Preparing for Future Floods: Leveraging Remotely Sensed Data, Modeling, and Social Science Information in a Multilayer Network Approach
Abstract
Many areas in the United States and around the world experience compound flooding events, in which a combination of pluvial, fluvial and coastal flooding affects residents in often dramatic ways, particularly in vulnerable communities. Parts of Texas (e.g., the cities of Beaumont in southeast Texas and Austin in central Texas), are examples of areas where vulnerable communities reside and are exposed to compound flooding events. Here, emergency responders and planners lack easy-to-use and computationally efficient tools for estimating flood inundation, highlighting high risk regions, and identifying the most impacted communities. In this presentation, we will show and discuss various applications that leverage remotely sensed data to quantify flood inundation during compound events and identify communities most at risk, using Beaumont and Austin as study areas. First, we will describe a computationally efficient approach for estimating flood inundation from compound flooding events. We will show the validation of our approach with high fidelity numerical modeling and show that we can obtain fairly accurate, fast, and computationally efficient solutions to flood inundation mapping. These inundation maps are static and capture flood inundation extent and depth under the worst conditions during an event. To quantify dynamic aspects of a flood event, for example a resident's access to various services and critical resources, we show results from a network based approach that leverages remotely sensed data and a cost function to quantify possible routes for reaching services during and immediately after flood events. By applying a cost minimization algorithm that accounts for the impacted road conditions, we can compute a household's change in travel costs. To highlight which communities are most at risk of flooding, we combine these technical methods with social science data by overlaying a social vulnerability index with our compound flood inundation maps. Additionally, we show results from the analysis of a recent stakeholder survey that captures current flood risk perception among the stakeholders in our study areas. Finally, to address the needs of emergency responders during flood events, we show an application called Pin2Flood that is based on our flood inundation maps and allows emergency responders to map the likely extent of a flood during an event. All these applications leverage remotely sensed data and a multilayer network-based approach to provide fast and easy to use flood inundation maps and estimate the social implications of these events.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2022
- Bibcode:
- 2022AGUFM.H46D..01P