Abstract: Agricultural activities are one of the important sources of greenhouse gas emissions and have a significant impact on global climate change. However, the spatiotemporal patterns of agricultural carbon emissions at the provincial level are not yet fully understood, especially in regions with diverse agricultural production modes, such as Yunnan Province in China. This article aims to systematically study the spatiotemporal characteristics, driving factors, and spatial spillover effects of agricultural carbon emissions in Yunnan Province from 2013 to 2022, providing scientific basis for formulating low-carbon agricultural policies and achieving sustainable agricultural development. This study used methods such as grey relational analysis, geographic detector, spatial autocorrelation test, standard deviation ellipse analysis, and spatial econometric model, combined with data from Yunnan Statistical Yearbook and China Rural Statistical Yearbook from 2013 to 2022, to quantitatively analyze the total amount, intensity, and spatiotemporal distribution of agricultural carbon emissions in Yunnan Province. The research results indicate that: (1) The total carbon emissions from agriculture show a trend of "first increasing and then decreasing", increasing from 12.4134 million tons in 2013 to 13.1095 million tons in 2016, and then decreasing year by year to 10.502 million tons in 2022, with a cumulative decrease of 19.9%. The carbon emission intensity continues to decline, from 0.406 tons per 10000 yuan in 2013 to 0.159 tons per 10000 yuan in 2022, a decrease of 60.8%, with an average annual decrease of 6.8%. (2) In terms of spatial distribution, agricultural carbon emissions show a pattern of "high in the east and low in the northwest", with Qujing, Kunming, and Honghe being the main high emission areas, while Nujiang and other northwestern regions have lower carbon emissions. The carbon emission intensity is characterized by "low in the middle and high in the surrounding areas", with Lijiang City and Honghe having higher carbon emission intensity. (3) Nitrogen fertilizer, agricultural film, and compound fertilizer are the main sources of agricultural carbon emissions, and agricultural output value (q=0.88), rural population (q=0.72), and GDP (q=0.69) are key factors driving spatial heterogeneity of carbon emissions. (4) The spatial spillover effect is significant. For every 1% increase in agricultural output value and rural population, the carbon emissions of adjacent areas increase by 0.009% and 0.013%, respectively; The growth of local GDP has a suppressive effect on carbon emissions in neighboring regions, with every 1% increase leading to a 0.001% reduction in carbon emissions in neighboring areas. This study uses spatial econometric models and geographic detectors to reveal the spatiotemporal differentiation patterns and cross regional linkage mechanisms of agricultural carbon emissions in Yunnan Province, and incorporates spatial spillover effects into the provincial scale carbon emission research framework. The research results provide a scientific basis for Yunnan Province to formulate differentiated low-carbon agricultural policies, especially for high emission areas such as Qujing, Kunming, and Honghe. It is recommended to promote precision fertilization, reduce the use of agricultural films, and adopt clean energy technologies; For low emission areas such as Nujiang, the low emission mode should continue to be maintained to reduce dependence on fertilizers and pesticides.