Skip to main content

A heuristic method for determining CO2 efficiency in transportation planning



CO2 emissions are generally considered the most important indicator to determine the global warming effects. Their evaluation in the case of a transportation infrastructure is generally not easy and could be achieved through a separate balance.


This paper introduces a new heuristic method for identifying the modifications in the CO2 emission balance, deriving from a variation in transportation supply. The focus is predominantly on the construction and on the operational phases, which are listed in all their main elements. The method compares the maintenance of the “do-nothing” option with a number of traffic scenarios resulting from the introduction of a new infrastructure and deriving from different policy measures.


The construction of the Brenner Base Tunnel is used as a case-study for the model, highlighting the role of an enlightened transport policy in the reduction of the CO2 emissions.


The extent and nature of the growing traffic demand in Europe pose several challenges for sustainable transportation, putting pressure on the decision-makers to provide new facilities for both passenger and freight transportation, and hence an ever increasing strain on infrastructure planning policy.

The decline in railway use favours the expansion of road mobility and its infrastructures. In this framework the overdependence on a limited number of routes has severe impacts on certain areas, generally without adequate compensation for the local communities, which call for measures to mitigate the negative impacts of traffic (congestion, severe crash incidents, large amounts of land wasted and pollution from all motorized traffic modes).

Among the polluting substances, greenhouse gases (GHGs) have steadily assumed a main role. Carbon dioxide (CO2), methane (CH4), nitrogen oxides (NOx), hydrofluorocarbons (HFC), perfluorinated compounds (PFC), sulphur hexafluoride (SF6) are the most important among them [39, 51]. However, as CO2 accounts for about 90% of global GHG emissions [12], it is often used as an overall indicator in heuristic methods.

The CO2 impacts concern the three environmental, social and economic dimensions of sustainability [11, 41, 50]. Referring to the first one, CO2 becomes one of the main causes of the global warming [14, 39] as soon as its level exceeds the threshold level of 350 parts per million by volume (ppmv) [18]. The linked environmental consequences are well known: weather changes, temperature increase, sea-level rise, harmful freeze–thaw cycles, precipitation changes, landslides, erosions in coastal areas etc.…

Social sustainability of CO2 impacts is normally referred to the remarkable influence of these consequences on human life. Among them, a reduction in the agricultural productivity and in the gross domestic product as well as a migration towards more productive areas are included, mostly in the development countries, thus confirming that the unmitigated climate change is incompatible with sustainable development [53]. Other consequences are visible in the life style and in the health of humans, including heat strokes, cardiovascular and respiratory problems [6, 22]. Finally, the economic aspect has been treated in a vast amount of studies [34, 53], including some radical theories that consider the global warming as the most important market failure ever seen [42]. Hence it derives that the determination of the future CO2 emissions is very relevant if related to the concept of sustainability and its three dimensions.

Even though our understanding of the physical mechanisms of the climate system has progressed rapidly, the use of this knowledge to support transportation decision making, manage risks, and engage stakeholders is still inadequate [45, 48]. Indeed, expressing global warming in monetary terms requires providing a consistent estimation of the effects of higher temperatures within the well-known uncertainty of linking CO2 and environmental damage. This is the biggest issue in dealing with global warming and the very aspect that sets it apart from other external costs (such as air and noise pollution). As known, the latter are generally converted into monetary units which are purported to express health expenses, property value reduction and other possible costs. Due to this, traditional techniques for monetization hardly apply to the context of CO2.

Quantifying CO2 impacts through the Net Present Value (NPV) is challenging, since impacts are difficult to estimate and their apportionment on an annual basis is problematical to apply. Moreover, no distinction between who receives benefits and to whom costs incur can normally be made, thus potentially waiving the principle of social equity (some segments of the community might receive all benefits at the expense of others). A successful attempt of estimating the impacts of greenhouse gas reduction policies through the use of the Benefit-Cost Analysis (BCA) is the use of the MERGE model [7]: the determination of the overall national production of GHGs and other polluting gases in different scenarios with this method showing the effectiveness of a policy that internalizes the costs of the global climate change and the local air pollution.

