Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

1 Introduction

Climate action in the urban transport sector presents a unique and daunting challenge. The transport sector accounted for 23% of global greenhouse gas (GHG) emissions in 2010 and remains one of the fastest growing sources of global emissions, despite advances in vehicle efficiency (Sims et al. 2014). In contrast with most other major sources of global emissions, fossil fuels remain the dominant final energy source in transport, with oil accounting for over 90% of the final energy demand (IEA 2016). The lack of progress with electricity decarbonization limits the prospects of potential emission, climate and air quality benefits of electric vehicles (UK Energy Research Centre 2016), and one of the key components of the decarbonization of vehicle travel to date—the shift to diesel powertrain—is now under significant scrutiny (UK Energy Research Centre 2016; Cames and Helmers 2013). Finally, and potentially of greatest importance, established transport networks and systems are costly, technically challenging and time-consuming to change once they are built, leaving current design and planning trends difficult to alter (Khreis et al. 2016). Transport investments over the coming 5 years will therefore substantially dictate the pathway of transport-related emissions for decades to come, adding more pressure for low-carbon transport options to be developed and implemented in the immediate future (Leather 2009; IEA 2015).

At the same time, the benefits of action are also substantial. The impact of climate change on cities is being felt. Rising seas, heat waves, air pollution and weather disasters made more violent by changing weather patterns are already costing millions of lives and billions in livelihoods each year (Harlan and Ruddell 2011). At the same time, research establishes that an economic case exists for many urban transport options that can mitigate climate change. Investments in a range of currently available public and private measures in the world’s urban areas could save 2.8 gigatonnes of GHG emissions annually by 2050—just less than the entire GHG emissions of India in 2011—or 7% of the 2011 global GHG emissions (WRI 2012), whilst providing direct net economic benefits between 2015 and 2050 of $10.5 trillion USD (Gouldson et al. 2015; Sudmant et al. 2016).

Perhaps even more compellingly, urban transport investments to mitigate climate change can have a substantial positive impact on public health in cities. In 2015, the Lancet Commission on Health and Climate Change emphasized that the response to climate change could be “the greatest global health opportunity of the 21st century” (Watts et al. 2015, p. 1). The health co-benefits of climate action in the transport sector can be realized through pathways of improved indoor and outdoor air quality , reduced exposures to noise, mitigation of the urban heat island effect, increased active travel and physical activity, reduced motor vehicle crashes, increased green space exposure and reduced social exclusion and inequalities. Up to 74% of air pollution in cities worldwide, and as much as 90% of urban air pollution in some cities of the developing world, may be attributed to motor vehicle emissions (Arup/C40 2014; UNEP 2014). Traffic-related air pollution is responsible for a large burden of global disease and at least 184,000 deaths in 2010 (Bhalla et al. 2014; Khreis et al. 2016). Transport is also responsible for much of the noise in urban areas (Foraster et al. 2011; Bell and Galatioto 2013; Zuo et al. 2014) with traffic density, distance from the location to the sidewalk and building density explaining over 73% of the variability of noise levels (Foraster et al. 2011). Lee et al. (2014) found that the total number of vehicle counts explained a substantial amount of variation in measured ambient noise in Atlanta (78%), Los Angeles (58%) and New York City (62%). The burden of disease attributable to traffic-related noise is comparable to that of air pollution (Hänninen et al. 2014; Tainio 2015; Mueller et al. 2017a, b). Over the last decades, declining active travel and increased vehicle use have been important contributors to overall declines in physical activity and increases in sedentary behaviour (Brownson et al. 2005; Ewing et al. 2003). According to the World Health Organization (2017), one in four adults and over 80% of the world’s adolescents are insufficiently physically active. Motor vehicle crashes cause over 1.5 million global deaths annually and 79.6 million healthy years of life lost (Bhalla et al. 2014). Indeed, after the first year of life through age 59, motor vehicle crashes rank amongst the top ten causes of global deaths (Bhalla et al. 2014) and are the number one global cause of death amongst those aged 15–29 years (World Health Organization 2015). Further deleterious impacts from the transport sector that have been linked with public health outcomes include exposure to local temperature increases (the so-called heat island effect), green space reduction and biodiversity loss (Khreis et al. 2016). The cost of these transport externalities is very high, especially in rapidly developing urban areas. For example, the costs of motorized transport’s congestion, air pollution, motor vehicle crashes, noise and climate change in Beijing are between 7.5% and 15.0% of its GDP (Creutzig and He 2009). Rapid technological changes and volatility in energy prices, which radically alter the economic case for investments, make predictions and scenario comparisons challenging. Nonetheless, and as we will show next, there is great opportunity for policymakers to develop transport roadmaps that jointly achieve climate change, health and economic objectives.

In this chapter, we provide a state-of-the-art review of the co-benefits of climate action on health at the urban level focusing on five urban transport actions: (1) compact land use planning to reduce motorized passenger travel demand, (2) passenger modal shift and improving transit efficiency, (3) electrification and passenger vehicle efficiency, (4) freight logistics and (5) freight vehicle efficiency and electrification. Health impacts from these actions occur via pathways of reduced air pollution, noise and temperature and increased green space and physical activity. We argue that presenting a more robust health case of low carbon action by assessing the co-benefits of these policy measures may unlock policy support and accelerate action.

