Economic valuation of Well-To-Wheel CO2 emissions from freight transport along the main transalpine corridors

Transportation Research Part D 47 (2016) 222–236

Contents lists available at ScienceDirect

Transportation Research Part D journal homepage: www.elsevier.com/locate/trd

Economic valuation of Well-To-Wheel CO2 emissions from freight transport along the main transalpine corridors Silvio Nocera ⇑, Federico Cavallaro IUAV University of Venice, Santa Croce 191, I-30135 Venice, Italy

a r t i c l e

i n f o

Article history:

Keywords: CO2 emissions Well-To-Wheel analysis Freight transport Transalpine corridors

a b s t r a c t CO2 emissions are one of the main externalities related to freight transport. Their evaluation is extremely difficult, due to the presence of several scientific and economic uncertainties. This paper discusses the approaches currently adopted by literature to deal with CO2, proposing a methodology based on a Well-To-Wheel quantification and an economic valuation deriving from a meta-regression. A freight transport analysis is then provided for one of the most critical areas of Europe, the Alps. Here, the different approaches adopted by the single nations determine divergent results in terms of modal shift towards rail and, consequently, CO2 emissions. An integrated and transnational strategy could lead to better results, avoiding detoured traffic and increasing the share of railway traffic. To this aim, the carbon impacts of three specific alpine-wide measures are evaluated: namely, Alpine Crossing Exchange, Emissions Trading and Differentiated Toll System. In comparison with business-as-usual scenario, the case study reveals a potential CO2 saving up to more than 600,000 tons and 38 M€ for the year 2030, thus providing policy makers with an integrative transnational tool able to evaluate the long-term carbon impact of their transport decisions. Ó 2016 Elsevier Ltd. All rights reserved.

Introduction One of the main elements to determine transport sustainability is the evaluation of the greenhouse gas (GHG) emissions (Sinha and Labi, 2007; Black, 2010): at European level, transport counts for 24.3% of total emissions (EU, 2014), which makes it the second most polluting sector after energy production. Policy makers are aware of this critical condition. Since the early 1990s, EU has constantly increased its efforts to reduce GHG emissions (EC, 2014a) and many sectors (e.g., agriculture, industry, buildings) have obtained encouraging results. However, this is not valid for transport, where GHGs have increased by about 22% in comparison to the 1990 levels (EU, 2014) and particularly for road freight transport, where emissions have increased by more than 35% (Enerdata, 2015). In freight transport operations, carbon dioxide (CO2) is the main component of GHGs and counts 93–95% of total emissions. CO2 emissions are thus a valid indicator to assess the global warming caused by freights (McKinnon and Piecyk, 2011). Among the possible approaches to limit CO2 emissions, the reduction of unitary values is considered relevant (EC, 2014b). Currently, a shared European methodology to calculate carbon pollution from freight transport does not exist. CO2 emissions of light commercial vehicles are already monitored (EU regulation 510/2011), but comparable standards are not available for Heavy Duty Vehicles (HDVs). According to McKinnon (2005), this is due to several factors, such as the capacity and the load ⇑ Corresponding author. E-mail address: [email protected] (S. Nocera). http://dx.doi.org/10.1016/j.trd.2016.06.004 1361-9209/Ó 2016 Elsevier Ltd. All rights reserved.

S. Nocera, F. Cavallaro / Transportation Research Part D 47 (2016) 222–236

223

factors of the vehicles, the adoption of parameters derived from international studies, the unity of measure (volume, weight or number of vehicles). To address this issue, EU is supporting the development of a simulation tool called the Vehicle Energy Consumption Calculation Tool (VECTO; Fontaras et al., 2013), which is expected to provide the first results in 2017. The technological development of vehicles is not enough to curb CO2 emissions: specific measures and policies to encourage a modal shift towards less polluting transport systems are necessary as well (Dray et al., 2012). This paper focusses on the assessment of several alternatives to reduce CO2 from freight transport in one of the most delicate areas of Europe, i.e. the Alpine arch. Here, emissions can be up to five times higher than in the plains due to the morphology and the presence of slopes (Alpnap, 2007). Hence, particular attention is required in planning adequate freight transport measures to reduce CO2 emissions. The Alpine arch is treated as a unique space in several transport studies (Reggiani et al., 2000; Dalla Palma et al., 2001; RappTrans AG and ProgTrans AG, 2004; Neuenschwander et al., 2011; Dörnenburg et al., 2015; Lückge et al., 2015), mostly focussing on the amount of freight and passengers monitored in the past and expected for the future. Only a few studies consider CO2 impacts: Ryan et al. (2005) assess HDV CO2 emissions and propose a scenario for the year 2025. iMonitraf! (2012) develops four scenarios for the year 2020 along the main five transalpine corridors. However, none of them provides an economic valuation of specific transport measures. To address this issue, this paper proposes a two-step process. First, Section ‘Methods to quantify CO2 transport emissions’ reviews the methods currently available to quantify CO2 emissions from freight transport, suggesting the adoption of a WellTo-Wheel approach that considers the entire energy process from fuel production to consumption. Second, an economic valuation is performed through a meta-regression analysis. Sections ‘Freight alpine transport’ and ‘Evaluation of future WTW CO2 emissions’ test this method to valuate the economic impact of CO2 emissions for the adoption of specific measures for freight transport along the main transalpine corridors. The paper concludes with some policy considerations that also highlight the role of a correct evaluation of GHG emissions deriving from freight transport.

Methods to quantify CO2 transport emissions The methods adopted to quantify the impact of transport CO2 emissions are rather heterogeneous, including different phases of the fuel production and emission. The most thorough method to analyse the complete process is the Life-Cycle-Assessment (LCA; A3PS, 2015). It covers the entire life cycle of a product, process or activity, encompassing the extracting and processing of raw materials; manufacturing, transportation and distribution; use, re-use, maintenance; recycling and final disposal. The application of traditional LCA methodologies are mostly product oriented: the aim of such analysis is to describe the processes necessary to obtain industrial and manufactured products (Cass and Mukherjee, 2011). Local specifications are normally not considered. Emphasis is on using estimated inventories and the assumption of uniform conditions. However, CO2 emissions cannot be included in such analysis because they are highly dependent on the context and the specific energy sources adopted. With the Well-To-Wheel approach (WTW; Edwards et al., 2014), the impact of a fuel can be determined through the description of specific pathways. These pathways are complete sets of assumptions about the resource used, including the primary energy source, the energy required for its extraction, transformations, transportation, fuel production and characteristics of the vehicle using the fuel. To guarantee a clear distinction between the emissions related to the primary energy source and those linked to the propelling technology, WTW is subdivided to the Well-To-Tank (WTT) and the Tank-To-Wheel (TTW) approaches. WTT describes the pathway necessary for the process of distributing fuels suitable for transport powertrains. Five main phases characterise this approach: production and conditioning of the energy, its transformation at source, transportation to market, transformation near the market and conditioning, and distribution of the finished fuels to the individual refuelling points (Edwards et al., 2013a). The energy required for fuel production can be expressed in terms of gCO2/MJ (or gCO2/kW h). For petrol, unitary emissions are calculated at about 13.05 gCO2/MJ (Edwards et al., 2013b). By adopting the adequate conversion factors, the results of the WTT analysis for petrol and diesel (Table 1a) determine a unitary emission of 532.59 and 447.43 gCO2/l respectively. According to the efficiency of the vehicle, it is possible to transform such a value in terms of gCO2/ km. For example, the average consumption of an HDV engine equals to 2.8 km/l, which determines a unitary WTT emission equal to 190 g/km, while the efficiency of an economy car (17 km/l) reduces WTT unitary emissions to 26.32 g/km. For railway transport, the process is more complex because the energy mix required to produce the electricity is the sum of different primary resources. For each of them, the provision (extraction and transport) of raw materials as well as the efficiency of energy production and distribution have to be included in the evaluation. The WTT electricity efficiency can be defined by multiplying the specific energy consumption required to transport 1 t of freight (function of the technology) by the CO2 emissions deriving from the production of the necessary electricity. In Table 1b, an example of a freight train of 1200 t is provided, according to the energy production of four different Alpine countries: Austria, France, Italy and Switzerland. In this last country, most of the electricity is produced by adopting renewable sources and thus the final value is significantly lower than for other countries like Italy, where most of the energy is produced by adopting coal power plants. TTW quantifies the unitary energy expended and the polluting substances emitted by a vehicle during its driving cycle. It includes evaporative and tailpipe emissions during the operation of the vehicle and it is obtained through specific emission models. The final values are determined by several factors, which can be classified into six main groups: travel, facility, driver, vehicle, fuel, environment (Table 2).

