Climate change had widespread and profound impacts on both natural and social ecological systems. Agricultural soil carbon storage, an important component of the terrestrial carbon pool, played a critical role in global climate change, sustainable agricultural development, and global food security. This study conducted a comprehensive literature review to investigate the multi-dimensional effects of climate change on agricultural soil carbon storage and evaluated the advantages, disadvantages, and applicability of various assessment methods for agricultural soil carbon storage. The results showed that the multi-dimensional effects of climate change included both single-factor effects and the combined effects of multiple climatic factors. Single-factor effects included rising temperatures, changes in precipitation patterns, increasing carbon dioxide concentrations, atmospheric nitrogen deposition, and various extreme weather events. These factors individually affected the input, output, and transformation processes of soil carbon in agricultural systems. For example, rising temperatures accelerated the decomposition of soil organic matter, leading to soil carbon loss, while elevated carbon dioxide concentrations enhanced plant photosynthesis, increasing carbon inputs from plant residues to the soil. The combined effects of multiple climatic factors, such as temperature-precipitation, temperature-precipitation-carbon dioxide, and carbon dioxide-nitrogen deposition-ozone, added complexity to understanding the effects of climate change on agricultural soil carbon storage. These interactions could have synergistic or antagonistic effects on soil carbon dynamics, highlighting the need for a comprehensive approach to studying the response of agricultural soil carbon to climate change. Three main categories of methods were identified to assess agricultural soil carbon storage under climate change: statistical analysis models, process-based simulation models, and hybrid models. Statistical analysis models, including linear regression and machine learning models, established relationships between environmental factors and soil carbon storage, revealing the influence of environmental variables on soil carbon. Process-based simulation models could be divided into process-oriented and microbial-oriented models, depending on whether the decomposition of microorganisms was explicitly represented. These models quantitatively described the interactions among climate, vegetation, soil, and other factors and predicted dynamic changes in agricultural soil carbon storage under future climate change scenarios. Hybrid models took advantage of each method to achieve greater simulation accuracy, reliability, and interpretability by combining two or more different types of models. Some common hybrid models included regression kriging model, process-oriented machine learning model, and so on. Finally, four key areas for future research were suggested. First, the soil carbon storage mechanisms driven by microorganisms should be studied in more detail, as they played an important role in soil carbon cycling. Second, the interactive effects of multiple climate change factors should be emphasized to capture the complexity of future climate change scenarios. Third, the development of hybrid models that integrated statistical analysis and process-based simulation was crucial to improve the accuracy and applicability of soil carbon storage assessments. Fourth, uncertainty analysis in model parameters, input data, and scaling methods should be improved to enhance the reliability of model predictions. In conclusion, this study presented a comprehensive climate-crop-soil research framework through systematic integration and analysis, contributing to the scientific understanding of the impact mechanisms of climate change on agricultural soil carbon storage. [ABSTRACT FROM AUTHOR]