Denitrification, the microbially mediated reduction of nitrate to di-nitrogen, is the most important process for returning excess fixed N to the atmosphere. Because of the concern over excess fixed N in the environment, denitrification has been proposed as one of the primary means of decreasing the transport of excess N, especially nitrate. In order to accomplish this, it has been proposed that large amounts of forested and other wetlands be re-established in major drainage basins, most notably the Mississippi River drainage. Although tested in some major ongoing experiments, the amount of denitrification to be expected in restored wetlands has not been modeled using a transport model that includes important factors controlling denitrification. We will present results using the Riparian Ecosystem Management Model (REMM) to determine the effects of soil carbon, water levels, denitrification rate constants, N loading, and forms of N loading on denitrification in modeled hardwood forest wetlands. REMM will be used to model two typical coastal plain wetland soils, one with high organic matter and frequent inundation (Kinston series) and one with lower organic matter and high water tables but little or no inundation (Alapaha series). Vegetation on the two soils will be fall deciduous hardwoods. Loadings of N will be varied within a range of 5 kg N/ha/yr to 100 kg N/ha/yr and will be introduced as either mostly nitrate, mostly ammonium, or a 50:50 mix of nitrate and ammonium. Denitrification rate constants for the simulations will be based on measured rates of denitrification potential. Results will be discussed relative to both measured rates of denitrification in these coastal plain soils and to anticipated rates of denitrification for major watershed water quality improvement projects.