GROWTH ANALYSIS IN BRASSICA USING A MECHANISTIC GROWTH-MODEL

The growth of three different Brassica oilseed species was studied in field experiments during three growing seasons, 1987-89, at Uppsala, Sweden. The results were analysed with a mechanistic growth simulation model developed for temporary grasslands at the Department of Crop Production Science in Uppsala. The model was based on a relative growth rate term, a development term and a weather index, including soil water balance. Parameters for temperature responses and plant available soil water store were assumed to be higher than those for grasslands. Parameters for the initial value of the relative growth rate and the development term were optimized from the data. High correlations were obtained between the observed and calculated dry weight, 96-99% of the variation, being accounted for by the model. In the low yielding B, campestris, growth analysis revealed that variation in yield between years was caused mainly by weather. In the high yielding B. juncea and B. napus, variation in growth potential (initial relative growth rate) was the main cause of yield variation.