224

S. Nocera, F. Cavallaro / Transportation Research Part D 47 (2016) 222–236

Table 1 Unitary CO2 emissions deriving from the WTT phase. Source: (a) Edwards et al. (2013b); (b) IEA (2010); (c) UIC (2012); (d) Ruffini et al. (2010). Fuel

WTT emissionsa gCO2/MJ

(a) ROAD Diesel 14.62 Petrol 13.05

(b) RAIL A CH F I

WTT emissions gCO2/kW h

Calorific value kW h/kg

WTT emissions gCO2/kg

WTT emissions gCO2/l

Efficiency km/l - km/kW h

WTT emissions gCO2/km

52.63 46.98

11.90 12.00

626.32 563.76

532.59 447.43

2.80 17.00

190.21 26.32

Production of electricityb kgCO2/kW h

Specific energy consumptionc kW h/tkm

Weightd t

2010 Emissions kgCO2/vkm

2010 Emissions gCO2/vkm

0.182756 0.027385 0.082717 0.398464

0.054 0.054 0.054 0.054

1200 1200 1200 1200

11.84 1.77 5.36 25.82

11,842.59 1774.55 5360.06 25,820.47

Table 2 Factors affecting Emissions of Vehicle Pollutants. Source: Sinha and Labi (2007). Factors affecting Emissions of Vehicle Pollutants Travel related

Engine operating mode: hot start, cold start, hot stabilized

Vehicle related

Age and mileage Maintenance condition Weight and size Engine power Fuel delivery system Emission control system

Environmental

Ambient temperature Altitude Temperature

Speed level Speed variation Facility related

Engineering features

Driver related

Behavioural changes

Fuel type used

According to the factors considered, Demir et al. (2011) identified six classes of emission models: instantaneous fuel consumption, four-mode elemental fuel consumption, running speed fuel consumption, comprehensive modal emission, methodology for calculating transport emissions and energy consumption (MEET), and computer program to calculate emissions from road transport. The choice of the most suitable model is a function of the scale and the temporal horizon of analysis. For example, the first three groups work well at micro level, when the fuel consumption per second has to be calculated. Such instantaneous calculations are a function of specific driving and traffic behaviours, which necessitate the modelling of each agent. This requires a great modelling effort and can be done only for delimited areas and time periods (Samaras et al., 2012). If the analysis deals with a larger scale or a broader timeframe, macro modelling has to be preferred. To this aim, the European Environment Agency suggests the use of COPERT (Gkatzoflias et al., 2012) as possible reference software that is also in line with the IPCC recommendations (IPCC, 2006). COPERT belongs to the group ‘‘computer program to calculate emissions from a road transport”. It is based on on-road measurements and the parameters it uses are extracted from real-life experiment. However, COPERT introduces two speed ranges for each vehicle class without taking neither road gradient nor acceleration into account. These are important aspects and oversimplification may cause significant variation from real values. For this reason, MEET models could represent a valid alternative. The Handbook Emission Factors for Road Transport v. 3.2 (HBEFA - INFRAS, 2014) is one of the most important software in this group. The emissions are calculated for specific type-vehicles, characterised by fuels (petrol and diesel), Euro classes (from Euro 0 to Euro 6), size and load. They vary according to the area (urban, rural), the type of road (motorway, primary, secondary, local, access), congestion levels (free flow, heavy, saturated, stop and go), and road gradient (classes between
6% and +6%). HBEFA provides yearly values from 1990 to 2035, which are not only limited to CO2 but also include the main GHG and criteria pollutants. Data is provided for each Alpine country, taking into account their current vehicular fleet and the expected technological development of the next years, developed by a specific national scenario. The next sections adopt a WTW method to analyse the GHG emissions deriving from freight transport along the transnational corridors of one of the most critical areas of Europe: the Alps. Freight alpine transport The transalpine corridors can be defined as the mountain stretches of all the main terrestrial infrastructures that connect Italy to the European countries that have a section of the Alpine arch within their national boundaries. There is no unequivocal definition of their boundaries. Based on the aim and the geographical boundaries of research, three different segments can be considered (UFT, 2013), which are called respectively ‘‘A” (from Moncenisio/Fréjus to Brenner), ‘‘B” (from Ventimiglia to Tarvisio) and ‘‘C” (from Ventimiglia to Vienna). In this section, segment ‘‘C” is considered (Fig. 1). This segment includes

S. Nocera, F. Cavallaro / Transportation Research Part D 47 (2016) 222–236

225

Fig. 1. Main transalpine freight corridors. Source: Maino and Cavallaro (2014), modified.

Ventimiglia, Montgenèvre, Mont-Cenis, Fréjus and Mont-Blanc (F-I); Grand St. Bernard, Simplon, Gotthard and San Bernardino (CH-I); Reschen, Brenner, Tarvisio, Felbertauern, Tauern, Schoberpass, Semmering and Wechsel (I-A). In 2013, the transnational freight transport along these axes was quantified at 93.7 Mio t (UFT, 2013): 55.9 by road (60%) and 37.8 by railway (40%). More than 51 Mio t were transported through the Austrian corridors, about 30 Mio t through Switzerland and 13 Mio t through France. Brenner Pass measured the highest freight traffic volume in 2013 for a single corridor (37.6 Mio t), followed by Gotthard (18.3 Mio t), Tarvisio (12.9 Mio t) and Ventimiglia (10.3 Mio t). A clear differentiation between countries emerges in terms of modal split. With almost 100% and 70% of the total amount of freight traffic respectively, most freight between France and Italy and between Austria and Italy is moved by road. On the other hand, Switzerland emerges as a clear counter-example with about 84% of the total freight traffic being transported by rail (Table 3). This is the result of long-lasting and clear political decision-making, which has aimed at the shifting of freight transport from road to rail. Accordingly, several push and pull measures have been introduced. The ‘‘modal shift package”, which synthetises this integrated approach, is a legally binding document that limits the number of yearly crossings (650,000) of the Swiss Alps by HDVs after the opening of the Gotthard Base tunnels. The values registered in 2013 are the result of a consolidated trend, visible in previous years, and a function of the European economic development. This trend has four distinct phases: generalised growth of traffic (+30%) from 1999 to 2007, crisis in 2008 and 2009 (
16.2%), recovery in 2010 and 2011 (+12.6%); decline between 2011 and 2013 (
2.8%). Considering the modal split, the economic crisis affected mostly rail traffic and those economic sectors (such as steel industry, chemistry and car production) that used rail as their favoured transport vector. In last years, and with Switzerland as the exception, the percentage of rail transport has decreased everywhere compared to the peak years registered in 2007 (Fréjus/Mont Cenis) and in 2010 (Brenner and Tarvisio) (Lückge et al., 2015). Freight transport through the Alps is the result of a complex interaction between European, national and regional transport policies. According to the White paper (EC, 2011), transport has to enable the economic growth and job creation in a sustainable way, remaining fully and competitively integrated in the world economy. The increase of road freight transport and the contextual reduction of rail transport are not in line with such sustainable approach. The solutions proposed by the EU to address this condition are multiple. They include, among others, modal shift towards less polluting systems and technological improvements. The improvement of existing technologies (mostly related to the engine efficiency, the heat recovery, the transmission, aerodynamics, tyres and auxiliaries) has the potential for unitary CO2 reduction of about 35% (Schroten et al., 2012). Further research on alternative fuels would also support this reduction. Some of them (e.g., methanol, natural gas, LPG) are already a valid alternative to fossil fuels, mostly for passenger transport; others (e.g., boron, Fischer-Tropsch

