Modeling Plant Competition in Monocultures of Barley and Oilseed Rape

Abstract

The inverse linear competition model and three non-linear hyperbolic models were tested on two years' results from field experiments with barley and oilseed rape. Densities ranged from 1/9 to 3 times the normal seed rate (i.e. 350 barley plants m⁻² and 100 oilseed rape plants m⁻²). A log transformation of both sides was necessary to ensure homogeneity of variance. The best precision of parameter estimates was found using the non-linear hyperbolic model: where w is the individual plant weight and N is the plant density. A _{N 0} is the dry matter production per plant at a density N ₀, and B is the potential biomass production per unit area under the actual growth conditions. The estimated values of parameters AA _{N 0} and B agreed well with the values actually measured. The computational, as well as the biological, implications of introducing a threshold density, N ₀ are discussed.