Data granularity for life cycle modelling at an urban scale

Abstract

Calculating emissions and related environmental impacts for buildings is typically a data-heavy and labour-intensive task. Widely used life cycle assessment (LCA) standards require meticulous modelling of multiple processes for each part within a product or a subassembly. This level of detailing demands time-consuming manual modelling and essentially renders full LCA of entire city blocks unrealistic. Within this context, this paper investigates how LCA results of modelling processes which involve a range of automated input data sources compare to those resulting from a highly detailed base case model. Findings show that models generated from data gathered from Google Street View and the U.S. Census produce the closest results to the base case model, with the lowest deviations occurring in embodied energy (0.06−6.0%) and global warming potential (0.7−4.8%) results. These findings imply that data with lower granularity can lead to precise LCA results, depending on the inventory and impact categories considered.