The use of semantic information in point-of-interest geospatial data to forecast and generate high precision maps of tourism carbon emissions

Abstract The issue of carbon emissions from tourism is a matter of urgent concern. The traditional methods of measuring carbon emissions from tourism, mainly bottom-up and top-down methods, have major limitations, and are poorly backward compatible and analyzable. POI (Point-of-interest) data with fine-grained categorization offers the possibility of downscaling tourism carbon emissions. In this paper, a 2km×2km map of tourism carbon emissions is generated using six types of tourism-related POI data, and the characteristics of the distribution of tourism carbon emissions in Chinese provinces and cities are discussed from the perspective of spatial distribution. This study confirms the feasibility of POI data in measuring tourism carbon emissions, characterizes the distribution of tourism carbon emissions in China and analyzes the distribution patterns of tourism carbon emissions within different cities, and identifies the contributing sources of tourism carbon emissions, which are all helpful for monitoring tourism carbon emissions and policy formulation. A variety of spatial analysis tools are used to support the conclusions of the study.