Relation of available soil nitrogen to rice yield

Thirty-one surface soil samples from nine years of rate and ratio rice fertilization experiments at different locations in Louisiana were analyzed for their available-nitrogen content, using six chemical and two biological extraction methods. The relationship between the available soil nitrogen and rough rice yields was then investigated.Total soil nitrogen varied widely, ranging from 540 to 5460 parts per million. The average percentage recovery of total soil nitrogen (representing the available soil nitrogen), as determined by the various methods used, also ranged widely, from as high as 16.74 percent, by the hot alkaline permanganate method, to the lowest value of 1.08 percent, by boiling soils in 0.01 M CaCl2 solution. Good correlation coefficients were obtained, especially among the anaerobic incubation and each of the other extraction methods used. The acid hydrolysis method correlated the least with the other extraction methods. Highly significant correlation coefficients were obtained between the rough rice yields from the untreated plots (plots that received no nitrogen) and the available soil nitrogen determined by all the extraction methods. The best correlation was obtained with the anaerobic incubation method (r = 0.622). The rough rice yield in plots receiving 56 or more kilograms of nitrogen per hectare did not exhibit a significant relationship with the available soil nitrogen. Relationships between the rough rice yields at zero and 28 kg/ha N and the available soil nitrogen, as determined by the anaerobic incubation method, were better described by curvilinear models than by linear ones. The relationship was also established between yield increase from application of 112 kg/ha N (the level of nitrogen generally considered near optimum for rice in Louisiana) and available soil nitrogen determined by the anaerobic incubation method. With some reasonable degree of accuracy, yield increase could be predicted from the graph signifying this relationship.