Module 4: Coastal Flooding

    In this week’s lab, I explored LiDAR data for before and after hurricane Sandy made landfall in New Jersey. I also used different elevation sources (USGS and LiDAR) to determine potential areas at risk for storm surge. Lastly, I then assessed the quality of the USGS elevation source for predicting storm surge threat. 

    Using LiDAR data for before and after hurricane Sandy made landfall, I was able to assess the destruction that Sandy caused. Using a raster calculator, I generated a raster that depicted the change in elevation between the before LiDAR generated layer and the after layer. From that output, I located areas where there was significant erosion of sand and areas where sand had accumulated following the storm. A buildings layer helped me identify the long term effects of the storm. Since the buildings layer was updated in 2019. I could see where houses were not rebuilt following the storm. See the map below to observe the impacts of hurricane Sandy on the New Jersey coastline. 

    Next, I used a DEM layer to delineate areas at risk for 2 meter storm surge within Cape May county, New Jersey. During this exercise, I worked through a workflow for generating a layer of at risk areas for storm surge using a DEM. While completing the workflow, I elected to use the con tool instead of the region group tool. 

    That same workflow was then utilized to conduct the same analysis along a stretch of Florida’s coast using two different elevation sources, USGS and LiDAR. After determining the areas at risk for a 1 meters storm surge, I compared the USGS and LiDAR outputs. I considered the LiDAR layer the accurate layer and determined the quality of the USGS layers output by comparing the number of buildings predicted to be affect by a 1 meter storm surge in each of the two layers. To do this, I used the select by attributes and select by location functions to screen for buildings intersecting each layer. Then, I compared the numbers. 

    To compare the quality of the USGS generated layer, I calculated the error of commission and omission. This was done by determining the building at risk and not at risk in the LiDAR and USGS layer. I created new fields in the buildings layer; the fields were usgs, lidar, and SurgeCode. I assigned a value of 1 to the buildings intersecting the USGS storm surge layer and a 0 for those that did not intersect. For the LiDAR layer, I assigned a value of 2 to the buildings intersecting the layer and a 0 for those that did not intersect. Then, I calculated the SurgeCode field by adding the usgs and lidar together. 

SurgeCode Key:

0 = not at risk

1 = at risk based only on USGS layer

2 = at risk based only on LiDAR layer

3 = at risk based on both layers

    Using the SurgeCode field, I was able to calculate the error of omission and commission. Based on the errors of omission and commission, the USGS elevation source was not a very accurate elevation source to use for storm surge predictions. The error of commission, in other words the buildings marked at risk in the USGS layer, but not marked at risk in the LiDAR layer, was particularly high. See the map below which highlights the differences between the USGS and LiDAR storm surge layers as well as the buildings that will be affected by each layer.