Search

Highlights • Passive crowdsourcing offers novel perspectives on human-environment relationships. • We systematically review applications of social media data in environmental science. • Large data volumes, scale of analysis and real-time monitoring are key strengths. • Challenges of data noise, biases, ethics and future availability need consideration. • We identify five key challenges for further development in the field. Abstract The analysis of data from social media and social networking sites may be instrumental in achieving a better understanding of human-environment interactions and in shaping future conservation and environmental management. In this study, we systematically map the application of social media data in environmental research. The quantitative review of 169 studies reveals that most studies focus on the analysis of people's behavior and perceptions of the environment, followed by environmental monitoring and applications in environmental planning and governance. The literature testifies to a very rapid growth in the field, with Twitter (52 studies) and Flickr (34 studies) being most frequently used as data sources. A growing number of studies combine data from multiple sites and jointly investigates multiple types of media. A broader, more qualitative review of the insights provided by the investigated studies suggests that while social media data offer unprecedented opportunities in terms of data volume, scale of analysis, and real-time monitoring, researchers are only starting to cope with the challenges of data's heterogeneity and noise levels, potential biases, ethics of data acquisition and use, and uncertainty about future data availability. Critical areas for the development of the field include integration of different types of information in data mashups, development of quality assurance procedures and ethical codes, improved integration with existing methods, and assurance of long-term, free and easy-to-access provision of public social media data for future environmental researchers. [ABSTRACT FROM AUTHOR]

Protected areas offer unique opportunities for recreation, but the non-market nature of these benefits presents a significant challenge when trying to represent value in the decision-making processes. The most common techniques to value recreation are based on resource-intensive primary surveys which are difficult to perform at a large scale or in remote locations. This is true in the case of Italy, where a large and diverse network of protected areas suffers from lack of data. Here, we offer an alternative data source for the valuation of recreation by integrating the metadata of geotagged photographs from social media into single-site, individual travel cost models for 67 Italian protected areas. Count data model results are generally consistent with standard economic and consumer demand theory for ordinary goods, with a zero-truncated Poisson model returning down sloping demand curves for 50 of 67 sites. A significant travel cost coefficient was returned for 33 sites (p -value <0.05) for which consumer surplus estimates were found in the range between €6.33 and €87.16, with a mean value per trip of €32.82. Although not without their own challenges, the results presented highlight the possibilities of new forms of spatial big data as a novel data source for environmental economists. [ABSTRACT FROM AUTHOR]

• Crowdsourced travel cost method validated against representative surveys. • Regional visitor-days are highly correlated between Flickr and survey data. • Value estimates highly correlated between Flickr and survey data. • MAE of €4.93 and MAPE of 22% for German national parks. • Flickr estimates outperform basic unit value transfer. Representing the economic value of recreation generated by natural ecosystems can provide a powerful incentive for their conservation. Demonstrating value, however, requires resource-intensive surveys or reliance on techniques which transfer value from other sites. Recent technological developments have contributed to an abundance of georeferenced public data as an alternative for exploring human-nature interactions, offering a potential substitute to costly and time-consuming visitor surveys and their commonly adopted alternative for the valuation of recreation, value transfer. Here, we (1) integrate data crowdsourced from geotagged photographs from social media into the travel cost method non-market valuation technique and (2) validate the results with value estimates generated using representative on-site surveys for German national parks and value transfer techniques. As expected from standard economic and consumer demand theory for ordinary goods, we find downsloping demand curves. Consumer surplus for access to the parks ranges between €16.54 and €34.90 (2016 rates). Value estimates are significantly correlated with those generated by on-site surveys, with a mean absolute error of €4.93 and a mean absolute percentage error of 22%, outperforming basic unit value transfer. We highlight the entailed, unprecedented opportunities to extend the scope and reduce costs of research seeking to value outdoor recreation. [ABSTRACT FROM AUTHOR]

Social media data are increasingly utilised as a low-cost alternative to visitor surveys in characterising nature-based recreation. However, the information available on individual users is limited and typically does not include provenance, restricting the potential applications and impact of the data. Here we investigate a methodology to estimate social media visitors' home locations at various spatial scales and apply it to the entire network of national parks in Germany. We compare predicted visitor provenance to representative onsite survey data and explore group-specific spatial and temporal patterns of recreation as characterised by users' geotagged photographs. Results show that photograph metadata can be used to assign home locations with accuracies between 62 and 89% depending on spatial scale implemented. Said social media-based predictions are reasonably well representative of the surveyed visitor structure in German national parks with Flickr visitor-days composed of 19% local, 62% non-local German and 19% international visits. • Photographs metadata can be used to assign home locations of Flickr users. • Home locations are predicted with accuracies between 62 and 89% depending on scale. • Flickr visits are 19% local, 62% non-local German and 19% international. • Flickr photographs can characterize seasonal, weekly and daily visitation patterns. • International visitors take more photographs per visit than local visitors. [ABSTRACT FROM AUTHOR]