Evaluation of some indices of potentially mineralizable nitrogen in soil

A series of soil N mineralization indices were evaluated using 153 samples chosen from arable fields representing a wide range of soil types, management practices, and climatic zones. These indices were compared against potentially mineralizable N (N-0) determined by aerobic incubation at 25 degrees C for 24 wk. Three different pools of mineralizable N were recognized: Pool I, the mineralization flush on rewetting in the first 2 wk; Pool II, gross N mineralization in the next 22 wk; and Pool III, the potentially mineralizable N, predicted from the fitted curve, that did not mineralize during the incubation period. Pool I was highly correlated with CaCl2-N, KCl-NH4, and KCl-NO3, which extract soil mineral N. Pool III was significantly correlated with ultraviolet absorbance of NaHCO3 extract at 205 and 260 nm (NaHCO3-205 and -260), Illinois soil N test, NaOH direct-distillation N, and hot KCl-NH4, which mostly extract hydrolyzable organic N. All indices except the mineral N based methods, phosphate-borate buffer method, and microbial biomass C were significantly related to N-0, which includes both Pools Il and III. The NaHCO3-260, NaOH direct-distillation N, and Illinois soil N test had the highest correlations with N-0 (r(2) = 0.74, 0.61, and 0. 51, respectively). Total organic C and N represent long-term changes in N-0 and were almost as effective in predicting N-0 as the other indices (r(2) = 0.60 and 0.67, respectively); however, they would be expected to be less sensitive to short-term changes in N-0 due to changes in soil management practices and history.