Abstract: Agricultural carbon management can share the spatial differentiation of the intra-provincial agricultural carbon compensation rates and their fine-scale identification. This study aims to explore the spatiotemporal distribution of the agricultural carbon compensation rates over the 13 prefecture-level cities, Jiangsu Province, China. A systematic investigation was also made to clarify the key imbalanced regions of the agricultural carbon compensation rates, indicating their regional differences and evolutionary trends. Furthermore, the driving factors were then determined to uncover the dominant influences on the carbon compensation rates and their spatiotemporal heterogeneity. The emission factor method was employed to calculate the agricultural carbon emissions and carbon sequestration in the study area from 2010 to 2023. A unified accounting framework was established to integrate crop production and livestock farming into the agricultural system of the carbon compensation rate. Theil index and kernel density functions were applied to explore the regional differences in the agricultural carbon compensation rates. While the Geodetector was utilized to identify the key driving factors. A geographically and temporally weighted regression (GTWR) model was finally combined to examine their spatiotemporal dynamics. Results indicate that: 1) There were staged variations in the agricultural carbon emissions in Jiangsu Province from 2010 to 2023. Specifically, crop production consistently served as the primary source, while the livestock emissions declined substantially after 2015. Carbon sequestration capacity was improved significantly, where the farmland carbon storage increased at an average annual rate of approximately 13%. Northern Jiangsu, as the primary grain-producing region, was consistently ranked first with both carbon emissions and carbon sequestration, which accounted for over half of the total, respectively. 2) The agricultural carbon compensation rate also demonstrated a consistent upward trend, where the kernel density distribution followed the bimodal functions. Theil index analysis revealed that there were significant disparities in the carbon compensation rates among southern, central, and northern Jiangsu. The inter-regional differences maintained a stable proportion above 70% of the total, indicating an increasing trend. There was the imbalanced development of the regional low-carbon cycle. 3) Geodetector analysis indicated that the crop production structure, the degree of openness, financial support, and rural economic development level were the multi-dimensional driving mechanisms of the agricultural carbon compensation rates. Among them, the crop production structure shared the strongest explanatory power. GTWR results revealed that there was spatiotemporal heterogeneity in the driving influencing factors: The financial support exhibited the spatial transfer and polarity reversal; The degree of openness demonstrated the continuous negative diffusion; The rural economic development shared the complex transitions of spatial polarity; While the crop structure was maintained, the relative stability with the emerging regional differentiation. Furthermore, the regionally coordinated system was constructed under the framework of horizontal benefit compensation in the major grain-producing regions. The agricultural carbon management was optimized with the financial support to implement the differentiated guidance for the openness development and crop structure, in order to enhance agricultural carbon compensation levels in Jiangsu Province. The finding can provide the scientific foundations for precise carbon management.