Volume XLII-2/W13
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLII-2/W13, 1213-1220, 2019
https://doi.org/10.5194/isprs-archives-XLII-2-W13-1213-2019
© Author(s) 2019. This work is distributed under
the Creative Commons Attribution 4.0 License.
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLII-2/W13, 1213-1220, 2019
https://doi.org/10.5194/isprs-archives-XLII-2-W13-1213-2019
© Author(s) 2019. This work is distributed under
the Creative Commons Attribution 4.0 License.

  05 Jun 2019

05 Jun 2019

A PRELIMINARY QUALITY ANALYSIS OF THE CLIMATE CHANGE INITIATIVE LAND COVER PRODUCTS FOR CONTINENTAL PORTUGAL

C. C. Fonte1,2, L. See3, M. Lesiv3, and S. Fritz3 C. C. Fonte et al.
  • 1Department of Mathematics, University of Coimbra, Coimbra, Portugal
  • 2Institute for Systems Engineering and Computers at Coimbra (INESCC), Coimbra, Portugal
  • 3International Institute for Applied Systems Analysis, Laxenburg, Austria

Keywords: Climate Change Initiative, Land Cover, Change detection, Quality, Omission errors, Commission errors

Abstract. The aim of this paper is to perform a preliminary analysis of the compatibility and quality of the available time series of land cover data available for continental Portugal, in particular, Climate Change Initiative Land Cover maps, which are available annually from 1992 to 2015; CORINE Land Cover and the Urban Atlas for 2006 and 2012; and the Portuguese Carta de Ocupação do Solo for 2007 and 2010. Changes were first identified per product between the different data sets for consecutive dates and then a comparison was made between products. This was followed by validation of two study areas using the COS and UA as reference products. The results show that increases in urbanization are visible in all pairs of products but that the amount of change varies. Moreover, some changes are not in the same direction but may be attributable to classes with small areas and the coarser resolution of the CCI LC maps compared to the other products. The CCI LC maps also overestimate the forest/natural vegetation class by 11–13%, which is also the largest class in Portugal.