Migratory tundra caribou are ecologically and culturally critical in the circumpolar North. However, they are declining almost everywhere in North America, likely due to natural variation exacerbated by climate change and human activities. Yet, the interconnectedness between climate, caribou, and human well-being has received little attention. To address this gap, we bridged Indigenous and scientific knowledge in a single model, using as an example the Porcupine caribou herd social-ecological system. Our analysis, involving 688 (fall season) and 616 (spring season) interviews conducted over nine years with 405 (fall season) and 390 (spring season) Indigenous hunters from nine communities, demonstrates that environmental conditions, large-scale temporal changes associated with caribou demography, and cultural practices affect hunters’ capacity to meet their needs in caribou. Our quantitative approach bolsters our understanding of the complex relationships between ecosystems and human welfare in environments exposed to rapid climate change and shows the benefits of long-term participatory research methods implemented by Indigenous and scientific partners.

Climate data. The seasonal weather variables present in this database were calculated based on daily weather variables specific to the seasonal ranges used by the PCH during the fall and spring. These daily weather variables are part of the CircumArctic RangiferMonitoring and Assessment (CARMA) network’s caribou range climate database (https://carma.caff.is/. The CARMA database was developed using NASA’s MERRA database, containing remotely sensed daily averaged climate data with a spatial resolution of 0.50° Lat. x 0.66° Long.

Caribou distribution data. The caribou distribution variables present in this database correspond to median caribou distances (m) to communities for the spring and fall seasons. These were calculated based on satellite tracking locations of adult Porcupine caribou females documented by U.S. and Canadian governmental agencies (see details at http://www.pcmb.ca/herd). During 2000–2008, 32 cows were monitored for a total of 7,428 locations. The average duration of individual monitoring was 3.5 years, but some animals were followed for up to 9 years. Frequency of locations varied across seasons, years, and individuals. On average, individuals were located every 6 days (range 1–251), except from mid-May to mid-July, when locations occurred approximately every second day. Caribou location data were considered highly representative of the PCH location given that this herd remains strongly aggregated.

We analysed caribou and community location data using the sp, rgdal and rgeos packages in R version 3.4.3. Locations were first imported in R and converted to the WGS84 coordinate system. We then calculated the weekly median distances between caribou locations and each of the 9 communities considered in our study. This allowed us to calculate the median of the weekly median distances for the fall (16 August-30 November) and spring (1 April-31 May) PCH seasons. Median seasonal distances between the PCH and the communities were highly correlated to minimum seasonal distances (r > 0.9). 

Authors were not allowed to publicly archive survey data from the ABEKS due to their sensitive nature relating to endangered species and human identity. Access to ABEKS data requires consent from each Indigenous community involved in the project and the completion of a data request form which can be accessed via https://www.arcticborderlands.org/services.

Nevertheless, we did include in the datafiles two .html code files (one for the fall season, one for the spring season) that describe the code used to perform the analysis, and the results obtained with the ABEKS data.

No specific software is required to open the .html files