The Use of Constructed Wetlands to Remove Nitrogen and Phosphorus from Pumped Shallow Groundwater

by Cook, Michael J.

Abstract (Summary)

ABSTRACTCook, Michael J., The Use of Constructed Wetlands to Remove Nitrogen and Phosphorus from Pumped Shallow Groundwater. (Under the direction of Robert O. Evans).Non-point pollution has received significant attention nationwide with seepage from lagoons one potential source. The research presented here discusses the hydrology and shallow groundwater quality associated with leakage from an old, unlined lagoon located in the Middle Coastal Plain of North Carolina. Ammonium-N concentrations from wells installed between the lagoon and a nearby stream averaged 121 mg/L, with the highest concentrations exceeding 170 mg/L. Mean annual NH4-N concentrations in the stream ranged from 10 to 25 mg/L indicating that the seepage plume was reaching the stream. A water control structure was installed down stream of the lagoon to reduce the hydraulic gradient towards the stream. In addition, the lagoon was closed out in April 2001. The hydraulic gradient has decreased from 0.023 m/m when the lagoon was in production to 0.0026 m/m since closure. Over a 33-month pre- and post- closure period, NH4-N concentrations in wells 15 m down gradient of the lagoon have decreased from 121 mg/L to 96 mg/L. A series of pumping wells were installed in the seepage plume to remove and route the contaminated groundwater to a 0.35 ha constructed wetland for treatment. Inflow and outflow of the wetland were continuously monitored to determine nutrient loading and reduction rates. Fourteen monthly mass balances were computed to compare the inflow and outflow of the wetland and to assess monthly nutrient reduction for TKN, NH4-N, NO3-N, TP and OP. Overall, greater than 79 % of the nitrogen and 26 % of the phosphorus were assimilated on a mass basis while concentrations decreased by more than 87 % across all nutrient species. Oxidation-reduction, air and water temperature, pH, and dissolved oxygen were measured weekly at several locations within the constructed wetland. Regression analyses were conducted to examine the relationship between these parameters and monthly nutrient reductions. Nutrient export from the wetland was positively correlated to water temperature (i.e., nutrient export increased as water temperature increased). In general, lower redox and DO were correlated to higher nutrient levels within the wetland and subsequently to higher export from the wetland. Caution should be taken on interpretation of these regression analyses as the conditions in the wetland changed over the course of the study. The loading rate was doubled at the beginning of the first full growing season (i.e., the loading rate was two times higher during the growing season than the previous dormant season). Another factor that likely impacted N and P assimilation is many plants in the upper portion of the wetland died in July and August of 2001. The hydrology of the site was evaluated using MODFLOW-GMS. MODFLOW was calibrated using water level data from the site and was used to evaluate the influence of water levels in the adjacent channelized stream on the movement of contaminants in the groundwater plume. Model results indicated a decrease in the hydraulic gradient from the former lagoon from 0.0045 m/m in the free drainage case to 0.0027 m/m in the controlled drainage case. From the gradient calculations, travel time of the seepage plume to the stream increased from 380 days free drainage scenario to 640 days in the controlled drainage case. MODFLOW analysis of the pumping wells indicated that 6.3 gpm was required in the free drainage mode to reverse the gradient from the stream and capture the seepage plume. In the controlled drainage mode, 4.7 gpm was required to reverse the direction of groundwater flow. A hydrologic analysis was also conducted to evaluate pumping requirements to mitigate an actively leaking lagoon. Simulations were performed using an interceptor drain (French drain) adjacent to the lagoon for collection of seepage discharge from the lagoon. Under controlled drainage, the interceptor drain collected 51.5 m3/day (9.4 gpm) while under free drainage 67.4 m3/day (12.4 gpm) would be collected. This analysis assumed a worst-case scenario where all wastewater deposited in the lagoon was lost to seepage. The research presented here provides strategies for clean-up of leaking lagoons or those lagoons targeted for closure. Overall, the wetland assimilated 383 kg of total nitrogen and 60 kg of total phosphorus during the 14 month study period. The wetland surface area was originally based on a pumping rate of 1.5 gpm which was the estimated seepage rate of the lagoon. In March 2001, the pumping rate was increased to 3.5 gpm to match growing season ET rates. Initial results indicated that the wetland assimilated this increased flux of nutrients; but as loading continued at this rate, nutrient concentrations in outflow began to increase, suggesting that this higher rate exceeded the assimilative capacity of the wetland. The MODFLOW analysis indicated that a pumping rate of 4.7 to 6.3 gpm was required to reverse the groundwater gradient from the stream. These pumping rates would require a wetland surface area three to four times larger than the wetland area used. The mass balance indicated that phosphorus mass exports were higher than imports for the final two months. From this, further study is needed to determine processes that are affecting removal of phosphorus. Plant uptake was responsible for only 9 % (35 kg) of the N and 18 % (11 kg) of the P assimilated. Oxidation-reduction reactions were judged to be the dominant nitrogen reduction mechanism. Considering the importance of plants in providing conditions conducive for nutrient assimilation, further investigations are needed to determine the cause of the plant die-off. The influence of temperature, dissolved oxygen and redox on nutrient assimilation should be further documented under conditions of constant loading.

Bibliographical Information:

Advisor:Robert Evans; R. Wayne Skaggs; Garry Grabow; Stephen Broome, Minor Representative

School:North Carolina State University

School Location:USA - North Carolina

Source Type:Master's Thesis

Keywords:biological and agricultural engineering


Date of Publication:01/31/2002

© 2009 All Rights Reserved.