I have three files. The first is a shapefile for country boundaries and the second and third are raster files with the global land covers for 2005 and 2009. I'm thus charged to calculate the areas for the different land uses in 5 countries in 2005 and 2009. The CRS for all files is World Mollweide EPSG:54009.

First, I selected the countries (one by one) that I wanted to deal with and I stored them as shapefiles... Afterwards, I clipped the two raster files using as mask each one of the five countries that I previously stored as shapefiles. I calculated then the areas for each one of the generated rasters (using r.report). So, my issue came out.

The problem is that the total area for each country in 2005 is not the same as the area for the respective country in 2009 (there are some differences like 100 square kilometres for some countries). I guess the problem started when clipped because some countries got more cells (I notice that when I zoom in and look at the borders of the generated rasters)...

So my question is, using QGIS, which other option could I use to calculate the areas more accurately? I tried to reproject, with the number of counts per value... but the differences are still too big...

  • Are all of your layres(vector and raster) have same projction? – SIslam Jan 17 '15 at 14:00
  • You can polygonize those clipped rasters and then clip again with mask (vector clipping vector). Bare in mind that the total area of country will be then even repective but still there will be differences in landover areas inside countries - due to differences in 2005 and 2009 rasters pixel size. – jzol Jan 17 '15 at 19:17
  • are you familiar with python+gdal? – user1269942 May 19 '15 at 0:55

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.