Purpose

Rarely considered in environmental assessment methods, potential land use impacts on a series of ecosystem services must be accounted for in widely used decision-making tools such as life cycle assessment (LCA). The main goal of this study is to provide an operational life cycle impact assessment characterization method that addresses land use impacts at a global scale by developing spatially differentiated characterization factors (CFs) and assessing the extent of their spatial variability using different regionalization levels.

Methods

The proposed method follows the recommendations of previous work and falls within the framework and principles for land use impact assessment established by the United Nations Environment Programme/Society of Environmental Toxicology and Chemistry Life Cycle Initiative. Based on the spatial approach suggested by Saad et al. (Int J Life Cycle Assess 16: 198–211, 2011), the intended impact pathways that are modeled pertain to impacts on ecosystem services damage potential and focus on three major ecosystem services: (1) erosion regulation potential, (2) freshwater regulation potential, and (3) water purification potential. Spatially-differentiated CFs were calculated for each biogeographic region of all three regionalization scale (Holdridge life regions, Holdridge life zones, and terrestrial biomes) along with a nonspatial world average level. In addition, seven land use types were assessed considering both land occupation and land transformation interventions.

Results and discussion

A comprehensive analysis of the results indicates that, when compared to all resolution schemes, the world generic averaged CF can deviate for various ecosystem types. In the case of groundwater recharge potential impacts, this range varied up to factors of 7, 4.7, and 3 when using the Holdridge life zones, the Holdridge regions, and the terrestrial biomes regionalization levels, respectively. This validates the importance of introducing a regionalized assessment and highlights how a finer scale increases the level of detail and consequently the discriminating power across several biogeographic regions, which could not have been captured using a coarser scale. In practice, the implementation of such regionalized CFs suggests that an LCA practitioner must identify the ecosystem in which land occupation or transformation activities occur in addition to the traditional inventory data required—namely, the land use activity and the inventory flow.

Conclusions

The variability of CFs across all three regionalization levels provides an indication of the uncertainty linked to nonspatial CFs. Among other assumptions and value choices made throughout the study, the use of ecological borders over political boundaries was deemed more relevant to the interpretation of environmental issues related to specific functional ecosystem behaviors.