Abstract: Taking the herbs of six typical forest stands in Dalou mountain as the research object, the differences and correlations in herbaceous load under the same and different forest types were analyzed, and stepwise regression analysis was performed with the forest characteristic factors to obtain a herbaceous load prediction model. The results showed that: the maximum herbaceous load in the broad-leaved mixed forest was 0.759 t/hm2, while the minimum was in the Cunninghamia lanceolata forest, which was only 0.058 t/hm2; The herbaceous load of Pinus massoniana forest is mainly related to the average thickness, while the herbaceous load of broad-leaved mixed forests is mainly influenced by the average coverage and stand density; The goodness of fit of the understory model of Pinus massoniana under the same forest type is 0.885, indicating a good fitting effect. The average thickness, canopy closure, and average ground diameter significantly negatively affect the herbaceous load; The goodness of fit of the model under the broad-leaved mixed forest is 0.973, indicating a good fitting effect. The average thickness significantly positively affects the herbaceous load. This study is of great significance for the study of herbaceous load and scientific forest fire management in Dalou mountain.

Key words: loading, herbaceous fuel, prediction model, forest stand characteristics