Simulation in pre-testing of rice genotypes in Tamil Nadu

Abstract

A long series of pre-testing trials of promising cultures at many locations is aprerequisite for the selection and release of a new rice variety. This involves considerable time and effort. To reduce time and expense, we sought to employ a crop growth simulation model that explains the interaction of environment and genotype.

Field experiments were conducted at Tamil Nadu Rice Research Institute (TNRRI), South India, during the 1990 wet season to establish the performance of six medium duration genotypes, in detail. Periodical sampling was done to establish the carbohydrate partitioning pattern of these cultivars, and the fraction stern reserves. Crop duration and yields of the 1990 TNRRI experiment were used to calibrate a few model parameters.

The calibrated model was then employed to predict the ranking of seven genotypes in 11 experiments at three locations (TNRRI, Coimbatore and Madurai) and four years (1987, 1988, 1989 and 1990). The ranking of the varieties was compared with that from field trials. In our first attempt, two genotypes were correctly predicted to be among the top three.

The paper discusses possibilities of using a model to compute the performance of pre-release rice genotypes at different locations of Tamil Nadu, using local weather data. We aim at reducing significantly the number of field experiments and the duration of the selection period.