Abstract: Forest fires, as a key disturbance factor in forest ecosystem succession, have their intensity and fire spread rate deeply regulated by the fuel load on the ground. Particularly in areas where the ground fuel layer is continuously distributed, the level of fuel load directly determines the fire risk level and fire behavior characteristics. Therefore, constructing high-precision fuel load prediction models has become a core link in improving the accuracy of forest fire risk forecasting. Based on domestic and international literature on forest ground fuel load, this paper reviews the research methods of fuel load, the application and verification of ground fuel load prediction models, and the factors affecting the accuracy of fuel load prediction models. At present, the research on the factors affecting the dynamic changes in ground fuel load in China is still not perfect. The paper highlights three major bottlenecks in current research: First, the coupling mechanism of multiple factors is insufficiently analyzed, and some key driving factors are not fully considered in model construction; second, the construction of basic databases is lagging, and the characteristic parameter system of typical fuel types has not yet been perfected; third, the regional scale adaptability has not been verified, and the promotion of models faces challenges of spatial heterogeneity. In response to the above issues, it is necessary in the future to focus on strengthening the construction of dynamic monitoring networks, multi-source data fusion analysis, spatial heterogeneity characterization, and multi-scale model integration research, so as to promote the development of forest fire risk prediction towards refinement and intelligence.