Targeting wetland restoration to cost-effectively reduce nitrogen loadings in the Gulf of Mexico

An estimated 50 percent of the nitrogen reaching the Gulf of Mexico is a byproduct of field crop production. Nitrogen entering the Gulf via the Mississippi River generates large algae blooms. At night, the algae absorb oxygen leaving oxygen concentrations levels unsuited to most aquatic life thus creating a ‘dead zone’ about the size of Massachusetts—though seasonal and annual variations in nitrogen inflows vary its size. United States Department of Agriculture (USDA) conservation programs have been credited with controlling, but not reducing, nitrogen inflows. That is, while corn production, the primary source of field nitrogen losses, has increased over the last two decades, nitrogen inflows to the Gulf have remained fairly constant. Much of the credit for controlling nitrogen loadings is given to the on-field nitrogen conservation practices funded through the Environmental Quality Incentive Program (EQIP), USDA’s largest conservation program. Two land retirement programs, the Conservation Reserve Program (CRP) and the Wetland Reserve Program (WRP), also reduce nitrogen inflows, primarily through nitrogen removal. Can land retirement programs be a cost-effective means of reducing nitrogen inflows? The objective of this research was to estimate the cost-effectiveness of reducing nitrogen loadings by restoring wetlands and to compare our results with prior estimates of the cost-effectiveness of encouraging the adoption of on-field nitrogen conservation practices. We used WRP contracts and other data to estimate a wetland cost function for each wetland region within our study area. With the cost functions and reasonable assumptions, we generated county-level estimates of the expected costs of restoring and preserving new wetlands. With adjusted R-squares of greater than 0.72, the functions are likely to have generated reasonable estimates. We generated estimates of the expected quantity of nitrogen a new wetland might remove by running 20-year simulations of a geographic information system (GIS) model, developed by Crompton et al., and data that include a variety of ecological and weather variables. Removal rates were generated for (hypothetical) new wetlands across the 100,671 grid points within our 183 million-acre study area. We then generate nitrogen-removal cost estimates at each GIS point by dividing the annualized wetland cost estimates by the estimated annual quantity of nitrogen a wetland is expected to remove. Results indicate that nitrogen removal costs range from $0.03 to $6.71 per pound. Half of these wetlands could remove nitrogen at less than $0.37 per pound. As one would expect, the most expensive areas to restore and preserve wetlands are in the areas with some of the highest corn yields hence highest-valued lands. But wetlands’ nitrogen removal rates (in pounds per wetland acre) in parts of these areas are also very high. As a result, some of the wetlands with relatively low nitrogen removal costs (in dollars per pound) lie in parts of Iowa, Illinois, Indiana, and Ohio. Wetlands cannot be restored at all GIS points. There are no GIS data that identify exactly where wetlands have been converted. Applying some conservative assumptions, we calculated the probability that a wetland might be restored within a GIS grid point and the associated quantity of nitrogen the wetlands might remove. Our calculation of the probability that a wetland might be restored is based on the likelihood that a converted wetland lies in the area and the probability that the converted wetland might be restorable. The National Wetland Inventory data indicate that there are currently approximately 10 million wetland acres in our study area. Based on data from the USDA Economic Research Service, we calculated that there were about 34 million wetland acres in our study area in the 1780s. Together, these suggest that there are about 24 million prior-converted wetland acres in our study area—about one out of every 7.5 acres was a wetland. Because of land development, mixed land ownerships, and similar factors, it is not likely that all converted wetlands can readily be restored. Additionally, it is not likely that all landowners are willing to participate in a Federal wetland conservation program (unless program payments are much greater than the market value of the land-use change). To incorporate both of these constraints, we assume that only 5, 10, and 20 percent (or 1.2, 2.4, and 4.8 million) of the prior-existing wetland acres might be restored. These represent 0.65, 1.3, and 2.3 percent of the total acres in the study area. Each GIS point represents 247 acres so that the expected number of prior existing wetland acres in each GIS unit area is 1.5 (e.g., 1.5=247*0.0065), 3.0, and 6.0. The assumptions and estimates laid out above allow us to calculate the probable number of wetland acres that can be restored at each level of nitrogen removal cost. Additionally, we can calculate the total quantity of nitrogen that would be removed below each level of cost. Given that 5 percent of the prior existing wetlands might be restored (1.2 million acres), more than 25 percent (or 0.3 million acres) have nitrogen removal costs below $0.17 per pound and 50 percent (or 0.6 million) have removal costs of less than $0.38 per pound. Restoring wetlands with nitrogen removal costs of less than $0.17 ($0.38) would reduce agricultural nitrogen inflows to the Gulf by 4,530 (8,900) tons per year--about 3 (6) percent of the average quantity of agricultural nitrogen entering the Gulf. If we assume that 10 (20) percent of the prior-existing wetland acreage might be restored, then, applying the same calculations, agricultural nitrogen inflow to the Gulf can be reduced by 9,060 (18,120) tons at less than $0.17 per pound and 17,000 (34,000) tons at less than $0.34 per pound. In earlier work, Petrolia and Gowda (2006) compared the cost-effectiveness of land-use changes and on-field conservation practices to reduce agricultural nitrogen losses. Their result indicates that, within their study area, a 20-percent reduction in nitrogen losses through adoption of on-field practices, the plugging of tile drainage systems, and moving cropland to hay and pasture, the average per-pound cost of nitrogen abatement cost is about $0.78. It appears that wetland restoration can be a cost-effect what of reducing agricultural nitrogen losses. One should be careful about drawing hard conclusions. For example, the Petrolia and Gowda (2006) analysis focused on a small region. Had we focused on the same region, we may have found wetland restoration to be less cost-effect