The naturalness index for the identification of natural areas on regional scale

Abstract

Researchers have been trying to measure the influence of humankind on our environment for several decades. Among the technical solutions for this goal, approaches for a rule-based fusion of different geospatial information layers have reached the greatest level of maturity, enabling the (semi-)automatic production of maps at resolutions of about 1 square kilometer. While those existing approaches aim at a global analysis of human influence, conservation efforts are usually implemented on a regional scale. To bridge the gap between a global, low-resolution analysis and a local, high-resolution analysis, with this paper, we propose the Naturalness Index (NI), which represents the naturalness of the Earth’s surface as a metric between 0 and 100 at a spatial resolution of 10 m per pixel. Our approach builds on the established Human Influence Index, but replaces different geospatial input layers with more recent, higher resolution counterparts. Using the cloud computing platform Google Earth Engine, regional maps at a resolution of 10 m × 10 m per pixel can be produced efficiently in a fully automatic manner. We demonstrate the functionality by creating maps for three different localities in Europe – Bavaria, Lapland, and Scotland. A comparison of the NI values achieved with our approach to official conservation areas as well as a correlation with existing similar maps indicate the validity of our approach.