Estimating seasonal fragrant rice production in Thailand using a spatial crop modelling and weather forecasting approach

Abstract

Fragrant rice is an important export commodity of Thailand and obtaining seasonal production estimates well in advance is important for marketing and stock management. Rice4cast is a software platform that has been developed to forecast rice yield several months prior to harvesting; it links a rice model with a Minimum Data Set (MDS) and Weather Research Forecast (WRF) data. The current study aimed to parameterize and evaluate the model and to demonstrate the use of the Rice4cast platform in forecasting seasonal KDML 105 rice yield and production with local data set. The study area encompassed 77 districts in Thailand, covering 0.94 of the total area of KDML 105 in the country. Minimum Data Sets for the 2013–2015 growing seasons were used for model parameterization and evaluation. The annual statistics from the Office of Agricultural Economics (OAE) were used as a reference basis and planted areas from the Geo-Informatics and Space Technology Development Agency (GISTDA) was used for production estimation. Model evaluation showed good to fairly good agreement between the predicted and reported OAE yield. Production forecasts, however, over-estimated the OAE values considerably, primarily because of the use of GISTDA planted areas that were larger than the harvested areas in the production estimates. Adjustment of the planted areas to account for damaged areas need to be explored further. Nevertheless, the results demonstrated the capability of yield predictions with the Rice4cast, making it a valuable tool for in-season estimates for fragrant rice yield and production.

Evaluation of CERES –Rice Model for Simulating Rice Yield and Phenophases

“…As a processbased model, CERES-Rice is having versatile applications in various research areas. These include irrigation responses [3], studies on cropping sequence [4,5], yield-gap analyses [6,7], yield forecasting [8], and climate change impact studies (Darikandeh et al, 2020 andDarikandeh et al, 2024). The model's flexibility makes it a powerful tool for addressing a wide range of agricultural research questions and practical applications.…”

Small scale method for estimation of genetic coefficients of photoperiod-insensitive rice using generalized likelihood uncertainty estimation

“…The newest version of DSSAT (version 4.7.5) can simulate 42 crops (Hoogenboom et al, 2019). Rice researchers use the CSM-CERES-Rice option in DSSAT to study the responses of rice to climate change (Gupta and Mishra, 2019;Nasir et al, 2020), to forecast KDML 105 fragrant rice yield (Kaeomuangmoon et al, 2020), to manage nitrogen (Zhang et al, 2018) and for yield gap analysis (Phakamas, 2015). Therefore, the use of crop simulation models is very important in agriculture where it provides valid information for decision making.…”

Identifying Suitable Genotypes for Different Cassava Production Environments—A Modeling Approach

“…The models in DSSAT are physiologically based and simulate daily crop photosynthesis, respiration, biomass partitioning, growth, and crop development as a function of weather conditions, soil properties, management practices, and cultivar characteristics or genetic coefficients [7,[10][11][12]. Potential applications of the DSSAT crop models for agricultural research have been reported for several approaches, for example, defining the suitable genotypes for peanut based on multienvironment yield trials [13][14][15], forecasting maize yield for the off-season in a subtropical environment [16], determining optimum management strategies for soybean [17], wheat [18], and rice [19], evaluating the impact of climate variability on wheat grain yield [20], examining the El Niño-Southern Oscillation effect on cotton yields at different planting dates and spatial aggregation levels [21], and estimating seasonal fragrant rice production in Thailand [22].…”