Graduation Year


Document Type




Degree Name

Doctor of Philosophy (Ph.D.)

Degree Granting Department

Geography, Environment and Planning

Major Professor

William J. Mitsch, Ph.D.

Co-Major Professor

Mark Cable Rains, Ph.D.

Committee Member

Thomas L. Crisman, Ph.D.

Committee Member

Jiyoung Lee, Ph.D.


Hydrology, Mass retention, Mesocosms, Suitability analysis, Wetland


Significant expansion of agricultural land use has been widely recognized for leading to global and regional negative environmental impacts, especially increased eutrophication of surface water systems for the last half-century. The landscape-scale environmental problem of overloading nutrients to lakes and streams by excessive fertilizer use and increased human-caused N-fixation is in urgent need of a sustainable landscape-scale solution. Wetlands have long been considered as an effective way to remove nutrients from surface water. However, the influence of regional seasonality and hydrologic conditions on agricultural runoff treatment wetlands still needs more investigation. A new approach, “wetlaculture,” has recently described as a landscape rotating between wetlands and agriculture. However, before practicing wetlaculture at large landscape scales, it is essential first to conduct comprehensive wetlaculture process studies, choose suitable wetland locations, provide appropriate wetland designs, and target appropriate nutrient retention goals. This study uses both mesocosm physical models and a landscape mathematical model to investigate the rates of nutrient retention in two major watersheds in Ohio (Ohio River Basin and Great Lakes Basin).

Two wetlaculture experiments with replicated mesocosms, one at Buckeye Lake, Ohio, which started in October 2017, and the other at Defiance, Ohio, which began in October 2018, are located at former swamp area and has an identical design of construction and hydrologic conditions. Each mesocosm compound experiment consisted of twenty-eight 380 L, 1-m2 RubbermaidTM tubs filled with local hydric soil. The sedge Schoenoplectus tabernaemontani was planted in each mesocosm tub at both sites. Each mesocosm compound was assigned to four hydrologic treatments involving two water depths (no standing water and ~10-cm of standing water) and two hydraulic loading rates (10 and 30 cm week-1). Nearby ditch/river water containing agricultural runoff was pumped weekly into an elevated water feed tank system, to provide weekly hydraulic loading rates to the mesocosms. Inflow and outflow water samples from each wetland mesocosm were collected and analyzed for soluble reactive phosphorus, total phosphorus (TP), nitrate+nitrite (NOx-N), and total Kjeldahl nitrogen (TKN), every other week during hydroperiods. Total nitrogen (TN) was estimated as the sum of TKN and NOx-N.

Chapter two and chapter three present the influence of hydrologic conditions on nutrient retention, and soil and plant development at Buckeye Lake wetlaculture mesocosm site (BLW) and Defiance wetlaculture mesocosm site (DW), respectively. In August 2019, 17 and 12 species, included the introduced S. tabernaemontani, were identified in the wetland mesocosms at BLW and DW, respectively. The wetland mesocosms at both sites soon became nutrient sinks. Over the study period, the average removal efficiencies of TP (71±1.3%) at DW was higher than the average removal efficiencies of TP (38±2.5%) at BLW. However, the average removal efficiencies of TN (41±1.6%) at DW was lower than the average removal efficiencies of TN (44±1%) at BLW. The combination of a high HLR (30 cm week-1) and 10 cm of standing water achieved the best nitrogen removal efficiencies. In comparison, the highest phosphorus removal occurred with the combination of a high HLR (30 cm week-1) and no standing water at BLW. while The combination of a high HLR (30 cm week -1) and 10 cm of standing water achieved the best phosphorus and nitrogen removal efficiencies at DW. For the DW site, net mass retention of phosphorus from the water was estimated to average 1.0 g P m-2 over two years in the wetland mesocosms with a high hydraulic loading rate while the highest estimated net nitrogen mass retention was 22 g N m-2 over two years. For the BLW site, the highest estimated net phosphorus and nitrogen mass retentions were 2.9 g P m-2 and 30 g N m-2 over three years, respectively. These mass retention rates of nitrogen and phosphorus, when extrapolated to a full year, compare well to rates measured by a long-term study carried out in central Ohio from 1994 to 2010.

There were no significant differences in soil phosphorus concentrations at either site before and after these studies. However, at BLW site, concentrations of soil carbon and nitrogen were significantly higher in 2019 than in 2016, increasing by 39% and 19% respectively through the three years of mesocosm experimentation.

Chapter four estimates the location for potential wetlaculture sites in the Western Lake Erie Basin (WLEB) and the Great Black Swamp (GBS) by using a GIS-based suitability analysis model. Wetlaculture places were limited to lands with actual cropland and hydric soils. Multiple criteria of topography, hydrology conditions and soil features for suitability were reclassified into a scoring system, ranged from 0 to 5. Three models with various weight index combinations were tested. Overall, the estimated area of highly suitable potential wetland/wetlaculture areas in the Western Lake Erie Basin and the Great Black Swamp areas are approximately 1000 km2 (3%) and 800 km2 (12%), respectively, much larger than the 400 km2 of wetlands that have been suggested as necessary to control algal blooms in Lake Erie.

This study provides valuable information for scaling up to pilot-scale demonstrations for restoring significant areas of wetlands from farmlands in the former Great Swamp to reduce nutrient loading from agricultural watersheds to lakes and streams in the Ohio River Basin.