PT  - JOURNAL ARTICLE
AU  - Francesca Mancini
AU  - George M. Coghill
AU  - David Lusseau
TI  - Using social media to quantify spatial and temporal dynamics of wildlife tourism activities
AID  - 10.1101/093112
DP  - 2016 Jan 01
TA  - bioRxiv
PG  - 093112
4099  - http://biorxiv.org/content/early/2016/12/11/093112.short
4100  - http://biorxiv.org/content/early/2016/12/11/093112.full
AB  - Wildlife tourism is a profitable industry that can affect the conservation status of targeted populations. Tourist behaviour plays a key role in the success of sustainable management strategies. Traditionally, visitor numbers are obtained through surveys, which are expensive and limited in coverage and resolution. Recently, data from social media have been used to quantify visitation. However, we do not know at which scale the use of this proxy is appropriate, especially outside protected areas. Here, we validated for the first time the use of a dataset obtained from the photo-sharing website Flickr as a proxy for wildlife tourism in Scotland.We used photos uploaded on Flickr to estimate visitation in the Cairngorms National Park (CNP) and compared this dataset to a time series of visitor numbers obtained from the CNP authority. Then, we compared the spatial distribution of photographs of birds, seals, whales and dolphins taken in Scotland and uploaded on Flickr to a dataset obtained from a 2014-15 Scotland-wide wildlife tourism survey.Wavelet analysis showed that the two time series are significantly correlated and synchronised. The results of the spatial validation showed that both the presence and the number of pictures uploaded on Flickr are correlated to survey data at different scales. Finally, kernel density maps of the wildlife pictures revealed spatio-temporal trends in wildlife watching hotspots that confirmed the validity of this dataset.Both temporal and spatial trends in the distribution of pictures uploaded on Flickr displayed similar patterns to those observed in datasets obtained using traditional methods. This was true for different spatial scales and for locations inside and outside protected areas. Therefore, this method allowed us to quantify visitation even in areas that are not monitored. In conclusion, despite limitations and challenges, data from social media offer great potential to study wildlife tourism at different spatial and temporal scales.