10.18710/5WNWRLMul, E.E.Mul0000-0001-7297-8081Norwegian Institute for Nature ResearchMurguzur, Francisco Javier AncinFrancisco Javier AncinMurguzur0000-0002-1435-5190UiT The Arctic University of NorwayReplication Data for: "Impact of the COVID-19 pandemic on human-nature relations in a remote nature-based tourism destination"DataverseNO2022Computer and Information ScienceSocial SciencesTourismCloud VisionNorthern NorwayContent analysisMul, EvertEvertMulNorwegian Institute for Nature ResearchUiT The Arctic University of NorwayUiT The Arctic University of NorwayUiT The Arctic University of Norway2022-01-022022-08-232023-11-01Textual data10.1371/journal.pone.02733548682132862509text/plaintext/csv1.2CC0 1.0This dataset was used for the article "Impact of the COVID-19 pandemic on human-nature relations in a remote nature-based tourism destination", published in PLOS ONE 2022. It contains information about 196.200 photos taken in Northern Norway that were uploaded to the online Flickr platform between 01-01-2000 and 31-03-2021. The photos were processed using Google's Cloud Vision algoritm. Each line in the dataset represents a photograph with a series of labels assigned by the algoritm, which describe the content of the photograph. Each line further includes ID numbers for the photo and the photographer, the date the photograph was taken, which area the photo was taken in and the photographers' origin (Norwegian or international).Paper abstract: Tourism and nature-based recreation has changed dramatically during the COVID-19 pandemic. Travel restrictions caused sharp declines in visitation numbers, particularly in remote areas, such as northern Norway. In addition, the pandemic may have altered human-nature relationships by changing visitor behaviour and preferences. We studied visitor numbers and behaviour in northern Norway, based on user-generated data, in the form of photographic material that was uploaded to the popular online platform Flickr. A total of 195.200 photographs, taken by 5.247 photographers were subjected to Google’s “Cloud Vision” automatic content analysis algorithm. The resulting collection of labels that were assigned to each photograph was analysed in structural topic models, using photography date (relative to the start of the pandemic measures in Norway) and reported or estimated photographers’ nationality as explanatory variables. Our results show that nature-based recreation relating to “mountains” and “winter” became more prevalent during the pandemic, amongst both domestic and international photographers. Shifts in preferences due to the pandemic outbreak strongly depended on nationality, with domestic visitors demonstrating a wide interest in topics while international visitors maintained their preference for nature-based experiences. Among those activities that suffered the most from decline in international tourism was northern lights and cruises as indicated by the topic models. On the other hand, images depicting mountains and flora and fauna increased their prevalence during the pandemic. Domestic visitors, on the other hand, spent more time in urban settings as a result of restrictions, which results in a higher prevalence of non-nature related images. Our results underscore the need to consider the dynamic nature of human-nature relationships. The contrast in flexibility to adapt to changing conditions and travel restrictions should be incorporated in collaborative efforts of municipalities and tour operators to develop sustainable local nature-based tourism products, particularly in remote areas.Rooglevision (R-package, based on Google's Cloud Vision)), Version: 0.0.1.1Tromsø3509065The Research Council of Norway280778