the wet season and 40.42 % in the dry season, whereas central China contributes 29.39 % in the wet season and 23.02 % in the dry season. The Hunshandake Sandy Land precipitation sample index (I_HSL) was defined, and a multiple linear model for monthly precipitation δ¹⁸O values was developed using temperature, relative humidity, and I_HSL: (δ¹⁸O = 0.320 T - 0.068RH + 0.022 I_HSL - 11.615). This model can explain 49 % of the monthly precipitation δ¹⁸O variability.These findings enhance the understanding of the water cycle in arid and semi-arid regions.