Graduation Year

2022

Document Type

Dissertation

Degree

Ph.D.

Degree Name

Doctor of Philosophy (Ph.D.)

Degree Granting Department

Civil and Environmental Engineering

Major Professor

Mark Ross, Ph.D.

Committee Member

Kenneth Trout, Ph.D.

Committee Member

Mahmood Nachabe, Ph.D.

Committee Member

Shawn Landry, Ph.D.

Committee Member

Mauricio Arias, Ph.D.

Committee Member

Aydin Sunol, Ph.D.

Keywords

hydrologic modeling, land use evapotranspiration, precipitation, vertical leakage

Abstract

Karst lakes are ubiquitous in ridge terrain settings in limestone aquifer coastal plain environments. In west-central Florida, these lakes are frequently connected to the Upper Floridan aquifer and have unique aquifer recharge characteristics. They have been selectively studied because they commonly have no or very limited surface water discharge and limited drainage areas, have appreciable surface water and groundwater interaction and leak to the deep aquifer. An innovative modeling approach was developed to collectively understand and more precisely quantify this vertical leakage, both from a lake-specific and regional water budget standpoint, for a 21-year study period (2000-2020). A unique, calibrated hydrological model, using a constructed monthly-varied aquifer potentiometric surface coupled to a surface and shallow groundwater simulation model was developed for this purpose. The model calibration for this objective was highly constrained by land use-based evapotranspiration (ET) rates and observed flows and stages. The 59-basin, 230 mi2 study area is located in central Polk County, Florida. The area consists of 88 named lakes with observed water levels, two (2) long-term USGS streamflow stations and limited internal recorded surface flows. The studied lakes were also sub-characterized into 79 ridge lakes and 9 valley lakes. Seven (7) of the valley basins in the study area contain no lake feature. The study found that the average lake area-weighted vertical leakage for the 79 ridge lakes was 26.1 inches/year, while the vertical leakage of the 9 valley lakes averaged only 14.1 inches/year. The ridge lake area represents 54% of the domain while the valley lakes area comprise only 46% of the study area. On average, the net basin vertical leakage rate from the ridge lakes to the Upper Floridan aquifer represents 2.6 inches of deep aquifer recharge (for a 52-inch annual rainfall period) for the basin overall, while the valley lakes contribute only 0.35 inches annually on average. Based on the study results, the ridge lakes contribute 7.5 times more groundwater recharge to the Upper Floridan aquifer than the valley lakes, are one of the larger land use forms in the basin and, while only 10% of the area, contribute nearly half of the overall deep recharge. The study found the vertical conductance of the lakes ranged from 1.39E-05 to 4.61E-03 day-1.

The methodology described in this dissertation for quantifying the surface water and groundwater interaction in the ridge and valley lake environments can be used to investigate vertical leakage and aquifer recharge in other karst lake environments, or any setting where lake and reservoir conditions have shallow aquifer inflows, surface runoff, and leak or gain fluxes from underlying deep aquifers. Furthermore, the highly calibrated hydrologic model developed for this study includes new insight and parameterization for modeling ET rates by land use, which can be useful in other continuous hydrologic modeling simulations to improve predictive results against observed flows and stages. New insight concerning long-term rainfall dataset construction and comparisons between Next Generation Weather Radar (NEXRAD) and point rain gauge records is also gleaned from this study. Overall, the findings presented in this study provide an innovative modeling approach and provide hydrologic modelers with valuable surface water and groundwater interaction rates, modeling parameters, ET targets and especially variability by land use, and rainfall construction information that can be used in future studies or for improved model calibration and water budget results.

Download

Share

COinS