Developing new cropland extent spatial data

Developing new cropland extent spatial data


Reliable global data sets of cropland extent would seem to be a key resource for many different purposes.  For example, the future agricultural potential of tropical areas will play a critical role in meeting the challenge of future food security.  While crop yield can be estimated with some reliability using a limited sample of ground observations, reliable crop acreages and their spatial distribution are rarely available. Different remote sensing products are available, but comparison work (see, for example, Fritz et al. (2010), “Comparison of global and regional land cover maps with statistical information for the agricultural domain in Africa”, International Journal of Remote Sensing 31 (9), 2237-2256), indicates that some products do better in some places and some in others.  The work here seeks to produce a hybrid product that combines the best of the available information.


The primary aim of this work is to hold a workshop that integrates the best land cover information that is currently available, as well as to build an improved land cover validation dataset. The workshop will be undertaken in close collaboration with the Group on Earth Observation (GEO) and the agriculture Communities of Practice (COP).  The workshop will link to a number of other activities which are already in place, namely the Cal/Val activities of the CEOS land cover validation sub-group, the FRA2010 remote sensing exercise, and the work done by the Living Standards Measurement Survey (LSMS).


Workshop organizers are sending participants a questionnaire asking them what type of data they will be able to provide and if they are willing to make the data available as part of the global datasets/hybrid product. These datasets will then be requested prior to the workshop and prepared so that the harmonization, integration and mosaicing can be facilitated both during and shortly after the workshop.


The ultimate outcome of the workshop will be an improved cropland map, as well as an increased number of validation sites (classified geo-tagged pictures with a timestamp), and eventually a peer-reviewed publication.  The data as well as the resulting hybrid map will be made available on  The organisers will also consider potentially extending the functionalities in order to track blogs and comments on the different overall qualities of the products which have gone into the hybrid map, and to capture overall expert opinion.


International Institute for Applied Systems Analysis (IIASA)


September 2010 – July 2011