Accurate prediction of the spatial distribution of soil organic matter （SOM） is crucial for improving the productivity of arable land， developing modern precision agriculture， regulating soil carbon balance， and protecting the ecological environment. Previous research has analyzed the horizontal spatial distribution characteristics and influencing factors of SOM in different regions. Still， few studies have examined the spatial distribution of SOM and its drivers from a vertical perspective. Taking the Yihe River Basin， a typical area in the northern soil and rocky mountain region， as the key research area， the random forest model was used to predict the spatial distribution of SOM content in the basin and to reveal the controlling factors affecting the differences in the spatial distribution of SOM. The results showed that： ① The coefficients of determination R2 of the four constructed random forest models for predicting soil depth SOM were 0.761， 0.651， 0.672， and 0.727， respectively， and the fitting accuracies R2 reached greater than 0.650， with a high accuracy in predicting the spatial distribution of SOM in the catchment. ② The spatial distribution of SOM content in the River Basin was characterized by a vertical decrease， with the average values of ω （SOM） in the 0-10， 10-20， 20-30， and 30-40 cm soil layers being 18.62， 12.71， 9.30， and 7.14 g·kg-1， respectively， which belonged to the fourth and fifth classes. The SOM content in different soil layers showed the spatial pattern of high in the east and north and low in the west and southwest. ③ Natural environmental factors， especially soil property factors， were the main controlling factors affecting the spatial variation of SOM in the River Basin， with total nitrogen （TN） having the greatest influence on the spatial distribution of SOM. Human activities have intensified the differentiation of SOM spatial distribution， with a trend of forest land （15.92 g·kg-1）>grassland （15.81 g·kg-1）>orchard （12.42 g·kg-1）>cultivated land （12.21 g·kg-1） for ω （SOM）. The study's results can provide scientific references for the rational use of land resources and the improvement of soil quality in northern China's soil and rocky mountain areas.
Keywords: Yihe River Basin; influencing factors; random forest; soil organic matter （SOM）; spatial distribution.