The Multi-Criteria Evaluation (MCE) is a promising approach when there are significant non-monetary and non-monetisable benefit-and-cost components to a proposed policy or project [11, 27, 28, 36]. This is also the case with GHGs and in particular with CO2. However, the assessment of the relative level of importance (usually referred to as “weight”) of GHGs with respect to other criteria is normally an issue, because the link between emissions and global effects is hard to estimate and quantify. This is obviously a key-step, which may yield flawed results if emissions are assessed using MCE.

In order to determine the CO2 contribute to the overall impacts, a separate balance analysis is required. This method has already been applied in many fields, namely the industrial [13, 19, 54] as well as the biological [25], agricultural [9] and renewable energy [26] ones.

CO2 balances kindle considerable interest also within transportation engineering. Von Rozycki [52] considers the variation in CO2 emissions related to the introduction of the high-speed railway line between Hanover and Würzburg. The study is detailed but it refers only to the environmental footprint of the railway without considering the concurrent trend in road traffic. Tuchschmid [49] proposes a method to quantify the emissions of several pollutant gases (including CO2) from the construction of high-speed lines in Europe. Being based on macro-scenarios at European level, the study assumes appropriate simplifications at such a scale, which however make the method inapplicable at infrastructure level. An interesting study is carried out to compare the emissions of CO2eq of the high-speed and traditional railway lines connecting London to locations in the North and West of the United Kingdom [31]. Results are based on scenarios for 2070. Also in this case the study dwells on the railway alone without considering the consequences that the new line could have on the road traffic.

Booz Allen Hamilton Ltd. [8] introduces a balance which compares the CO2 emissions deriving from the London-Edinburgh and the London-Manchester high speed lines with their corresponding air lines over a period of 60 years. The forecast of the CO2 emissions is provided by the calculation deriving from the construction and the operation phases, the latter being based on the emissions module (i.e. the forecast of future specific emissions) and the demand module (i.e. the forecast of future demand).

Finally, the engineering company of the Italian State Railways [23] proposes a method to forecast the CO2 emissions that would be generated by a new railway line between Rho and Gallarate. This study is based on well-defined technical and regulatory assumptions and takes into account various stages of the project (preliminary, final, execution, “as built” phases). However, since it lacks a clear description of the methodology, it can hardly be applied elsewhere.

All the afore-mentioned studies being recent, the current interest in the issue among infrastructural planners is confirmed. Nonetheless, most of the methods quoted do not seem flexible enough to be used for assessing the carbon footprint of different transportation systems.

This paper presents a new heuristic method to that end (thoroughly described in section 2), which will be tested on the Brenner Base Tunnel (BBT) currently under construction (section 3).

CO2 balance: description of the method

A CO2 balance is an analysis intended to assess the impact of a given project or activity over time, for the purpose of climate protection and the prevention of detrimental effects on human health. In civil engineering, it consists of four macro-phases over a long time-frame, namely design, construction, operation and decommissioning (Fig. 1). Each of them must be taken into account in order to assess the overall scope.

Fig. 1

The phases of a process that yield CO2 emissions

The goal of such an analysis is to quantify a system’s incoming and outgoing CO2 emissions in the various phases. The CO2 balance of a certain project X in a given time frame is positive if the overall CO2 emissions produced (P) are lower than the ones generated by the “do-nothing” scenario (N). Formally, this assumption can be expressed as follows:


If BALCO2(X) >0, then N > P: the CO2 emissions of the project are lower than those of the “do-nothing” scenario and hence there is a potential gain for the community. On the contrary, if BALCO2(X) <0, then the new project system leads to a rise in emissions and its implementation should be discouraged.

