Shifting perspectives: A comparison of travel-time-based and carbon-based accessibility landscapes

Undoubtedly, climate change and its mitigation have emerged as main topics in public discourse. While accessibility planning is recognized for supporting sustainable urban and transport development in general, the specific challenge of reducing transportrelated greenhouse gas emissions has rarely been directly addressed. Traditionally, accessibility is operationalized in line with the user perception of the transport system. Travel-time-based measures are considered to be closely linked with travel behavior theory, whereas CO2 emissions are not necessarily a major determinant of travel decisions. Given the changed prioritization of objectives, additional emphasis should be placed on the environmental costs of travel rather than solely the user costs. Accessibility analysis could account for this shift in perspectives by using CO2 emissions instead of travel time in the underlying cost function. While losing predictive power in terms of travel behavior compared to other implementations of accessibility, carbon-based accessibility analysis enables a normative understanding of travel behavior as it ought to be. An application in the Munich region visualizes the differences between travel-time-based and carbonbased accessibility by location, transport mode, and specification of the accessibility measure. The emerging accessibility landscapes illustrate the ability of carbon-based accessibility analysis to provide new insights into land use and transport systems from a different perspective. Based on this exercise, several use cases in the context of low-carbon mobility planning are discussed and pathways to further develop and test the method in cooperation with decision-makers are outlined. 346 JOURNAL OF TRANSPORT AND LAND USE 14.1 ceeded in decreasing emissions (EEA, 2019; US EPA, 2019). Environmental objectives seem to be in conflict with the social and economic benefits linked to mobility (Banister, 2011). Accessibility, determined by the joint characteristics of the land use and transport systems, could be a suitable concept to address this challenge. The first reason for employing accessibility to plan for low carbon mobility options is its intrinsic capability to integrate land use and transport planning. Dense and mixed use urban development can contribute to the goal of greenhouse gas emission reductions, especially if oriented towards public transport systems (Banister, 2011; Schwanen, Banister, & Anable, 2011). Thus, consideration of land use configurations and policies is indispensable in promoting sustainable transport (Loo & Tsoi, 2018). Increased vehicle efficiency will not solve the issue of transport-related emissions if separation of urban functions, suburbanization, and car dependence prevail (Chapman, 2007). Multimodal mobility behavior, increasingly enabled by innovative mobility services, will not suffice if the level of travel activity, in particular trip distance, continues to grow (Heinen & Mattioli, 2019). Through the introduction of a land use dimension, accessibility helps to distinguish between the need to reach opportunities as an end and the need to travel as a means. Tackling climate change requires efforts on multiple scales, from local to global (Marsden, Ferreira, Bache, Flinders, & Bartle, 2014; Ostrom, 2010), and depends on the involvement and interaction of multiple actors (Geels, 2012). In order to reduce transport-related emissions, not only the impacts of potential interventions and policies need to be assessed, but also the issue of implementation needs to be addressed (Lewis, Zako, Biddle, & Isbell, 2018). Even if political decision-makers have succeeded in defining a suitable framework for low carbon mobility, public awareness, acceptance, and commitment are equally important (Banister, 2008). Against this background, the second reason for exploring accessibility-based planning approaches to address climate change is their ability to support decision-making on multiple levels, both spatially and institutionally. Firstly, accessibility analysis is applicable on various geographical planning scales (Papa, Silva, Te Brömmelstroet, & Hull, 2016). Secondly, accessibility can – given an appropriate implementation – contribute to enhancing discussion and decision-making of stakeholders across different institutions, disciplines, and levels of expertise (Te Brömmelstroet, Curtis, Larsson, & Milakis, 2016; Wulfhorst, Büttner, & Ji, 2017). Complex tools might be needed for further in-depth analysis, but simpler tools, featuring high transparency and communication value, are most suitable to explore alternative scenarios in strategic planning (Ford, Dawson, Blythe, & Barr, 2018; Te Brömmelstroet, 2010). While the underlying accessibility metrics might be of varying complexity, accessibility instruments often produce visual outputs in map format (Papa et al., 2016), which tend to improve understandability and communicability (Büttner, Ji, & Wulfhorst, 2019; Curtis & Scheurer, 2010). This potential is not yet fully exploited, since emissions are seldom explicitly considered in accessibility analysis and planning. Environmental objectives are often addressed indirectly, for example when trying to minimize the gap between accessibility by car and accessibility by other modes that are considered to be more sustainable (Salonen & Toivonen, 2013). Accessibility measures in these and many other applications are based on the uses and perceptions of the people. Consequently, travel costs are operationalized as internal user costs, typically represented by travel time (Cui & Levinson, 2018). In contrast, emissions are not necessarily a major determinant of individual travel decisions, but represent a normative, politically defined constraint to travel activities. The plea for reinventing seemingly invariable concepts, thus enabling new rather than habitual ways of thinking in the context of climate change (Schwanen, 2019), might as well be transferred to the accessibility concept. In this paper, we propose an alternative conceptualization of accessibility, dominated by an environmental perspective instead of a user perspective. More precisely, travel time is replaced by CO2 emissions as the relevant travel cost. Mul347 Shifting perspectives: A comparison of travel-time-based and carbon-based accessibility landscapes tiple studies have compared accessibility implementations based on different cost components (Büttner, 2017; Cui & Levinson, 2018; El-Geneidy et al., 2016), impedance functions (Higgins, 2019; Vale & Pereira, 2016), behavioral foundations (Páez, Scott, & Morency, 2012) or indicator types (Kwan, 1998). However, to the best of our knowledge, the partially conflicting perspectives of the user and the environment have never been directly compared. In order to determine whether this reinvention provides new insights compared to traditional implementations, both approaches are compared and contrasted using the Munich region as a case study. A review of the theoretical considerations underlying the accessibility concept and its operationalization follows in section 2, the presentation of the implementations in section 3, and a discussion of the application potential of carbon-based accessibility in section 4. Conclusions and future research paths are outlined in section 5. 2 Perspectives on accessibility In this paper, accessibility is defined as the number of opportunities within acceptable reach of a given place (Te Brömmelstroet, Curtis, Larsson, & Milakis, 2016), where acceptable could refer to either a user perspective (section 2.1) or an environmental perspective (section 2.2). The objective of the analysis determines the relevant perspective as well as the appropriate operationalization of accessibility. 2.1 The user perspective The user perspective is centered on how (potential) travelers experience accessibility. There are different manifestations of this viewpoint in how accessibility is conceptualized and measured. In fact, it can be related to all four components of accessibility, as defined by Geurs and Van Wee (2004): the land use, transportation, temporal, and individual component. One central aspect of the land use component is the spatial distribution of destinations, representing relevant activities or opportunities (Handy & Niemeier, 1997; Páez et al., 2012). Different types of opportunities can be analyzed, most of which are assumed to provide some benefit to individuals (e.g., job opportunities). Destination potentials can be weighted by their attractiveness or value for the user and classified according to their characteristics, which make them particularly relevant (or irrelevant) for a specific group of travelers. Much-cited papers describe the transportation component as determining the effort (Geurs & Van Wee, 2004) or ease (Handy & Niemeier, 1997) of traveling for an individual. Consequently, travel costs in accessibility measures are often purely internal, typically measured in travel time or generalized costs, as experienced by the traveler (Cheng & Bertolini, 2013). Such implementations are useful to analyze the attractiveness and affordability of different transport modes for the user (El-Geneidy et al., 2016). A person’s range of accessible opportunities might be reduced due to the limited time available in between activities that are fixed in space and time. The temporal component of accessibility represents these individual spatial-temporal constraints (Geurs & Van Wee, 2004). Handy and Niemeier (1997) criticize loose consideration of the user perspective, arguing that accessibility measures should be determined by the uses and perceptions of the travelers, rather than the assumptions of the analyst. This comes along with a need for more disaggregate measures, focusing on (groups of) individuals as the unit of analysis. The individual component of accessibility acknowledges that persons have different characteristics, capabilities, and preferences (Geu