2 Co-benefits of Climate Action on Health

2.1 Urban Planning and Reduced Passenger Travel Demand

2.1.1 Air Pollution

Urban planning to reduce passenger travel demand is a key measure that cities can adopt to improve air quality and public health and achieve significant economic savings (Ling-Yun and Lu-Yi 2016; Bartholomew 2007; Stone et al. 2007; Reisi et al. 2016; Guttikunda and Mohan 2014; Frumkin 2002; Grabow et al. 2012; Giles-Corti et al. 2016; Conlan et al. 2016; Stevenson et al. 2016). Designing cities to be compact and mixed-use can lead to shorter distances and easier access to work, school, and other activities and therefore reduce the need for passenger car travel (Guttikunda and Mohan 2014). Reductions in passenger car travel demand are often also accompanied by modal shifts towards more sustainable transport means, such as walking, cycling and the use of public transport. Conversely, the rapid expansion of metropolitan areas, or urban sprawl, and the resulting un-mixed land use and low-density development patterns reinforce the need and convenience for extensive road networks and private car travel (Frumkin 2002). The literature is supportive of this narrative; in the following key articles are described.

Ewing et al. (2008a, b) found that high density can reduce vehicle kilometres by 40% per capita and comparisons of urban centres showed that dense, highly connected urban centres like Hong Kong produce only 1/3 of the carbon emissions per capita of European cities, whilst European cities produce only 1/5 the carbon emissions of sprawling poorly connected cities like Houston (Rode et al. 2013). Recent reviews and large-scale health impact assessments concluded that urban planning measures, unlike other transport policy instruments aimed at reducing traffic-related air pollution (e.g. freight management), have the potential to realise air quality improvements over the longer term and provide additional benefits related to relieving congestion, improving the quality of places and increasing population physical activity levels, all of which associated with many health and wellbeing benefits (Conlan et al. 2016; Giles-Corti et al. 2016; Stevenson et al. 2016).

Many studies have provided quantification of the expected air quality and health benefits from such measures, although these were almost exclusively based on health impacts assessment modelling exercises. For example, reducing the kilometres travelled by Chinese residents by 5% and 10% via increasing cycling would lead to around 1.56% and 3.11% decrease in annual average concentrations of SO2, respectively, and up to 2.80%, 6.18% and 5.86% decrease in NO2, PM2.5 and PM10, respectively. If these reductions in demand were accompanied with a shift towards public transit, the estimated benefits were higher. The number of associated preventable deaths from air pollution-related disease per year was estimated to range from 569,000 to 4,516,000, depending on the scenario being tested. The estimated health improvements including the reduction in total mortality, cardiovascular and respiratory hospital admissions and asthma attacks would save 3433.25 to 27,337.1027 billion Yuan (Ling-Yun and Lu-Yi 2016).

Grabow et al. (2012) suggested that the elimination of automobile round-trips ≤8 km in 11 metropolitan areas in the upper Midwestern United States would reduce PM2.5 by 0.1 μg/m3 and although summer ozone (O3) would slightly increase in cities, it would decline regionally, resulting in net health benefits of $4.94 billion/year (95% confidence interval (CI), $0.2 billion, $13.5 billion). If 50% of the eliminated trips were made by bicycle, the health benefits would increase significantly due to increased levels of physical activity reducing mortality by 1295 deaths/year (95% CI: 912–1636). The combined health benefits of improved air quality and increased physical activity were estimated to exceed $8 billion/year (Grabow et al. 2012).

Reisi et al. (2016) selected multiple sustainability indicators including vehicle emissions and mortality effects of air pollution and evaluated these under three urban planning scenarios: a base case scenario based on governmental plans in Melbourne for 2030, activity centres scenario based on compact urban development patterns and a fringe focus scenario based on expansive urban development patterns. The activity centres scenario resulted in the least GHG and other emissions, as well as a reduction of mortality when compared to the other two scenarios.

In a health impact assessment of six cities, land use changes were modelled to reflect a compact city in which land use density and diversity were increased and distances to public transport were reduced to drive a modal shift from private vehicles to walking, cycling and public transport. The modelled compact city scenario resulted in health benefits for all cities (for diabetes, cardiovascular disease and respiratory disease) with overall health gains of 420–826 disability-adjusted life years (DALY) per 100,000 people. However, for moderate to highly motorized cities, such as Melbourne, London and Boston, the compact city scenario predicted a small increase in road trauma for cyclists and pedestrians (health loss of between 34 and 41 DALYs per 100,000 people) (Stevenson et al. 2016).

Overall, these studies demonstrate that compact and mixed-use urban planning which can reduce passenger travel demand will improve air quality and public health in urban areas.

2.1.2 Noise

Whilst relatively less research has focused on the effect of urban planning and reduced passenger travel demand on noise and human health, impacts are increasingly recognized. Creutzig et al. (2012) provided scenarios of increasingly ambitious policy packages, reducing GHG emissions from urban transport by up to 80% from 2010 to 2040. Based on stakeholder interviews and data analysis, the main target was a modal shift from motorized individual transport to public transit and nonmotorized individual transport (walking and cycling) in the four European cities of Barcelona, Malmo, Sofia and Freiburg. The authors reported significant concurrent co-benefits of better air quality, reduced noise, less traffic-related injuries and deaths, increased physical activity, alongside less congestion and monetary fuel savings. For the most ambitious scenario explored which included multiple measures and a large shift towards nonmotorized individual transport (referred to as push scenario), a reduction between 10% and 29% in noise levels in cities was reported. The push scenario included congestion charging and aggressive land use policies restricting new development to car-free areas with bicycle infrastructure and public transport access, amongst others. The scenario included policies that were only considered by some stakeholders, and although these were mostly not under consideration by local authorities, they were judged to be plausible.

There is a clear need to fill large gaps in the literature on the effects of urban planning and reduced passenger travel demand on noise and associated health effects.

