The Effects of Spatial Resolution on the Maize acreage estimation by Remote Sensing

Crop acreage estimation is essential to forecast crop production using remote sensing. The different spatial resolution of remotely sensed data directly affects the accuracy of crop acreage estimation. It is necessary and valuable to study the effect of resolution on crop acreage estimation, from both qualitative and quantitative points of view. Therefore, this paper analysed the resolution effect on the accuracy of acreage estimation by using CBERS-02B imagery. Spatial statistics methods and manifold accuracy evaluation indices were used respectively to analyse the data with different spatial resolutions and crop proportion statistics. The study results indicate that decreased spatial resolution will lead to reduced regional accuracy in addition to increased standard deviation, RMSE and bias due to the augmentation of mixed pixels. A replacement of higher resolution data by lower resolution data will have an important impact on the derived crop proportions. The regional accuracy of crop statistics can remain higher than 88%, when the crop proportion is higher than 40%. In summary, the higher resolution of the imagery can lead to increased average regional accuracy. The results of this paper also provide academic and experimental reference to resolve the problem of data selection in crop acreage estimation by remote sensing.