Presented at ASPRS Mid-South Region Conference 2026.
Project Context
The Koshi River Basin is one of the most flood-prone regions in Nepal and northern India, with recurring flood events that cause major social and economic losses [1]. Existing flood-risk literature and workflows often emphasize static predictors such as elevation, slope, and drainage, but they do not consistently account for one critical dynamic factor in Koshi: sediment behavior [2]. This project was designed to complement those methods by testing whether satellite-only spectral information can estimate SSC reliably enough to support more robust flood-risk modeling.
Key Constraints
- Event-driven sediment behavior in a steep Himalayan setting
- Limited usable scenes due to revisit timing and cloud cover
- Sensor differences (Landsat vs Sentinel) and resolution mismatch
- Cross-period transfer remains harder than within-period fitting