In the case of the data Land Corine I noticed some discrepancies. Some surfaces are normally covered with orchards or forests. When maps of land use are made (after the data Corine Land) these areas are cities or industrial areas. Is there a misinterpretation of GIS software? What else generates these inaccuracies, the difference between reality and the data Land Corine? Are the data Land Corine safe enough to be used in geographical analysis?
I think there are studies that specifies the precision of the CLC for particular areas / countries. For example, the accuracy of CLC for Poland is estimated from 92.7%. Everything depends on the precision that you assumed for your studies. But Andreas Paul is right some errors are inevitable.
you can find more details about the errors on the EEA Website http://www.eea.europa.eu/publications/technical_report_2006_7
The quality of the classification depends on the class that is described and on the countries. Globally, CLC passed the 85% overall accuracy target that was set up, but for some classes it is down to 70 %. Furthermore, you must account for the minimum mapping unit of CLC (small patches are included into larger ones if they are below this threshold, so that small villages might have disappeared.
At this stage, CLC remains one of the best consistent database accross Europe, but National projects are often of higher quality. On the other hand, the Copernicus dataset are soon to be released and those will provide additionnal and more precise information with 5 high resolution layers http://land.copernicus.eu/pan-european/high-resolution-layers/view
I have found a similar question on researchgate, so envoy the research...
https://www.researchgate.net/post/Can_anyone_suggest_an_efficient_method_to_evaluate_habitat_quality_in_agricultural_landscape?_sg=YJScIElyNa7mxSx5PfFSX8%2FR%2FlrRhS3MhkupeeLuDAm%2FFDxMoxkvd8IXcmW9msy1_4mJng9ySBxHeQTlOTzS5%2FAB1jfeZoD8IFD%2FL%2B52oYIRb%2B0VX5J2918klOZkIpELb_KZ0QLDx7jhwyG%2BaCNCBVulXnSyoF17j2xLTXuKmvoknXed3y47QV8iqlaoe0YITc#view=52ecec45d039b15b708b4602
Dear Ana-Maria if you intend to use these data for regional analyses (large scales usually above 1:100000), then it is safe. Now, about the differences, you have identify them on the field or between topo maps and the Corine data? Because sometimes what was on topo maps did not exist anymoore on the field, or in the case of some changes for example from orchard to urban (rare cases) it is about the old orchard located in the vicinity of urban areas, etc. Such differences may have multiple causes (human error in the interpretation process, old maps, poor quality of some parts of satellite inage, etc). The interpretation was done manually not authomatically and for a better understanding of the process I would recomend you to read the methodology (see the attachment). You will understand the differences.
- Badges
- Science topic
- Similar topics
- Geoscience
- Geoinformatics
- Geographic Information System