Characterizing forest recovery effects on streamflow using process-based hydrological models and multisource remote sensing

Changes in vegetation structure and cover due to forests regrowing after stand-replacing disturbance can lead to dynamic changes in streamflow. Process-based hydrological models are valuable tools for understanding streamflow response to disturbance, however rely on accurate parameterization, which is often limited by data availability. As a result, few studies incorporate forest recovery dynamics thereby limiting our understanding of the long-term effects of disturbance and recovery on streamflow. To address this, first, I characterized forest recovery of leaf area index (LAI), a key parameter in many process-based hydrological models, using two remote sensing datasets: a monthly 30m satellite-derived LAI time series and an airborne laser scanning (ALS)-derived LAI dataset. These data are used to assess how variation in forest recovery parameterization of LAI for a process-based hydrological model affects modelled streamflow estimates. Second, these LAI data, as well as ALS-derived canopy height and canopy cover, are used to develop a temporally high-resolution parameterization of forest recovery. Using this new forest recovery parameterization, hydrological model simulations are run and streamflow response to disturbance across elevations and stages of forest recovery is assessed. The results showed that ALS-derived LAI differed by up to 40% from the satellite product, which underestimated LAI in dense forests. Five years post-disturbance, LAI reached one third of pre-disturbance values. After 10 years, LAI recovery ranged from 45-73% across ecozones. The LAI parameterizations derived from the remote sensing data resulted in notable differences in daily and mean annual flow, as well as 2- and 20-year peak flows. The results also demonstrated that 10-30% disturbance in both upper and lower elevation regions of the catchment increased 2- and 20-year peak flows, with relative change increasing with disturbance elevation and spatial extent as well as return period. Elevated peak flows persisted for over two decades, with substantial recovery coinciding with forest regrowth after 20–30 years. These findings highlight the importance of representing vegetation recovery dynamics in modelling the long-term hydrological impacts of forest disturbance.