Transforming the use of citizen science data for biodiversity conservation at different scales

Obtaining large amount of data on species’ characteristics such as their distribution, abundance or movement patterns is not only important for scientists to better understand species’ ecology but it is also fundamentally important for policymakers and environmental managers because it provides a knowledge platform to ensure successful biodiversity conservation. Citizen science (i.e. the involvement of volunteers who collect and/or process data as part of a scientific inquiry) data has recently gained attention among researchers as it can help researchers tackle research questions that otherwise could not be addressed without the involvement of large numbers of professional data collectors. Nevertheless, because volunteer recorders are highly motivated to encounter interesting wildlife the spatial and temporal patterns of records are not random or systematic and hence very different from the kind of rigorous random sampling protocol that scientists are used to work with. The aim of this thesis is to understand the citizen science observation generation process and then assess the performance of citizen science observations to infer ecological properties at different ecological scales. This thesis consists of five articles. The first study takes the example of hunters as a special case of citizen scientists and assesses their importance for biodiversity monitoring. In articles 2 and 3 I compile distribution maps of large mammals in Europe using citizen science data in tandem with other source of data to study species’ ecology at a macro scales (distribution). Finally, articles 4 and 5 show that if potential biases are accounted for using appropriate statistical methods, citizen science observations can give a good approximation of species’ ecology at micro scales (habitat patches). From this thesis two main conclusions emerge. Firstly, to transform how citizen science is used it is critical to understand the data generation process underlying the creation of the geo-localised citizen science observations in order to fully grasp the extent of the potential biases in citizen science data. Then citizen science data can be used at multiple scales if biases are accounted for using proper methodology. It is important to realize that the methodologies used to account for biases in citizen science data have to be scaled to the research question and to the data available as results could lead to misleading conclusions about the species’ ecology. Overall, the future of citizen science remains very bright and this thesis contributes to further understanding and making better use of it