Date of Award


Document Type


Degree Name

Master of Science (MS)


Plant and Soil Science

First Advisor

Drizo, Aleksandra


Treating elevated nutrients, suspended solids, oxygen demanding materials, heavy metals and chemical fertilizers and pesticides in agricultural wastewaters is necessary to protect surface and ground waters. Constructed wetlands (CWs) are an increasingly important technology to remediate wastewaters and reduce negative impacts on water quality in agricultural settings. Treatment of high strength effluents typical of agricultural operations results in the production of methane (CH4), a potent greenhouse trace gas. The objective of this study was to evaluate CH4 emissions from two subsurface flow (SSF) CWs (223 m2 each) treating dairy wastewater. The CWs were implemented at the University of Vermont Paul Miller Dairy Farm in 2003 as an alternative nutrient management approach for treating mixed dairy farm effluent (barnyard runoff and milk parlor waste) in a cold, northern climate. In 2006, static collars were installed throughout the inlet, mid and outlet zones of two CWs (aerated (CW1) and a non-aerated (CW2)) connected in-series, and gas samples were collected via non-steady state chambers (19.75 L) over a nine-month period (Feb-Oct 2007). Methane flux densities were variable throughout the nine-month study period, ranging from 0.026 to 339 and 0.008 to 165 mg m-2 h-1 in CW1 and CW2, respectively. The average daily CH4 flux of CW1 and CW2 were 1475 and 552 mg m-2 d-1, respectively. Average CH4 flux of CW1 was nearly threefold greater than that of CW2 (p = .0387) across all three seasons. The in-series design may have confounded differences in CH4 flux between CWs by limiting differences in dissolved oxygen and by accentuating differences in carbon loading. Methane flux densities revealed strong spatial and seasonal variation within CWs. Emissions generally decreased from inlet to outlet in both CWs. Average CW1 CH4 flux of the inlet zone was nearly threefold greater than mid zone and over tenfold greater than flux at the outlet, while fluxes for CW2 zones were not statistically different. Methane flux of CW1 was nearly fifteen fold greater than CW2 during the fall, representing the only season during which flux was statistically different (p = .0082) between CWs. Fluxes differed significantly between seasons for both CW1 (p = .0034) and CW2 (p = .0002). CH4 emissions were greatest during the spring season in both CWs, attributed to a consistently high water table observed during this season. Vegetation was excluded from chambers during GHG monitoring, and considering that the presence of vascular plants is an important factor influencing CH4 flux, the potential CH4 emissions reported in our study could be greatly underestimated. However, our reported average CH4 fluxes are comparable to published data from SSF dairy treatment CWs. We estimate average and maximum daily emissions from the entire CW system (892 m2) at approximately 1.11 and 6.33 kg CH4 d-1, respectively, yielding an annual average and maximum flux of 8.51 and 48.5 MtCO2-e y-1, respectively.