Parameter Estimation of Improved Rice Varieties in the Philippines Using Gencalc, Glue and Nmcga.

Rice (Oryza sativa L.) is the most vital nourishment for half of the total population, particularly in Asia. Increasing population with growing pressure on food demands gave impetus for proper decision making in agriculture. Optimum resource use and selecting improved rice varieties are major means of increasing yield. Field trials of new varieties through agronomic experiments offers development of input management strategies to enhance input use efficiency. However, the data generated through traditional agronomic research methods is not sufficient to fulfill the need for information to resource management in agriculture. For this, crop simulation models provide a vital platform to design multiple strategies for management in order to enhance rice productivity. To improve accuracy in model predictions, proper estimation of all model parameters is required. Estimation of cultivar coefficients is a prominent and basic step to define growth and development in order to introduce a new cultivar through crop simulation models. Here, parameter estimations for two rice varieties in Philippines, Inbred PSB Rc82 and hybrid Mestizo 20, were done using GENCALC and Generalized Likelihood Uncertainty Estimation (GLUE). The inherent uncertainty in a set of parameters in GENCALC is resolved in GLUE, a Bayesian Monte Carlo technique. GLUE computational estimation takes a lot of time, and hence an alternative yet robust and quick method, Noisy Monte Carlo Genetic Algorithm (NMCGA) was used which deals with first two moments of the parameters and optimizes after fitness evaluation. The objective of this study was to establish a comparative study for identifying suitable cultivar coefficients from a minimum set of experimental data using GENCALC, GLUE, and NMCGA. Moreover, we envisage developing best management practices to optimize resource use efficiency and maximize production under different nitrogen and water management practices.