Modeling revealed the effect of root dynamics on the water adaptability of phreatophytes

Abstract

Given their high root plasticity, phreatophytes have flexible water use strategies and can dynamically adjust their rooting depth for the effective uptake of water from soils and shallow aquifers. By this strategy, phreatophytes are strongly ecologically resilient to water stress and thus are commonly grown in drylands. In this study, we used a modified soil–hydrological model, Hydrus-1D, with the implementation of a dynamic root scheme to analyze the role of the distribution and dynamics of roots in simulating evapotranspiration (ET) of phreatophytes with declining groundwater levels (GWLs). The results showed that the root mean square error (RMSE) between simulated and observed ET was reduced by approximately 39% using a model with a more authentic root distribution than the generic root profile. Static root schemes cannot portray the adaptation of phreatophytes to GWL changes well, but their performance can be enhanced by introducing a water compensation scheme, which, however, has a weakness related to the difficulties in parameter calibration. The ET was well retrieved by the model with the dynamic root scheme. Its RMSE was 40% to 70% less than those of the static models, and the Nash–Sutcliffe efficiency reached 0.94, demonstrating the importance of root dynamics in simulating phreatophytic root water uptake. In addition, multiscenario estimations showed nonlinear responses of phreatophyte ET to the rate of GWL decrease (RGWD); that is, as the RGWD accelerated, plant adaptation showed three different stages: no water stress, adaptable and unadaptable. The resistance of plants to water stress decreases with decreasing root growth rate. The identified key ecological thresholds for the RGWD provide a reference for ecological protection in arid areas. We highlight the importance of root processes in the plant response to water stress and suggest that more attention should be given to the root adaptation process in Earth system models.