Forecasting Inundation Extents and Flood Impact in the Amazon Basin

Abstract:

As flood frequency and duration increase throughout the world, more effective flood warnings are needed to improve response before and recovery after an event. Group on Earth Observation’s (GEO) Global Water Sustainability (GEOGloWS) provides 15-day streamflow forecasts for river reaches all over the world. While valuable, forecast hydrographs on their own have limited potential for local flood mitigation and response. However, these forecasts can be used as boundary conditions to create flood maps that in turn can forecast flood impact and better inform response. As part of a project in NASA-SERVIR to build capacity in flood warning, we create flood maps in multiple locations throughout the Amazon basin using the Height Above Nearest Drainage (HAND) and Sedimentation and River Hydraulics – Two-Dimension (SRH-2D) methods. We then compare the inundation extents using Synthetic Aperture Radar (SAR) satellite images. The flood maps are compared to determine the value added when switching from a simple “bathtub” method like HAND to a more complex and computationally rigorous 2D model that can better account for backwater effects caused by infrastructure. This study also seeks to create a pattern for generating an SRH-2D model that can be replicated in other basins around the world. These flood mapping techniques are driven by the GEOGloWS streamflow forecasts to create an enhanced flood forecasting system that allows decision-makers and stakeholders to better view and understand flood extents, timing, and impact in order to more effectively protect people and property.