Enhancing life cycle impact assessment from climate science: Review of recent findings and recommendations for application to LCA

Abstract

Since the Global Warming Potential (GWP) was first presented in the Intergovernmental Panel on Climate Change (IPCC) First Assessment Report, the metric has been scrutinized and alternative metrics have been suggested. The IPCC Fifth Assessment Report gives a scientific assessment of the main recent findings from climate metrics research and provides the most up-to-date values for a subset of metrics and time horizons. The objectives of this paper are to perform a systematic review of available midpoint metrics (i.e. using an indicator situated in the middle of the cause-effect chain from emissions to climate change) for well-mixed greenhouse gases and near-term climate forcers based on the current literature, to provide recommendations for the development and use of characterization factors for climate change in life cycle assessment (LCA), and to identify research needs. This work is part of the ‘Global Guidance on Environmental Life Cycle Impact Assessment’ project held by the UNEP/SETAC Life Cycle Initiative and is intended to support a consensus finding workshop. In an LCA context, it can make sense to use several complementary metrics that serve different purposes, and from there get an understanding about the robustness of the LCA study to different perspectives and metrics. We propose a step-by-step approach to test the sensitivity of LCA results to different modelling choices and provide recommendations for specific issues such as the consideration of climate-carbon feedbacks and the inclusion of pollutants with cooling effects (negative metric values).

Introduction

Life cycle assessment (LCA) is a decision support tool that estimates the potential environmental impacts of any product system over its entire life cycle. It is commonly used to guide environmental policies and programs, to inform consumers’ choices through environmental labeling and declarations, and to help industries reduce the environmental impact of their activities or design more sustainable products, amongst others (ISO 14044, 2006).

The first step in an LCA – after defining the goal and scope – is to develop an inventory of all environmental emissions from, and natural resource inputs to, each unit process in the system. The total environmental inputs and outputs from all activities are called the life cycle inventory (LCI). In life cycle impact assessment (LCIA), these environmental flows are classified according to the type of environmental impact they cause, and multiplied by characterization factors (CF) that express their contribution to that indicator. CFs are developed using environmental models that estimate the relative or absolute effect of each flow on a selected indicator, which is a quantifiable representation of an impact category. LCA practitioners usually select a specific LCIA method that proposes a series of CFs for different types of environmental impact (ISO 14044, 2006).

Emissions of CO₂ and other greenhouse gases (GHGs), aerosols, and ozone precursors are affecting the climate system as illustrated by the cause-effect chain presented in Fig. 1. In current LCIA methods, CFs for the climate change impact category are usually proposed only for well-mixed greenhouse gases (WMGHG), using Global Warming Potential (GWP) values published in the Intergovernmental Panel on Climate Change (IPCC) assessment reports. Other anthropogenic causes of global warming such as near-term climate forcers (NTCF) or albedo changes are currently not considered in LCA (Levasseur, 2015). The important difference between WMGHGs and NTCFs is their lifetime. WMGHGs have atmospheric lifetimes long enough to be well mixed throughout the troposphere, and their climatic impact does not depend on the location of emissions. WMGHGs include CO₂, N₂O, CH₄, SF₆ and many halogenated species. By contrast, NTCFs have atmospheric lifetimes of less than one year so that their climatic impact depends on the emission location. NTCFs include ozone and aerosols, or their precursors, and some halogenated species that are not WMGHGs (Myhre et al., 2013).

Researchers have shown that cumulative emissions of WMGHG with a lifetime greater than 50–100 years dominate the peak warming (Smith et al., 2012). However, reducing emissions of NTCFs and WMGHGs with shorter lifetimes could reduce the rate of climate warming over the next few decades and, if emission reductions are sustained, also lower the peak temperature attained (Myhre et al., 2011, Penner et al., 2010, Rogelj et al., 2014, Shindell et al., 2012, Smith et al., 2012). If net CO₂ emissions do not decline significantly and eventually reach zero, mitigation of short-lived species will only postpone but not avoid the breaching of a temperature threshold in line with those adopted within the UNFCCC process (Allen et al., 2016, Bowerman et al., 2013).

There are two different types of CFs depending on the position of the selected indicator in the cause-effect chain (see Fig. 1). Midpoint CFs refer to effects at an earlier stage of the cause-effect chain such as radiative forcing or temperature, while endpoint CFs are derived from relatively more complex mechanisms (with increased uncertainties) for translating emissions into impacts on human health (e.g. disability-adjusted life years caused by climate change) and ecosystems (e.g. potential disappeared fraction of species because of climate change) (Levasseur, 2015). All current LCIA methods offer midpoint CFs using GWPs published by the IPCC. The only distinctions between LCIA methods on this matter are the choice of time horizon and the issue year of the IPCC Assessment Report. Most LCIA methods use a 100-year time horizon to be in line with the time horizon selected for the application in the 1997 Kyoto Protocol, while a very few others use a 20- or 500-year time horizon (Levasseur, 2015). For instance, the ReCiPe method uses time horizons of 20, 100 and 500 years respectively for the individualist, hierarchist and egalitarian perspectives (Goedkoop et al., 2013). Users must choose between one of these perspectives to set the default value for some modeling choices.

Since the GWP was first presented in the IPCC First Assessment Report, the metric has been scrutinized and alternative metrics have been suggested. GWP was intended to clarify the relative contributions to global warming of different countries and different activities to help develop cost-effective emission policies at both national and international levels (Lashof and Ahuja, 1990). However, Shine (2009) reminds us that the GWP concept was initially a simple approach adopted in part to illustrate the difficulties encountered when developing a single metric to assess climate impacts associated with GHG emissions of gases with very different physical and chemical properties. There exists a plethora of other metrics based on physical and biogeochemical aspects of climate change (e.g. Gillet and Matthews, 2010, Lauder et al., 2013, Peters et al., 2011a, Shine et al., 2005, Shine et al., 2015, Smith et al., 2012, Sterner et al., 2014, Tanaka et al., 2009, Wigley, 1998), and a large range of metrics where aspects of economics are also taken into account (e.g. Eckaus, 1992, Johansson, 2012, Manne and Richels, 2001, Rilley and Richards, 1993). In recent years, the issue of metrics has received increased political attention and several publications have addressed concerns regarding the use of appropriate climate metrics in an LCA context (e.g. Peters et al., 2011b, UNFCCC, 2012, UNFCCC, 2014). The IPCC Fifth Assessment Report (5thAR) gives a scientific assessment of the main recent findings from physical climate metrics research and provides the most up-to-date values for a subset of metrics and time horizons (GWP and Global Temperature change Potential (GTP); see below). Crucially, the latest IPCC assessment emphasises that the choice of emission metric and time horizon depends on type of application and policy context and no single metric is optimal for all policy goals (IPCC, 2014a).

The objectives of this paper are to perform a systematic review of available midpoint metrics for WMGHGs and NTCFs based on the current literature, to provide recommendations for the development and use of climate change CFs in LCA, and to identify research needs. We primarily discuss research findings on metrics presented in the IPCC 5thAR, which emphasized GWP and GTP, and under which circumstances these metrics could be applied to improve current climate change midpoint characterization factors in LCA. This work is part of the ‘Global Guidance on Environmental Life Cycle Impact Assessment’ project held by the UNEP/SETAC Life Cycle Initiative and is intended to support a consensus finding workshop.