A critical review of some approaches to modelling nitrogen mineralization

摘要

A number of approaches have been used to model and thus describe N mineralization kinetics in soils. In this paper, we review and present a comparative evaluation of some approaches to modelling adopted for this purpose. These may broadly be categorized as: (1) simple functional approaches to predict net N mineralization and, (2) mechanistic approaches that include a description of microbial biomass processes to predict long-term C and N turnover in soils. The following conclusions emerged. First, the simple functional models try to quantify one or more active fractions of organic matter with associated rate constants to predict net N mineralization. A minimum of two organic N fractions are considered essential to significantly contribute towards N mineralization. Second, the mechanistic models have the advantage that model algorithms represent the basic mechanisms believed to influence mineralization. The input parameters of the models could be obtained independently of the test data. However, it is difficult to validate the models as some of the presumed functional pools cannot be quantified by physical, chemical and biological techniques. Consequently, the models have to be calibrated by adjusting the rate coefficients and pool sizes (site-specific) to fit the measured data. Generally, the performance of a model depends on site-specific calibration. In future, a major challenge will be to free simulations from site-specific calibration and to devise experimental methods for providing initial values to run a model. The amount of N mineralized during laboratory incubation experiments and the parameter estimates for different models depend on methods used to pretreat the soil prior to incubation and their exposure to experimental conditions (such as temperature, soil moisture content) during the incubation. To obtain comparable values amongst studies it is, therefore, necessary to standardize methods of sample pretreatment as well as incubation conditions.