Modelling C and N turnover through the microbial biomass in soil

Many mathematical descriptions of C and N transformations in soils have been developed in the last decade, but only a few explicitly model the activity and mass of soil organisms. Great difficulties still exist in establishing basic parameters governing the kinetics of microbial turnover. The present state of the art is discussed briefly. The model of Van Veen and Frissel on C transformations and related mineralization and immobilization of N has been developed further based on laboratory and field data obtained with different Australian soils. Firstly, case studies show the large effects of the frequency of drying and rewetting of soils on the decomposition of organic matter and on the turnover of biomass. Secondly, the more refined model embraces the concept that soils have characteristic capacities to preserve both organic matter and microorganisms. Preservation of microorganisms could result from protection against predation and/or from amelioration of harsh environmental conditions. Biomass formed in excess of a soil's preservation capacity is assumed to die at a relatively high rate. Furthermore, biomass and its immediate organic products of decay are considered to form mainly a closed system from which only small proportions of the products leak out as stabilized materials. These concepts have been tested with data from laboratory experiments in which ¹⁴C-and ¹⁵N-labelled substrates and bacteria were added to a clay and sandy soil. Net mineralization of C and N (labelled and unlabelled) and changes in the total and labelled biomass as determined by the chloroform fumigation technique allowed for a thorough testing of these concepts in the manner in which they were included in the model. The fits between experimental observations and model outputs were very close. The model indicated that the contrasting metabolism of both C and N in a clay versus a sandy soil could largely be explained by differences in the capacities of the two soils to preserve microorganisms. The ability of a simulation model to describe accurately not only short-term events, e.g. N cycling during one growing season, but also the same processes over, say a decade, is an important criterion in assessing its predictive power. In this paper some of the results will be discussed of testing the model, developed from a consideration of the aforementioned laboratory studies, for its accuracy in describing the decomposition of plant residues in an 8-year field experiment.