Temperature is a key factor affecting agricultural production. Exploring the relationship between extreme temperature and agricultural carbon emissions is of great significance for accelerating the achievement of emission peak and carbon neutrality and sustainable development. Based on the panel data of 110 prefecture-level cities in the Yangtze River Economic Belt， the bidirectional fixed effect model was used to empirically test the impact， heterogeneity and functionary channel of extreme temperature on agricultural carbon emissions on the basis of measuring the extreme temperature index and the total amount of agricultural carbon emissions. The results shows that extreme temperature can significantly inhibit the growth of agricultural carbon emissions in the Yangtze River Economic Belt. There is a single threshold for environmental pollution in the process of extreme low temperature affecting agricultural carbon emissions， where the inhibitory effect of extreme cold on agricultural carbon emissions presents a marginal decline when the annual average concentration of inhalable particulate matter is higher than 45 micrograms per cubic meter. The research conclusions remain robust when using methods such as replacing explained variables， explaining variables and sub-sample test for re-estimation. Heterogeneity analysis reveals that the inhibitory effect of extreme temperature on agricultural carbon emissions was more significant in the middle reaches of the Yangtze River Economic Belt， as well as in regions with developed and medium agriculture. Further research finds that agricultural technology progress is the functionary channel through which extreme temperature affects agricultural carbon emissions in the Yangtze River Economic Belt. Therefore， it is necessary to promote agricultural carbon emission reduction through agricultural technological progress， reduce carbon emissions caused by agricultural chemical input， and develop characteristic low-carbon agricultural production according to local conditions.