2025 Volume 25 Pages 47-56
Direct seeding of rice is increasingly used to grow the most suitable cultivars under the best cropping schedule in the face of changing climatic and weather conditions. Although crop modeling can be used to optimize these factors, models developed for transplanted rice are not directly applicable, owing to differences in crop development due to weather conditions. Here, we present a method for estimating a parameter set for a crop development model applicable to direct-seeded rice from a set developed for transplanted rice. We hypothesized that model parameters C and D, which are related to the initial development index, can describe the difference in crop development between direct seeding and transplanting. The method estimates the values of C and D for each practice independently using shared parameters A, Th, and G, which are related to the temperature response of development and the days required for panicle initiation and heading in transplanted rice. We used data from field trials conducted in Iwate Prefecture between 2015 and 2019 with ‘Ginga no Shizuku’ to establish and verify the method. The estimated parameter set was accurate within a root mean square error (RMSE) range of 1.08 to 3.85 days at the heading stage and 2.96 to 3.11 days at the panicle initiation stage. When we tested the validity of the method by growing ‘Ten no Tsubu’ in both practices, the estimated parameter accuracy was 2.06 to 3.37 days RMSE at heading. We conclude that our method supplies a suitable parameter set for crop development models of direct-sown rice with an acceptable degree of accuracy. It will contribute to the estimation of optimal combinations of cultivars and cropping schedules for direct-sown rice under climate change, using historical field trial data and estimated parameter sets for transplanted rice.
