Relationships among main soil properties and three N availability indices

Abstract

A biological (3 and 6 weeks aerobic incubation) and a chemical method [successive extractions with cold 0.1 (H1-N) and 0.5M HCl (H2-N)] were applied to 21 soils to determine: a) the potentially mineralizable-N; b) the most useful soil variables for predicting soil-N availability; and c) their usefulness for predicting N uptake by a greenhouse wheat crop. At t = 3, both net N mineralized (NNM) and net N mineralization rate (NNMR) were correlated: a) positively with SOM- and CEC-related variables; and b) positively with soil δ15N and negatively with soil pH, suggesting that N-mineralization, dominated by nitrification, is associated with NO3 −-N losses and soil acidification. At t = 6, all previously discussed variables were important for NNM, but not for NNMR, mainly controlled by the available-P content. The importance of H1-N increased with N2-inputs and decreased with NO3 − losses and soil-N. Relationships of H1-N and H2-N with soil CEC and texture showed the strong relations among nutrients content, biological activity and N mineralization, and the recalcitrance of clay-bounded SOM. Soil-N correlations with wheat-N in absolute amount (positive) and as percentage of soil-N (negative) showed an important available-N supply by N-rich soils, despite their slow N turnover. All best regression models for wheat-N included 1-2 main available nutrients. The percentage of soil-N exported to plant biomass was negatively correlated with non-crystalline Al compounds and soil δ15N. Mineralized-N and wheat-N pools did not share many correlations with soil properties and seemed to come from different sources; consequently, the mineralized-N, which only explained a quarter of wheat-N variance, was not more useful than soil-N for predicting it. Wheat-N and hydrolysable-N only shared a positive correlation with soil-N, highlighting that they are two mainly unrelated N pools. Nevertheless, half of wheat-N variation was explained by its negative relationship with the percentage of soil-N as (H1+H2)-N; a possible explanation is that (H1+H2)-N is biologically labile and it constitutes an important soil-N fraction only when a limiting factor for microbial N mineralization or plant growth allowed its cumulation.