CO2 emissions for the construction of a new transportation infrastructure

In the field of transportation engineering, the CO2 balance could be efficiently used in order to evaluate the environmental sustainability of a modified or newly-implemented traffic system or facility. In this last case, this method should take into account the indirect impact that such infrastructures might have in terms of modal distribution of traffic.

To this end a comparison should be made between those scenarios following the building of the planned infrastructure and others, in which no new construction is undertaken. Looking at the phases in Fig. 1, from an operational viewpoint, the emissions resulting both from the design and the decommissioning phases are generally considered as negligible, since the former are of scarce relevance (0,03% in comparison with overall construction phase, [23]) and the latter concern interventions supposed to have a long life.

Also the construction phase seems to play a minor role in the process. In the calibration case chosen in this paper, it accounts only for about 4% overall (Fig. 8) – smaller than the expected error of the methodology to calculate operation emissions. The balance presented should however be applied to a larger amount of cases to make this conclusion generalizable. For this reason, the impacts of the construction phase are examined in short in section 2.2 of this paper.

In this framework, since construction is not to be considered in the “do-nothing” scenario, the terms contained in formula (1) can be expressed as follows:



P :

are the total CO2 emissions related to the project analysed

C P :

are the total CO2 emissions resulting from the construction phase of the project

O P :

are the total CO2 emissions resulting from the operative phase of the project

N :

are the total CO2 emissions related to the “do-nothing” scenario

O N :

are the total CO2 emissions resulting from the operative phase of the “do-nothing” scenario.

The emissions resulting from the construction (C) and operation (O) phases may be further detailed in formulas (4) and (5) (Fig. 2):

Fig. 2

CO2 emissions during the construction and the operation phases of a new infrastructure


C ex :

are the CO2 emissions produced by excavation and filling operations

C tr :

are the CO2 emissions produced by materials transportation operations

C pr :

are the CO2 emissions produced by construction materials production operations

C cn :

are the CO2 emissions produced by operations to run the working site

O vh :

are the CO2 emissions produced by vehicles

O up :

are the CO2 emissions produced by maintenance of the infrastructure

O io :

are the CO2 emissions produced by idle operation.

Assuming 1 as the first year of operation of the new infrastructure and n as the last year to be included in the balance, letting m be the year in which construction of the infrastructure is accomplished (with m ≤ n), formulas (2) and (3) can be written as (6) and (7). The choice of the time horizon [1;n] must be well weighed, being based on several factors such as the expected life time of the infrastructure, the reliability of future forecasts etc.


Sections 2.2 and 2.3 discuss in detail how the terms C and O can be obtained.

Construction phase

In the construction phase analysis, the calculation of CO2 emissions is largely linked to the energy consumption, as anthropic CO2 is released during all energy production and transformation processes that involve combustion. The calculation method proposed is based on a bottom-up approach: it takes into account the final energy consumption of the construction phase and the quantity of materials needed to build the infrastructure. Such data are converted into CO2 emissions through appropriate factors.

Once the most significant elements of the entire construction process are defined (Fig. 2), these are then broken down into single operations in order to set up the calculation through simplifying assumptions, thus distinguishing relevant aspects from negligible ones.

The accuracy and validity of the results mostly depend on the quality of the data available. Therefore, it is crucial that the entire analysis be based on data as consistent and accurate as possible and capable of covering all the sectors considered. As regards the construction phase, the most useful instruments are the analysis of the available design documentation, an on-going dialogue with planners, direct observations and surveys.

However, since the analysis is often carried out when the project is still in the assessment phase, there is limited knowledge of the actions and techniques used for construction. To offset this situation, estimates need to be made based on analogy, on the study of the construction techniques, on the information collected from specialised firms and on scientific literature.

Once the final energy consumption for the single construction operations and the quantities of the materials used are obtained, these are converted into CO2 emissions through the following formula (8) for the construction phase C:



h :

is the specific energy source

f h :

is the final energy consumption obtained from the energy source h

χ h :

indicates the CO2 emission factors for the energy source h

k :

is the type of material

q k :

is the quantity of material k

φ k :

indicates the CO2 emission factors for the material k.

