Evaluation and prediction of hydrology and nitrate-nitrogen transport in tile-drained watersheds

Thumbnail Image
Ikenberry, Charles
Major Professor
Michelle L. Soupir
Committee Member
Journal Title
Journal ISSN
Volume Title
Research Projects
Organizational Units
Organizational Unit
Agricultural and Biosystems Engineering

Since 1905, the Department of Agricultural Engineering, now the Department of Agricultural and Biosystems Engineering (ABE), has been a leader in providing engineering solutions to agricultural problems in the United States and the world. The department’s original mission was to mechanize agriculture. That mission has evolved to encompass a global view of the entire food production system–the wise management of natural resources in the production, processing, storage, handling, and use of food fiber and other biological products.

In 1905 Agricultural Engineering was recognized as a subdivision of the Department of Agronomy, and in 1907 it was recognized as a unique department. It was renamed the Department of Agricultural and Biosystems Engineering in 1990. The department merged with the Department of Industrial Education and Technology in 2004.

Dates of Existence

Historical Names

  • Department of Agricultural Engineering (1907–1990)

Related Units

Journal Issue
Is Version Of

Implementation of artificial subsurface drainage (tile drainage) for cultivation of row crops in poorly-drained areas of the Upper Midwest of the United States has enabled the region to be one of the most agriculturally productive areas of the world; but has also resulted in loss of wetland ecosystems, altered hydrology, and increased transport of nitrate-nitrogen (NO3-N) to surface water. The direct link between subsurface tile drainage and transport of nonpoint-source pollutants, particularly NO3-N, to surface waters is a primary concern for downstream drinking water supplies and hypoxia in the Gulf of Mexico. The studies described in this dissertation include evaluation of NO3-N export from small, tile-drained watersheds typical of agricultural drainage districts on the Des Moines Lobe ecoregion of Iowa, evaluation of watershed-scale simulation of hydrology and NO3-N transport at the daily interval using the Soil and Water Assessment Tool (SWAT), investigation of important nitrogen pathways and processes simulated in SWAT, and the evaluation and improvement of SWAT algorithms for simulating water quality treatment wetlands in this landscape.

Specific objectives of the first study were to quantify hydrology and NO3-N export patterns from three tile-drained catchments and the downstream river over a 5-yr period, compare results to prior plot-, field-, and watershed-scale studies, and discuss implications for water quality improvement in these landscapes. The tile-drained catchments had an annual average water yield of 247 mm yr-1, a flow-weighted NO3-N concentration of 17.1 mg L-1, and an average NO3-N loss of nearly 40 kg ha-1 yr-1. Overall, water yields were consistent with prior tile drainage studies in Iowa and the upper Midwest, but associated NO3-N concentrations and losses were among the highest reported for plot studies and higher than those found in other small watersheds. More than 97% of the NO3-N export occurs during the highest 50% of flows at both the small catchment and river basin scales. Findings solidified the importance of working at the drainage district scale to achieve NO3-N reductions necessary to meet water quality goals. They also point to the need for implementing strategies that address both hydrology and nitrogen supply in tile-drained landscapes.

The objectives of the second study were to develop and calibrate SWAT models for small, tile-drained watersheds, evaluate model performance for pathway-specific flow and NO3-N simulation at monthly and daily intervals, and document important intermediate processes and N-fluxes. For simulation in the KS and AL watersheds, Nash-Sutcliffe Efficiency (NSE) values were 0.79 and 0.71, respectively, for monthly water yield (WYLD); 0.55 and 0.66 for monthly subsurface flow (SSF); and 0.72 and 0.60 for monthly NO3-N load (using the modified NO3-N lagging algorithms). However, calibration efforts were extensive and detailed monitoring data allowing such efforts are not typically available. Simulation of daily surface runoff (SURQ) and SSF proved more challenging and were generally not satisfactory (NSE < 0.50) with the exception of daily SURQ in the KS watershed, for which NSE was 0.55 and percent bias (PBIAS) was -10.0%. Simulation of daily NO3-N concentration was not satisfactory even after modifying algorithms to lag NO3-N transport via tile flow. For daily NO3-N concentration the KS watershed NSE was 0.20 and AL watershed NSE was -1.12, indicating that simulation in the AL model was less accurate than using the average concentration. Important soil NO3-N processes such as mineralization, denitrification, and plant uptake are often overlooked in watershed modeling studies, but should be evaluated and reported as standard practice. These processes are highly variable and difficult to measure. Better parameterization methods are needed, and related inputs and soil-N fluxes should be constrained within reasonable ranges.

The objectives of the third study were to modify wetland algorithms in SWAT by adapting proven CREP wetland models, compare model performance using both original SWAT algorithms and modified wetland equations, and evaluate the ramifications of watershed and tile drain simulation errors on prediction of NO3-N in Iowa CREP wetlands. The modified equations improved simulation of hydrology and NO3-N in the wetlands, with NSE values of 0.88 to 0.99 for daily load predictions, and PBIAS values generally less than 6%. The applicability of the modified equations to wetlands without detailed monitoring data was improved over the original SWAT equations due to more objectively-informed parameterization, reduced need for hydrologic calibration, and incorporation of an irreducible nutrient concentration and temperature correction factor. The NO3-N removal rate (NSETLR) is the critical input parameter for NO3-N reduction and strongly influences model performance. Isolating the KS wetland from the watershed resulted in an overall NSE of 0.98 and PBIAS of 2.6% for NO3-N load at the wetland outlet. When the wetland was integrated with the watershed simulation using existing soil and tile NO3-N algorithms, the NSE decreased to 0.30 and PBIAS increased to 53.3%, indicating that simulation of the BMP is limited by the ability of the model to reflect short-term fluctuations in flow and NO3-N concentration.

Fri Jan 01 00:00:00 UTC 2016