Mon, Aug 15, 2022: 5:00 PM-6:30 PM
ESA Exhibit Hall
Background/Question/MethodseButterfly, an online butterfly community science platform, has been growing across North America from a small team located in Montreal, to an international community with more than half a million records over the last ten years. eButterfly's ever-growing database has been used to evidence large-scale shifts in butterfly phenology, such as the Northern expansion of the Giant Swallowtail (Papilio cresphontes) and range expansion of the invasive European common blue (Polyommatus icarus) across North America and the relevance of these landscape-level phenomena is only expected to increase with climate change. On our ten-year anniversary, eButterfly aims to expand globally and, to do so, several challenges need to be overcome regarding Data Quality, Data Availability, and finding the right balance between the platform's Level of Complexity and Appeal to Community Scientists.In this talk, we want to share with the ESA/CSEE community, all the learning points we've had along the way, so other conservation community science initiatives may be created and grow efficiently.
Results/ConclusionsThroughout the eButterfly's global expansion, we identified three challenges, 1) Data Quality, 2) Data Availability, and 3) the Appeal Vs Complexity Challenge. We implemented a Checklist-based system to control for effort to ensure data quality, following Darwin Core Standards with different privacy levels. Additionally, we created the first harmonized global butterfly taxonomy to structure our database. Finally, vetting butterflies worldwide is a massive task, which motivated us to create a weighed crowdsourced identification system with AI pre-vetting.Data Availability is at the core of our work; we ensured that anyone could access our database, configuring it to upload the entire sampling event (not only presence/absence) every 24 hours to GBIF. In addition, we created eBLabs, a place where researchers can access our raw dataset and a repository for them to upload all their analyses, articles, slideshows, apps, and didactic materials. Finally, being a community science platform, motivating people to use our platform is crucial. To do so, we set ourselves the vision of being more than a data repository, but a thriving Learning/teaching butterflying community with a user-friendly interface, intuitive, flexible, and features such as a mobile app, a discussion forum, and an Image recognition Artificial intelligence algorithm.
Results/ConclusionsThroughout the eButterfly's global expansion, we identified three challenges, 1) Data Quality, 2) Data Availability, and 3) the Appeal Vs Complexity Challenge. We implemented a Checklist-based system to control for effort to ensure data quality, following Darwin Core Standards with different privacy levels. Additionally, we created the first harmonized global butterfly taxonomy to structure our database. Finally, vetting butterflies worldwide is a massive task, which motivated us to create a weighed crowdsourced identification system with AI pre-vetting.Data Availability is at the core of our work; we ensured that anyone could access our database, configuring it to upload the entire sampling event (not only presence/absence) every 24 hours to GBIF. In addition, we created eBLabs, a place where researchers can access our raw dataset and a repository for them to upload all their analyses, articles, slideshows, apps, and didactic materials. Finally, being a community science platform, motivating people to use our platform is crucial. To do so, we set ourselves the vision of being more than a data repository, but a thriving Learning/teaching butterflying community with a user-friendly interface, intuitive, flexible, and features such as a mobile app, a discussion forum, and an Image recognition Artificial intelligence algorithm.