The emission factors can be estimated using regularly updated methods approved by international scientific bodies such as the Intergovernmental Panel for Climate Change [21]. Databases available in literature [16, 35, 43] contain different values depending on the geographical area considered and on the calculation method used. The conversion factors χ h for energy consumption vary depending on whether the CO2 emissions refer to the share of final energy consumed or to the share of primary energy involved in the process.Footnote 1 Similarly, as regards the conversion factors φ k for the construction materials, these vary whether or not the emissions of the primary energyFootnote 2 are considered. Consistency is essential when the emission factors to be used are selected: one should always specify whether the calculation is based on primary or final energy. Moreover, area-specific factors related to the geographical setting of the infrastructure should be preferred.

Operation phase

Operation-related CO2 emissions must be calculated considering both options of maintaining of the status quo and modifying the existing transportation supply (e.g. by building a new infrastructure). The key elements of the operation phase have already been identified as the number and type of vehicles, maintenance and idle operation (Fig. 2).

One must also determine the territorial scale of reference based on the scope and the repercussions of the infrastructure under consideration. These issues give rise to a series of methodological questions that are normally difficult to work out.

The scenario analysis offers an effective way to address the problem: after forecasting the most likely evolution over time, the pathways leading thereto are explored, as illustrated in Fig. 3.

Fig. 3

Outline of the scenario method

Estimated trends differ depending on the evolution of the considered parameters. Among these, the most important ones concern the socio-economic conditions (population, GDP), social and technological development and transportation policy (market organization, taxes and tariffs, infrastructure policies, adoption of requirements and bans). These factors determine the growth rates in the various scenarios. Therefore, future traffic flows can be quantified on the basis of historical data, provided that the latter are reliable and of good quality. The next step consists in calculating the CO2 emissions generated by such scenarios. To do that, the amount of specific emissions per kilometre travelled is multiplied by the number of transiting vehicles according to the formula (9):



O vh , 1, n :

have previously been defined

e i :

is the specific emission of vehicles and trains in the year considered

v i :

is the number of vehicles transiting in the year considered

d i :

is the distance covered by vehicle v i .

The difficulty here lies in determining the trend of the specific traffic-related CO2 emissions (e i ), especially when the time horizon considered is quite extensive. Literature provides several methods for dealing with this issue (Corinair-IPCC [15], Copert [17] and Ecoinvent [43]). Infras [20] and Tremove [47] are two of the most commonly used software products to forecast future specific emissions on the basis of simple input data (country, region, network, period, fuel and vehicle type, vehicle technology and pollutant type) [20, 47]. Differently from other methodologies that calculate emissions related to the covered distances or use only the tank-to-wheel emissions (i.e. Infras, Copert and Corinair), Tremove considers both the well-to-tank and tank-to-wheel emissions. Even if further researches in the field of specific emissions are required, CO2 emission models generally provide more accurate results than other forecasts on polluting gases [40], thus making the results more consistent.

The case-study: the Brenner base tunnel

The method previously described has been applied to the case study of the Brenner Base Tunnel (BBT), a 55 km long railway infrastructure that, once completed, will connect Austria (Innsbruck) with Italy (Fortezza/Franzensfeste).

The BBT belongs to the central sector of the TEN-T Corridor 1 that connects Berlin with Palermo with a 2,200 km long high-speed railway line. Due to the presence of the Alps, one of the most critical stretches of this line is the Brenner corridor between Kufstein, a town close to the Austrian-German border, and Verona (Fig. 4).

Fig. 4

Munich-Verona railway line. Realization of the access routes. Source: [10], elaborated

The Brenner corridor is commonly divided into three sections:

  • Northern access route (Munich – Kufstein, Kufstein – Kundl, Kundl – Baumkirchen);

  • Brenner Base Tunnel (Innsbruck – Fortezza/Franzensfeste) with the Innsbruck bypass;

  • Southern access route (Fortezza/Franzensfeste – Ponte Gardena/Waidbruck, Bolzano/Bozen bypass, Trento bypass, Verona approach).

