We investigated the effect of different measurement intervals on the sensitivity of detecting yarding detours using a digital elevation model (DEM). These intervals were shorter than the mesh size of the DEM used for the analysis. First, we defined the proportion of total yarding area in which the yarding detour coefficient is >1.0 as the yarding detour rate Pdy, which is used as an index to evaluate increases in yarding distance. Here, the yarding detour coefficient is the ratio of the extra distance to the shortest yarding distance. Subsequently, we took the absolute value of the exponential part of the regression equation using the power function of the yarding detour coefficient distribution form to represent ways in which large yarding detours can occur |β|. The measurement intervals were set as d = 50, 25, 20, 10, 5, and 1 m. The results showed that Pdy decreased when d increased; however, the tendency was not linear and Pdy was almost at the same level when 5≦ d ≦ 20 m. |β| decreased at d ≧25 m; however there are no large change in the |β| value at d < 25 m. This indicates that d < 25 m is optimum. Considering that the calculation time increases with the decrease in the measurement interval, d =20 m was considered optimal for efficient measurement of the yarding detour using DEM.