MaxEnt modelling of the potential distribution areas of cultural ecosystem services using social media data and GIS

Abstract

This study uses photographs on social media to spatially model the potential distribution of user preferences for cultural ecosystem services (CES). The areas within the administrative boundaries of the province of Isparta in Turkey’s Mediterranean region constitute the study area. Hundred and sixty-six photographs with geographical coordinates, taken between the years 2012–2018 and shared on photograph sharing platform Flickr, were linked to CES, and the CES provided in the study area were identified and categorised. The species distribution model was used in the study, and the natural and cultural assets in the study area were taken as environmental independent variables. The study used MaxEnt and geographical information systems integrally. For every CES, hotspot areas were identified and the degrees of significance of environmental variables for generating CES potential were determined. The highest level of CES provision in the study area was for recreation. The most important environmental variables for determining CES distribution were roads, religious places and distance to historical and cultural areas, identified by degrees of proximity using Euclidian distances. Among the significant conclusions of the study are overlapping outcomes for closely related CES (such as aesthetic values and recreational values) and the relationship between the outcomes and the natural and cultural assets in the area (such as water surfaces, green fields.) The study is thought to contribute to the extant literature in terms of spatially assessing the intangible benefits of ecosystem services and land use decision-making.