I am working on a river watershed on QGIS and I need a land cover map.

I have a good raster of land use data with 15 m resolution for year 2015, but it covers around the 90 % of the basin.

I have also global land cover raster maps with a resolution of around 300 meters for year 2012 and 2013.

Is there a way to generate a raster data covering the missing 10% of the 15 m raster by using some geostatistics on the 15 m available data and the 300 m raster?

In the case I would assume the land use of 2015 equal to the 2012 and 2013.

  • 1
    Since land cover is a categorical raster, I don't think that geostatistics is the correct approach. You can rescale the 300 meter resolution down to 15 meter resolution, but you won't be gaining anything: the areas will be "blocky", matching the values in the original 300 meter pixels. – Micha Jul 17 '16 at 10:50
  • You could use association rules to relate land use in the 300 meters resolution to these found in the 15 meters data. However, it would involve a lot of work in other statistical software (R, etc.) and would require some cross validation with data you already have at both resolution. I believe it won't be robust and it would be better to consider investing in completing the missing data, or analyze the 90% of the area covered by high resolution data. – dof1985 Jul 17 '16 at 12:57

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.