Abstract. Pre-disaster planning and mitigation necessitate detailed spatial information about flood hazards and their associated risks. In the US, the Federal Emergency Management Agency (FEMA) Special Flood Hazard Area (SFHA) provides important information about areas subject to flooding during the 1 % riverine or coastal event. The binary nature of flood hazard maps obscures the distribution of property risk inside of the SFHA and the residual risk outside of the SFHA, which can undermine mitigation efforts. Machine learning techniques provide an alternative approach to estimating flood hazards across large spatial scales at low computational expense. This study presents a pilot study for the Texas Gulf Coast region using random forest classification to predict flood probability across a 30 523 km2 area. Using a record of National Flood Insurance Program (NFIP) claims dating back to 1976 and high-resolution geospatial data, we generate a continuous flood hazard map for 12 US Geological Survey (USGS) eight-digit hydrologic unit code (HUC) watersheds. Results indicate that the random forest model predicts flooding with a high sensitivity (area under the curve, AUC: 0.895), especially compared to the existing FEMA regulatory floodplain. Our model identifies 649 000 structures with at least a 1 % annual chance of flooding, roughly 3 times more than are currently identified by FEMA as flood-prone.
The 100-year floodplain is the traditional indicator of flood risk and the area in which specific flood mitigation requirements are required to occur in the United States. However, recent studies have indicated that there is a growing disconnect between the 100-year floodplain and the location of actual losses. As a result, there is a strong need to understand what is undermining the efficacy of the 100-year floodplain and to generate a more accurate depiction of flood risk. However, there have been few studies that examine the characteristics of insured flood claims occurring outside the 100-year floodplain and how more advanced hydrologic models may improve flood risk delineation. This study addresses this issue by cross-validating a fairly new distributed hydrologic flood inundation model and the Federal Emergency Management Association's 100-year floodplain with historical, parcel-level insured flood losses in two subbasins near Houston, Texas. Results illustrate that spatially distributed hydrologic models greatly improve floodplain delineation, provide important insights on the drivers of flood damage outside of the floodplain, and offer alternative ways to more effectively communicate flood risk.
Although the 100-year floodplain is the traditional indicator of risk from flooding and a catalyst for mitigation decisions in the United States, increasing evidence indicates that this boundary is not sufficient in representing actual economic losses caused by floods. Although studies have demonstrated that up to 50% of losses occur outside floodplain boundaries, as of this writing it is believed little or no research has been conducted on the precise spatial characteristics of these losses or offers an alternative approach for depicting flood exposure at the local level. This perceived lack of inquiry is addressed by spatially examining the pattern of insured flood loss within the Clear Creek watershed near Houston as a first step in better understanding the relationship between floodplain boundaries and actual loss. First, property damage claims are mapped under the National Flood Insurance Program over an 11-year period from 1999 to 2009 and then these points of loss are analyzed in relation to the 100-year floodplain and other landscape-level proximity variables. Second, spatial cluster analysis of damage points is used to generate an alternative delineation for representing flood risk and associated loss across the landscape. Results provide important insights into the spatial reality of flood damage across coastal watersheds that can help local decision makers and homeowners better understand the risk of property damage from flooding events.
Characteristics of the built environment and overall local-level land use patterns are increasingly being attributed to greater surface runoff, flooding and resulting economic losses from flood events. Specific configurations of impervious surfaces and land cover may be as important to determining a community's flood risk as baseline environmental conditions. This study addresses this issue by statistically examining the impacts of adjacent land use and land cover (LULC) on flood damage recorded on parcels within a coastal watershed in southeast Texas. We analyse empirical models to identify the influence of different LULCs surrounding over 7900 properties claiming insured flood losses from 1999–2009. Results indicate that specific types of surrounding LULCs impact observed flood losses and provide guidance on how neighbourhoods can be developed more resiliently over the long term.
Abstract Despite the increasing economic losses from floods in coastal communities, little observational research has been done at a fine spatial scale to identify the relative influence of residential location in predicting adverse economic impacts. In response, this study conducts a parcel‐level analysis of flood losses to identify the influence of specific location‐based variables on property damage from multiple flood events. We statistically isolate the effect of multiple location‐based characteristics on insured flood claims associated with two major coastal storms for over 7813 properties within the Clear Creek watershed southeast of Houston, Texas. Results indicate that location‐based variables are among the strongest predictors for both, where seemingly subtle shifts in location add up to large dollar losses from flooding. These findings provide an increased understanding of the role of physical location within a flood‐vulnerable region and how residential choice can be a major factor in exacerbating actual property loss.
Abstract Major coastal flooding events over the last decade have led decision makers in the United States to favor structural engineering solutions as a means to protect vulnerable coastal communities from the adverse impacts of future storms. While a resistance‐based approach to flood mitigation involving large‐scale construction works may be a central component of a regional flood risk reduction strategy, it is equally important to consider the role of land use and land cover ( LULC ) patterns in protecting communities from floods. To date, little observational research has been conducted to quantify the effects of various LULC configurations on the amount of property damage occurring across coastal regions over time. In response, we statistically examine the impacts of LULC on observed flood damage across 2,692 watersheds bordering the Gulf of Mexico. Specifically, we analyze statistical linear regression models to isolate the influence of multiple LULC categories on over 372,000 insured flood losses claimed under the National Flood Insurance Program per year from 2001 to 2008. Results indicate that percent increase in palustrine wetlands is the equivalent to, on average, a $13,975 reduction in insured flood losses per year, per watershed. These and other results provide important insights to policy makers on how protecting specific types of LULC can help reduce adverse impacts to local communities.
'/%3e%3c/svg%3e)