Caragana korshinskii and Tamarix ramosissima are pioneer shrubs for water and soil conservation and windbreak and sand fixation in arid and semi-arid areas. Understanding the water use patterns of C. korshinskii and T. ramosissima and their response to rainfall is of great importance for their survival in regions where drought occurs. In this work we present the monitoring results of stable isotopes in soil water (depths from 0 to 200 cm), twig xylem water of juvenile, intermediate, and adult C. korshinskii and T. ramosissima. The monitoring campaign took place in western Chinese Loess Plateau from July to October 2020. During the same period, we also measured relevant environmental and meteorological variables and soil water content. The results showed that juvenile and of intermediate age C. korshinskii both mainly absorb water from the surface (0–10 cm) and shallow (10–40 cm) soil layers, but adult C. korshinskii use mainly water from the deep soil layers. Juvenile and of intermediate age T. ramosissima extract water from deep soil layers, while adult T. ramosissima use mainly water from middle (40–100 cm) soil depths. Both plant species increase the proportion of surface and shallow soil layer water after precipitation. This increase is more pronounced and faster for the C. korshinskii of intermediate age rather than for juvenile and adult plants. On the contrary, it is the absorption of surface and shallow soil water from juvenile and of intermediate T. ramosissima plants that fluctuates more after precipitation than from adult plants. Our findings provide a reference for vegetation restoration and ecological management of the western Chinese Loess Plateau.

The stable hydrogen and oxygen isotopes as well as their correlation in precipitation have been widely investigated for the understanding of various hydrological processes. Monthly precipitation data were usually recommended in order to establish a linear relationship between the stable hydrogen and oxygen isotope ratios (also known as local meteoric water lines or LMWL for a specific location); however, the LMWL based on daily (or event-based) precipitation data is usually different from that using monthly data. Based on 83 sampling stations across the world from 2000 to 2017, local meteoric water lines were calculated using daily (or event-based) precipitation data (n=9354) and corresponding monthly data (n=1895), respectively; multiple regression methods were used, including ordinary least squares, reduced major axis and major axis regressions as well as their precipitation-weighted counterparts. The global meteoric water line from daily data is δ2H = (7.72 ± 0.02) δ18O + (6.84 ± 0.15) (n=9354, r2=0.96) and from monthly data is δ2H = (7.81 ± 0.04) δ18O+(7.61 ± 0.32) (n=1895, r2=0.96). The stations used in this study were grouped into five climate types, according to the Köppen Climate classification. The precipitation-weighted regression may increase the long-term receptiveness of LMWL using daily-based (or event-based) samples, not only for arid regions, but also for cold regions. When only relatively short-term isotopic records in event-based precipitation samples are available, which is usual in modern hydrological studies, the weighted regression (especially precipitation weighted ordinary least squares regression, PWLSR) is helpful to create a respective local meteoric water line.