Using a plant hydraulic model to design more resilient rehabilitated landscapes in arid ecosystems

Abstract

Mining is a major driver of dryland disturbance and degradation, and there is a growing need for effective and resilient methods for restoration of former mine sites. An important restoration goal is preventing water from accessing mine waste, thus avoiding mobilization and transport of contaminants. Evapotranspiration (ET) covers are soil covers where vegetation manages the water balance to minimize leakage into underlying waste, with potential co-benefits of restoring ecological function and fixing carbon. However, cover designs often overlook potentially complex interactions between plant physiology and physical design parameters (cover depth, soil properties, etc.) that affect plant water fluxes, particularly in water-limited environments. To better understand how physiologically mediated dynamics impact cover performance, we develop an ET cover model that mechanistically describes plant-environment interactions through a plant hydraulics framework. We use the model to determine how soil cover depth, a fundamental design parameter, interacts with physiology to impact leakage, plant stress/mortality, and carbon sequestration. The model is parameterized using data from a prior study of plant water relations in engineered cover systems of varying depths. When run under historical rainfall trajectories, the model shows that significant plant water stress was ubiquitous across cover depths and was most frequent in shallower covers, where it was accompanied by higher leakage and lower net carbon assimilation. Precipitation variation had an important role in driving outcomes, and hydraulic impairment of vegetation played a role in higher leakage and lower net carbon assimilation. Design approaches that account for plant physiological processes have the potential to yield more effective and resilient systems, and we present a framework for incorporating these critical feedbacks into the design process.