104-10 Examining the Effects of Precipitation on Flood Severity in a Small Watershed in Central Kentucky

Session: A Showcase of Undergraduate Research in Hydrogeology (Posters)

Presenting Author:

Authors:

Abstract:

Severe flooding is a growing concern in Kentucky, with a total of 294 flood-related events reported between February 2024 and February 2025. In April 2025, another major flood event caused significant damage along the Kentucky River in central Kentucky near the study watershed. Although it is clear that communities near major rivers are vulnerable to flooding, the responses of smaller watersheds (often farther from population centers) to extreme rainfall are poorly understood. Furthermore, flooding results from complex interactions among soil, precipitation, vegetation, and land use, with each factor seldom examined in isolation. This study aims to investigate the sensitivity of a small central Kentucky watershed solely to variable precipitation.  

Since there are no rain gages in the study watershed, Kentucky Mesonet precipitation data from May 2020 to May 2025 were reviewed for four neighboring counties to identify a significant storm event. Data from the maximum event (July 2021) were downloaded from each station, and inverse distance weighting was used to estimate precipitation at 15-minute intervals within the watershed. Using an established flood modeling workflow and the estimated precipitation, a model of the watershed was developed. This model enabled testing of variable precipitation totals and temporal intervals, using hourly data for the July 2021 event, a February 2023 storm, and a projected storm for the year 2100 as the basis for comparison with the estimated 15-minute precipitation data.  

Key findings indicate that the temporal resolution of precipitation data has a significant impact on flood model results. The model with 15-minute data generated higher flood heights than the hourly data, despite both data sources reporting the same total precipitation. Summer storms were more intense due to shorter heavy rainfall durations, with the July storm causing higher floods than the February storm. Although the 2100 storm had more rainfall than July 2021, flood heights matched the July 15-minute data, indicating hourly data make for efficient models but a less accurate constraint on floods than models run with closer to real-time precipitation. Model results suggest that precipitation timing may be more important than total rainfall in determining flood severity. Thus, flooding may persist as a significant issue even when future storm events are brief.   

Geological Society of America Abstracts with Program. Vol. 57, No. 6, 2025

doi: 10.1130/abs/2025AM-6550

© Copyright 2025 The Geological Society of America (GSA), all rights reserved.