Abstract: The aim is to provide a basis for studying spatial differentiation and interannual variability of rice quality. The data from geographical and periodical seeding experiments on different rice varieties were used to normalize the rice quality. The sliding variable-step analysis was used to search the key period for climate factors, such as average temperature, maximum temperature, minimum temperature, sunshine duration and daily temperature range, affecting rice quality elements; and utilizing a stepwise regression analysis to establish a climatic-ecological comprehensive relation model of rice quality factors. The results showed that: (1) the key periods and key factors affecting rice quality were different along with different components of rice quality, some components were generally affected by the highest temperature, lowest temperature, average temperature, sunshine, and the daily temperature range, some components were only affected by a few factors, and the individual component was only affected by the daily temperature range; (2) most of the key periods being influenced were at the full heading-maturing stage, but there were also quite a few components that lasted from 25 days before the heading stage to the filling and fruiting stage, some of the influence patterns were parabolic, and others were straight; (3) the established climatic-ecological comprehensive model of each quality factor could be used to provide technical methods for evaluating the climate impacts among different years and different regions, and provide a scientific basis for higher quality rice production layout; (4) the assessment showed that the spatial distribution of rice protein in the Sichuan Basin increased from northeast to southeast during the past 20 years, the spatial distribution of milled rice rate was high in the west and low in the east.