A Framework for Mechanistic Flood Inundation Forecasting at the Metropolitan Scale. Issue 10 (29th September 2022)

• Record Type: Journal Article
• Authors: Schubert, Jochen E.; Luke, Adam; AghaKouchak, Amir; Sanders, Brett F.
• Is Part Of: Water resources research. Volume 58:Issue 10(2022)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2022-09-29
• Subjects: Forecasting -- flooding -- urban -- Hurricane Harvey -- inundation -- Houston; Hydrology -- Periodicals; 333.91
• DOI: 10.1029/2021WR031279
• Languages: English
• ISSNs: 0043-1397

Abstract

Abstract: Urban flooding from extreme precipitation and storm surge is a growing threat to cities, and detailed forecasts of urban inundation are needed for emergency response. We present a mechanistic framework to simulate flood inundation over metropolitan‐wide areas at fine resolution (3 m). A dual‐grid shallow‐water model is used to overcome computational bottlenecks, and an application to Hurricane Harvey focused on pluvial flooding provides a multi‐dimensional assessment of predictive skill. A hindcast model is shown to simulate peak stage across 41 stream gages with a mean absolute error (MAE) of 0.63 m, and hourly stage levels over a 5‐day period with a median MAE and Nash‐Sutcliffe Efficiency (NSE) of 0.74 m and 0.55, respectively. Peak flood level across 228 high water marks (HWMs) were captured with an MAE of 0.69 m. A forecast model forced by Quantitative Precipitation Forecast data is shown to be only marginally less accurate than the hindcast model. Peak stage is simulated with an MAE of 0.86 m, hourly stage is captured with a median MAE and NSE of 0.90 m and 0.41, respectively, and HWMs are captured with an MAE of 0.77 m. The forecast system also achieves hit rates of 90% and 73% predicting distress calls and FEMA damage claims, respectively, based on simulated flood depth. These results demonstrate the potential to operationally forecast pluvial flood inundation in the U.S. with the timeliness and accuracy needed for early warning, and we also highlight future research needs. Plain Language Summary: Major cities across the U.S. and globally are experiencing severe flooding that impacts large populations of people, disrupts economies and livelihoods, and causes extensive damage. While short‐term weather forecasts are now able to predict the extreme precipitation, storm surge, and/or streamflow which create conditions conducive to urban flooding, forecasting of local flood inundation on a street‐by‐street or house‐by‐house basis is not generally available. Here, we present a new modeling system capable of making street‐level forecasts of flood inundation with lead times of hours to several days. We report the level of accuracy in terms of hydrologic skill and the ability to predict distress and damage within the built environment. We also show that the modeling system runs sufficiently fast to support timely decision‐making. This study reports information that cities across the U.S. and elsewhere can use to develop forecast systems useful for damage avoidance and public safety. Key Points: A mechanistic framework is presented for flood inundation forecasting at 3 m resolution and metropolitan scale Hurricane Harvey forecast shows sub‐meter accuracy for high water marks and executes 26 times faster than real‐time Framework demonstrates capacity to forecast human impacts including distress and damage