Assessment of fixed nitrogen emission from land cover/use using remote sensing imagery and biogeochemical modeling

This thesis is an interdisciplinary study that combines remote sensing imagery, biogeochemical modeling, and ecology to address the contribution of fixed nitrogen to the atmosphere from corn agriculture. Agricultural land covers and associated farming practices provide a pathway for fixed nitrogen to enter the atmosphere where its effects are transported globally. It is estimated that agriculture is the dominant source of nitrous oxide to the atmosphere, contributing over 70% of the estimated anthropogenic release annually.; To characterize the emission of nitrogen as well as additional emission over what would have occurred from a natural cover, a combination of in situ and remotely sensed parameters were used to drive biogeochemical models. The DeNitrification and DeComposition (DNDC) model was used to produce estimates of NO, N₂O, and NH₃ emission from a corn field. A unique feature of this model is its ability to accept an external source of Leaf Area Index (LAI) and biomass. It was found that emission estimates in DNDC were very sensitive during denitrification to changes in these two variables. This is attributed to soil moisture levels at the threshold between nitrification and denitrification after precipitation events.; Remote sensing has the ability to provide actual measures of vegetation both spatially and temporally; as opposed to model parameterizations which represent more general conditions. Advanced Very High Resolution Radiometer (AVHRR) imagery was obtained daily over the course of the 1997, 1998, 1999, and 2000 growing seasons and processed into 10-day maximum value NDVI composites. For further noise reduction in the NDVI time series and to interpolate the NDVI time series based upon the satellite data, local regression was found to be a superior technique with optimal selection of input parameters. To convert from NDVI to LAI and biomass, models were formulated using in situ measurements taken throughout four growing seasons. High correlations of 0.97 and 0.96 were obtained for the NDVI-LAI and NDVI-biomass models, respectively.; Since DNDC does not model natural cover types, the DAYCENT model was used to estimate emissions from tall grass prairie, the natural cover type of the study site. It was found that N₂O emission from agriculture released approximately 3–4 times the amount of N₂O as tall grass prairie due to the more intensive land use associated with agriculture. This additional emission, when extrapolated for corn cover in the United States, comprises between 5% and 8% of the annual anthropogenic emission in the United States. With world populations growing, the conversion of natural ecosystems to agricultural land will become an even more significant source of fixed nitrogen to the atmosphere.