A high-accuracy method for filling voids on remotely sensed XCO2 surfaces and its verification

The method for high accuracy surface modeling (HASM), inverse distance weighting (IDW) and ordinary Kriging (OK) are used to fill voids on XCO2 surfaces of GOSAT and SCIAMACHY. Inner voids and boundary voids are artificially made by cleaning out the downloaded data, where there are no voids, in different latitude belts of the northern hemisphere and the southern hemisphere at random. The cleaned-out data in the artificial voids are selected as the 'true' values for verification. The results demonstrated that HASM always has the highest accuracy compared with the classical methods of IDW and OK, whether voids are inner ones or boundary ones, data sets are from GOSAT or from SCIAMACHY, and the void areas are larger or smaller. HASM is an alternative approach to filling voids on XCO2 surfaces from the satellites.