A statistical model for finding the influence of weather factors on Thai Golden Honey mango flowering in Phitsanulok Province

Abstract

        A multiple regression model using factor analysis was used to predict the flowering of 'Namdokmai Sithong' mangoes in Phitsanulok province. Data concerned with the average number of flowerings per tree in eleven sample gardens in two sub-districts of Wang Thong district and three sub-districts of Noen Maprang district, and climate data for the same area including average temperature, average air pressure, average relative. humidity, and average rainfall were collected over two periods: February to March 2017 and September to December 2018. The data were applied to determine a multiple regression model based on Principal Component Analysis (PCA). The results showed that y ^{^} =21.623913-0.542305FS ₁ -8.285513FS ₂was the predictive equation or regression equation for the average number of flowerings/tree. From the equation, the average number of flowerings/tree increased when the first weather factor and the second weather factor decreased. This implied that climate was an important factor influencing the average number of flowerings/tree. It also suggested climate factors 1 and 2 could account for the 48.5 percent variation in the mean number of flowerings/plant. Furthermore, the researchers created a website, http://www.mangoforecast.com/ which can be used to predict the average number of flowerings/tree and to collect climate data from interested farmers. The data will be used to improve the forecasting equation for the average number of flowerings/tree and to achieve greater accuracy. In addition, the data will be used to determine the optimal climate factors influencing flowering in present and future Phitsanulok province.