2.1.3 Physical Activity

Multiple climate policies targeting the transport sector have the potential to increase physical activity levels and provide significant health benefits, even after consideration of increased air pollution exposure and motor vehicle crashes (Mueller et al. 2015). Urban planning to reduce passenger travel demand and increase transit mode share has been identified as a key measure which cities can adopt to increase population physical activity and improve public health Footnote 1 (D’Haese et al. 2015; Ewing et al. 2014; Frank et al. 2005; Cohen et al. 2006; Sugiyama et al. 2012; Buehler and Pucher 2012; Wong et al. 2011; Committee on Environmental Health 2009; Saelens et al. 2003; Lee and Moudon 2004; Rodriguez et al. 2006; Salon 2016; Stevenson et al. 2016). The following describes the key articles we identified in the literature.

Sugiyama et al. (2012) systematically reviewed 46 quantitative studies examining the associations between walking, as an active travel mean (utilitarian walking), and multiple built environment factors including the presence and proximity of destinations and sidewalks, connectivity, aesthetics and traffic on and safety of the walking routes. Half of the studies came from North America, 11 from Australia, 8 from Europe, 3 from South America and 1 from Japan. The literature synthesized in this review consistently showed (80% of studies) that the presence and proximity of retail and service destinations are conducive to adults walking. Other factors including well-connected streets and the availability of sidewalks also facilitated active travel.

Additional reviews support the findings of Sugiyama et al. (2012). Day et al. (2006) focused their review on 42 empirical studies conducted in China and similarly showed that active travel was most strongly associated with the proximity of non-residential locations. The literature reviewed supported an association between land use mix (proximal non-residential locations) and active travel in Chinese cities. In a review of 65 studies of children from North America, Europe, Australia and Asia, D’Haese et al. (2015) found that walkability, density and accessibility were associated with active travel to school. Buehler and Pucher (2012) reviewed the literature on the advantages and disadvantages of higher urban densities and showed that numerous studies pinpointed the public health benefits stemming from more walkable and cycling-friendly environments alongside decreasing the number and distance of vehicle trips. Similarly, in a systematic review synthesizing 14 studies by Wong et al. (2011), the authors concluded that increasing distance is negatively associated with children’s active travel to schools. Lee and Moudon (2004) showed that long distances between destinations and poor accessibility to recreational facilities are the most cited barriers to physical activity, whilst Cohen et al. (2006) showed that every mile a girl lived further to her school translated into 13 fewer weekly minutes of metabolic activity and that the time spent commuting was responsible for this reduction.

Increased urban sprawl and further distances between origins and destinations were also associated with decreased physical activity and increased obesity rates in adults (Ewing et al. 2014; Frank et al. 2005). In a review synthesizing studies conducting neighbourhood comparisons of walkability and physical activity, Saelens et al. (2003) showed that residents of high-walkable neighbourhoods reported two times more walking trips/week as compared to residents of low-walkable neighbourhoods (3.1 vs. 1.4 trips). The difference magnitude between high-walkable and low-walkable neighbourhoods ranged from −0.1 to 5.7 walk trips and was partially dependent on the purpose of the trip, with walking to work being consistently more likely in high-walkable compared to low-walkable neighbourhoods . This difference translated into 1–2 km of active travel or about 15–30 min more walking per week. Similarly, Rodriguez et al. (2006) showed that residents of walkable urban neighbourhoods are making twice as many trips walking and cycling compared to residents of suburban neighbourhoods, logging 40–55 extra minutes of weekly physical activity. Like previous observations, these differences were mainly driven by utilitarian travel, rather than leisure travel. Residents of the walkable urban neighbourhoods also travelled fewer vehicle miles. Salon (2016) showed that pedestrian and cyclist road use in urban census tracts is double that in suburban census tracts, which in turn is an order of magnitude greater than that in rural census tracts.

Besides reducing passenger travel demand, urban planning that promotes compact and mixed land use patterns can also make transit more viable and efficient (Boyko and Cooper 2011) and contributes to increasing public transit demand (Buehler and Pucher 2012). When coordinated in packages of other complementary policies (e.g. attractive transit fares, high taxes, restrictions on car usage, etc.), compact cities can increase the likelihood to use public transport by five times (Buehler and Pucher 2012). American people who use public transit spend a median of 19 min a day walking to and from transit, and almost 30% of transit users achieve the daily recommended physical activity levels of ≥30 min (Besser and Dannenberg 2005). There is also evidence that people from ethnic minorities and lower socio-economic classes are more likely to harvest more health benefits of physical activity as these groups were more likely to spend ≥30 min daily commuting to and from transit (Besser and Dannenberg 2005).

Overall, there are plenty of studies and evidence reviews which demonstrate that compact and mixed-use urban planning can reduce passenger travel demand and promote active travel modes, increasing population physical activity and therefore improving public health in urban areas.

2.1.4 Motor Vehicle Crashes

Increased traffic flows are associated with increased motor vehicle crashes (Nakahara et al. 2011; Yiannakoulias and Scott 2013); therefore, urban planning measures to reduce traffic and vehicle kilometres travelled are likely to reduce road deaths and injuries (Ewing and Hamidi 2015). Further, more sustainable urban planning patterns accommodating walking, bicycling and transit-friendly neighbourhoods are expected to lead to reduced passenger car demand and use, which in turn has been associated with reduced motor vehicle crashes (Wei and Lovegrove 2012).

Research has shown that traffic volumes and road densities are the main determinants of motor vehicle crash frequency. Dumbaugh and Li (2010) estimated that each additional mile of thoroughfare is associated with a 9.3% increase in motor vehicle-pedestrian crashes. A reduction of 30% in traffic volumes is associated with a 35% reduction in the number of pedestrians injured in motor vehicle crashes and a 50% reduction in the average risk of a pedestrian collision (Miranda-Moreno et al. 2011). Lowering traffic flows was linked to a decline in child pedestrian deaths in New Zealand (Roberts et al. 1992), and when flow reductions were achieved as part of the London congestion charging, they were linked to a substantial reduction in both the number of crashes and crash rates (Green et al. 2016). The London congestion charging zone was associated with 44 less motor vehicle crashes per month which represented a 35% decline.

