International Industrial Ecology Day 2021

The accumulation of long-lived material stocks as buildings and infrastructure is a major driver of resource extraction and energy use, thereby contributing to increasing global GHG emissions. At the end of their lifetimes, these stocks are also a challenge for waste management and a potential secondary resource. Spatial patterns of stocks in the built environment are, thus, an important research frontier in socio-economic metabolism research.

We present a novel and collaborative workflow for high-resolution stock-driven bottom-up mapping of material stocks at national scale and beyond. This workflow uses freely available, globally consistent satellite data, and other geodata to generate national maps of material stocks in buildings and infrastructure at a spatial resolution of up to 10m. We separately map features of the built environment – surface area of infrastructure, as well as surface area, height, volume and type of buildings – based on geodata-processing and machine learning algorithms. We transform these into spatially explicit stock maps through a collaborative effort involving national and topic experts to develop material intensity factors compatible with the approach. The workflow has been recently established in a proof-of-concept study, using optical Sentinel-2 and radar Sentinel-1 imagery, as well as the OpenStreetMap database to map stocks for Austria and Germany (Haberl et al. 2021 ES&T). We currently transfer the workflow to map material stocks in the UK, the USA, Japan and Uganda.

We illustrate and discuss the advantages and challenges of our approach in terms of the required disciplinary, technical and interdisciplinary know-how, the substantial potentials for collaborative efforts, as well as the completeness and detail achieved. We provide insights on potential uses of intermediate mapping products and final stocks maps as well as the transferability of our approach to other study regions, and exemplarily illuminate how using high-resolution satellite imagery differs and complements other bottom-up stocks mapping approaches.