Author(s): Ioana Popescu, Thaine H. Assumpcão, Ioana Popescu
Linked Author(s): Ioana Popescu, Thaine H. Assumpção, Ioana Popescu
Keywords: Crowdsourcing; Citizen science; Hydrodynamic modelling; Wetland ecosystem; Danube Delta;
Abstract: Crowdsourcing of environmental data has recently been proposed as a possible alternative to augment and enrich available datasets for managing environmental systems. In water-related studies, it could supplement the data available from existing monitoring networks, including remote sensing. This work presents experiences from research in collecting and processing crowdsourced data for use in flood modelling studies. The work has been carried out in an ongoing European research project (of the H2020 Research Programme), where data collected by citizens are used to support the development, calibration and validation of hydrodynamic models used for flood analysis. The data are gathered by dedicated game-like mobile phone app in the form of images and videos that are later post-processed to provide data on e.g. land use/land cover, river geometry, water levels and water velocities. Given the source of these data, large uncertainties may be expected and questions arise regarding their quality and the ways in which they can be used in the modelling process. The experiences presented here are from the area, where citizens’ data gathering campaigns have been organized. The case study is in the Danube Delta, Romania, which is still preserved in quite natural conditions and serves as a very important ecosystem, especially for various bird species, including migratory ones. Flooding patterns in the delta are important for supporting this ecosystem and their analysis is supported by a 1D-2D hydrodynamic model. The model is first developed with existing data (without the contribution of crowdsourced data), and then they are improved by data obtained from dedicated citizens’ data collection campaigns. Although the initial results are promising many challenges remain open in crowdsourcing approaches, regarding gathering sufficient amount of data at right time and on right locations.
Year: 2019