Density and ‘compactness’ are shown to be negatively correlated with motor vehicle crashes and fatalities. Ewing et al. (2003) found that all traffic fatalities and pedestrians’ fatalities decrease by 1.49% and 1.47%, respectively, with each 1% increase in a compactness index. Godwin and Price (2016) argued that the distinct pattern of low-density urban areas in Southeast USA is likely to be leading to rare and more dangerous walking and cycling resulting in decreased road safety. Yeo et al. (2015) conducted a path analysis to examine the causal linkages between urban sprawl, vehicle miles travelled and traffic fatalities, drawing on data from 147 urbanized areas in the USA. The authors, in line with previous research (Ewing et al. 2014), found that there was an indirect positive effect of urban sprawl on fatalities which occurred through increases in vehicle miles travelled, alongside a more influential direct effect possibly occurring through increased traffic speeds and emergency medical service delays.

However, literature also shows that urban planning interventions which are established to reduce passenger car demand may have different effects on road safety, perhaps mediated by the increase in active travel levels. In a systematic review of 85 quantitative studies investigating the effect of built environment on childhood walking and pedestrian injuries, Rothman et al. (2014) found that higher road density and higher traffic speeds increase injury incidence and severity. Land use mix and proximity to services and facilities were also found to increase injuries incidence and severity but were associated with increased walking. The authors discussed that these urban environment characteristics may not be inherently dangerous but rather be markers for increased exposure to traffic in general. Two urban measures were consistently found associated with increased walking and a reduction in child pedestrian injuries, namely, traffic calming and the diversity of land use including the presence of playgrounds, recreation, parks and open spaces. Further studies similarly showed that land use mix and proximity to retail, although favourably promoting walking and cycling, are associated with increased crash risks in children and adults (Cho et al. 2009; Elias and Shiftan 2014; Yu 2014; Lee et al. 2013), whilst the results from other studies were mixed (Blazquez et al. 2016). Policymakers therefore need to be cautious when applying urban planning measures and take into account contextual factors and the need for complementary measures to protect those shifting or choosing active travel modes.

2.2 Passenger Mode Shifts

2.2.1 Air Pollution

Passenger mode shifts towards more sustainable travel means and improving transit efficiency can also lead to significant public health improvements, not only through improvements in air quality but also importantly through the increase in physical activity as people walk, cycle and move to catch public transport (Nieuwenhuijsen and Khreis 2016; Pathak and Shukla 2016; Rojas-Rueda et al. 2012, 2013; Woodcock et al. 2013, 2009; Sabel et al. 2016; Xia et al. 2015; McKinley et al. 2005; Yang et al. 2016; Rabl and De Nazelle 2012; Mueller et al. 2017a, b). Some of the studies identified in the previous section also fall under this category but were kept in the section whose searches located them. Similarly, some of the studies we identify next on modal shifts and transit efficiency fall under the next category of vehicle efficiency measures and indicate health impacts through pathways other than air pollution (e.g. changes in physical activity).

Nieuwenhuijsen and Khreis (2016) evaluated the radical concept of car-free cities, a model primarily driven by the need to reduce GHG emissions but has impacts on public health. The authors suggested great benefits in terms of reduction in not only air pollution (up to a 40% reduction in NO2 levels on car-free days) but also noise and heat island effects and potential increases in green space and physical activity. Three health impact assessments cited in the review of Nieuwenhuijsen and Khreis estimated small air quality improvements and health benefits from the replacement of private car journeys by active or public transport (Rojas-Rueda et al. 2012, 2013; Woodcock et al. 2013). For example, 76 annual deaths and 127 cases of diabetes, 44 of cardiovascular diseases, 30 of dementia, 16 minor injuries, 0.14 major injuries, 11 of breast cancer, 3 of colon cancer, 7 of low birth weight and 6 of preterm birth can be prevented each year, if 40% of long-duration car trips were substituted by public transport and cycling. The largest health benefits estimated were in association with increased physical activity, and then with the reduced air pollution. The reductions in air pollution were not as large as was expected because the contribution of private cars was minor compared to trucks, buses and motorbikes, which were not included in the assessment.

Sabel et al. (2016) estimated that the introduction of a new metro in Thessaloniki, Greece, would reduce local deaths from air pollution by about 20%. Xia et al. (2015) estimated that shifting of 40% of vehicle kilometres travelled to alternative transport in Adelaide, South Australia, would reduce annual average PM2.5 by a small margin of 0.4 μg/m3, preventing 13 deaths a year and 118 DALYS. Woodcock et al. (2009) conducted a health impact assessment of alternative transport scenarios in London, UK, and Delhi, India. The use of lower-carbon-emission motor vehicles, increased active travel and a combination of the two scenarios were tested. The increase in active travel and less use of motor vehicles (i.e. modal shifts) had larger health benefits per million people (7332 DALYs in London and 12,516 in Delhi in 1 year) than the benefits from the increased use of lower-emission motor vehicles (160 DALYs in London and 1696 in Delhi). The combination of active travel and lower-emission motor vehicles resulted in the largest health benefits (7439 DALYs in London and 12,995 in Delhi). Most of the preventable premature deaths were estimated to result from increased physical activity, followed by the reduction of air pollution exposures.

