Flood Warning: A Generalized Approach to Forecast the Impacts of Flooding Events Using ArcGIS Pro, QGIS, and Python
Floods are the most common global natural disaster, and 1 billion people live in floodplains worldwide adding to the impactful damage that inundation causes. Disaster managers strive to mitigate damages to their communities but need to know what the impact of a potential flood may be. GEOGloWS ECMWF Streamflow Services estimates forecasted streamflow around the world. These forecasted streamflow's can be used to create predicted flood extent maps using Height Above Nearest Drainage (HAND) or Sedimentation and River Hydraulics - Two Dimension (SRH-2D). Another method to obtain a flood map is using Setinel-1 satellite Synthetic Aperture Radar (SAR) imagery. Flood maps alone will not demonstrate the impact of the flood, but some exposure data will provide needed impact metrics. In this research, I wanted to produce a general geoprocessing method for stakeholders to compute flood impact metrics over any flood extent map using any exposure dataset. Additionally, I sought to create similar geoprocessing workflows in ArcGIS Pro, QGIS, and stand-alone Python script so that the stakeholders can choose the best suited method that correlates with their access and familiarity. The general geoprocessing workflow was tested using three different global exposure datasets (Agriculture, Infrastructure, and Population). The three different geoprocessing implementations were tested in three areas that are of concern in the greater NASA SERVIR organization using the same flood map and exposure datasets for each area. This research produced a feasible, sustainable, successful, generalized geoprocessing workflow that computes flood impact metrics from a flood map and global exposure datasets. The global datasets can be interchanged with higher resolution exposure datasets specific to an area of interest generating more accurate results. The three geoprocessing methods performed similarly. The results were slightly different when the exposure dataset was a raster file as the conversion from raster to vector format produced differences in rounding values and programming implementation. However, this research's findings are such that the three geoprocessing methods are comparable and that any of the three geoprocessing implementations will produce reasonably similar flood impact results. Ongoing work by the Brigham Young University (BYU) Hydroinformatics lab is to create a Tethys web application that will allow stakeholders to view the flood map and flood impact of areas of interest. Future work may include investigating the workflow workability on a global scale, discovering and implementing global exposure data sources of better resolution, researching more data metrics that can contribute to a more robust flood impact results, and increasing the accuracy of flood impact results when compared among ArcGIS Pro, QGIS, and Python.
College and Department
Ira A. Fulton College of Engineering and Technology; Civil and Environmental Engineering
BYU ScholarsArchive Citation
Smith, Robert Evan, "Flood Warning: A Generalized Approach to Forecast the Impacts of Flooding Events Using ArcGIS Pro, QGIS, and Python" (2022). Theses and Dissertations. 9362.
flood impact, flood map, GEOGloWS, HAND, SRH-2D, SAR, geoprocessing