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.

保护区提供了独特的娱乐机会，但这些利益的非市场性质在试图在决策过程中体现价值时提出了重大挑战。评估娱乐价值的最常见技术是基于资源密集型的初级调查，这些调查难以在大规模或偏远地区进行。意大利的情况就是如此，那里庞大而多样化的保护区网络因缺乏数据而受到影响。在这里，我们通过将来自社交媒体的地理标记照片的元数据集成到 67 个意大利保护区的单一站点、个人旅行成本模型中，为娱乐估值提供了另一种数据源。计数数据模型结果与普通商品的标准经济和消费者需求理论基本一致，使用零截断泊松模型返回 67 个站点中的 50 个的倾斜需求曲线。33 个站点返回了显着的差旅成本系数（p值 <0.05），消费者剩余估计值在 6.33 欧元到 87.16 欧元之间，每次旅行的平均值为 32.82 欧元。尽管并非没有自己的挑战，但所呈现的结果强调了新形式的空间大数据作为环境经济学家的新型数据源的可能性。