McKinley et al. (2005) quantified cost and health benefits from a subset of air pollution control measures in Mexico City. The control measures tested were taxi fleet renovation, metro expansion and use of new hybrid buses replacing diesel buses. Avoided cases of 11 health outcomes, including premature mortality, chronic bronchitis, hospitalizations and emergency room visits for cardiovascular and respiratory disease, and minor restricted activity days (MRAD) were quantified in association to reductions in O3 and PM10. The measures were found to have air pollution reductions of approximately 1% for PM10 and 3% for O3. The associated health benefits were substantial, and their sum over the three measures was greater than the measures’ investment costs. For the individual scenarios, the benefit to cost ratio was 3.3 for the taxi renovation measure, 0.7 for the metro expansion measure and 1.3 for the new hybrid buses measure. The taxi fleet renovation was the most appealing option, but this was contextual because of the size and age of the taxi fleet in Mexico City.

Overall, these studies demonstrate that passenger modal shifts away from the private car and towards active travel modes and public transit will improve air quality and yield public health benefits in urban areas.

2.2.2 Noise

Private vehicles are one of the largest single sources of noise pollution in urban areas. Transport measures that shift transport demand away from motorized vehicles and towards public transport or nonmotorized options therefore present a major opportunity for impacts on noise pollution.

In their health impact assessment exercise, James et al. (2014) projected that a proposed fare increases and service cuts to public transport which would shift 48,600 people from public transport to driving in the Boston region would result in lost time due to congestion, increased air pollution, lower levels of physical activity, additional motor vehicle crashes, increased exposure to high noise levels and increased greenhouse gas emissions. In their scenario representing fare increases by up to 35% and service reductions affecting 53–64 million trips each year, an additional 2000 people on average were estimated to become exposed to over 60 dB of noise per day, which would be detrimental to health.

Sabel et al. (2016) modelled the impact of urban climate change mitigation transport measures in five European and two Chinese cities. The changes in exposure to noise due to the investigated transport measures were modelled in three of the seven cities, and the results suggested that promoting electric cars and reducing the use of personal cars had a very limited effect on reducing noise levels. In part, this was due to the high proportions of traffic on non-urban roads such as motorways which were not subjected to the investigated local traffic reduction policy measures (e.g. up to 35% of motorway traffic in Rotterdam). On the other hand, the introduction of a metro in Thessaloniki was predicted to more significantly reduce noise levels.

Rabl and De Nazelle (2012) presented an estimate of the potential health impacts due to a shift from car to cycling or walking, by evaluating the effects relating to changes in air pollution exposures and accident risk and citing costs for other impacts, namely, from noise to congestion. A driver who switches to cycling for a commute of 5 km (one way) 5 days/week for 46 weeks/year would experience health benefits from the reduction in noise which are worth about €1700 (at a cost of 0.76 h/km). The results for walking (2.5 km) were similar. Applying these estimates to a specific example of a policy measure in Paris, the Velib bike sharing system, the population benefits of the reduction in noise were estimated at €69.9 million/year.

Research of the effects of modal shift on noise pollution on cities, though limited in scope, presents a strong case that climate actions can yield substantial co-benefits.

2.2.3 Physical Activity

Promoting passenger modal shift from the private car towards active travel means, including walking and cycling, also has many well-documented health benefits through increased physical activity. These benefits were explored in multiple epidemiological studies, health impact assessments and systematic reviews (Lubans et al. 2011; Saunders et al. 2013; Ogilvie et al. 2004; Mueller et al. 2015; Flint et al. 2016; Matthews et al. 2007; Hu et al. 2007, 2003; Hamer and Chida 2009; Woodcock et al. 2009, 2013; Xia et al. 2015; Rojas-Rueda et al. 2012, 2013; Rabl and De Nazelle 2012; Stevenson et al. 2016).

One of most well regarded such studies followed 47,840 Finnish people, aged 25–64 years old, for an average of 19 years. The results showed that the risk of coronary heart disease was significantly decreased with increasing occupational, leisure time or active commuting-related physical activity (Hu et al. 2007). Daily commuting on foot or by bike was associated with a 20% reduction in risk of new coronary heart disease amongst women, after adjusting for a comprehensive set of confounders. Similarly, walking or cycling to work for ≥30 min/day was associated with a 36% reduction in risk of type 2 diabetes in both women and men, even after adjusting for a comprehensive set of confounders (Hu et al. 2003).

Flint et al. (2016) used longitudinal data for 5861 individuals from the UK Biobank to test the effect of transition from car to active (walking and cycling) or public transport on objectively measured body mass index. After adjusting for a comprehensive set of confounders including baseline body mass index, age, sex, ethnicity, income and education, a mode shift from the car to active or public transport was associated with a 0.30 kg/m3 decrease in body mass index. Conversely, a mode shift from active or public transport to car commuting was associated with a 0.32 kg/m3 increase in body mass index. These results were almost identical to a similar study using self-reported body mass index (Martin et al. 2015).

Finally, a recent health impact assessment evaluated the potential effects of a compact city model on physical activity levels in six cities: Melbourne, Australia; Boston, MA, USA; London, UK; Copenhagen, Denmark; São Paulo, Brazil; and Delhi, India (Stevenson et al. 2016). The compact city model tested resulted in increased active travel and public transport use which translated into increases in travel-related physical activity ranging from +72.1% metabolic equivalent per week (in the most motorized city; Melbourne) to +18.5% in the rapidly motorizing Delhi.

2.2.4 Motor Vehicle Crashes

Modal shifts can also reduce the number of vehicles on the road, thereby reducing the opportunities for motor vehicle crashes. At the same time, cycling and walking are more vulnerable transport modes. Research on increases in public and nonmotorized transit generally finds reductions in total crashes, but as we will show next, interventions need to be carefully planned to avoid unintended consequences and protect individuals switching to more vulnerable transport modes.

