2009 BES Annual Meeting Presentation and Poster Abstracts
Simulation on lawn N fertilization using socioeconomic data in household level in Baltimore city and county, Maryland
Co-Authors: Shuxia Wu, Lawrence E. Band, Josh Bazuin, Jim Fraser, Weiqi Zhou, J. Morgan Grove
Abstract: Household lawn nitrogen application rate and amount are calculated based on the 496 interviews in Baltimore City and Baltimore County, Maryland. The results indicated that about half of the sampled households state they fertilize their lawns, and about half of those that fertilize self-apply, with the other half employing commercial lawn companies. The results for the self-application group show a very wide range of lawn N fertilization application rate between 10.4 and 679.4 kg/ha/yr, with a mean of 115.8 kg/ha/yr, while the N fertilization amount is between 0.1 and 23.9 kg/yr, and mean is 5.0 kg/yr. The N fertilization is also calculated on subdivision level due to large error on individual households. With global multiple-linear regression models as well as geographically weighted regression model (GWR), both of the lawn N fertilizer application rate and amount are predicted using household socioeconomic data from the survey and the Maryland Parcel database. Results indicated that GWR method provides better estimates and goodness-of-fit than the global multiple-linear regression method, and the simulation of lawn N fertilization application amount is better than lawn N fertilization application rate. Simulation results are also compared on subdivision level.