A Global Analysis of the Spatial Footprints of Storm Surges.

Abstract:

Understanding the spatial dependences of storm surges on coasts is highly relevant to derive effective risk management and protection plans, since the impact of storm surges may be aggravated as a result of the simultaneous economic and personal damages. However, very few studies included the spatial correlation of storm surges on coasts. Here, we perform the first global analysis of the spatial footprints of storm surges, using observed and simulated storm surge data. Three different techniques are applied to quantify the spatial footprints: K-Means clustering algorithm, percentage of co-occurrence, and joint probability analysis. We assess in more detail the storm surge footprints along the north-western Atlantic coast using an updated reanalysis data set which specifically accounts for tropical cyclone storm track information. Results indicate that contiguous and spatially unconnected coastal stretches are often affected by the same storm surge event. The spatial footprint sizes differ around the globe, partially conditioned by the geography of the coastline. We find an agreement between some spatial footprints and individual extreme storm surge events, mostly caused by notable named storms. For each spatial footprint, we obtain the reference series which can be employed as indicators of storm surge behavior for spatially coherent coastlines. The knowledge gained through the spatial footprint analysis presented here facilitates improved spatial planning and resource allocation for disaster response, including in areas with sparse sea level observations.