In a systematic review including 21 health impact assessment studies of the effects of passenger modal shift on motor vehicle crashes, Mueller et al. (2015) reported that 14 studies estimated an increase in motor vehicle crashes, 6 studies estimated a decrease in motor vehicle crashes and 1 study estimated no change in motor vehicle crashes with the increase in active travel. However, only eight studies accounted for the non-linear traffic incident risk attributed to the increased safety that comes with the increased number of active travellers, the so-called safety in numbers effect (Jacobsen 2003).

Jacobsen (2003) showed a consistent inverse relation between the amount of walking and cycling and the likelihood of a pedestrian and cyclist being involved in a motor vehicle crash. The author argued that a community doubling its walking can expect a 32% reduction in cycling injuries. Contrasting with this result, another review by Götschi et al. (2016), which considered studies modelling the safety impacts of cycling, showed that a modal shift from car travel to cycling is estimated to increase the number of cyclists’ fatalities and seriously injured. The authors, however, also showed that the public health benefits stemming from increases in physical activity well outweigh the estimated risks, for example, shifting a 10-km commute from car to bike will result in an average cost of €50 per person/year from fatal crashes but an average saving of €1300 per person/year from physical activity (Rabl and De Nazelle 2012). Götschi et al. (2016), in line with other investigators (Rabl and De Nazelle 2012), also concluded that impact modelling of cycling crash risks is a highly case- and context-specific matter. Furthermore, the increases in cyclists’ and pedestrians’ fatalities and injuries following a mode shift from the car might be a representation of a period of increased vulnerable road users’ collisions and injuries which will cease when the cycling mode split approaches the 20% level (Wei and Lovegrove 2012).

From a different angle, Tainio et al. (2014) estimated the years lived disabled or injured for pedestrians, cyclists and car occupants and found that injured pedestrians and cyclists sustain 9.4 and 12.8 years lived disabled or injured whilst car occupants sustain a higher period of 18.4 years due to the severities of injuries. The authors estimated that a person who would switch from car use to cycling in an urban area would, on average, have 40% less severe injuries.

Public transit offers similar benefits. Public transport is substantially safer than car travel, both for passengers and for the public (Litman 2011; Kenworthy and Laube 2002; Beck et al. 2007). Research shows that cities with the highest share of public transport users have the lowest share of traffic fatalities (Litman 2012). Cities built on a combination of transit and walking could also mitigate motor vehicle crashes. For comparison, between 2010 and 2014, the four largest Dutch cities, Amsterdam, The Hague, Rotterdam and Utrecht (all with a very high bicycle modal share by international standards) recorded 2.0 road deaths per 100,000 populations (SWOV 2016), whilst Hong Kong and Paris, both centred on mass transit (Sun et al. 2014), recorded 1.5 and 1.6 road deaths per 100,000 populations, respectively (Transport Department Hong Kong 2016; Préfecture de Police 2013). In a more recent analysis, public transit was shown to have 1/10 the per mile traffic casualty rate when compared to car travel (American Public Transportation Association 2016).

Overall, these studies demonstrate that passenger modal shifts away from the private car and towards active travel modes and public transit will improve traffic safety and yield public health benefits in urban areas, although contextual factors are specifically important.

2.3 Passenger Car Efficiency and Electrification

2.3.1 Air Pollution

A small number of studies investigated the air quality and health impacts of improving passenger car efficiency and electrification (Xue et al. 2015; Pathak and Shukla 2016; Timmers and Achten 2016; Ji et al. 2012; Soret et al. 2014).

Xue et al. 2015 showed that although increasing new energy vehicle proportions including hybrids, CNG and electric vehicles will improve air quality in Xiamen, China, the greatest contribution to air pollution reduction comes from other policy scenarios including alternative fuels and diesel truck decreases contributing to nearly 30% of air pollution reductions, followed by an option to control the intensity of private vehicles.

Incentivizing the electrification of passenger cars is often proposed as a sustainable approach to urban mobility and economic development (Ji et al. 2012). The literature, however, suggests that this may be a simplistic view and that electric vehicles may not result in presumed air quality and health benefits, unless their uptake is optimistically large, and they are complemented by non-exhaust PM reduction measures. A state-of-the-art review by Timmers and Achten (2016) investigated the effects of fleet electrification on non-exhaust PM emissions and found that total PM10 emissions from electric vehicles are likely to be higher than their non-electric counterparts, due to non-exhaust PM being increased by the higher weight of electric vehicles. Electric vehicles are, on average, 280 kg or 24% heavier than their non-electric counterparts. Vehicle weight is directly proportional to non-exhaust PM emissions attributable to traffic, as road abrasion and tyre wear, brake wear and resuspension are all influenced by the vehicle’s weight. Tyre, brake and road wear increase by 50% when comparing a medium (1600 kg) and small (1200 kg) car. Large cars (2000 kg) emitted more than double the amount of PM10. Further, the reduction in PM2.5 emissions from electric vehicles was estimated to be small (1–3%). Research on the health effects of non-exhaust PM is new and perhaps not sufficiently explored, but there are numerous studies, both epidemiological and toxicological, indicating distinct adverse health effects of non-exhaust PM that warrant consideration (Gasser et al. 2009; Gehring et al. 2015).

In a large-scale health impact assessment conducted by Ji et al. (2012), the health impacts of the use of conventional vehicles and electric vehicles in 34 major Chinese cities were modelled. PM2.5 emissions from electric cars were estimated to be higher than conventional gasoline vehicles, resulting in more preventable deaths, even when (un)proximity to the emission sources was accounted for. The total annual excess deaths in Shanghai were 9 as attributable to gasoline cars and 26 as attributable to electric cars. When compared to diesel cars, electric cars were shown to provide health benefits, yet the main reason behind these beneficial impacts was not the reduction of total PM2.5 emissions from electric cars, but the fact that most of these emissions occurred at power plants instead, away from human receptors. The total annual excess deaths in Shanghai were 90 as attributable to diesel cars and 26 as attributable to electric cars. There was evidence, however, that the urban use of electric vehicles will move adverse exposures and health impacts to rural, non-electric-car users and potentially lower socioeconomic populations.