226

S. Nocera, F. Cavallaro / Transportation Research Part D 47 (2016) 222–236

Table 3 Road and rail transport along the main transalpine axes. Source: UFT (2013). Country

Corridor

Road Mt

Transalpine freight transport - Year 2013 F Ventimiglia 10.30 Montgenèvre 0.10 Fréjus/Mont-Cenis 1.00 Mont-Blanc 1.60 Subtot France 13.00 CH Grand-St-Bernard 0.30 Simplon 0.30 Gotthard 6.20 San Bernardino 1.00 Subtot 7.80

Rail Mt

Total Mt

Country

Corridor

Road Mt

Rail Mt

Total Mt

– – – – – – 9.70 12.20 – 21.90

10.30 0.10 1.00 1.60 13.00 0.30 10.00 18.40 1.00 29.70

AT

Reschen Brenner Tarvisio Felbertauern Tauern Schoberpass Semmering Wechsel Subtot

0.60 25.90 8.70 – 8.40 4.50 0.10 4.00 52.20

– 11.70 4.20 – 3.00 0.80 3.00 – 22.70

0.60 37.60 12.90 – 11.40 5.30 3.10 4.00 74.90

fuel, p-series, solar fuel) constitute a niche market, due to lack of infrastructures and high costs of production and distribution. The Alpine area is very active in this field: the A22 Brenner highway will be one of the first European roads with hydrogen fuel stations every 100 km (A22, 2012). However, currently this is a pioneering project and fossil fuels are expected to continue playing the dominant role in the coming years (Weiss et al., 2012). As mentioned in the introduction, technological development alone remains inadequate if not accompanied by policies that encourage modal shift (Nocera and Cavallaro, 2011, 2016). These measures, called ‘‘push” and ‘‘pull”, are intended to shift freight from more to less polluting transport systems, analysing financial instruments (e.g. taxes, charges and tolls), technical and regulatory constraints (e.g. orders and bans), as well as the improvement of the attractiveness of existing alternatives to road vehicles. The effects of such measures are complex and they do not always reach the expected results in terms of modal shift (Cappelli and Nocera, 2006): some measures, taken at different scales or in different geographical areas can be conflicting, thus thwarting the potential benefits of the single measure. The current layout is the result of policies mostly applied at national or regional level, which do not grant a balanced growth of all systems. For example, Köll et al. (2007) estimated that about 20% of the freight road transport along the Brenner corridor is detoured traffic from the Gotthard axis: due to the lower unitary costs, carriers prefer to cover longer but cheaper trips, with all the well-known consequences in terms of transport externalities, including CO2 emissions. Indeed, to obtain long lasting results, an integrated and structural approach is necessary, which includes the entire Alpine territory. The next section deals with this specific issue, by investigating the carbon potentialities of some transnational transport measures. Evaluation of future WTW CO2 emissions Description of the scenarios The future traffic forecasts along the Alpine corridors are based on the Albatras study (Neuenschwander et al., 2011), whose aim is to contribute to the discussion about a common policy for transalpine freight transport by proposing alternative scenarios. Four main scenarios for the year 2030 have been developed1, based on the introduction of specific transport measures to limit polluting emissions: Business As Usual, Alpine Crossing Exchange, Alpine Emissions Trading System and Differentiated Toll Systems. In the following sections, only the essential aspects of such scenarios are presented and analysed for the year 2030; interested readers may refer to Neuenschwander et al. (2011) for further details. Business As Usual (BAU) is the evolution of the current transport condition along the Alps. The first decade of this century revealed a generalised increase of freight transport, which was only limited by the economic crisis of 2008. A moderate recovery followed the recession period, with highest growth on the Central and Eastern routes of the Alpine arch (Switzerland and Austria) and a slight decrease in France. The scenario is divided into two distinct parts: by 2020, the economic crisis determines a lower growth; no new policy instruments are introduced. Rail subsidies are expected to decrease both in Switzerland and in Austria, and to be abolished in France. Lötschberg and Gotthard base tunnels are operative. This will lead to a moderate growth, particularly on the Central and Eastern routes of the Alpine arch. After 2020, the subsidies for rail freight transport are abolished in all three countries; Brenner and Mont Cenis base tunnels will be operative. Figures regarding BAU can be summarised as follows: transalpine freight transport volume increases from 208 M t/y in 2004 (reference year for the calibration) to 314 M t/y in 2030 (+51%). Such an increase is equal to 27% on A-I railways and 52% on CH-I railways by 2020, thanks to the newly operative Lötschberg and Gotthard base tunnels. By 2030, this growth 1 In each scenario, the ‘‘restrictive” and the ‘‘tolerant” alternatives are analysed. The former case implies thresholds that are more binding, while the latter is less rigid. In this assessment, we consider only the ‘‘restrictive” option, hypothesising that all the measures are applied along all the main corridors of the Alpine arch.

