In situ annual nitrogen mineralization predicted by simple soil properties and short-period field incubation

A typical method to determine nitrogen mineralization is year-round field incubation of undisturbed soil cores. As this technique is very laborious, the aim of the present study was to establish less time consuming methods to assess the annual N mineralization. Options considered were reducing the period of subsequent field incubations to less than a full year, and prediction using simple (direct) soil measurements. In situ annual N mineralization was measured in 74 semi-natural, unfertilized sites under year-round vegetation cover by sequential incubation of undisturbed soil cores during 2 full years. The sites classify into 14 very different plant community types and encompass a wide range of edaphic conditions. A high correlation with the annual N mineralization was maintained when sequential incubations were restricted to the period early March-mid August (r=0.98). Shorter periods may also be used, as long as they lie within the period early March-mid August and cover 3 months at least. The average annual N mineralization could be predicted accurately using the soil mineral nitrogen pool (N-min), moisture content and pH as the predictor variables. In a multiple regression model based on log-transformed values, the R-2 was 0.835. Samples (preferably bulked) are best taken in early March, just before the start of the growing season. Higher N-min contents (moisture and pH assumed constant) indicated higher annual N mineralization rates. Higher moisture contents (at constant N-min and pH) generally indicated lower annual rates. Soil pH showed a quadratic relation (at fixed N-min and moisture) with maximum annual mineralization occurring at a pH-CaCl2 value of 5.5-5.6. N-min appeared to be the most important predictor variable. Even within plant community types (with a reduced range of conditions), its relationship with the annual N mineralization persisted. Frequently studied factors, such as total soil N, organic matter content, C:N ratio, soluble organic N and particle size composition, only showed low and insignificant correlations, although stronger relationships have sometimes been reported. It is assumed this discrepancy is caused by the smaller range of conditions usually studied.