Community Research and Development Information Service - CORDIS


CrowdLand Report Summary

Project ID: 617754
Funded under: FP7-IDEAS-ERC
Country: Austria

Mid-Term Report Summary - CROWDLAND (Harnessing the power of crowdsourcing to improve land cover and land-use information)

The overall aim of the CrowdLand project is to demonstrate the potential of using crowdsourcing and citizen science for the collection of data on land cover and land use using mobile phones. The FotoQuest Austria app was developed and tested during the summer of 2015, which is a serious game that encourages players to go out into the landscape and undertake quests. Each quest consists of photographing the landscape in multiple directions and entering the type of land cover and land use. The protocol is loosely based on the authoritative data collection of the LUCAS (Land Use Cover Area Sample frame) survey, which takes place every three years across EU member states. The idea was to see whether citizens can help to complement the LUCAS survey by providing data on a more regular basis. When the crowdsourced data were compared to LUCAS, the results showed that citizens are able to document the landscape in a useful manner, and has great potential for enhancing LUCAS, in particular, in environments that are easy to recognize (e.g. built up areas, agricultural land), as well as for calibration of remotely-sensed land cover products. Methods and tools developed in Austria are currently being tested in other European countries, as well as in Kenya.

CrowdLand has also focused on a tool for the rapid sorting of images, supporting the development of the cross-platform game Picture Pile. The game has run for about 14 months and has led to the collection of more than 5 million observations on the presence of tree cover loss in Tanzania and Indonesia. The data from the game are currently being analyzed. CrowdLand has also contributed to strengthening the flagship Geo-Wiki visualization and data collection platform ( and the underlying database.

In addition to the development of tools and the running of data collection campaigns, research in the CrowdLand project has addressed the quality of crowdsourced data and methods of data fusion. An in-depth comparative analysis of crowdsourced and authoritative data has been undertaken, which indicates that the crowd is able to make meaningful and useful contributions to monitoring land cover and land use. The project has also investigated the application of innovative data fusion methods, which have contributed to the production of hybrid global land cover maps.

To date, the CrowdLand project has produced 12 journal articles, 5 conference papers and 2 book chapters.

Reported by

Follow us on: RSS Facebook Twitter YouTube Managed by the EU Publications Office Top