S. Nocera, F. Cavallaro / Transportation Research Part D 47 (2016) 222–236

227

is more pronounced: +44% on A-I, +67% on CH-I and +17% on F-I. The infrastructural renewal, together with the adoption of some specific transport policies, leads to a general decrease of road modal split from 70% to 62%. In absolute terms, Austrian corridors have the highest share of total transalpine freight transport, both on road and rail. The role of the Brenner axis continues to be dominant; on the other hand, French corridors present lower increases. Alpine Crossing Exchange (ACE) is a system to limit the transnational road transport and, consequently, to reduce CO2 emissions. Originally conceived for Switzerland, it is expanded to all the Alpine corridors. ACE is based on two main elements: the Alpine Crossing Permits (ACPs) and the Alpine Crossing Units (ACUs). The former are documents (limited in number) required for a specific passage over an Alpine crossing. They are assigned to a specific vehicle and are not tradable. ACPs can be obtained in two different ways: either directly assigned (non-directly tradable ACPs) or by a conversion of the ACUs, which can be bought and sold on an electronic platform. Based on the length of the trip and the characteristics of the vehicle (size, emission class, loading), a different number of ACUs is converted to a single ACP. A defined number of ACUs can be freely traded. Their period of validity should be limited to 12 months, so that it is possible to manage the yearly traffic. The differentiation of tradable ACU from non-tradable ACP grants a certain flexibility. In this Alpine-wide dimension, ACPs can be used for all Alpine routes within the assigned area or countries; in specific cases, they can be used also as tools aiming to rebalance passages along the different corridors. Furthermore, special exemptions are possible for local and short distance traffic, which can receive a preferential treatment in order to avoid traffic obstruction between nearby economic areas on both sides of the Alps. In 2030, the unitary prices for ACP are fixed at 263 € for A-I, 269 € for CH-I, and 345 € for F-I (the value here is particularly high because the modal shift is lower). ACE reduces the volume of transalpine road freight transport by around 34% compared to BAU 2030: from 195 to 130 M t/y and a contextual comparable increase in rail transport. The following changes are expected for road transport:
23% A-I,
57% for both CH-I and F-I. 64.8 M t/y of the total reduction in transalpine road freight transport is shifted towards rail infrastructures. Despite the assumed opening of the new Mont Cenis base tunnel, the Gotthard rail corridor continues to be more attractive. The introduction of ACE reduces the road modal split of total transalpine freight transport by 41% in 2030. Alpine Emission Trading System (AETS) is strictly connected to the international environmental policy targets for reducing CO2 emissions, thus also leading to a reduction of capacity on transalpine road corridors. Conceptually, AETS is the inverse of ACE. This method, proposed in Austria and extendable to all Alpine corridors, is based on the adoption of emission certificates: for each unit of CO2 emitted, one certificate is required. For every time period, only a fixed number of emission certificates is available: this is the result of a political choice about the maximum threshold of emissions. These certificates are traded on the market by realizing an emission certificate trading exchange, similar to that adopted in other contexts (e.g. industries, air transport). The vehicle owner who wants to make a transalpine journey has to buy certificates according to the emission category of the vehicle and the distance travelled. The modal shift granted by this measure is a function of the CO2 reduction target chosen at political level. Referring to 2030, up to 40% of CO2 reduction is expected, according to the recent EU proposal (EC, 2015). This leads to a unitary price of certificates equal to about 0.70 €/km, which causes a decrease in total transalpine road freight transport volume of around 29% compared to BAU 2030 (from 195 to 139 M t/y), mostly acquired by rail transport (55.7 M t/y). Like in other scenarios, the reduction of road transport volume is different between corridors:
24% on A–I,
47% on CH–I and
37% on F–I corridors. The road modal split of total transalpine freight transport can be reduced by the introduction of the AETS by 44%. Differentiated Toll Systems (TOLL+) is a measure developed in France to determine fairer transport prices and to use the available physical capacity (including safety aspects) efficiently. To achieve these aims, TOLL+ internalises the external effects of road freight transport in terms of air pollution (here also including CO2), noise and congestion, by implementing the ‘‘polluter pays” principle (Directive 1999/62/EC). In the TOLL+ concept, the external costs would be added as a surcharge on the already existing toll rate. Furthermore, it proposes a differentiation of toll rates according to the congestion of the road: higher prices would be paid at peak times, while more convenient rates would be charged for the rest of the day. In contrast to other systems, the price of the ‘‘passage permit” is the charged toll rate, which means lower implementation costs than ACE and AETS. Furthermore, the price per km is fixed (and known) in advance. For the year 2030, TOLL+ is fixed at 0.80 €/km. This produces a road modal split equal to 43% of total freight transport (value in line with AETS). In comparison to BAU 2030, transalpine road traffic decreases by 32% (from 195 to 133 M t/y), with 61 M t/y shifted from road to rail:
26% along A-I,
52% along CH-I and
41% along F-I. To summarise the results of the four scenarios previously described, the amount of tons transported via road and rail along the main transalpine corridors in the year 2030 has been divided between transnational corridors (Table 4). Traffic volumes between Austria and Italy continue to be dominant (more than 60% in every scenario), due to the pre-eminence of the Brenner axis. On the other hand, connections between France and Italy continue to be weak (less than 20%), despite the introduction of the Mont-Cenis high-speed railway. Finally, the high quantity of freight exchange between Italy and Switzerland is mostly due to the rail transport on the new Gotthard High Speed/High Capacity (HS/HC) railway line. Quantification of future CO2 emissions The goods expressed in t/year can be transformed into vehicles by choosing a reference type vehicle. For road transport, we have selected an articulated lorry with a total weight of 32 t, a load capacity equal to 23 tnet and powered by diesel with a Euro 6 engine (Iveco, 2008). The maximum speed is 80 km/h. Considering a precautionary fill factor of 0.6, which also takes

228

BAU 2030 Corridor

ACE 2030 Rail Mt

Road Mt

Total Mt

Corridor

A–I CH - I F–I

68.29 36.97 13.67

133.50 20.78 40.80

201.78 0.06 54.46

A-I CH - I F-I

Total

118.93

195.07

314.00

Total

AETS 2030 Rail Mt

Road Mt

Total Mt

Trip cost €

Corridor

215 217 281

A-I CH - I F-I Total

80.30 45.57 22.93

92.53 8.85 17.24

172.83 54.42 40.17

148.81

118.61

267.42

TOLL+ 2030 Rail Mt

Road Mt

Total Mt

Trip cost €/km

Corridor

Rail Mt

95.89 53.48 25.27

101.85 10.97 25.92

197.74 64.45 51.19

0.70 0.70 0.70

A-I CH - I F-I

98.40 54.89 26.67

174.63

138.74

313.37

Total

179.96

Road Mt

Total Mt

Trip cost €/km

99.28 10.01 24.12

197.67 64.90 50.79

0.80 0.80 0.80

133.40

313.36

S. Nocera, F. Cavallaro / Transportation Research Part D 47 (2016) 222–236

Table 4 Road and rail transnational transport in different scenarios. Source: Neuenschwander et al. (2011).

S. Nocera, F. Cavallaro / Transportation Research Part D 47 (2016) 222–236

229

into consideration empty legs and partially laden trips, the average load is equal to 13.8 tnet. For transalpine and longdistance road transport, these assumptions, even if simplified, may be considered realistic. First, today almost all HDVs use diesel and this fuel is expected to play a dominant role also for the next years (Weiss et al., 2012). Second, EURO VI is the last class introduced (years 2013–2014) and it is still not certain when (and if) a class Euro VII will be introduced, or if it will be substituted or integrated with non-fossil fuels. Third, an articulated lorry can be considered the reference vehicle of long-distance and transnational journeys for logistics reason. Fourth, the average maximum speed is fixed at 80 km/h: according to the Transport protocol of the Alpine Convention (2015), no future changes along the existing main transalpine roads are expected in order to limit environmental externalities. For rail transport, a distinction has to be made, according to the type of vehicle adopted. Two trains are considered according to the type of railway infrastructure (either traditional line or new HS/HC line). For this analysis, we refer to a study elaborated specifically for the Brenner Corridor (Ruffini et al., 2010), but expandable to all Alpine axes. Two locomotives are necessary for the trains that run along a traditional line. Such trains guarantee a maximum speed of 100 km/h, an overall weight of 1200 t and a capacity of 597 tnet. Trains running along a new HS/HC line are characterised by one locomotive, a maximum speed of 100 km/h, an overall weight of 1200 t and a capacity of 664 t. According to the definition of scenarios, in 2020 the Lötschberg and Gotthard base tunnels will be operative and in 2030 also the Mont Cenis and Brenner base tunnels will be operative. As far as the length of the different corridors is concerned, we refer to the infrastructure comprised in the Alpine Convention perimeter. The lengths span from 305 km for the Montgenèvre to 60 km for the Tarvisio axis (Table 5). The unitary CO2 emissions of the different transport modes are the last factor. Regarding the WTT phase, values for HDVs and trains have been discussed in Section ‘Quantification of future CO2 emissions’. For train transport, the values for the year 2030 are difficult to quantify because they are a function of future national energy mixes. In this case, we adopt the current energy mixes presented in Table 1, hypothesising an improvement of efficiency by 10% by 2030 and assuming an average value of energy mix between Italy and the other transalpine country crossed by a corridor. In TTW phase, there are no rail emissions because all the transalpine railway lines are electrified. Road freight transport emissions can be calculated according to the nature of the trip. In the case study presented here, trips are of medium to long range and a constant speed is maintained. Hence, it is advisable to adopt a method that is based upon the direct measurement of values, disregarding the aspects related to the stop and go that are typical for short or urban routes. To this aim, HBEFA (Infras, 2014; see Section 2) is an adequate software program: it has already been used with appreciable results in other Alpine transport studies (Colberg et al., 2005; Hueglin et al., 2006; Cavallaro et al., 2013), proving its reliability for the Alpine territory. For this type of analysis, we have considered a condition of free traffic flow along the main transnational highways, with an average speed of the HDVs equal to 80 km/h. As far as the slope is concerned, we have selected weighted average values, which are the sum of different traffic gradient classes and consider both the ascending and the descending parts of the highways. Data regarding emissions in specific countries is presented in Table 6. It is now possible to determine the expected WTW CO2 emissions in the different scenarios by multiplying the number of vehicles by the unitary emissions and by the distances covered. In Table 7 and Fig. 2, aggregate data per nation is presented, while an analytical reconstruction of the single corridors is available in Appendix A. With about 2.30 Mio t, the scenario with the highest emissions in 2030 is BAU, followed by AETS and TOLL+ (at about 1.98 and 1.95 Mio t). The most virtuous scenario is ACE, with only 1.69 Mio t (
26.5% in comparison to BAU). Economic valuation of CO2 reduction After the quantification phase, the transalpine CO2 emissions produced by freight transport must be valuated economically. To do this, we adopt a model developed in Nocera et al. (2015a) and refined in Nocera et al. (2015b). The model is based on a meta-analysis of about 600 studies using either FIAM, NCGEM or CGEM2. Four main classes describe each study: general information (name and year of the study, GHG value); scenario (model adopted, reference scenario, temperature increase, concentration, temporal horizon and geographic scale); economic impacts (GDP variation, discount rate, equity weight, damage function); physical impacts considered (among sea level rise, energy use, agricultural impact, water supply, health impact, ecosystem and biodiversity, extreme weather events and major events/large scale discontinuity). Formula 1 represents the meta-analytic regression model:

