In this module I learned about the many ways storm surge and flooding from natural disaster can affect an area. I learned how GIS is used to map changes from such natural disaster and how this information can be used by policy makers and the general public to make our flood-prone communities safer.
In the lab assignment for this module I performed three analyses. One looked at elevation changes post-hurricane Sandy in Mantoloking, NJ, the next utilized DEM's to map flood zones based on a 2m. storm surge in the state of New Jersey, and the other involved using different types of DEM's to infer flood zones in Collier County, FL using general flood behavior assumptions.
In the first analysis I learned to convert .las files to TIN's (triangular irregular networks) using the TIN to Raster tool in ArcGIS Pro. I then converted the two layers, containing DEM's for the area both before and after hurricane Sandy, to rasters using the TIN to Raster tool. Next, I used the Raster Calculator to subtract the pre-Sandy layer from the post-Sandy layer to provide me with DEM that shows elevation changes after the hurricane struck. In conjunction with a polygon shapefile containing existing buildings in the area, I was able to look at how this area's communities were affected by the floods.
In the image above, one can see a raster providing the elevation change model. The darker the red the greater the erosion due to storm surge, while the darkest blues show areas of buildup perhaps caused by sand accretion or building debris.
To map a 2m. flood zone for the state of New Jersey, I was given a DEM created by a mosaic of rasters combined via the Int tool and a boundary shapefile of the state. I began using the Reclassify feature on the DEM to show breaks for values below or at 2 meters and the rest above 2 meters. By converting this raster to a polygon using the Raster to Polygon tool, I was able to use selection queries to isolate the areas affected by a 2m. surge and create a separate polygon depicting this area. Given this data, I was able to look at how much (percentage) of a certain county (Cape May) would be affected by the surge.
The final analysis for this assignment was the most involved and most detailed. I was provided with a DEM produced by liDAR for Collier County, FL as well as one produced by the USGS using traditional methods. Using similar processes to the second analysis, I created new storm surge rasters at an elevation of 1 meter for each DEM. In doing so, the rasters also provided data for a large area of low-lying, disconnected areas that I was to assume would not be affected by the surge. To exclude these areas from the flood zone, I used the
Region Group tool to isolate large regions of raster data. Using selection query I then selected the areas of desired flood prone area and created new layers isolating them. I then converted the rasters polygons to begin the next step of the analysis. The next step of the analysis called me to determine the impact of the 1 meter surge on buildings in the area. I needed to assign a code to each building to do so. In the building layer, I created two new fields and determined that a code of 1 would mean that the building existed in each flood zone. Adversely, a code of 0 would mean it was not affected by the flood. To populate these fields correctly, I used
Selection by Location with
Intersect as the parameter. By selecting the buildings properly, I then used
Calculate Field in the building attribute table to assign the codes. Once I had the codes, I used
Select by Attributes to get raw counts of number of buildings by type that existed in each flood zone. With this data, I performed accuracy analysis by errors of omission and commission. The result of which I will post below:
Comments
Post a Comment