Georeferencing time series of Canadian vertebrates in the Living Planet Database

Data rescue intern:
Olivia Rahn

The Living Planet Index tracks long term population trends in vertebrate populations across the globe- these population trends are published as part of the WWF’s Living Planet Report. The database is curated and managed by multiple organizations, including the London Zoological Society. While the database is extensive and has been used for many different analyses of vertebrate populations, many records in the dataset are either not georeferenced or are not georeferenced using a standardized protocol. This limits the scope of analyses that can be done using the dataset- for example, analyses that integrate local climate data or land-use change data with population trends require accurate georeferencing. 

​Throughout the course of this internship, graduate students Mary Paz Mañé and Olivia Rahn worked with Valentina Marconi, Louise McCrae, and Stephanie Deinet from the London Zoological Society to develop a protocol to georeference Canadian vertebrate records in the LPI Database. The first step in this process was identifying the different data types associated with each record- some population trend data is sourced from journal articles with associated maps or specific coordinates, some have broad descriptions of study areas, and other records have no geospatial information associated with them at all. The interns then identified different types of publicly available data that could be used to help georeferenced records, primarily by gathering polygons for provincial and national parks, protected areas, and administrative boundaries in Canada. This information was used to create a protocol that walks users through how to georeferenced records using both the information provided in the original source and the spatial data gathered by the interns. The interns also wrote a series of R scripts that can be used to calculate new spatial information to add to the LPI dataset, including newly calculated centroids for each population record. 
 
The georeferencing protocol and R scripts associated with it will be publicly available through the Open Science Framework (OSF). The protocol will be used in the future to work towards georeferencing all Canadian records in the LPI dataset, and also to inform georeferencing protocols that can be applied to the entire global dataset.