lnðcost2010Þ ¼ a þ bi X i þ ui

ð1Þ

The dependent variable is the vector of the cost of emission values. The subscript i assumes values of observations (from 1 to 699), a is the constant term, bi are the coefficients of the vector of explanatory variables X and u is the vector of residuals. The equation is expressed as a natural log to account for the right skewedness of the different economic impacts of climate change. This implies that undesirable attributes are more likely than desirable ones (Nocera and Tonin, 2014). Furthermore, the coefficients of the regression model can be interpreted more easily: a one percent change in the value of the independent 2 Fully Integrated Assessment Models (FIAMs) are models that include an economic growth/dynamics (energy sector comprised), damage and climate modules. Non-Computable General Equilibrium Models (NCGEM) are models that include only the climate and damage modules (occasionally, they consist of an energy module as well but without an economic optimisation procedure and adopting scenarios provided by third parties). Finally, Computable General Equilibrium Models (CGEM) are models that focus the economic optimisation procedure on a greater number of sectors but do not include a climate module.

230

S. Nocera, F. Cavallaro / Transportation Research Part D 47 (2016) 222–236

Table 5 Length of the alpine stretch of the transnational rail and road infrastructures. Source: Suter (1999). Country

Corridor

Specification

Rail (km)

Length of transalpine infrastructures A-I Reschen Bludenz - Bolzano

Road (km)

Country

Corridor

Specification

0

227

CH-I

Gr. St. Bernard Simplon

Martigny - Ivrea

0

134

Sion - Stresa Thun – Stresa Luzern – Chiasso Chur – Chiasso

170

141

181 0

190 161

A-I

Brenner

Border D/A - Bolzano

195

186

CH-I

A-I A-I

Felbertauern Tauern

St. Johann - Lienz Region Salzburg - Spittal

0 151

78 137

CH-I CH-I

A-I A-I

Schoberpass Semmering

162 138

145 121

F-I F-I

A-I

Wechsel

140

184

A-I

Tarvisio

Windischgarsten - Graz Region Wiener Neustadt - St. Michael Region Wiener Neustadt Region Fürstenfeld Gemona - Tarvisio

60

60

Rail (km)

Road (km)

Region Geneva - Ivrea Region Chambéry Region Torino AS area

0 205

202 200

F-I

Gotthard San Bernardino Mont-Blanc MtCenis/ Fréjus Montgenèvre

305

305

F-I

Ventimiglia

Savona - Nice

142

141

Table 6 Unitary CO2 emissions of rail and road transport, year 2030. Country

Vehicle

WTT emissions g/km

TTW emissions g/km

WTW emissions g/km

Vehicle

WTT emissions g/km

TTW emissions g/km

WTW emissions g/km

A-I CH-I F-I

Train Train Train

16,948 12,418 14,031

0 0 0

16,948 12,418 14,031

HDV HDV HDV

190 190 190

656 600 710

846 790 900

Note: (a) WTT train emissions from IEA, 2010; (b) TTW road emissions from Infras, 2014.

variable leads to a b⁄100% change in the dependent one. The meta-analytic regression (formula 2) gives the final unitary value of 1t of GHG3 emitted, considering the variables we judged meaningful to explain the estimates’ variation: the nature of the model (FIAM, CGE), the adoption of the equity weighting (ew), the year of publication (ypub), the peer review nature of the study (peer_rev), the Pure Rate of Time Preference (prtp), the number of categories (comb), the geographic scale (geogsc), the increase of temperature and concentration (temp, ppmv), and the adaptation.

lnðcos t2010Þ ¼ 9:60 þ 1:09  fiam þ 0:69  cge þ 0:40  ew
0:006  ypub
0:47  peer rev
0:60  prtp þ
0:26  comb þ 0:03  geogsc þ 0:13  temp þ 0:001  ppmv þ 0:22  adaptation

ð2Þ

This formula can be adopted to valuate the CO2 emissions from the scenarios described above. We assume different conditions in terms of temperature increase and ppmv concentration, according to the projections made by IPCC (2013), holding all other variables constant. On the one hand, BAU scenario is the prosecution of the current trend, for which we hypothesise an increase of temperature by about 3.2 °C by 2050 and a growth in CO2eq concentration up to 590 ppm by volume (ppmv). On the other hand, AETS and TOLL+ are assumed to be part of a more virtuous global strategy, aimed at limiting CO2eq concentration up to 490 ppmv and temperature increase by 2.4 °C. Finally, ACE is the most virtuous scenario: it is expected to produce an increase of only 2 °C and a growth in CO2eq concentration up to 445 ppmv. With these assumptions, the model presented in formula (2) leads to the final values of €36.68t/CO2 for BAU, €29.46t/CO2 for AETS and TOLL+ and €26.55 t/CO2 for ACE (Table 8). By multiplying these unitary economic values by the emissions calculated in Table 7, it is possible to determine the economic impact of CO2 emissions in different scenarios for the year 2030. With about 84.5 M€, BAU scenario presents the higher costs, while costs of ACE are almost halved (44.9 M€). With 58.3 and 57.5 M€, AETS and TOLL+ are in middle position (Table 9). Findings and discussion of results The transalpine carbon analysis presented in this article can suggest some useful considerations for policy makers. First, data reveals the major role of road emissions: in BAU, it is almost four times higher than rail (66.8 M€ against 17.7 M€) and contributes about 79% of the overall CO2 costs. Other scenarios, by lowering the impact of road transport in favour of train, contribute to the decrease of CO2 economic impacts. 3 CO2eq is the indicator of the Global Warming Potential (GWP) produced by GHGs, i.e. the concentration of CO2 that would cause the same level of radiative forcing as that caused by the GHGs. CO2 is the reference unit value of GWP (1tCO2 = 1tCO2eq) and it is legitimate to adopt this indicator only for the CO2 component.

Country

BAU

ACE

Rail t

Road t

Total t

Rail t

A-I CH-I F-I

300,511 123,741 58,337

1,155,242 208,482 457,363

1,455,753 332,223 515,700

356,419 152,393 96,969

Total

482,589

1,821,087

2,303,676

605,781

AETS Road t

Total t

Rail t

811,830 88,058 185,378

1,168,249 240,451 282,346

426,805 178,843 106,964

1,085,265

1,691,046

712,612

TOLL+ Road t

Total t

Rail t

Road t

Total t

872,276 111,628 283,761

1,299,081 290,471 390,724

438,248 183,564 112,834

852,840 101,969 263,293

1,291,089 285,533 376,127

1,267,664

1,980,276

734,646

1,218,103

1,952,748

S. Nocera, F. Cavallaro / Transportation Research Part D 47 (2016) 222–236

Table 7 WTW CO2 emissions in different scenarios, year 2030.

231

232

S. Nocera, F. Cavallaro / Transportation Research Part D 47 (2016) 222–236

Fig. 2. WTW CO2 emissions along main transalpine corridors, year 2030.