Overall, the available evidence demonstrated that electric vehicles may only have a small impact on improving air quality and population public health and that conclusions will vary depending on the reference/comparison scenario (e.g. comparing to gasoline versus diesel cars) and whether rural populations are considered in the analysis. The remaining studies which investigated the air quality impacts of improving passenger car efficiency or electrification provided no estimates for the potential health impacts associated with these policies (Pathak and Shukla 2016; Soret et al. 2014).

2.3.2 Motor Vehicle Crashes

In relation to the potential effects of fleet electrification on road safety, the only theoretical link currently explored in the literature relates to the effects of increasing vehicle weight due to electrification and decreasing vehicle noise increasing non-detection (Timmers and Achten 2016). Although increased vehicle weight and stiffness may increase the safety of car occupants (Torrao et al. 2016), the opposite is expected for vulnerable and other road users involved in motor vehicle crashes with higher weighted vehicles. For example, Anderson and Auffhammer (2013) estimated that the baseline risk of fatalities increases by 47% when being hit by a vehicle that is 1000 pounds heavier. Studies also suggested that the reduction in noise from electric vehicles may lead to increased motor vehicle crashes because of non-detection, especially to the blind and visually impaired (Mendonça et al. 2013; Verheijen and Jabben 2010; Jabben et al. 2012).

2.4 Freight Logistics Improvements and Freight Vehicle Efficiency and Electrification

2.4.1 Air Pollution

Lee et al. (2012) assessed the air quality and health impacts attributable to a clean truck program in the Alameda corridor, USA, which progressively banned the older and most polluting trucks. The truck replacement was estimated to have reduced NOx and PM emissions by 48% and 55%, respectively, within a 7-year period. The health benefits from the associated reduction of PM2.5 only were equivalent to $428.2 million, but these estimates only incorporated two age groups (age 30–65, >65), and two health endpoints (mortality and chronic bronchitis), and are therefore likely underestimated. To contextualize the results, the authors demonstrated that the payback period for replacing all drayage trucks in the study area (11,000 trucks with an assumed new truck costs $150,000) is less than 4 years.

The published research for this policy measure is scarce. There is clearly a need for more research regarding the effects of freight policies on air quality and public health, but there also is sufficient current evidence showing that freight is a major contributor to local and regional air quality problems. These problems and associated adverse health impacts can be mitigated by upgrading freight fleets and increasing their efficiency (Guttikunda and Mohan 2014; Conlan et al. 2016).

3 Discussion

Co-benefits of climate mitigation actions in the transport sector include improved outdoor air quality, increased physical activity, reduced ambient noise and reduced motor vehicle crashes, all of which are associated with reduced morbidity and premature mortality. Generally, there was a lack of costing estimates in the identified literature, and only a few studies monetized the value of the estimated co-benefits. This creates challenges for comparative assessments and for the transfer or good practice in planning and policy.

The most important co-benefits we identified from the literature seem to be the increases in physical activity at the population level, which predominantly result from passenger modal shifts to active travel and from urban planning measures that reduce car travel and promote walking, cycling and public transit use. The least obvious co-benefits, which can also be risks depending on technical improvements, contextual factors and the investigated reference scenarios, seem to be the co-benefits of electrification which may occur through improved outdoor air quality.

Our findings suggest that policymakers should actively pursue urban planning and land use elements to reduce passenger car travel and at the same time support a modal shift away from the private car towards walking, cycling and public transit. Such measures should be complemented with safety increasing measures to avoid unintended consequences by reducing motor vehicle crashes and protecting existing and new pedestrians and cyclists. Further, new interventions should be carefully examined as they can feasibly introduce an additional way by which wealthy neighbourhoods become more fragmented from poorer areas. Compact cities and dense and diverse land use planning can also facilitate transit-oriented development, further adding to the net benefits from such interventions and providing other economic benefits. Despite being expensive and technically challenging to implement in already developed cities, urban planning and land use policies have the potential to realize longer-term and sustainable health improvements and provide additional benefits related to congestion, retail and quality of public space (Timilsina and Dulal 2010; Conlan et al. 2016). Unfortunately, as it stands, transport planning is generally not well integrated with land use planning . This integration is essential—if system transformations are to be made towards sustainable and healthy developments (Khreis et al. 2017).

Although the increases in physical activity from more compact and more connected urban planning patterns may seem small (e.g. 15–55 min of physical activity per week based on the evidence overviewed above), these differences should be considered important and relevant. Unlike interventions which target individuals or specific population segments to increase their physical activity levels, changes in the urban environment are a universal intervention that reaches the whole population of the targeted area and therefore increases population level physical activity rather than select individuals who are motivated and participate in tailored interventions. Further, changes in the urban environment are long lasting, whilst behaviour change campaigns and programs are rarely maintained (Saelens et al. 2003). The benefits of higher urban densities go beyond the climate and public health benefits we discussed and include increasing the potential for urban agriculture which in turn reduces food miles and improves local food security, improving the housing choices for all residents, reducing social exclusion, increasing the opportunities for creative social interaction, reducing crime rates and improving economic efficiency and employment opportunities, amongst others. There are, however, documented disadvantages of increasing density which highlight the need for an integrated policy approach to overcome these potential negative consequences. These include exacerbating traffic congestion and parking problems, reducing an area’s capacity to absorb rainfall, reducing green space and negatively impacting the economic development of surrounding rural areas.

