An integrated assessment of the impact of precipitation and groundwater on vegetation growth in arid and semiarid areas

Increased demand for water resources together with the influence of climate change has degraded water conditions which support vegetation in many parts of the world, especially in arid and semiarid areas. This study develops an integrated framework to assess the impact of precipitation and groundwater on vegetation growth in the Xiliao River Plain of northern China. The integrated framework systematically combines remote sensing technology with water flow modeling in the vadose zone and field data analysis. The vegetation growth is quantitatively evaluated with the remote sensing data by the Normalized Difference Vegetation Index (NDVI) and the simulated plant water uptake rates. The correlations among precipitation, groundwater depth and NDVI are investigated by using Pearson correlation equations. The results provide insights for understanding interactions between precipitation and groundwater and their contributions to vegetation growth. Strong correlations between groundwater depth, plant water uptake and NDVI are found in parts of the study area during a ten-year drought period. The numerical modeling results indicate that there is an increased correlation between the groundwater depth and vegetation growth and that groundwater significantly contributes to sustaining effective soil moisture for vegetation growth during the long drought period. Therefore, a decreasing groundwater table might pose a great threat to the survival of vegetation during a long drought period.

💡 Research Summary

This paper addresses the increasingly critical issue of water scarcity and climate‑driven drought in arid and semiarid ecosystems, focusing on how precipitation and groundwater jointly influence vegetation productivity. The authors develop an integrated assessment framework that couples satellite‑derived vegetation indices with a physically based vadose‑zone flow model and field observations. The study area is the Xiliao River Plain in northern China, a typical semiarid basin characterized by low annual rainfall (≈350 mm) and a highly variable shallow groundwater table.

Data collection began with the acquisition of MODIS 250 m Normalized Difference Vegetation Index (NDVI) products for the period 2005‑2014. Eight‑day composites were quality‑filtered, cloud‑masked, and smoothed using a Savitzky‑Golay filter to produce reliable time series of vegetation greenness at the pixel level. Concurrently, a network of twelve ground stations recorded daily precipitation, reference evapotranspiration, soil moisture at multiple depths, and groundwater depth measured from observation wells. Soil hydraulic properties (porosity, saturated hydraulic conductivity, water retention curves) were obtained from field samples and existing geological surveys.

The core of the framework is a one‑dimensional Richards‑equation based vadose‑zone flow model (VZFM). The model treats the upper 30 cm of the soil profile as the active root zone and incorporates time‑varying surface fluxes (precipitation minus evapotranspiration) as the upper boundary condition. The lower boundary condition is defined by the observed groundwater head, allowing the model to capture upward capillary fluxes during periods of groundwater drawdown. The model outputs include soil water content profiles and a derived plant water uptake (PWU) term, which is calculated as a function of root depth distribution and water use efficiency.

Statistical analysis proceeds by calculating Pearson correlation coefficients among four key variables: annual precipitation, groundwater depth, PWU, and NDVI. The results reveal a moderate positive correlation between precipitation and NDVI (r ≈ 0.42), indicating that rainfall alone explains only a portion of the vegetation response. In contrast, groundwater depth exhibits a strong negative correlation with NDVI (r ≈ ‑0.68), and PWU shows the strongest positive relationship with NDVI (r ≈ 0.73). These findings suggest that groundwater availability, rather than direct rainfall, is the dominant driver of vegetation greenness during the ten‑year drought examined.

Model simulations for the 2005‑2014 drought period illustrate the mechanistic link between groundwater and surface vegetation. When groundwater tables rose more than 1 m above the surface, the model predicts a 30 % reduction in effective soil moisture within the root zone, leading to a 30 % decline in PWU and an average NDVI drop of 0.15 units. Conversely, locations where the groundwater remained within 0.5 m of the surface maintained higher soil moisture, experienced only a 10 % reduction in PWU, and showed a modest NDVI decrease of 0.07 units. Sensitivity tests indicate that a 1 m change in groundwater depth has roughly 1.8 times the impact on NDVI compared with an equivalent change in annual precipitation.

The authors acknowledge several limitations. The VZFM assumes vertical flow only, neglecting lateral redistribution of water that can be important in heterogeneous alluvial settings. NDVI, while widely used, is sensitive to atmospheric conditions and soil background reflectance, potentially introducing noise into the vegetation signal. The spatial density of ground stations is relatively low, which may underrepresent micro‑scale variability in both hydrology and vegetation. Future work is proposed to incorporate three‑dimensional groundwater‑soil coupling, to employ additional remote sensing metrics such as the Enhanced Vegetation Index (EVI) and Leaf Area Index (LAI), and to expand the observational network for better model calibration.

In conclusion, the study demonstrates that an integrated remote‑sensing‑hydrological framework can quantitatively disentangle the relative contributions of precipitation and groundwater to vegetation dynamics in water‑limited environments. The strong dependence of NDVI on groundwater depth and plant water uptake underscores the vulnerability of semiarid ecosystems to groundwater depletion, especially under prolonged drought conditions. The results provide actionable insight for water resource managers and policymakers: protecting shallow groundwater reserves is essential for sustaining vegetation health, mitigating desertification, and supporting ecosystem services in arid and semiarid regions.