Submitted to: Journal of Sustainable Watershed Science & Management
Publication Type: Peer reviewed journal
Publication Acceptance Date: 8/10/2010
Publication Date: 2/5/2011
Citation: Liu, G.D., Li, Y.C., Migliaccio, K.W., Ouyang, Y., Alva, A.K. 2011. Identification of Factors Most Important for Ammonia Emission from Fertilized Soils for Potato Production Using Principle Component Analysis. Journal of Sustainable Watershed Science & Management. 1:21-30. Interpretive Summary: Ammonia emission from agricultural sources, i.e. crops and livestock productions, contribute to about 90% of the total ammonia emissions. a number of factors, including soil properties and nitrogen sources, influence the rate and magnitude of ammonia emission. Ranking the importance of these factors provide basis to develop strategies to minimize the ammonia emissions. In this study principal component analysis (PCA) technique was applied to identify the ranking of factors influencing the ammonia emissions using data from an experiment with four soils, five nitrogen sources, two soil water regimes, three incubation temperatures, and five measurements during the course of 28 days incubation. Soil factors, i.e. particle size distribution and bulk density, contributed to 42% of the variance in ammonia emission. The factors next in ranking were fertilizer source (15%), soil pH (12%), and temperature and soil water regime (9%). Thus, with application of principal component analysis statistics to a large data set (of 1800 observations) confirmed that soil factors are the major contributor to ammonia emissions followed by the nitrogen source.
Technical Abstract: The goal of this study was to identify the primary factors affecting ammonia (NH3) emission from fertilized soils in potato production fields using principal component analysis (PCA). A dataset consisting of 14 different variables and 1800 NH3 emission rates was geometrically classified for analysis. The data used in this analysis originated from an incubation experiment using four major potato production soils (two from Washington and two from Florida), five N sources, two soil water regimes, three incubation temperatures and five measurements during the course of 28 days of incubation. Results indicated that NH3 emission rates could be classified into five distinct emission zones. Soil particle size distribution and bulk density accounted for an average of 47% of the total variance in NH3 emission in the five zones. Fertilizer sources accounted for 15%, soil pH 12%, and temperature and soil water regime 9% of the total variance. Thus, the NH3 emissions could be potentially reduced with optimal management of soils, fertilizer type, and soil water regime in agricultural practices.