Quantifying the time-varying period and time lag features of groundwater response: Dynamic impacts of precipitation-fed groundwater recharge
Understanding dynamic groundwater responses to precipitation is critical for sustainable water resource management. Conventional methods relying on stationary hydrological assumptions fail to capture the time-varying and spatially heterogeneous features of groundwater responses, potentially leading...
Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-07-01
|
| Series: | Ecological Indicators |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S1470160X25005783 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Understanding dynamic groundwater responses to precipitation is critical for sustainable water resource management. Conventional methods relying on stationary hydrological assumptions fail to capture the time-varying and spatially heterogeneous features of groundwater responses, potentially leading to mechanistic misinterpretations. To address these gaps, this study develops Coefficient Weighted-Wavelet Coherence Analysis (CW-WTC) from wavelet coherence to quantify time-varying periods and lags, integrating it with spatiotemporal Independent Component Analysis (stICA) to construct a framework that isolates precipitation-dominated response pattern and validates effectiveness via simulations. Applied to the Heilongjiang Basin, the framework reveals groundwater response periods to precipitation concentrate at 10–14 months, with > 50 % of the basin exhibiting significant decreasing trends (Mann-Kendall test, p < 0.01), indicating intensified oscillatory dynamics. Spatially, time lags form a northeast-southwest divide: 3-month lags in the northeast part and up to 8-month lags in the southwest part, with 99 % of the basin showing significant lag shortening (p < 0.01), reflecting declining groundwater buffering capacity, elevated drought and wetland degradation risk. Driven analysis via Geodetector and XGBoost-SHAP reveals precipitation as the primary driver of the response period and time lag features, with its influence amplified by synergies with soil type, aquifer medium type, potential evapotranspiration, porosity and elevation. Notably, soil type surpasses precipitation in governing time lag distribution, and demonstrating critical synergy with aquifer medium. Bipolar SHAP value distributions for these two factors highlight their role in spatial lag zonation, revealing the regulatory importance of soil-aquifer integration in infiltration processes. In summary, this study develops a framework to quantify groundwater storage response dynamics, improving assessments under climate-human pressures. Combined with driven analysis, it uncovers mechanistic links between response features and environmental controls, facilitating the development of robust, ecologically informed water resource strategies to address declining groundwater resilience and protect dependent ecosystems. |
|---|---|
| ISSN: | 1470-160X |