Dissertation
Development of a tile module for GHOST and its application
University of Iowa
Doctor of Philosophy (PhD), University of Iowa
Autumn 2020
DOI: 10.17077/etd.005702
Abstract
Subsurface tile drainage networks are widely used in the Midwest United States to improve crop production in poorly drained soils by lowering groundwater levels. In addition to changing surface and subsurface flow patterns, tiles are the primary pathway for nutrient loss from agricultural fields, thus contributing to deteriorated water quality and to environmental catastrophes such as the Gulf of Mexico hypoxia zone. While the agronomic benefit of subsurface drainage is widely known, the hydrological impact of tile at watershed-scale on flooding has remained conjectural. To quantify the impact of tiles on watershed hydrology, especially during heavy storms, a reliable numerical representation at watershed-scale is needed. In this study, a robust, conceptually-based tile module was developed and incorporated into the “Generic Hydrologic Overland Subsurface Toolkit” (GHOST), a fully integrated process-based hydrological model developed by IIHR-Hydroscience & Engineering at the University of Iowa in fulfillment of the Iowa Watershed Approach (IWA) project’s hydrologic assessments.
Including the new tile module, GHOST was used to simulate the Boone River Watershed from January 1st, 2002, until December 31st, 2018. The model was calibrated by comparing observed and simulated flows at Webster City and Goldfield USGS gages. Tile flow field measurements at the outlet of two drainage districts located in Lyons Creek, a sub-basin of the Boone River Watershed, were compared with simulated daily average tile flow.
The calibrated Boone River Watershed model under historical conditions verified the model’s ability to distribute flow accurately amongst hydrological components. According to the annual water balance analysis, tile flow contributes more than 80% of streamflow and baseflow. Moreover, the model showed high sensitivity to the tile depth parameter, consistent with the expected strong relationship between the groundwater level and watershed hydrologic response. Based on sensitivity analysis, shallow tile depths tend to decrease baseflow and increase surface runoff.
The impact of subsurface tile drainage systems on watershed hydrology was further investigated by simulating the Boone River Watershed using the calibrated parameters while turning off the tile module. Comparing the simulated tiled and no-tile Boone River Watershed indicated that, on average, tiles increase annual flow by 35% and decrease subsurface flow and surface runoff flow by 46% and 87%, respectively. Tiles also increase the runoff depth throughout the growing season. The average runoff depth in July, August, and September are subjected to the largest increase by an average of 22 mm.
2002-18 simulation results show tiles often decrease peak flows during heavy rainfalls. In a tile drained-watershed, rainfall is distributed between surface runoff and tile flow, resulting in reduced peaks. However, in an undrained watershed, rainfall mostly turns into surface runoff, causing a slightly flashier response with a higher peak value. In dry conditions, the limited storage capacity of tiled soil results in less subsurface flow and more surface runoff, which can intensify peak flow. Undrained soil absorbs more water during rainfall, which, in subsequent dryer weather, maintains streamflow.
Boone River Watershed hydrologic responses were examined for two hypothetical climate change scenarios in which the precipitation time series were modified using the Climate Science Special Report’s projections of 21st century climate change impact on precipitation intensity and frequency. The first scenario assumed a 10% increase in hourly precipitation of the top 5% wettest days, while the second scenario assumed a 20% increase in same. Results indicate no significant change in the annual balance ratios. Intensifying precipitation by 10% and 20% caused 10% and 30% increases in peak flow, respectively. The no-tile simulation peaks increased by 10% under the first scenario and 20% under the second scenario. Overall, changes in the no-tile simulation were smaller than changes in the tiled simulation.
This study presents a detailed description of the hydrology of heavily tile-drained agricultural watersheds. Representing tile conceptually in the GHOST tile module resolves a significant challenge in numerical representation of tile at watershed-scale—the absence of data regarding pipe location and depth. Results of the Boone River Watershed model provided a more comprehensive assessment of the role of tile in altering watershed hydrology, especially during growing seasons. The systematic hydrologic characterization of subsurface tile drainage networks on a large scale and over a long period can enhance flood mitigation and water quality efforts.
Details
- Title: Subtitle
- Development of a tile module for GHOST and its application
- Creators
- Maral Razmand
- Contributors
- Larry J Weber (Advisor)Marcela Politano (Advisor)Antonio Arenas Amado (Committee Member)Witold F Krajewski (Committee Member)Allen Bradley (Committee Member)Michelle Scherer (Committee Member)
- Resource Type
- Dissertation
- Degree Awarded
- Doctor of Philosophy (PhD), University of Iowa
- Degree in
- Civil and Environmental Engineering
- Date degree season
- Autumn 2020
- DOI
- 10.17077/etd.005702
- Publisher
- University of Iowa
- Number of pages
- xxiv, 223 pages
- Copyright
- Copyright 2020 Maral Razmand
- Language
- English
- Description illustrations
- color illustrations, color maps
- Description bibliographic
- Includes bibliographical references (pages 213-223).
- Public Abstract (ETD)
Agriculture in the Midwest United States relies on subsurface tile drainage to maintain groundwater levels and improve conditions for crop growth. However, extensive evidence links tile drainage to deteriorated water quality, contaminated drinking water, and to the creation of the Gulf of Mexico hypoxia zone.
This study quantifies the impact of subsurface tile drainage on watershed hydrology using the numerical model “Generic Hydrologic Overland Subsurface Toolkit” (GHOST), a numerical model developed by IIHR-Hydroscience & Engineering, the University of Iowa. We developed a robust tile module for GHOST and then used the newly augmented GHOST to simulate the Boone River Watershed from 2002 to 2018, studying the impact of subsurface tile drainage on watershed hydrology during various climatological conditions. Our analysis shows that, in tile-drained agricultural lands, drainage pipe depth significantly influences tile’s hydrological impact during heavy storms.
Simulation results suggest that tile drainage can increase annual streamflow by 35% and decrease surface runoff by 87%. The results also suggest that tiling can also increase surface runoff during growing season. Most important, tile is shown to be able to lower flow during heavy storms but increase flow during dryer weather.
Lastly, we studied the impact of climate-change-induced intensified rainfalls on the Boone River Watershed through two hypothetical scenarios. Our results suggest that intensified rainfalls have a more significant impact on flow values in tiled watersheds than in no-tile watersheds. This study aims to improve the sustainability of agricultural tiling, enhancing management of water quantity and quality.
- Academic Unit
- Civil and Environmental Engineering
- Record Identifier
- 9984035892802771
Metrics
67 File views/ downloads
187 Record Views