Use of simulation modelling to define environmental constraints to increased productivity and rice based cropping systems in Sri-Lanka and the Philippines

This project complements another in the same Program - No 8330, based at the Australian National University. The latter is focusing on socio-economic factors responsible for the difference between potential productivity and farm performance in rice-based systems in the two countries. Because environmental factors may also affect the choice and productivity of cropping patterns, the results of socio-economic studies many apply only to the seasons in which they were collected.

To overcome the difficulty, this project will use base data analysis and a simulation model of existing farming systems and of improved technology options to define above environmental factors. It will then run the model with a long sequence of weather data - say, 20 years - to determine the long-term stability of possible new cropping patterns. The model will also be used to test whether patterns that have succeeded in one area should do so in others.

Both project will use data from trials on farms and research stations to quantify the constraints to productivity imposed by such factors as nutrition, cultivars, water balance, weeds, pests and timing of operations. While the well-qualified national agricultural authorities in Sri Lanka and the Philippines will conduct these experiments, the CSIRO group will take part in their design and planning to ensure that they incorporate two levels of each important factor determining yield - the farmers' level land a (generally higher) level recommended to give maximum profit. Accurate records of the crop environment will be kept, with emphasis on the water balance of farmers' and experimental fields.

Data preparation and modelling will take place in Canberra, using a previously developed computer model of a rained rice-based cropping system that simulates the growth of crops grown in sequence on an area specified by its hydrology and landscape position. The model already has a 'library' of crop descriptions. Since it was written, a group at Los Banos has collected, checked and encoded a large set of rainfall and other weather data for 100 locations in the Philippines, covering an average of 35 years. These data are already in a suitable form for use with the model, and it is hoped that the available Sri Lankan weather records will be similarly compatible.

When run with the weather data, the simulation model will provide cross-sectional and time-series estimates of productivity for selected cropping patterns. In addition to determining the likely stability of a pattern and its long-term risk at any one location, the results could estimate yields of individual crops and therefore likely profits, and the number of crops possible in a season.

Because of their collaborative nature, each project will benefit the other. Access to experimental and farm yields will allow the CSIRO group to improve the accuracy of the model, which in turn should be able to eliminate any confusion due to environmental constraints in the socio-economic results. Together, they can form the basis of the introduction of more intensive cropping systems in both the developing countries.