FotoQuest Go: A Citizen Science Approach to the Collection of In-Situ Land Cover and Land Use Data for Calibration and Validation

Fritz S, Sturn T, Karner M, Moorthy I, See L, Laso Bayas JC, & Fraisl D (2019). FotoQuest Go: A Citizen Science Approach to the Collection of In-Situ Land Cover and Land Use Data for Calibration and Validation. In: Digital Earth Observation, 01-04 July 2019, Salzburg, Austria.

[img]
Preview
Text
EARSel Symposium - Fritz et al FotoQuest Go.pdf - Presentation
Available under License Creative Commons Attribution Non-commercial.

Download (11MB) | Preview

Abstract

The Land Use/Cover Area frame Survey (LUCAS) is a harmonized data collection exercise on land cover and land use, which employs a systematic sample across EU member countries. The survey is undertaken every three years by trained surveyors and is a rich data set of land use and land cover, including geotagged photographs. LUCAS has been used to validate the CORINE land cover map, which is generated for EU member countries every 6 years, and it represents one of the only publicly available in situ data sets for the calibration and validation of products derived from Earth Observation for Europe. However, the LUCAS exercise is undertaken at a considerable cost to the taxpayer. Given that citizen science is becoming more popular, i.e., the involvement of citizens in scientific research including data collection, we set out to determine whether citizens could help in gathering in situ data on land use and land cover. Advantages of this approach include data collection that is at a denser sample in some areas, the potential for more up-to-date information, since LUCAS is only carried out every 3 years, and as a cost effective way to complement and enrich LUCAS data collection.

To test out this idea of land use and land cover data collection by citizens, the FotoQuest Go app was developed. FotoQuest Go is one of many tools that are part of the H2020-funded LandSense Citizen Observatory for land cover and land use. FotoQuest Go (shown in Figure 1) leads any citizen taking part in our crowdsourcing campaigns to pre-specified locations shown on the map. In some cases, these locations overlap with LUCAS points so that quality assurance can be undertaken, comparing the land cover and land use data from the citizens with that of the professional surveyors. As a location on the map is reached, users are asked to take 4 photographs in 4 cardinal directions away from the location and one at the actual point. The map guides the users, e.g., only allowing them to take a photograph if the compass direction is S, N, E or W, and providing advice regarding how the photos should be taken, e.g. two-thirds land and one-third sky. The citizens are then asked to classify the land cover using a simple, visual decision tree, followed by the land use. The app has been designed to be easy-to-use. For example, it is visually attractive and intuitive as the map interface provides guidance on reaching locations, and the app helps users in taking optimal photographs. The decision tree for determining land use and land cover has also been designed in a simple user-friendly fashion.

A number of different campaigns have been run with FotoQuest Go, where incentives for participation have ranged from prizes at the end of the campaign to small, monetary rewards for each point captured that was deemed to be of sufficient quality. These gamification elements have helped to motivate the crowd and make the crowdsourcing experience more fun. An analysis of the data showed good agreement between the citizens and the surveyors at the LUCAS locations when considering high level land cover classes, e.g. forest, urban, water, etc., i.e., accuracies greater than 80%. Thus, using an app such as FotoQuest Go, citizens can collect land cover and land use data that could be used for calibration and validation of land cover and land use maps. Moreover, many geotagged photographs have been collected, which could additionally be interpreted and used for calibration and validation purposes.

More recently, the main functionality contained within FotoQuest Go has been moved into the PAYSAGES mobile app for crowdsourcing data on land cover and land use. The idea is to involve citizens in the validation and improvement of the French land cover map developed by the French Mapping Agency (IGN). The PAYSAGES app has been developed within the H2020 LandSense Citizen Observatory as part of a demonstration case in urban areas. The app will be used in data collection campaigns during the summer of 2019.

Item Type: Conference or Workshop Item (Paper)
Uncontrolled Keywords: land cover, land use, in situ data collection, validation, LUCAS
Research Programs: Ecosystems Services and Management (ESM)
Depositing User: Luke Kirwan
Date Deposited: 03 Jul 2019 06:24
Last Modified: 03 Jul 2019 06:24
URI: http://pure.iiasa.ac.at/15980

Actions (login required)

View Item View Item

International Institute for Applied Systems Analysis (IIASA)
Schlossplatz 1, A-2361 Laxenburg, Austria
Phone: (+43 2236) 807 0 Fax:(+43 2236) 71 313