Organic matter dynamics and N mineralization in grassland soils.

The aims of this study are i) to improve our understanding of the interactions between soil texturelsoil structure, soil organic matter, soil biota and mineralization in grassland soils, ii) to develop a procedure that yields soil organic matter fractions that can be determined directly and can be used in soil organic matter models, iii) to develop a model that predicts the long-term dynamics of soil organic matter, iv) to develop a simple model that can be used by farmers and advisers to predict the non-fertilizer N supply of grassland soils, and v) to quantify the effect of the non-fertilizer N supply of grassland soils on the optimum N fertilizer application rate.In mineral soils there is a positive relationship between the amount of soil organic N and the clay + silt content, and a negative relationship between the percentage of soil N mineralizing and the clay + silt content. For soil C the relationships are less clear, as a result of the presence of charcoal in the sandy soils (inert C). The degree of physical protection of organic matter in soil increases with the clay and silt content of the soil. In sandy soils, organic matter apparently becomes physically protected only by the adsorption to or coating by clay and silt particles, while in fine-textured soils, organic matter is also protected by its location in small pores and aggregates. Each soil has a maximum capacity to preserve organic C and N by association with clay and silt particles. The degree of saturation of the protective capacity of a soil with soil organic matter, and not soil texture per se, affects the decomposition rate of applied residue C. The biomass of bacteria is closely correlated with pores with neck size diameters between 0.2 and 1.2 μm and nematodes with pores with neck size diameters between 30 and 90 μm, suggesting that most of the nematodes are spatially separated from most of the bacteria in the soil. Food web calculations indicate that the observed C and N mineralization rates can not be explained from differences in microfaunal activity, but must be caused by the observed but hitherto unexplained differences in C:N ratios of the microbes between fine- and coarse-textured soils.A simple procedure is developed that separates soil organic matter into size and density fractions, using silica suspensions as heavy liquids. The fractions differ in decomposition rate and can be used in organic matter dynamics models. Grass-derived C incorporated into the soil is transferred from soluble and light macroorganic matter fractions to intermediate and heavy macroorganic matter fractions and accumulates in microaggregates. In all fractions grass-derived C decomposes faster than soil-derived C.Two models are presented. The first model predicts the long-term changes in soil organic matter. It includes the observation that the degree of saturation of the protective capacity determines the degree of physical protection of residue C. The second model is an empirical relationship that can be used by farmers and advisers to estimate the Non Fertilizer Nitrogen Supply (NFNS) of grassland soils. For mineral soils NFNS can be estimated from the difference between the actual soil organic N content and the content under equilibrium conditions; for peat soils NFNS can be estimated from correlation with the average deepest groundwater table. An annual increase in NFNS of 100 kg per ha on mineral soils results in an annual decrease of the optimum N application rate of 80 kg N per ha on mown grassland