Modelling the competition for nitrogen between plants and microflora as a function of soil heterogeneity

Plant-microbe competition for inorganic N is a common phenomenon in soils, and the traditional view is that microorganisms are much stronger competitors than are plant roots. We challenged this view, hypothesizing that the balance between the two competitors is strongly dependent on the spatial heterogeneity of the soil. We constructed a model to explore this hypothesis. The model was structured to simulate a laboratory experiment where N-limited plants were grown in a soil containing two types of microbial "hotspots"; with high potential for either net N-immobilization (straw particles) or net N mineralization (clover particles). In this experiment (Wang and Bakken, 1997), plant roots appeared to compete successfully with microorganisms for inorganic N, and their competitiveness depended strongly on the distance (3, 6, 9 or 12 mm) between the layers of straw and clover material. Most model parameters were taken from the literature, whilst some parameters were estimated by least square model optimization using selected treatments within the experiment. The parameterized model was then statistically evaluated against treatments not used in the parameterization. The model efficiently simulated the observed transient inorganic N accumulation in unplanted soil and its dependency on distance between hotspots (modelling efficiency, EF = 0.611). It also captured very well the plant N uptake and its dependency of the distance between hotspots (EF = 0.860). The modelling exercise underscored the importance of soil heterogeneity in determining the outcome of the competition between plant roots and microorganisms for inorganic N in soil. Spatial segregation of hotspots with net N-immobilization and net N mineralization is likely to be the rule rather than the exception in natural soils. This has a profound impact on the N dynamics of soil plant systems, and it should be taken into account for the interpretation of experiments (such as 15N pool dilution experiments) as well as in general models for the C and N dynamics of soil-plant systems.