Table 8 Cost prediction of CO2 emissions. Estimated coeff.

BAU

ACE

Cost prediction of GHG emissions according to the changes in temperature and concentrations Const 17.1483 FIAM 1.0917 1 1 cge 0.6867 0 0 ewn 0.3977 1 1 ypub
0.0058 2006 2006 peer_rev
0.4740 0 0 prtp
0.5932 3.00 3.00 combined
0.2636 3 3 geogrscale
0.0288 0 0 temp 0.1298 3.2 2.0 ppmv 0.0012 590 445 adaptation 0.2211 1 1 €36.68

Unitary economic cost of GHG emissions (€/tCO2)

€26.55

AETS

TOLL+

1 0 1 2006 0 3.00 3 0 2.4 490 1

1 0 1 2006 0 3.00 3 0 2.4 490 1

€29.46

€29.46

Table 9 Cost of WTW CO2 emissions along main transalpine corridors, year 2030. BAU Rail M€

ACE

Total M€

Rail M€

Road M€

Total M€

Cost of WTW CO2 emissions along main transalpine corridors, year 2030 A-I 11.02 42.37 53.40 9.46 21.55 31.02 CH-I 4.54 7.65 12.19 4.05 2.34 6.38 F-I 2.14 16.78 18.92 2.57 4.92 7.50

12.57 5.27 3.15

25.70 3.29 8.36

38.27 8.56 11.51

12.91 5.41 3.32

25.12 3.00 7.76

38.04 8.41 11.08

Total

20.99

37.35

58.34

21.64

35.89

57.53

84.50

Rail M€

16.08

Road M€

28.81

Total M€

TOLL+ Road M€

66.80

Total M€

AETS Rail M€

17.70

Road M€

44.90

Second, such analysis highlights the importance of considering the whole WTW process. By limiting the assessment only to the WTT phase, economic values are 51.90 M€ in BAU, 22.37 M€ in ACE, 29.01 M€ in AETS and 27.88 M€ in TOLL+. The comparison between the two methods of evaluation reveals that the incidence of the WTT component constitutes 39% in BAU and about 50% in the other scenarios. This component is not always considered in the environmental assessments of new transport policies, thus underestimating their potential environmental impact.

S. Nocera, F. Cavallaro / Transportation Research Part D 47 (2016) 222–236

233

Third, the adoption of transport measures in the Alpine area is efficient only if not limited to a single infrastructural corridor. A partial approach could cause variation in transport quantities or in modal shift for the single corridors, but it will not have significant impact on the overall Alpine scale. The Swiss case is emblematic: the national policy for freight transport implemented several relevant measures, such as the distance-related heavy vehicle fee (ARE, 2015), the AlpTransit high speed/high capacity railway network (Zuber, 1997) and the Rail 2000 program (SBB, 2004). This integrated use of pushand pull-measures leads to good results at the national level, where rail transport constitutes the vast majority of all transalpine freight journeys (see Section ‘Freight alpine transport’). However, in an Alpine perspective, this has caused detrimental consequences. Indeed, in many cases, such limited approach favours the detour of traffic rather than the modal shift: at Brenner, for example, a good 20% of HDV traffic is detoured from the Gotthard. This leads to an increase in the length of a trip and, consequently, a growth in CO2 emissions. Only the adoption of integrated Alpine measures, such as those presented in this section, can contribute to a more balanced growth along the different corridors and a more efficient answer to a global issue such as CO2 emissions. The analysis presented in this paper refers only to the year 2030. To evaluate the long-term effects of these measures, the temporal horizon should be prolonged to a broader period (30 or 50 years). Such extension would contribute to an improved comprehension of the climate impacts produced by the transport policies at transalpine level. In this case, a deeper analysis of the different scenarios is required, with a specific assessment of the numerous scientific and economic uncertainties regarding future CO2 emissions. In Nocera and Cavallaro (2014a, 2014b) and Nocera et al. (2015a), we discussed these issues thoroughly, identifying some key aspects in the future travel demand, modal split, energy mix and technological development. The future travel demand is a particularly critical issue: a rigorous ex-post analysis about the construction of new infrastructures (Flyvbjerg et al., 2005) demonstrated not only that its accuracy has not improved over the years, but also that vehicle forecasts lose accurateness over time. A valid solution to address this issue is to use assumptions that are more realistic as well as empirically based assessments of risk, rather than to improve models. Indeed, macro modelling seems to be most suitable when dealing with the international level and long-term evaluations (Linton et al., 2015) even if some interesting implications regarding driving behaviour and micro modelling (Samaras et al., 2012) and mesoscopic modelling (Flügel et al., 2014) may be also considered, but mostly at the urban level. Whichever method policy makers and modellers decide to adopt, they have to consider the vast range of uncertainty related to the economic valuation of CO2 emissions. This suggests that the adoption of sophisticated methods may be an elegant but useless subtlety, due to the large endogenous uncertainty that affects the process. Finally, a more detailed specification about the vehicles circulating, their fuel and Euro classes, as well as a more precise description of the infrastructural network (e.g., considering the slope and the congestion) may have to be provided. Referring to the technological development, a study on clean transport systems, commissioned by the Directorate-General for Mobility and Transport of the European Commission (EC DGMT, 2011) indicates a relevant growth of vehicles powered by alternative sources, such as electricity, hydrogen or bio fuels. This increase, which EU expects to occur mostly between 2030 and 2050, has to be adequately assessed and linked to the continental energy policies, in order to design realistic scenarios.

Conclusions In this paper, we have quantified and valuated economically the WTW CO2 impact of freight transport along the main Alpine roads and rail infrastructures following the possible introduction of specific measures (Crossing Exchange, Emission Trading, and Differentiated Toll Systems). The results in terms of CO2 reduction and economic savings have been presented and interpreted, revealing the efficiency granted by the Alpine Exchange System and its significant difference in economic terms from BAU (
40 M€ for the only year 2030). The method is based on some assumptions about future travel demand, energy mixes and economic impacts of CO2 emissions, which are partially taken from other scientific studies. This means that a degree of epistemic uncertainty, which cannot be eliminated, affects the final value (Salling et al., 2007). Despite these issues, this method can be adopted to valuate the carbon impact of specific measures in other territorial contexts (not necessarily mountainous), and for other transport modes (air, water), in case accurate data about energy mixes and freight transport is provided. Furthermore, this analysis can be extended to other forms of transport externality, including accidents, air pollution, noise, and other costs such as nature and landscape, soil and water pollution, sensitive areas, up- and downstream processes (Maibach et al., 2008), in order to make the analysis more comprehensive. Finally, we are aware that these transnational transport measures are not easily implementable, both for conflicting political interests and for technical reasons (not least of which are the implementation costs and the managing process). However, after the signing of the Paris agreement (UN, 2015), climate change and CO2 emissions can no longer be addressed in generic terms. Indeed, they have to be discussed operatively, with the introduction of concrete and transnational measures or policies. The European Union is still far from an adequate answer, at least to the issues regarding freight transport. The most recent amendment of the Eurovignette directive (EP, 2011) does not include CO2 as one of the external-cost charges that concurs to determine the final unitary toll, and the external costs of air pollution caused by HDVs do not include CO2 (EEA, 2013). These issues have to be reconsidered strategically, in order to redefine the potential contribution of freight transport to the reduction of CO2 emissions and to invert a trend that –despite many proposals and international agreements- has not seen significant changes in the last 20 years.

