ISSN:
1573-5036
Source:
Springer Online Journal Archives 1860-2000
Topics:
Agriculture, Forestry, Horticulture, Fishery, Domestic Science, Nutrition
Notes:
Summary A mathematical model is described which can predict the abundance of microorganisms in the rhizosphere (as μg microbial dry weight/cm3 soil) in relation to distance from the root surface and time since the root started exuding substrate. The growth rate of the microorganisms at each point in the soil is assumed to be controlled by the concentration of soluble organic substrate. The concentration of substrate changes due to (1) its production by the root and diffusion through the soil, (2) its production in the soil by breakdown of insoluble organic matter, and (3) its use by the microorganisms. Values for all of the required input parameters have been obtained from the literature. The model predicts that a high population density will develop near the root surface, but the density will fall off steeply with increasing distance from the root. At the root surface microbial growth continues for many days, provided exudation by the root continues at a steady rate, but further away the population reaches a peak and then declines. This is because the amount of substrate reaching the outer soil is no longer adequate to support the maintenance requirement of the population. Starting with a microbial concentration of 2 μg/cm3, and using what are considered to be average values for other input parameters, the microbial concentrations predicted after 10 days are 1509 μg/cm3 at the root surface, and 2.2 μg/cm3 at 1.8 mm from the root. The model also predicts the substrate concentrations in the soil: these reach a maximum within the first day and then decline, reaching by 10 days values not very different from those in root-free soil. The model is used to predict the effect on microbial and substrate concentrations of changes in soil water content, root density, root exudation rate, initial microbial concentration and microbial response to substrate concentration. Where the predictions of the model can be tested against observed data there is good agreement. re]19760308
Type of Medium:
Electronic Resource
URL:
http://dx.doi.org/10.1007/BF00015157
Permalink
