1991 Volume 7 Issue 2 Pages 11-23
Although the standard growing season of pasture plants in the Konsen Area in the north-eastern part of Hokkaido occurs from April to October, it changes year by year depending on the meteorological conditions. In this study the seasonal changes in 10-day mean air-temperature monitored over a period of 62 years were divided into 7 typical patterns, for which plant growth was calculated based on a system model describing temporal changes in the pasture herbage biomass. Although long term meteorological predictions cannot be easily obtained, it may be possible to pick up one or two, as predicted pattern, out of the 7 patterns of seasonal changes in meteorological conditions, and herbage biomass can be predicted using the patterns. 1. At the Hokkaido Prefectural Konsen Agricultural Experiment Station, annual meteorological data have been accumulated for 62 years since 1928. Using the 10-day mean air temperature data collected during the period from April to October, which corresponds to the growing season of pasture plants, the 62 years were divided into 7 year groups by applying a numerical classification method (k-means). 2. A complex system model describing seasonal changes in pasture plant biomass at Nishinasuno, Central Japan, had already been developed. This Nishinasuno model was revised to fit the data obtained at the Konsen Agricultural Experiment Station. Using this new model, the seasonal changes in plant biomass were calculated for each of the 7 meteorological patterns. 3. By fitting a simple logistic growth curve to the seasonal plant biomass changes analysed above, seasonal changes in the growth coefficient, r, were estimated. Farmers and cooperatives of farmers can use this simple logistic model and the r series obtained in the above calculations instead of the complex system model. The logistic model is expressed by the equation: dx/dt = rx (1-x/K)-a, where x denotes the herbage biomass at time t, K is the carrying capacity and a the herbage consumption by cattle.
