Soil Organic Matter and Soil Fertility

Decades of soil organic matter research have indicated the need for maintaining soil organic matter (SOM) concentrations, especially in highly weathered tropical soils. The nutrient source and sink functions of SOM have been studied extensively with an emphasis on N dynamics for the former and on charge characteristics for the latter. Nevertheless, insufficient knowledge seems to exist on exactly how much organic matter is minimally needed to preserve soil fertility, how this situation can be reached, and what quality SOM should possess. This chapter will review the state of knowledge related to these issues, targeting new approaches or concepts generated during the 1990s. Knowledge on the dynamics of SOM in relation to factors such as clay content and mineralogy and soil pH, factors known to vary widely within tropical soils, will also be reviewed. At first, several issues—related to the “source” function of SOM—will be discussed, including size separation approaches to better predict nutrient release and/or availability in situations where crop demand for nutrients is largely fulfilled by organic sources. In view of the recent interest in replenishing soil P status in sub-Saharan Africa, the relation between organic matter additions and P-fertility will be dealt with in some detail. Secondly, some state-of-the-art aspects of the SOM “reservoir” function will be discussed. Residue inputs of different quantity and quality are decisive for microbial immobilization of N and P but also determine charge characteristics of SOM fractions, providing a tool to modify soil cation exchange capacity (CEC) and/or anion exchange capacity (AEC). Finally, while the above functions of SOM are widely accepted and considered essential for a good soil quality, information on how to achieve, or maintain a suitable SOM content with all desirable features of nutrient release and buffering is much more scanty. Therefore, some modeling exercises, built on concepts generated in temperate areas have been presented. Together with a relatively new set of methods (i.e., 13C-isotope studies) models with predictive power in a tropical environment or more specifically adapted to weathered soils are constructed, addressing organic matter inputs in realistic cropping systems.