May 13, 2020

Extracting and quantifying composition and configuration of landscapes is the starting point for landscape ecology. While this is often done on a local scale using landscape indices selected to quantify a particular feature of landscape pattern, the systematic study of landscape compositions and configurations over the entire landmass is lacking. The goal of this study is to create a global inventory of spatial composition and configuration of landscapes for several time frames and to show how these properties change over time. For this purpose, we have used the annual ESA CCI Land Cover data covering years 1992-2015 and C3S Global Land Cover covering years 2016-2018. The composition and configuration of landscapes were quantified using the information theory-based indices derived from co-occurrence matrices of land cover categories. The global landmass was partitioned without overlap into 30 x 30 kilometers square blocks resulting in a set of about 150,000 mesoscale landscapes. For each of these landscapes, five information theory indices (marginal entropy, conditional entropy, joint entropy, mutual information, and relative mutual information) were calculated in each time frame. The entire data set was used to find a high level of correlation between some of the five indices. As a result, only two indices, marginal entropy (describing the diversity of composition) and relative mutual information (describing configuration/aggregation), are sufficient to provide a general description of a landscape. The results show that, on average, the values of marginal entropy had risen (landscapes are increasingly diverse) while the values of relative mutual information had decreased (landscapes are increasingly disaggregated). However, changes in landscape patterns varied between locations depending on geography and the predominant thematic content. Some areas even showed trends opposite to that described by a global average. The next step in the analysis will involve relating the state and trends of landscape changes to human and environmental factors.

Slides: https://nowosad.github.io/iale-na_20/