
Research Background
Groundwater is a vital component of the water cycle in alpine regions, and its recharge processes are strongly regulated by freeze–thaw (FT) cycles. Under climate warming, permafrost on the Qinghai–Tibet Plateau (QTP) is degrading and the active layer is deepening, altering the storage and transport conditions of soil water and thereby reshaping the hydrological linkages among precipitation, snowmelt water, soil water and groundwater. Nevertheless, systematic understanding remains limited regarding how groundwater recharge sources shift across different FT stages and how their migration pathways change.
To address this question, the research team led by Prof. Xiaoyan Li at Beijing Normal University selected the Qinghai Lake Basin (QLB) as the study area. Combining stable hydrogen and oxygen isotope tracing with the MixSIAR Bayesian mixing model, the team quantitatively resolved groundwater recharge sources across different FT stages and revealed how FT processes influence soil hydrological connectivity and groundwater recharge pathways.

Figure 1. (a) Location of the QLB on the Qinghai–Tibet Plateau (QTP). (b) Spatiotemporal distribution of permafrost within the QLB; grey denotes seasonally frozen ground and orange denotes permafrost. (c) Location of the QLB and the distribution of sampling sites. (d) Groundwater sampling. (e) Precipitation sampling. (f) Monthly soil sampling at adjacent sites.
Research Methods
Study design:The study area is the Qinghai Lake Basin in the northeastern QTP. From May to October 2025, the team established 15 groundwater sampling sites, 17 soil sampling sites and 11 precipitation sampling sites across the basin, collecting 481 samples in total — 90 groundwater, 306 soil and 85 precipitation samples. Soil samples were taken from the 0–30 cm, 30–60 cm and 60–90 cm layers.
Based on air temperature and soil temperature, the study period was divided into three stages: May–June as the thawing period, July–September as the completely thawed period, and October as the freezing period.
Sample collection: Monthly field campaigns were conducted in the QLB, yielding 481 samples of precipitation, groundwater and soil at different depths, with δ²H and δ¹⁸O determined for all water samples. Soil water was extracted using the LI‑2100 Automated Vacuum Water Extraction System (Beijing LICA United Technology Limited). Based on low-pressure vacuum distillation and freezing principles, the system extracts soil water while controlling the extraction process to minimize the influence of evaporation and isotope fractionation on analytical results, thereby providing samples for subsequent stable hydrogen and oxygen isotope analysis. In addition, soil water content (SWC) was determined by the 105 °C oven-drying method to characterize moisture variation across soil layers at different depths.
Statistical analysis: One-way analysis of variance (ANOVA) was used to compare differences in soil water indicators (δ¹⁸O, δ²H, lc-excess and SWC) among soil layers, with the least significant difference (LSD) method applied for multiple comparisons.

Figure 2. Variations in δ²H (a), δ¹⁸O (b) and lc-excess (c) of precipitation, groundwater and soil water at different depths during the freeze–thaw (FT) periods.

Figure 3. Contributions of rainfall, snowmelt and soil water at different depths to groundwater in the hinterland of the Qinghai–Tibet Plateau (QLB) during the FT periods (a, b: thawing period; c, d, e: completely thawed period; f: freezing period).

Figure 4. Recharge contributions of rainfall, snowmelt and soil water at different depths to groundwater during the FT periods. (a) Middle and upper reaches. (b) Downstream reach.

Figure 5. Schematic summary of the recharge contributions of rainfall, snowmelt and soil water at different depths to groundwater during the FT periods.
Key Findings
Soil water at different depths exhibits distinct isotopic signatures, and deep soil water shows strong isotopic similarity to groundwater.
Groundwater recharge does not depend primarily on contemporaneous precipitation, but is closely linked to the storage and transport of soil water at different depths.
Freeze–thaw processes regulate water storage, migration and groundwater recharge by modifying active-layer thickness, soil water content and soil-profile connectivity.
Groundwater recharge pathways are governed not only by FT processes but also by permafrost type, topographic conditions and groundwater table depth.
Conclusion
This study reveals how freeze–thaw processes regulate the recharge sources and migration pathways of shallow groundwater in the Qinghai Lake Basin. As permafrost continues to degrade under climate warming, differences among groundwater recharge sources may gradually narrow, whereas differences among recharge pathways may further widen. Future work can integrate long-term in situ monitoring with multi-tracer techniques to characterize the pathways by which soil water recharges groundwater more precisely, providing stronger scientific support for the sustainable management of groundwater resources in alpine regions.
Journal:Hydrology and Earth System Sciences (Impact factor: 7.0)
Institutions:Beijing Normal University and collaborators
Study site:Qinghai Lake Basin
Instrument:LI‑2100 Automated Vacuum Water Extraction System
DOI: https://doi.org/10.5194/hess-30-6039-2026