< Back to previous page

Project

A Statistical Framework for the Analysis of Ecological Data Collected via Citizen Science

Recent global biodiversity declines have prompted increased investments in research towards the understanding of ecosystem dynamics and spatio-temporal trends in species distributions. These complex natural phenomena necessitate the long-term assessment of geographically expansive regions, for which citizen science is considered to be a cost-effective data collection approach. However, study protocols are often absent in citizen science, or not adhered to, such that bias that originates from so-called opportunistic samples invalidates statistical inference. In this project, a statistical framework is developed to correctly analyse spatio-temporal data collected by citizens, in order to evaluate biodiversity in the province of Limburg (Belgium) and link it to environmental dynamics caused by climate change and habitat fragmentation. We collaborate with LIKONA (Limburgse Koepel voor Natuurstudie), a renowned citizen science collective, (i) to develop statistical methods that correct for opportunistic sampling, and (ii) to construct adaptive spatial randomisation processes, based on a geostatistical golden standard in study designs, which we integrate in a mobile app for citizen scientists. Through this, we aim to improve the ecological knowledge of spatio-temporal trends in the distribution of key species, in order to set up guidelines for nature conservation in Limburg and beyond.

Date:15 Feb 2021 →  Today
Keywords:Opportunistic sampling models, Biodiversity and species distribution modelling, Spatio-temporal statistics, One Health
Disciplines:Statistics, Modelling and simulation, Conservation and biodiversity
Project type:PhD project