Water Quality Improvement and Methane Emissions from Tropical and Temperate Wetlands

Wetlands are important ecosystems in our landscape because of the broad array of ecosystem services they provide to humans and the environment. Wetlands have unique biotic and abiotic chemical interactions among soil, water, and vegetation that, combined with long retention times that are characteristic of wetlands, allow for nutrients, metals, and organic pollutants to be removed from the water column, resulting in cleaner water. The same characteristics that make wetlands so efficient at improving water quality also provide anaerobic conditions and organic substrate that is optimal for methanogenesis, the microbial production of the greenhouse gas (GHG) methane (CH₄). The objective of this dissertation is to investigate the biogeochemistry, specifically water quality improvement and CH₄ emissions, of natural and created wetlands in tropical and temperate climates.

Five tropical treatment wetlands dominated by floating aquatic plants and constructed to deal with a variety of wastewaters were compared for their effectiveness in treating organic matter and nutrients in the Parismina River Basin in eastern Costa Rica. Wastewaters were from a dairy farm, a dairy processing plant, a banana paper plant, and a landfill. Four of the five wetland systems were effective in reducing nutrient levels of effluents before water was discharged into rivers. Ammonia nitrogen (N) levels in water entering most wetlands were considerably higher than ambient (i.e., riverine) levels; concentrations were reduced by as much as 92% in the wetlands, which retained, at a maximum, more than 166 g NH₄-N m^-2 y^-1. Nitrate N removal occurred in low concentrations in the inflows (less than 1 mg-N L^-1). Phosphate phosphorus (P) was effectively reduced through the wetlands (92 and 45% reductions through dairy farm wetlands, 83% reduction through banana paper wetlands, and 80% reduction through dairy processing wetlands). Retention of phosphate ranged from 0.1 to 10.7 g-P m^-2 year^-1 in the treatment wetlands. Dissolved oxygen in the wetland outflows were ≤2 mg L^-1 in three of the sampled wetlands, most likely a result of the abundant free-floating macrophytes that sheltered the water from diffusion and shaded aquatic productivity. The efficacy of these created wetlands to treat effluents from different sources varied, and modified wetland designs or active management may be necessary to further improve water quality. Recommendations on tropical wetland design and management are presented, as are suggestions for implementing an ecological engineering approach with farmers in Central America.

Wetlands are one of the largest natural sources of the greenhouse gas CH₄ released to the atmosphere. Despite the fact that a large percentage of wetlands occur in tropical latitudes, CH₄ emissions from natural tropical wetlands are not well understood. The objective this research was to compare wetland CH₄ emissions from three natural tropical wetlands located in different climatic and ecological areas of Costa Rica. There were three distinct tropical wetland ecosystems: 1) a humid flow-through wetland slough with high mean annual temperatures (25.9 °C) and precipitation (3700 mm yr^-1); 2) a stagnant rainforest wetland with high mean annual temperatures (24.9 °C) and precipitation (4400 mm yr^-1); and 3) a seasonally wet riverine wetland with very high mean annual temperatures (28.2 °C) and lower mean annual precipitation (1800 mm yr^-1). CH₄ emission rates measured from sequential gas samples using non-steady state plastic chambers during 6 sampling periods over a 29-month period from 2006-2009 were higher than most rates previously reported for tropical wetlands. Means (medians) were 356 (116), 906 (145), and 1004 (371) mg CH₄-C m^-2 d^-1 for the three sites, with highest rates (p = 0.000) occurring at the seasonally flooded wetland site compared to the humid sites. Highest CH₄ emissions occurred when water levels were between 30 and 50 cm. We estimate that Costa Rican wetlands produce about 1.3 Tg yr^-1 of CH₄, or approximately 1 percent of global tropical wetland emissions. Elevated CH₄ emissions at the seasonally wet/warmer wetland site suggests that some humid tropical wetlands of Central America may emit more CH₄ if temperatures increase and precipitation decreases with climate change.

There have been few studies of CH₄ emissions in created and restored wetlands. We measured seasonal and spatial patterns of CH₄ emissions over a two-year period (2006-08) from two 12 to 14-year-old created wetlands in central Ohio, one initially planted and the other allowed to self-colonize, to determine how season, hydrology, and the original wetland creation approach influence those emissions. Median (mean) spring/summer CH₄ emissions for the planted and self-colonized wetlands were 56 (84) and 111 (287) mg CH₄-C m^-2 d^-1 for Wetland 1 and Wetland 2 respectively, while autumn and winter emissions were considerably lower (11 (28) and 24 (66) CH₄-C m^-2 d^-1, respectively). Overall, the two created wetlands were different with respect to CH₄ emissions, with the plant self-colonized wetland emitting higher annual CH₄ emissions (median and mean emissions of 19 and 68 g CH₄-C m^-2 y^-1, respectively) than the planted wetland (median and mean emissions were 6 and 17 g CH₄-C m^-2 y^-1, respectively). Since hydrology and soil/water temperature were identical for the two wetlands, we hypothesize that differences in carbon accumulation due to higher net primary productivity in the self-colonized wetland may be causing higher CH₄ emissions in that wetland. Net primary productivity in the self-colonized wetland was higher 7 out of 11 years prior to the study. Mean CH₄ emissions from the two created wetlands were 21 and 83% of the CH₄ emission of 82 g CH₄-C m^-2 y^-1 measured in a natural wetland in Ohio with similar hydrologic patterns. Annual CH₄ emissions increased at a higher rate in the planted wetland than in the self-colonized wetland over a four-year period with increases of 4 and 16 g CH₄-C m^-2 y^-1 in the planted and self-colonized wetland, respectively. CH₄ emissions from created wetlands in their early decades may depend as much or more on the methods used to create the wetlands (e.g. planting vs. natural colonization) as on the hydrogeomorphic conditions of the wetlands.