Manuscript received November 23, 2022; revised January 10, 2023; accepted February 16, 2023; published October 26, 2023.
Abstract—The effects of climate change, including severe droughts, fires, and extreme weather events, are increasing every year. One of the most dangerous natural disasters around the world is flooding. Remote sensing and machine learning offer opportunities to utilize remotely collected data for analysis and predictive modeling purposes. Using Python programming methods, Sentinel-1 Synthetic Aperture Radar (SAR) data was collected for water pixel detection from Google Earth Engine and combined with National Oceanic and Atmospheric Administration (NOAA) precipitation data to understand seasonal flood events between 2015 and early 2022. Models were developed to predict flooding in the Lower Cape Fear and greater Wilmington, North Carolina area. One of the challenges in this method is the lack of imagery data recorded within the Region of Interest and time frame as well as the interplay between different data types, programming methods, and outputs. Outputs include data tables and multiple methods for data validation. Overall, accuracy for resulting models was high, with an artificial neural network for binary classification returning an accuracy value greater than 90%.
Index Terms—Flood detection, machine learning, predictive modeling, SAR
D. McMoran currently works as a Data Scientist for Bowman Consulting based out of Virginia, USA
A. Langevin works as an Associate Software Engineer at nCino, inc in Wilmington North Carolina, USA
A. Whittaker works as a Software Engineer at CGI Federal in Durham, North Carolina, USA
P. Poosapati worked at Capgemini Technologies Limited as a Cloud Developer in Azure with 6 Years of Experience, USA
N. Pricope currently works as a Professor of Geography and Geospatial Science at the University of North Carolina Wilmington, Wilmington, North Carolina, USA.
G. Dogan is currently an Assistant Professor in the Computer Science department at the University of North Carolina Wilmington (UNCW), where she founded the Applied AI Lab, USA
*Correspondence: dogangulus@gmail.com (D.M.)
Cite: D. McMoran, A. Langevin, A. Whittaker, P. Poosapati, N. Pricope, and G. Dogan, "Flood Prediction in the Lower Cape Fear River Using SAR Based Water Extraction," International Journal of Modeling and Optimization vol. 13, no. 4, pp. 129-133, 2023.
Copyright © 2023 by the authors. This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (CC BY 4.0).
Copyright © 2008-2024. International Journal of Modeling and Optimization. All rights reserved.
E-mail: ijmo@iacsitp.com