Improving flood hazard datasets using a low-complexity, probabilistic floodplain mapping approach.

Area of Science:

• Hydrology
• Geospatial analysis
• Climate change adaptation

Background:

• Existing flood hazard maps are insufficient for communicating evolving flood risks due to climate change.
• Traditional floodplain mapping is data-intensive and difficult to scale, while low-complexity models lack process representation.
• There is a need for efficient and accurate methods to map flood inundation extents.

Purpose of the Study:

• To formalize and extend the Height Above Nearest Drainage (HAND) approach into the probHAND model.
• To incorporate uncertainty analysis into low-complexity floodplain mapping.
• To produce probabilistic floodplain maps for large areas using publicly available datasets.

Main Methods:

• Developed the probHAND model, an extension of the HAND approach with uncertainty analysis.
• Utilized publicly available datasets for rapid mapping over large areas.
• Applied the model to the Lake Champlain Basin, Vermont, USA for an example application.

Main Results:

• The probHAND model can generate probabilistic floodplain maps efficiently.
• Uncertainties in parameter inputs and process simplifications led to an average 40% change in inundated areas.
• The model's probable inundation extent effectively captured observed and modeled flood extents.

Conclusions:

• Low-complexity models like probHAND, when accounting for uncertainties, are sufficient for flood risk and conservation mapping.
• Investing in accurate topographic datasets and flood frequency analyses is crucial for improving flood hazard datasets.
• Reducing model output variability increases the certainty of flood inundation extents.