234

Appendix A. WTW CO2 emissions in the single corridors, year 2030

Rail t CO2

ACE Road t CO2

Total t CO2

WTW CO2 emissions in the single corridors, year 2030 A-I Reschen 0 28,306 28,306 Brenner 87,382 395,673 483,054 Felbertauern 0 5845 5845 Tauern 69,819 160,228 230,047 Schoberpass 56,568 219,171 275,739 Semmering 86,742 77,190 163,932 Wechsel 0 173,290 173,290 Tarvisio 0 95,541 95,541

AETS

TOLL+

Rail t CO2

Road t CO2

Total t CO2

Rail t CO2

Road t CO2

Total t CO2

Rail t CO2

Road t CO2

Total t CO2

0 118,176 0 92,800 58,482 86,962 0 0

10,627 166,500 2710 47,883 254,941 83,418 195,895 49,857

10,627 284,676 2710 140,683 313,423 170,379 195,895 49,857

0 148,523 0 110,757 67,431 100,094 0 0

7191 128,787 2385 59,644 292,790 96,237 224,172 61,070

7191 277,310 2385 170,401 360,222 196,330 224,172 61,070

0 153,893 0 114,650 68,370 101,336 0 0

5925 109,469 2127 51,023 300,245 98,164 229,527 56,360

5925 263,362 2127 165,672 368,614 199,500 229,527 56,360

CH - I

Gr. St. Bernard Simplon Gotthard San Bernardino

0 21,647 102,094 0

9296 27,529 149,098 22,559

9296 49,176 251,192 22,559

0 28,845 123,548 0

4103 13,817 61,162 8976

4103 42,662 184,710 8976

0 33,608 145,235 0

4318 10,758 84,728 11,823

4318 44,366 229,963 11,823

0 34,778 148,786 0

3889 9434 77,975 10,671

3889 44,212 226,761 10,671

F-I

Mont-Blanc MtCenis/Fréjus Montgenèvre Ventimiglia

0 55,397 0 2940

61,260 204,843 5989 185,271

61,260 260,240 5989 188,212

0 88,982 0 7986

16,932 70,660 1631 96,154

16,932 159,642 1631 104,140

0 98,603 0 8360

36,751 111,143 2766 133,101

36,751 209,747 2766 141,462

0 103,776 0 9058

33,852 101,463 2447 125,532

33,852 205,239 2447 134,589

Total

482,589

1,821,087

2,303,676

605,781

1,085,265

1,691,046

712,612

1,267,664

1,980,276

734,646

1,218,103

1,952,748

S. Nocera, F. Cavallaro / Transportation Research Part D 47 (2016) 222–236

BAU

S. Nocera, F. Cavallaro / Transportation Research Part D 47 (2016) 222–236

235

References A22, Autostrada del Brennero S.p.A., 2012. La via dell’idrogeno Online at: (22.01.2016). A3PS, Austrian Association for Advanced Propulsion Systems, 2015. Eco-Mobility 2025plus. Roadmap Online at: (22.01.2016). Alpine Convention, 2015. Protocol on the Implementation of the 1991 Alpine Convention in the Field of Transport Online at: (22.01.2016). Alpnap, 2007. Living Along a Transit Route: Air Pollution, Noise and Health in the Alps Online at: (22.01.2016). ARE, Federal Office for Spatial Development, 2015. Distance-Related Heavy Vehicle Fee (HVF) Online at: (22.06.2015). Black, W.R., 2010. Sustainable Transportation: Problems and Solutions. The Guilford Press, New York. Cappelli, A., Nocera, S., 2006. Freight modal split models: data base, calibration problem and urban application. In: Brebbia, C.A., Dolezel, V. (Eds.), Urban Transport XII – Urban Transport and the Environment in 21st Century. The Wit Press, Southampton, UK. http://dx.doi.org/10.2495/UT060371. Cass, D., Mukherjee, A., 2011. Calculation of Greenhouse Gas Emissions for Highway Construction Operations by Using a Hybrid Life-cycle Assessment Approach: Case Study for Pavement Operations. J. Constr. Eng. Manage. 137 (11). Cavallaro, F., Maino, F., Morelli, V., 2013. A new method for forecasting CO2 operation emission along an infrastructure corridor. Eur. Transport/Trasporti Europei 55 (4). Colberg, C.A., Tona, B., Stahel, W.A., Meier, M., Staehelin, J., 2005. Comparison of a road traffic emission model (HBEFA) with emissions derived from measurements in the Gubrist road tunnel, Switzerland. Atmos. Environ. 39 (26), 4703–4714. Dalla Palma, R., Chevroulet, T., de Tiliere, G., Perret, F.L., Jaccard, P.A., Bays, P., 2001. Transalpine Freight Transport System 2005–2020 Scenarios Online at: (22.01.2016). Demir, E., Bektas, T., Laporte, G., 2011. A comparative analysis of several vehicle emission models for road freight transport. Transport Res. Part D: Transport Environ. 16 (5), 347–357. Dörnenburg, K., Haas, T., Leonardi, G., Wenzel, L., Legat, M., Debyser, C., Fusseis, W., 2015. Observation et analyse des flux de transports de marchandises transalpins Rapport annuel 2013 Online at: (25.01.2016). Dray, L.M., Schäfer, A., Ben-Akiva, M.E., 2012. Technology limits for reducing EU transport sector CO2 emissions. Environ. Sci. Technol. 46 (2012), 4734–4741. EC DGMT, European Commission, Directorate-General for Mobility and Transport, 2011. Study on Clean Transport Systems. Project 2010 n. 419-1-Unit Move B4, Final Report Online at: (26.02.2015). EC, European Commission, 2011. White Paper: Roadmap to a Single European Transport Area – Towards a Competitive and Resource Efficient Transport System Online at: (26.07.2015). EC, European Commission, 2014a. The European Union Explained: Climate Action. Publications Office of the European Union, Luxembourg. EC and European Commission, 2014b. Communication from the Commission to the Council and the European Parliament. Strategy for Reducing Heavy-Duty Vehicles’ Fuel Consumption and CO2 Emissions, Online at: (26.07.2015). EC, European Commission, 2015. 2030 Framework for Climate and Energy Policies Online at: (26.02.2015). Edwards, R., Hass, H., Larivé, J.F., Lonza, L., Maas, H., Rickeard, D., 2014. Well-to-Wheels Report Version 4.a. JEC Well-to-Wheels Analysis. Publications Office of the European Union, Luxembourg. Edwards, R., Larivé, J.F., Rickeard, D., Weindorf, W., 2013a. Well-to-Tank Appendix 4 – Version 4.0. Description, Results and Input Data per Pathway. Publications Office of the European Union, Luxembourg. Edwards, R., Larivé, J.F., Rickeard, D., Weindorf, W., 2013b. Well-to-Tank Appendix 2 – Version 4.0. Summary of Energy and GHG Balance of Individual Pathways. Publications Office of the European Union, Luxembourg. EEA, European Environment Agency, 2013. Road User Charges for Heavy Goods Vehicles (HGV) Online at: . http://dx.doi.org/10.2800/70164 (26.07.2015). Enerdata, 2015. Odyssee mure project Online at: (26.07.15). EP, European Parliament, 2011. Directive 2011/76/EU of the European Parliament and of the Council of 27 September 2011 Amending Directive 1999/62/EC on the Charging of Heavy Goods Vehicles for the Use of Certain Infrastructures Online at: (26.07.2015). EU, European Union, 2014. EU Energy in Figures, Statistical Pocketbook 2014. Publications Office of the European Union, Luxembourg. Flyvbjerg, B., Skamris Holm, M.K., Buhl, S.L., 2005. How (In)accurate Are Demand Forecasts in Public Works Projects? J. Am. Plan. Assoc., Spring 71 (2). Flügel, S., Flötteröd, G., Kwong, C.K., Steinsland, C., 2014. Evaluation of methods for calculating traffic assignment and travel times in congested urban areas with strategic transport models Online at: (21.01.2016). Fontaras, G., Rexeis, M., Dilara, P., Hausberger, S. et al., 2013. The Development of a Simulation Tool for Monitoring Heavy-Duty Vehicle CO2 Emissions and Fuel Consumption in Europe, SAE Technical Paper 2013-24-0150. http://dx.doi.org/10.4271/2013-24-0150. Gkatzoflias, D., Kouridis, C., Ntziachristos, L., Samaras, Z., 2012. COPERT 4, Computer Programme to Calculate Emissions from Road Transport. User Manual (version 9.0) Online at: (16.01.2016). Hueglin, C., Buchmann, B., Weber, R.O., 2006. Long-term observation of real-world road traffic emission factors on a motorway in Switzerland. Atmos. Environ. 40 (20), 3696–3709. IEA, International Energy Agency, 2010. CO2 Emissions from Fuel Combustion. OECD/IEA, Paris. iMonitraf! consortium, 2012. Alpine Transit Traffic – Policy Scenarios 2020 Online at: (25.01.2016). Infras, 2014. The Handbook Emission Factors for Road Transport (HBEFA) Online at: (25.07.2015). IPCC, Intergovernative Panel for Climate Change, 2006. Intergovernative Panel for Climate Change, IPCC Guidelines for National Greenhouse Gas Inventories, Prepared by the National Greenhouse Gas Inventories Programme. In: Eggleston, H.S., Buendia, L., Miwa, K., Ngara, T., Tanabe, K. (Eds.). IGES, Japan. IPCC, Intergovernative Panel for Climate Change, 2013. Quantifying Uncertainties in Practice Online under: (25.01.2016). Iveco, 2008. Iveco Stralis Activeday AD-N320S42Y/PS. Technical Sheet. Köll, H., Lange, S., Ruffini, F.V., 2007. Detours of trans-alpine goods transport by road, Revue de Géographie Alpine. J. Alpine Res. [En ligne], 65–76. Linton, C., Grant-Muller, S., Gale, W.F., 2015. Approaches and techniques for modelling CO2 emissions from road transport. Transport Rev. 35 (4), 533–553. Lückge, H., Heldstab, J., Cavallaro, F., Muscella, C., Kistler, R., Joos-Widmer, N., 2015. iMONITRAF! Annual Report 2014. Moving Ahead on Common Measures: Toll Plus in Focus Online under: (26.07.2015). Maibach, M.C., Schreyer, D., Sutter, H.P., van Essen, B.H., Boon, R., Smokers, A., Schroten, C., Doll, B., Pawlowska, M., Bak, M., 2008. Handbook on Estimation of External Costs in the Transport Sector Internalisation. Measures and Policies for All External Cost of Transport (IMPACT) Version 1.1. Delft, CE. Maino, F., Cavallaro, F., 2014. Il contributo delle grandi infrastrutture ferroviarie alla riduzione delle emissioni di CO2: il caso della Galleria di Base del Brennero. EURAC, Bolzano/Bozen. McKinnon, A., 2005. CO2 Emissions from Freight Transport: An Analysis of UK Data Online under: (26.07.2015).