The entire line will be upgraded in different phases: the works are currently at a more advanced stage in the Northern access route than in the Southern one.

The BBT represents the central sector that links the two parts of the line. The construction phase of the main tunnels began in April 2011 and is planned to be completed by 2022, barring any setbacks.

In calculating the CO2 emissions of the construction and operation phases, two different spatial scopes were used. While construction-related emissions refer to the realization of the infrastructure (i.e. the 55 km long tunnel: paragraph 3.1), operation-related emissions consider the transnational impact of the tunnel and its effects on the entire line. Therefore, the traffic scenarios (paragraph 3.2) refer to the line as a whole, including all Italian and Austrian sectors between Kufstein and Verona.

Construction phase

The BBT infrastructure is schematized in Fig. 5.

Fig. 5

Overview of the Brenner Base Tunnel system. Source: [5], elaborated

Main tubes, exploratory tunnels, side tunnels, multi-purpose areas, lateral access and the interconnections with Fortezza and the Innsbruck bypass are the components considered during the calculation phase. In the following analysis, most of the data and values derive from the final design drawn up by BBT SE [4]. Figure 2 lists the main process elements that should be taken into account in the CO2 balance of the construction of a tunnel. The largest energy consumption and consequently the CO2 emission peak are normally generated during the excavation phase, the transportation of the spoil to the deposits, the production of the materials used to coat the tunnels and the running of the working sites [33]. These elements have been specified into the sub-process elements listed in Table 1.

Table 1 Main process elements of the analysis. Source: [33], elaborated

In the case of the BBT, the tunnel is excavated using the conventional method with excavators and rock blasting, and the mechanical method with the Tunnel Boring Machine (TBM). The method to be used is chosen on the basis of the geological and geotechnical surveys and the cross section, as well as on the length and gradient of the tunnel section to be built. The calculation estimates mainly the energy consumption of the machines used. In the conventional excavation, also the CO2 released during rock blasting is factored in. Likewise, in the mechanical excavation, the CO2 emissions resulting from the production of TBMs and their transportation from the production to the working sites are included.

The materials used to build the infrastructure are mainly cement and steel, with the latter used mostly for anchoring, building bridges and the production of reinforced concrete. Emissions due to the use of plastic material for the piping and other finishing materials are assumed to be negligible.

The transportation of the construction materials and the spoil to the deposits is then considered: a distinction is made between belt and truck method on account of the different energy consumption of the two forms of transportation.

The CO2 emissions related to the construction site include those generated by the tunnel ventilation and cooling systems, the water treatment plants, the mechanical shops and offices. Finally, the contribution of the lighting of the tunnels and external areas are considered.

After calculating the energy consumption [33], values are converted into CO2 emissions by applying the following emission factors X h , expressed in kg CO2/kWh:

  • X g  = 0.202 for energy obtained from natural gas [21];

  • X d  = 0.267 for energy obtained from diesel oil [21];

  • X el,IT  = 0.435 for electricity generated in Italy [44];

  • X el,A  = 0.216 for electricity generated in Austria [44].

The emissions resulting from the combustion of crude oil derivatives are calculated by referring to the guidelines provided by the IPCC [21], while the calculation of the emissions deriving from electricity are based on Terna’s studies [44]. Since about 60% of the infrastructure is located in Austria and the remaining 40% is in Italy, the combined average of the energy mix of both countries was considered.

Specific emission factors (φ k ) are then provided for the construction materials. The data about concrete are provided by BBT SE, while the data about steel are provided by the Munich-based FFE research centre [16], as specific values are not available for Italy and Austria. Emission factors are expressed in kg CO2/t:

  • φ steel, structural  = 1.980 [16];

  • φ steel, machines  = 1.449 [16];

  • φ portland cement  = 622;

  • φ pozzolanic cement  = 576.

