Changes in rice yield in China under future climate scenarios

Abstract

The Levinsohn-Petrin consistent semi-parametric estimation method was used to empirically analyze input-output panel data on rice yields in 30 Chinese provinces from 1990 to 2019 and to predict the level of rice yields at the end of the 21st century. The results show that, in addition to natural disasters and objective natural conditions, rice yields depend mainly on pesticide application, that China's current rice production and operation model is insufficient to match the rapidly growing level of agricultural mechanization, and that promoting large-scale production in rice fields is an effective way to address the waste of agricultural machinery resources. From the prediction results, the model has stronger prediction ability for the central-eastern and southern production areas of China, and very low prediction ability for the northern and western areas. The prediction of rice yield levels in China shows that rice yields in China at the end of the 21st century increase by 9.81% under the RCP8.5 climate scenario. This paper puts forward targeted suggestions for improving rice yields. The government should strengthen the promotion and training of agricultural technology and increase the channels for farmers to learn rice cultivation techniques; promote large-scale rice production and operation; and improve the education level of rice growers.