Identifying the Rice Yield Determinants Among Comprehensive Factors

Paddy production in Japan is currently undergoing a transition, moving away from the former acreage reduction policies of the 1970s to improve the sector’s efficiency and competitiveness. Meanwhile, agricultural production corporationsAgricultural production corporation and the adoption of ICTICT (information and communication technology) and GAPGAP (good agricultural practice) have been steadily increasing over last decades. This chapter aimed to identify the determinants of paddy yield measured by smart combine harvesterSmart combine harvester within large-scale farms. The sample included 351 paddy fields from a farm corporation scaled over 113 hectares, located in the Kanto region of Japan. The candidate determinants included the continuous variables of field area and condition evaluation scores, transplanting or sowing time, and amount of nitrogen, as well as the stage-specific growth indicators for chlorophyll contain, number of panicles, plant height, and leaf plate valueLPV (leaf plate value). Meanwhile, three discrete variables including variety, cultivation regimeCultivation regime, and soil type were also adopted. Empirical analysis was conducted using a multivariate linear regression, with logarithmic transformations of the continuous variables. Within the continuous variables, transplanting or sowing time was identified as possessing the largest absolute standardized regression coefficient, and thus be the most important determinant. The negative coefficient indicated that earlier transplanting or sowing benefits vegetative growth, thus panicle number and plant height in heading stage, which were identified as positively significant together with field area, and amount of nitrogen. Within the discrete determinants, Akidawara was measured as a productive variety, while the well-drained and submerged direct sowing were identified as negatively affecting yield.