Both factors (X h and φ k ) estimate the CO2 emissions resulting from the consumption of primary energy, taking into account the emissions related to the entire process and not only to the final energy, as described in footnotes 1 and 2.

Energy consumption and material quantities are converted into CO2 emissions using CO2 emission factors (formula 8). The total amount of CO2 associated with the BBT construction phase is about 2,280 kt (Table 2).

Table 2 CO2 emissions in the construction phase of the BBT

Operation phase

Recalling Fig. 2, emissions in the operation phase arise from vehicles, maintenance and idle operation. The last two factors are not calculated as their overall emissions are considered negligible in comparison with the vehicle emissions.

The calculation is based on a study by ProgTrans [37], which analyses the evolution of transportation along the Brenner axis (i.e. the Brenner rail line and the Brenner highway) up to 2030 in terms of average annual variations in freight and passenger traffic, by the development of six different scenarios.

The future traffic demand

In order to determine the future traffic demand and to provide the possible growth rates, ProgTrans adopts a classical four-stage model [30]. An area which includes 37 countries (all EU countries – except for Malta and Cyprus – , Switzerland, Norway and South-East European countries) is analysed: most of them are considered on NUTS 0-level, except for Austria, Germany, Italy, France, Netherlands, Belgium and Switzerland (NUTS 2) and the Alpine area (NUTS 3).

The traffic generation is calculated depending on five main parameters: socioeconomic tendencies, technological development, transportation policies, social changes (for passengers) and development of transportation economy and logistics (for goods).Footnote 3

The distribution is provided by the use of an origin–destination matrix, which includes 296 traffic cells. The modal choice is based on the generalized costs, which include both the distance costs and the travel costs. The former are meant as the sum of the costs necessary to transport a freight train or a HGV (freight transportation) or a person (passenger transportation) for a kilometre; the latter include fixed costs for the exercise of a transportation system (HGVs, freight trains, passenger trains or cars), energy and tracing costs and tolls.Footnote 4

Finally, the route assignment is developed for both road and rail traffics. The former is obtained through TRIBUT [2], a capacity restraint method based on a bicriterion path search algorithm which includes times and costs as parameters. The latter is calculated through a procedure which takes into account the aggregate travel time of four different systems [37]: three of them are related to the passenger transportation (namely high speed, intercity/eurocity and regional trains), while the fourth is linked with good transportation (freight trains).

At the end of this method, six different scenarios are determined. Only three of them are included in the balance, as the assumptions of the other three are deemed either unlikely (two of them consider the new Gotthard Base Tunnel not in use – even though it is currently at an advanced stage of construction) or incomplete (forecasts considered freight traffic only). Three scenarios are therefore taken into account, in this paper called “minimum”, “trend” and “consensus”, whose main points are shown in Tables 3 and 4.

Table 3 Measures to discourage the use of road transportation adopted in “minimum”, “trend” and “consensus” scenarios. Source: [32], elaborated
Table 4 Measures to encourage the use of rail transportation adopted in “minimum”, “trend” and “consensus” scenarios. Source: [32], elaborated

The “minimum” scenario coincides with the maintenance of the do-nothing, i.e. the BBT is not built, while both the “trend” and “consensus” scenarios are based on the construction of the tunnel.

The difference between these last two scenarios lies in the transportation policies. The continuation of the trend of the last decade is considered in the “trend” scenario, namely a market liberalisation that boosts road traffic. This policy should encourage the development of rail transportation. At the same time, the absence of measures to discourage the use of road vehicles should cause also a demand increase of HGVs and cars. It follows that an higher amount of overall traffic is generated, if compared with the two other scenarios.

