Philine Gaffron, TUHH - Hamburg University of Technology


The paper compares the results of an environmental justice analysis of road transport emisssion using different spatial units as a basis for defining affected sub-groups of the population.


European directives stipulate regular air quality (e.g. PM10, NO2) and noise assessments in urban areas and for roads that exceed certain size or traffic volumes respectively (cf. 2008/50/EC on air quality and 2002/49/EC on noise). If set emission limits are violated, air quality or noise abatement plans must be drawn up and implemented.

However, in addition to obtaining an overall picture of where emissions from (road) transport are problematic from a human health perspective, it is also of great interest to find out who is actually affected by such emissions. Different socio-economic sub-groups are differently able to cope with the health risks and effects associated with high emission burdens and it has been shown that often, more vulnerable groups bear higher loads – in addition to being disadvantaged in other ways. It is thus both interesting and relevant to analyse emissions from an environmental justice perspective and differentiate levels of affectedness between different population groups.

The findings obtained from such analyses can help to prioritize abatement measures and address issues of environmental inequities, which along with being highly problematic from a social justice perspective also have very real public health and thus social cost implications.

Environmental justice (EJ) analyses often focus on residential populations and use statistical units such as census areas or administrative boundaries as a basis for distinguishing differently composed sub-populations. It has already been shown that the results from such analyses can differ – sometimes substantially – depending on the spatial units that they are based on. This paper will show that even when the smallest spatial units are chosen for which population data is available – in this case US census blocks, which in urban areas often cover just one city block – additional information on the distribution of land uses within these blocks can still serve to distinguish more accurately how different population groups are differently affected. Urban road transport emissions as characterized by PM2.5 are the focus of this analysis. Emissions modelling is based on road traffic volumes as shown by a regional travel demand model (SACSIM) and an emissions inventory (here: Californian EMFAC 2011) and the EJ analysis takes into account only those parts of the population who are found within a certain distance from the road network. This approach avoids the assumption that anyone living beyond this distance is not affected at all by transport emissions. The population of each census block is reallocated to land use parcels (using a dasymetric mapping approach. This helps ensure that no population is allocated to land parcels within a block that do not actually have any residential use. Results from this analysis will be contrasted with what the findings would be if an even distribution of the population throughout each census block was assumed.

This paper is based on research carried out at the University of California Davis that was funded by the German Research Foundation (Deutsche Forschungsgemeinschaft). The spatial focus thus lies on the region around the Californian capital Sacramento but the differences in EJ analysis results that arise from including land-use information can occur in any other study area. And if decisions on emission mitigation measures take into account the results of such analyses, which it is argued they should, then the highest practicable level of precision must be desirable.


Association for European Transport