Dwarf bamboos are major understory plants in subalpine and temperate forests of Japan. Several microphysical factors are known to influence their coarse-scale distribution, including slope angle, convexity, and proportion of gravel on the soil surface (hereafter gravel cover), but the effects of these factors on local variations in distribution are unclear. To elucidate these effects we examined relationships between local variations in Sasa palmata distribution and the 3 mentioned microphysical environmental factors within a 20- × 20-m plot divided into 0.5- × 0.5-m quadrats in a cool-temperate deciduous broadleaf forest, applying intrinsic conditional autoregressive models at 3 spatial resolutions (0.5 × 0.5 m, 1.0 × 1.0 m, and 2.0 × 2.0 m). The models show that the 3 microphysical factors, known to affect the distribution of S. palmata at coarser scales, also affect its distribution at fine scales. Slope angle and gravel cover had negative effects, while convexity had positive effects on its abundance. However, the strength of their effects depended on the spatial resolution: convexity had the strongest effects at 2-m scale, while slope angle and gravel cover had stronger effects at the finer scales. These findings provide indications of scale-dependent effects of environmental factors that may contribute both to fundamental understanding of forest dynamics and to efforts to enhance the regeneration and conservation of target species.