In our rapidly urbanizing world, sustainable transportation presents a major challenge. Transportation decisions have considerable direct impacts on urban society, both positive and negative, for example through changes in transit times and economic productivity, urban connectivity, tailpipe emissions and attendant air quality concerns, traffic accidents, and noise pollution. Much research has been dedicated to quantifying these direct impacts for various transportation modes. Transportation planning decisions also result in a variety of indirect environmental and human health impacts, a portion of which can accrue outside of the transit service area and so outside of the local decision-making process. Integrated modeling of direct and indirect impacts over the life cycle of different transportation modes provides decision support that is more comprehensive and less prone to triggering unintended consequences than a sole focus on direct tailpipe emissions.

The recent work of Chester et al (2013) in this journal makes important contributions to this research by examining the environmental implications of introducing bus rapid transit and light rail in Los Angeles using life cycle assessment (LCA). Transport in the LA region is dominated by automobile trips, and the authors show that potential shifts to either bus or train modes would reduce energy use and emissions of criteria air pollutants, on an average passenger mile travelled basis. This work compares not just the use of each vehicle, but also upstream impacts from its manufacturing and maintenance, as well as the construction and maintenance of the entire infrastructure required for each mode. Previous work by the lead author (Chester and Horvath 2009), has shown that these non-operational sources and largely non-local can dominate life cycle impacts from transportation, again on an average (or attributional) basis, for example increasing rail-related GHG emissions by >150% over just operational emissions.

While average results are valuable in comparing transport modes generally, they are less representative of local planning decisions, where the focus is on understanding the consequences of new infrastructure and how it might affect traffic, community impacts, and environmental aspects going forward. Chester et al (2013) also present their results using consequential LCA, which provides more detailed insights about the marginal effects of the specific rapid bus and light rail lines under study. The trade-offs between the additional resources required to install the public transit infrastructure (the "resource debt") and the environmental advantages during the operation of these modes can be considered explicitly in terms of environmental impact payback periods, which vary with the type of environmental impact being considered. For example, bus rapid transit incurs a relatively small carbon debt associated with the GHG emissions of manufacturing new buses and installing transit infrastructure and pays this debt off almost immediately, while it takes half a century for the light rail line to pay off the "smog debt" of its required infrastructure. This payback period approach, ubiquitous in life cycle costing, has been useful for communicating the magnitude of unintended environmental consequences from other resource and land management decisions, e.g., the release of soil carbon from land conversion to bioenergy crops (Fargione et al 2008), and will likely grow in prevalence as consequential LCA is used for decision support.

The locations of projected emissions is just as important to decision-making as their magnitudes, as policy-making bodies seek to understand effects in their jurisdictions; however, life cycle impact assessment methods typically aggregate results by impact category rather than by source or sink location. Chester et al (2013) address this issue by providing both local (within Los Angeles) and total emissions results, with accompanying local-only payback periods. Much more challenging is the geographic mapping of impacts that these emissions will cause, given the many point and mobile sources of air pollutants over the entire transportation life cycle. Integration of LCA with high-resolution data sets is an active area of model development (Mutel and Hellweg 2009) and will provide site- and population-specific information for impacts ranging from water quality to biodiversity to human respiratory health.

Another complex challenge in modelling environmental impacts of transportation (and cities in general) is the long run, interdependent relationship between transportation technologies and urban form. LCA modelling has tended to assume a fixed pattern of settlements and demand for mobility and then examined changes to a particular technology or practice within the transportation system, such as electric or hybrid vehicles or improved pavement materials. New transit options or other travel demand management strategies might induce mode switching or reduced trips, but the overall pattern of where people live and work is generally assumed in these models to be constant in the short run. In contrast, the automobile has been influencing land-use patterns for a century, and it is the resulting geographic structure that determines the baseline need for transportation, and thus drives the use of material and energy resources used in transportation systems (Kunstler 1994).

We have seen that cities with high population densities tend to have lower tailpipe emissions from transportation (Kennedy et al 2009). Recent studies have modelled how changes in urban land-use or zoning changes the geographic structure of transportation demand and then used LCA to determine the environmental benefits of such policies. For example, Mashayekh et al (2012) summarized travel demand reductions projected from several studies of compact, smart growth, and brownfield in-fill development strategies to find benefits ranging up to 75% reductions in life cycle GHG and air pollutant emissions. A related study in Toronto on life cycle energy use and GHG emissions for high- and low-density development strategies found a ∼60% difference in GHG emissions, largely due to transportation (Norman et al 2006). System dynamics and agent-based models may complement LCA in capturing long-term effects of transportation strategies as they are inherently dynamic (Stepp et al 2009), and can internalize spatially resolved decisions about where to settle and work (Waddell 2002).

Transportation planning decisions have both direct and indirect, spatially distributed, often long-term effects on our health and our environment. The accompanying work by Chester et al (2013) provides a well-documented case study that highlights the potential of LCA as a rich source of decision support.

For references see Life cycle assessment in support of sustainable transportation at environmentalresearchweb's sister website ERL.

Related link