Floods devastate communities. Predicting flood events helps protect them.
Some of the most important technology we have to predict flooding is Hydrologic and Hydraulic (H&H) modeling. However, these technologies face challenges that vary by region, impacting the ability of certain communities to predict and mitigate flood risks.
One crucial aspect of these technologies is stream geometry: the width, depth, and cross-sectional flow area of a waterway. The cross-sectional area of a river is calculated using the discharge rate, or the volume of water that passes over a period of time. But detailed river monitoring data isn’t always easy to come by–how do you determine stream geometry without years of flow rate data?
While most alternatives rely on meteorological data like rainfall, hydrological engineers at the Institute for Resilient Infrastructure Systems (IRIS) propose a technique that can predict watershed characteristics with just simple field measurements.
Using a statistical method called ordinary least squares (OLS) regression, the authors of a new Environmental Modelling and Software publication determined sixteen watershed characteristics using exclusively morphological characteristics, such as stream depth and width, which can be obtained through simple field surveys rather than detailed long-term datasets.
The researchers emphasized the utility of these methods in areas that lack long-term monitoring data, using Puerto Rico as an example to test the method. They also point out the need for region-specific models, accordingly breaking the island into multiple areas for a more accurate result. The team is hopeful that their innovative approach can be applied in similar contexts.
The subject of accessible flood modeling hits close to home: flooding is a major natural threat across the Caribbean and South America, and the paper’s authorship team consists of IRIS researchers from Columbia and Puerto Rico. The Compound Inundation Team for Resilient Applications (CITRA) takes a systematic approach to improving flood modeling technology to help improve flood prediction in communities around the world.