Multi-faceted Approach to Assess Reservoir Sedimentation on Sub-annual Scales

Abstract Submission: Sediment trapping is a critical issue affecting the long-term functionality and capacity of multi-purpose reservoirs, affecting water supplies, recreation, hydropower production, and flood control, in addition to impacting ecological systems. This research aims to quantify the storage capacity reduction in reservoirs due to sedimentation through multiple sensing techniques on sub-annual scales. Traditional approaches to assess sediment accumulation rely on repeated bathymetric surveys over time, which provide valuable historical data on how much reservoir storage has decreased. The Hydrographic Survey Program managed by the Texas Water Development Board (TWDB) has produced nearly 200 surveys on large, multi-purpose reservoirs since the early 1990s. However, these data cannot provide frequent coverage to describe seasonal or interannual change in sediment deposition and its impacts on reservoir storage capacity. To address this challenge, records based on multi-satellite remote sensing data (e.g., Landsat, Sentinel-2, Cubesat) that describe water surface extent are integrated with elevation data and reservoir operations to estimate changes in reservoir storage, complementing the historical sedimentation surveys at a subset of reservoirs throughout Texas. By leveraging machine learning (ML) based image classifications of long-term high spatial resolution images, we demonstrate the degree of change that can be observed on different time scales, including pre- and post-rainfall events and sediment management (e.g., flushing or dredging). The multi-faceted approach allows for a comparison between direct survey methods and remote sensing data, enabling better tracking of sediment deposition trends that advance beyond coarse estimates of annual capacity loss. Ultimately, this effort aims to provide a more accurate and cost-effective framework for managing sedimentation in reservoirs, thereby improving the sustainability and performance of multi-purpose dams.

Learning Objectives/Expected Outcome (Optional) :