The latter (“consensus” scenario) foresees a series of actions to foster the growth of railway traffic and encourage the simultaneous reduction of road traffic. The following measures are introduced in order to reach the complete internalization of external costs: in the field of tax policy, the increase of tolls (Austrian tolls are assumed to reach the Swiss ones) for all the types of vehicles and the funding of the railway mode. The introduction of an eco-tax for mineral oil and the reduction of transportation subsidies are also forecasted. Related to the infrastructural policy, the improvement of the high-speed rail line (with subsequent reduction of travel time), the development of the rolling highway and the modernization of the materials are scheduled. In relation to the road empowerment, only the most critic bottlenecks are solved (no further construction of road segments between the Alpine highways).

Finally, a scenario which includes only restriction to the circulation of the vehicles is considered unrealistic and therefore not analysed here: it foresees only constrictions to the free circulation, thus being against the important principle of “social equity” promoted by the EU in the development of its transportation policy.

Originally up to 2030, the forecasts were extended to 2035, thus determining a time horizon for the Brenner Base Tunnel of 25 years (Table 5). The values shown are mean annual growth rates.

Table 5 Overview of results of the mean annual variations of traffic along the Brenner axis. Source: [37], elaborated

As above mentioned, the growth of rail traffic (both for passengers and goods) in the “consensus” and “trend” scenarios is due to the introduction of the BBT and the high speed rail line. They allow the use of new tracks, thus implementing the overall railway capacity and rationalizing the line already existing. The growth rate of freight rail transportation is higher than the passenger one. This result can be explained by the nature of BBT, which is supposed to introduce about 400 new trains per day: according to the traffic simulations developed by BBT SE, about 75% of them will be introduced for freight transportation and the leftover for passengers [3]. This is the reason for which the expected growth rates of rail freight transportation are much higher than passengers’. Indirectly, it justifies also the difference between road passenger traffic of “consensus” and “trend” scenarios.

The number of vehicles circulating up to 2035 is then obtained by multiplying the rates listed above by the historical data [1, 29; Table 6]. The analysis considers both freight and passenger traffic on road and rail. By way of example, Table 7 shows the data for 2035 alone.

Table 6 Traffic demand along the Brenner axis in 2008. Sources: [1, 29], elaborated
Table 7 Traffic along the Brenner axis. Forecast for 2035 [vehicles/year]. Source: [38]; elaborated

The future specific emissions

The Infras Handbook and the Tremove software applications are used to determine the average specific emissions for road and railway traffic respectively [20, 47]. Both software applications provide the trend of the specific emissions over five-year periods (Table 8).

Table 8 Calculation of the trend in CO2 emissions for road traffic. Source: [20, 47], elaborated

Tremove software includes the following parameters: country, region, network, period, fuel type, vehicle type, vehicle technology, pollutant type. Each of them requires a choice among several possibilities: for example, referring to the vehicle technology, the choice is among passenger train, bus, car, heavy duty truck >32 t, heavy duty truck 6–32 t, heavy duty truck 3.5–7.5 t, heavy duty truck 7.5–16 t, light duty truck, moped, motorcycle, van, plane, freight train, inland ship, metro/tram [24].

Infras handbook considers the following parameters: country, vehicle type, polluting gases, year, fleet composition, network, level of service, speed limit, parameters for hot emissions factors and cold start access [46]. The evaluation of the specific emissions is based on the type-vehicle. Two trains (one for each railway line) are considered for the freight transportation: the one going through the existing line has two locomotives, a max speed of 100 km/h, an overall weight of 1,200 t. The other one has only one locomotive and same attributes. The train for passengers is an Intercity express type 1 (length: 200 m, average number of passengers: 400; gross weight: 435 t). Data are extracted from an exercise document of the Monaco-Verona line, provided by BBT SE.

For the road, the standard vehicle for goods is an HGV of 32 t. Average speed is 80 km/h and it is powered by diesel. The class of the engine depends on the year considered. The standard vehicle for passengers is a car with an engine of 1,600 cc. Maximum speed is 130 km/h and it is powered by diesel. Also in this case, the class of the engine depends on the year considered.

