Abstract: Choosing Linglong Mountain with varied relief and its surrounding areas as the experimental zone, the accuracy of elevation simulated by two types of free digital elevation model (DEM) including Shuttle Radar Topography Mission 3 (SRTM3) and the Advanced Spaceborne Thermal Emission and Reflection Radiometer Global Digital Elevation Model (ASTER GDEM) was analyzed by means of applying two quantitative evaluation methods, namely the root mean square errors (RMSE) and the accumulation products (AP), and one subjective evaluation method, i.e. visual inspection of reconstructing contours through DEM to testify the application of DEM to small regions in forestry sector. The results showed that forest cover in the experimental zone had some influence on elevation accuracy simulated by DEM, the value of RMSE and AP by SRTM3 was 11534 m and 0177, whereas the value of RMSE and AP by ASTER GDEM was 10459m and 0166, respectively. Though the accuracy of elevation simulated by both SRTM3 and ASTER GDEM in the study area met the demand published officially, the elevation accuracy of ASTER GDEM was higher than that of SRTM3. The results from the visual inspection of reconstructing contours through DEM suggested that the contours extracted by SRTM3 and ASTER could better describe the practical status of the low mountainous terrain in the experimental area, but the interpretation for the river valley or plain area was not so precise, and the deflection of contours derived from SRTM3 was a bit larger than that of derived from ASTER GDEM. Comprehensively concluding that the data from ASTER GDEM were preferably recommended apply to Linglong Mountain and its surrounding areas to simulate the elevation contour.