全球变化知识资源中心

In most crop growth models used for estimating yields, much information, including environmental data and the cultivars planted, must be obtained in advance. Such models are difficult to use when this information is lacking. The aim of the present research was to evaluate the ability of methods to estimate leaf biomass with limited information. We used crop biomass data for maize, rice, and soybean that were gathered continuously at locations throughout Japan between 1951 and 1980 and compared several models that used different methods for estimating the biomass of each plant tissue of a crop species. The results showed that allometry-type models, which predict the biomass of a plant tissue using the relative relationship between the tissue biomass and another crop parameter such as total biomass, were more robust than partitioning-type models, which predict the increase in biomass of each tissue per unit time from the daily partitioning of photosynthate. In cases when it is difficult to acquire detailed crop information, such as for global-scale predictions, the use of an allometry-type model allows more robust estimates to be obtained.