Abstract: Bamboo forest is a forest resource with high economic value, high efficiency and accuracy of monitoring of its spatial distribution provides data support for bamboo resource utilization and ecological protection. Compared to the high cost of traditional manual surveys, remote sensing image extraction of bamboo forests has the advantages of high efficiency and low cost. However, the spectral similarity possessed by bamboo forests and other vegetation poses challenges to the accurate extraction on remote sensing images. Taking Jingning County, Zhejiang Province as the research area, this paper utilized Random Forest (RF) algorithm to extract bamboo forest distribution on the basis of Landsat 9 images, Digital Elevation Model (DEM) and forest resource inventory data. The results showed that the total area of bamboo forest extraction was 13 340 hm², with an overall accuracy of 92.17% and a Kappa coefficient of 0.84. Overlay analysis of the extracted results with precipitation, temperature, and terrain data found that bamboo forests in Jingning County were mainly distributed in areas with an altitude of 500～1 200 meters, an average annual precipitation of 1 560 ~ 2 000 millimeters, and an average annual surface temperature of 19～25 °C. The bamboo forest distribution extraction in this method is of accuracy and security, which can provide reference for monitoring bamboo forests in mountainous areas.