Remote sensing based data products for large scale life cycle inventory models

The inclusion of indirect lifecycle-based environmental impacts (Scope 3) of products is becoming increasingly important for companies as well as for investors, consumers and other stakeholders in value chains. Life Cycle Assessment can provide valuable decision support for companies, investors and consumers. However, the availability of representative, consistent and reliable regionalised Life Cycle Inventory models is still lacking. In the preparation of life cycle assessments, the collection of the necessary data is a time-consuming and difficult step. This large effort is especially a problem if one wants to compare different large companies or different product groups. In addition to the large effort for data collection, in almost all cases the cooperation of the respective companies is required, and the underlying data are therefore mostly collected by the companies themselves, which limits the reliability and makes them susceptible to greenwashing. The aim of this project was to identify possible approaches to reduce the effort for data collection and at the same time the dependency on data collected by the companies themselves. To achieve this, we explored approaches to link remote sensing data from satellite imagery to LCI models. In the last ten years the availability of satellite imagery has increased significantly, and data providers have developed advanced remote sensing-based data products. Together with remote sensing and geoinformation systems experts we identified remote sensing-based data sources, which can be used for life cycle inventory models. The focus was on three economic sectors, (1) agriculture, (2) mining and (3) energy supply. However, satellite images cannot be used directly for LCA studies, but have to be processed and evaluated in order to derive information which can be used within LCA inventories, such as land use changes or transportation distances. There are several commercial providers which offer already processed remote sensing