QUANTITATIVE RELATIONSHIP BETWEEN NET NITROGEN MINERALIZATION AND MOISTURE CONTENT OF SOILS

The 0- to 15- and 15- to 30-cm depths of five cultivated Queensland soils and 32 virgin and cultivated western Canadian soils were incubated at a range of moisture contents for 14 days at 35 °C. In most soils, net nitrogen mineralization was linearly related to moisture content in the available range (−0.03 to −4.0 MPa). Optimum moisture for net nitrogen mineralization corresponded to a soil pore water potential of between −0.01 and −0.03 MPa, while that at which no net nitrogen mineralization occurred was close to −4.0 MPa. Initially, the regression for each soil was normalized by the method of Stanford and co-workers, but this technique proved unsatisfactory. However, by normalizing against available moisture (between −0.03 and −4.0 MPa) we succeeded in grouping the soils. The response of all soils could be described by a model of the form: y = bx + (1 − b)x2 constrained to pass through x0, y0 and x1, y1when both axes were scaled between 0 and 1. In the equation, y is net nitrogen mineralized expressed as a proportion of the maximum rate; x is normalized moisture content; and the subscripts max and 0 refer to soil pore water potentials of −0.03 and −4.0 MPa, respectively. Almost all data conformed to this model. Most soils had a linear response (i.e., b = 1.0), but nine Canadian and one Queensland soil showed a curvilinear response. The expression worked well for the nine Canadian soils and b was estimated statistically to range between 1.22 and 2.18.