There was considerable evidence for an improvement in air quality with the mitigation measures we reviewed, and thereby improvements in health. Most of the benefits come from reducing car use and increasing public and active transport.Technological advances such as electrification have been suggested to improve air quality . However, the literature we identified for electrification is scarce, but some key points emerged. First, the air quality (and associated health) benefits of electric cars need more rigorous quantification, but current evidence suggests that these might be negligible (in the case of PM2.5) or might even have the opposite direction (in the case of PM10). Second, the reduction of PM2.5 emissions from electric vehicles is also likely to disappear as exhaust emission standards become stricter over time (Timmers and Achten 2016). Particularly, the non-exhaust PM component of air pollution is expected to increase with the adaptation of electric vehicles, primarily driven by the increase in vehicle weight due to electrification. Whilst both emissions control regulation and electrification plans could lead to reductions in exhaust emissions, non-exhaust emissions from road vehicles remain unabated and need more attention (Thorpe and Harrison 2008). Some views expect battery technology to quickly improve making electric vehicles lighter, yet the overall trend over the last decade was a steady increase in vehicle weight in almost all segments (Timmers and Achten 2016). Third, there was evidence that there may be air quality benefits of electric vehicles when compared to diesel vehicles, but this was not true when compared to gasoline vehicles. Improvements in gasoline technology could also lead to important air quality and health benefits, yet the advancement of gasoline engines has received considerably less attention during the past 20 years, when, for example, Europe shifted its research capacity towards the diesel powertrain (Cames and Helmers 2013).

Despite diesel-fuelled freight vehicles being the largest motor vehicle contributors to air pollution within and between cities (Guttikunda and Mohan 2014; Conlan et al. 2016), there is significantly less research on freight interventions to improve air quality and health outcomes. In this research, the main uncertainty is related to uncertainties in the emission factors used which are likely to have underestimated the air quality benefits as these are generally optimistic and not reflective of real-world driving and conditions. The current evidence base suggests that the replacement of old trucks by newer, more efficient models can lead to significant air quality and health benefits, with savings that exceed the cost of investments.

The studies on noise are scarce. These studies used a health impact assessment methodology and mainly focused on the effects of a mode shift. The estimates for noise were crude and were not based on detailed modelling or measurement exercises. In all the studies we found, noise was not the main exposure that was being explored, but was rather a secondary outcome. The literature on the potential health impacts of noise is very limited and did not incorporate any scenario testing to quantify the health effects of specific policy measures or interventions (Hänninen et al. 2014; Mueller et al. 2017a, b; Fritschi et al. 2011). There is a clear need to fill large gaps in the literature on the effects of urban planning; reduced passenger travel demand, vehicles efficiency and electrification; and freight logistic improvements on noise and associated health effects. There is also a need to provide health estimates under different scenarios and for different policy measures.

Although the research is currently lagging behind, policymakers need to be aware that noise can have large adverse health impacts, which are comparable to those of air pollution (in the case of premature mortality) (Mueller et al. 2017a), or exceeding those of air pollution (in the case of morbidity) (Mueller et al. 2017b). Such impacts are currently not seriously considered in policy or in academic circles, as shown by the results of our searches and review. Further, there was evidence in the literature we identified that the scale of the positive impacts of traffic noise reduction measures may be limited to the scale of interventions, where local actions can have little effect due to national sources such as motorway traffic not being subjected to local traffic reduction measures. This highlights the advantage of complementing local policies with national policies. Unintended consequences of traffic noise reduction measures include the potential for increases in motor vehicle crashes, because of non-detection. Various studies have suggested that electric cars or hybrid cars are favourably quieter than conventional cars but that this may lead to increased motor vehicle crashes because of non-detection, especially in the blind and visually impaired (Mendonça et al. 2013; Verheijen and Jabben 2010; Jabben et al. 2012). For this reason, some countries like the USA and Japan are considering minimum noise requirements for such vehicles; however, this can restrict the noise reduction advantages (Verheijen and Jabben 2010).

The increases in motor vehicle crashes with increasing mixed land use may be explained by the increasing walking or cycling population which becomes exposed to a well-established higher crash risk. Policies that encourage densification, mixed land use and increase transit supply are likely to increase pedestrian and potentially cyclists’ activity. With no supplementary safety strategies in place, these policies may indirectly increase motor vehicle crashes (Miranda-Moreno et al. 2011). Similarly, compact areas with low-speed, but high-conflict traffic environments, can increase crash exposures (Ewing and Hamidi 2015). Policies that encourage modal shifts from the private car, although clearly demonstrating net public health benefits, can increase the risk of pedestrians and cyclists being involved in motor vehicle crashes . These results, however, were highly context specific and differ substantially depending on the study area, including its motorization levels, baseline active travel commuters and safety measures in place for pedestrians and cyclists. Furthermore, there are numerous methodological limitations which should be considered in this literature including the fact that most studies do not account for the ‘safety in numbers’ effect and the weaknesses in the indicators used to measure crashes (numerators) and corresponding exposures (denominators) (Mueller et al. 2015; Götschi et al. 2016).

Overall, we show that climate change mitigation measures in the transport sector have a great potential to improve public health in urban areas whilst mitigating climate change and its distal impacts. The scale and magnitude of the health benefits differed across the policy measures and were wider and highest in the case of the two policy measures of (1) compact land use planning to reduce motorized passenger travel demand and (2) encouraging passenger modal shift away from the private car and improving transit efficiency. We conclude that climate change action represents a great opportunity for policymakers to develop transport roadmaps that jointly achieve climate change objectives and improve public health in cities.