Assessment of vegetation response to flow scenarios using a coupled eco-hydrological model in a semi-arid wetland

Semi-arid wetland vegetation is increasingly threatened by limited water availability. Managing environmental flows can help reduce impacts of river regulation and drought, but effective implementation requires timely and predictive quantification of vegetation response. To address this, we develop a scenario analysis framework that enables the continuous quantification of vegetation response under alternative environmental flow scenarios and accounts for predictive uncertainty. The framework allows for probabilistic comparisons of vegetation outcomes and the robust identification of critical thresholds for environmental flow intervention. We demonstrate this framework by applying a coupled eco-hydrological model to simulate inundation extent and vegetation condition in the Narran Lakes, Murray-Darling Basin in Australia, over 58 years. The scenario analysis reveals three key outcomes. First, at annual or five-year scales, there is no clear evidence of vegetation condition decline over the last twenty years, but prolonged drought periods have increased the duration of vegetation stress. Second, by explicitly accounting for predictive uncertainty, we identified the critical vegetation threshold: during the growing season, a leaf area index below 0.4-0.6 indicates a critical state that requires environmental flows to improve vegetation condition. Third, scenario analysis indicates that additional environmental water is needed to improve vegetation from critical conditions, although improvement in condition reaches a saturation point at around 50,000 ML for a flow event in the Narran Lakes.