236

S. Nocera, F. Cavallaro / Transportation Research Part D 47 (2016) 222–236

McKinnon, A., Piecyk, A., 2011. Measuring and Managing CO2 Emissions of European Chemical Transport Online at: (26.01.2016). Neuenschwander, R., Scheuchzer, P., Lieb, C., Newton, S., Kawabata, Y., Oehry, B., Jordi, P., Haas, L., Herry, M., Sedlacek, N., 2011. Transport Research and Training, ALBATRAS – Alignment of the Heavy Traffic Management Instruments ACE, AETS and TOLL+ on a Comparable Scientific, Technical and Operational Level Taking into Account the Introduction of Different Thresholds in Order to Analyze Transport Flow Impacts on Alpine Routes Online at: (27.06.2015). Nocera, S., Cavallaro, F., 2011. Policy effectiveness for containing CO2 emissions in transportation. Procedia-Soc. Behav. Sci. 20, 703–713. Nocera, S., Cavallaro, F., 2014a. A methodological framework for the economic evaluation of CO2 emissions from Transportation. J. Adv. Transp. 48, 138–164. http://dx.doi.org/10.1002/atr.1249. Nocera, S., Cavallaro, F., 2014b. The ancillary role of CO2 reduction in urban transport plans. Transport. Res. Procedia 3, 760–769. http://dx.doi.org/10.1016/j. trpro.2014.10.055. Nocera, S., Cavallaro, F., 2016. The competitiveness of alternative transport fuels for CO2 emissions. Transp. Policy, 50, 1–14, ISSN 0967-070X. http://dx.doi. org/10.1016/j.tranpol.2016.05.013. Nocera, S., Tonin, S., 2014. A joint probability density function for reducing the uncertainty of marginal social cost of carbon evaluation in transport planning. In: de Sousa, J.F., Rossi, R. (Eds.), Computer-based Modelling and Optimization in Transportation, Advances in Intelligent Systems and Computing, vol. 262. http://dx.doi.org/10.1007/978-3-319-04630-3_9. Nocera, S., Tonin, S., Cavallaro, F., 2015a. The economic impact of greenhouse gas abatement through a meta-analysis: valuation, consequences and implications in terms of transport policy. Transp. Policy 37, 31–43. http://dx.doi.org/10.1016/j.tranpol.2014.10.004. Nocera, S., Tonin, S., Cavallaro, F., 2015b. Carbon estimation and urban mobility plans: opportunities in a context of austerity. Res. Transport. Econ. 51, 71– 82. http://dx.doi.org/10.1016/j.retrec.2015.07.009. RappTrans AG, Progtrans AG, 2004. Alpenquerender Verkehr 2020. Entwicklung und Beurteilung des alpenquerenden Verkehrs mit Fokus auf die GotthardAchse Online at: (26.01.2016). Reggiani, A., Cattaneo, S., Janic, M., Nijkamp, P., 2000. Freight transport in Europe: policy issues and future scenarios on trans-border alpine connections. IATSS Res. 24 (1), 48–59. Ruffini, F., Maino, F., Nocera, S., 2010. Sostenibilità del tunnel di Base del Brennero in rapporto alle emissioni di CO2. Research Report, Bolzano, Italy. Ryan, J., Lückge, H., Heldstab, J., Maibach, M., 2005. Monitraf Synthesis Report: Activities and Outcome Online at: (26.01.2016). Salling, K.B., Leleur, S., Jensen, A.V., 2007. Modelling decision support and uncertainty for large transport infrastructure projects: the CLG-DSS model of the Øresund Fixed Link. Decis. Support Syst. 43, 1539–1547. Samaras, Z., Ntziachristos, L., Burzio, G., Toffolo, S., Tatschl, R., Mertz, J., Monzon, A., 2012. Development of a methodology and tool to evaluate the impact of ICT measures on road transport emissions. Procedia – Soc. Behav. Sci. 48 (2012), 3418–3427. SBB and Schweizerischen Bundesbahnen, 2004. Sulla giusta via. Ferrovia 2000 Online at: (26.05.2011). Schroten, A., Warringa, G., Bles, M., 2012. Marginal Abatement Cost Curves for Heavy Duty Vehicles. Background Report. Delft, CE Delft. Sinha, K.C., Labi, S., 2007. Transportation Decision Making – Principles of Project Evaluation and Programming. John Wiley and Sons, New York. Suter, S., 1999. Pricing European Transport Systems, Online at: (27.06.2015). UFT, Ufficio Federale Dei Trasporti, 2013. Alpinfo 2013, Traffico merci su strada e ferrovia attraverso le Alpi. Berna Online at: (27.06.2015). UIC, International Union of Railways, 2012. Energy Consumption and CO2 Emissions. 1990–2010 European Data Overview. International Union of Railways, Paris. UN, United Nations, 2015. Adoption of the Paris Agreement Online at: (22.12.2015). Weiss, M., Patel, M.K., Junginger, M., Perujo, A., Bonnel, P., van Grootveld, G., 2012. On the electrification of road transport-learning rates and price forecasts for hybrid-electric and battery-electric vehicles. Energy Policy 48, 374–393. Zuber, P., 1997. The AlpTransit scheme: new railway lines in base tunnels under the swiss alps. Tunn. Undergr. Space Technol. 12 (3), 357–360.