Abstract: In this paper RS sample plots and high resolution images are used to interpret them for the types of stands and forest land uses,and then employ the probability estimation approach to extend the scale of results from plot to population.In the end,the Germany NFI(National Forest Inventory)statistical results are utilized to test the predicted results.With the method,the paper sets 334 RS plots with the scale of 1km×1km in Germany,afterwards,it also obtains high resolution images to be interpreted and make a sampling prediction.The results show that for Germany,the relative error of land area is 2.43%;that of forest area 0.59%;that of forest coverage ratio 2.96%.In addition,the detailed information of provincial forest resources is also obtained.In summary of the aforementioned results,it is safe to say that the approach is feasible and practical,the result reliable,and factors sufficient.Besides,it also realizes the active monitoring of overseas forest resources and provides a new approach for it.

Key words: oversea forest resource monitoring, RS sample plot, sampling inventory theory