The constant decreasing values presented in Table 8 are due to the progress in the technology field, which grants the development of more efficient engines and vehicles.

The future overall emissions

The total emissions in the various scenarios were obtained for the years 2010–2035 by multiplying the number of vehicles by the mean specific emissions and by the distances travelled (formula 9).

Distances were calculated using for road, for the Italian sector of the railway line and for the Austrian one. From these calculations, the road distance for the Italian rail and road stretches (Verona-Brenner) are, respectively, 220 and 226 km; for the Austrian one (Brenner-Kufstein), respectively 111 and 106.

As soon as all these data are known it is possible to determine the yearly overall CO2 emissions by using the method described in section 2 (Table 9).

Table 9 CO2 emissions produced by road and railway traffic. Kufstein-Verona section. Source: [38], elaborated

Table 9 shows that the overall lowest CO2 emissions (about 43,150 kt) are forecasted in the “consensus” scenario, followed by the “minimum” and “trend” scenarios, with about 46,700 kt and 48,800 kt respectively.

As the year of reference increases, the difference in the emissions between the scenario without tunnel and the two scenarios with tunnel increases as well. In 2035, the “consensus” scenario produces about 180 kt CO2 emissions less than the “minimum” scenario, while the “trend” scenario implies an increase in emissions of about 200 kt (Fig. 6).

Fig. 6

CO2 emissions in the various scenarios: yearly variations. Source: [38], elaborated

Three aspects need here a further explanation.

First, although the BBT is expected to be working only by 2022, the variations in CO2 emissions between the “consensus” scenario on the one side, and the “trend” and “minimum” scenarios on the other, are already visible starting from 2010 on. This is mainly due to a supposed different behaviour of freight operators: ProgTrans forecasts [37] consider the traffic support policy (as described above) to be strictly linked to the opening of the tunnel. Some of these measures are supposed to be put into force already in 2010 in order to obtain a gradual but constant shift from road to rail (“consensus” scenario) or a gradual growth of traffic rail (“trend” scenario). Both these conditions are supposed to be fully realized in the first operation year of the tunnel (believed 2022). In this year, at these traffic growth rates the current Brenner railway line by itself would not have the capacity to serve the demand of neither of these two scenarios.

Second, “minimum” and “trend” scenarios show the same rise between 2010 and 2015. This should be considered an expected result, as both call for the same traffic growth forecasts (Table 5).

Third, the overall trend in the “consensus” and “minimum” scenarios seems to have a similar shape in the years 2016–2035 (Fig. 6). This is mostly an undesired visual impression: a rough data comparison can confirm this statement.Footnote 5 This means that the two curves are divergent and the difference in CO2 emissions is progressively increasing.

The overall CO2 emissions for the years 2010–2035 deriving from the operation phase (Fig. 7) confirm these considerations, showing how the main differences in the three scenarios are broader after the opening of the BBT (2022).

Fig. 7

CO2 emissions in the various scenarios: aggregate variations 2010–2035

Result comparison

Table 9 and Fig. 8 show the impacts of the BBT on CO2 emissions, which are supposed to be about 46,700 kt in the do-nothing case (“minimum” scenario). If the tunnel is realized, future emissions will depend on the policies adopted. If the support measures for the railway are adequate, the road traffic is likely to decrease and the overall CO2 emissions are expected to be lower than in the “minimum” scenario. They will amount to about 45,400 kt, equal to the sum of the emissions resulting from the construction of the tunnel (2,280 kt) and from the “consensus” scenario (about 43,150 kt). Therefore, recalling formula (1), BALCO2(BBT) >0.

Fig. 8

CO2 emissions in the various scenarios, as a sum of construction and operation phases. Years: 2010–2035

On the other hand, if policies adopted continue the current trend, rail and especially road traffic will further increase. As a consequence, also emissions are expected to grow and reach about 51,100 kt, which is the sum of the emissions resulting from the tunnel construction (2,280 kt) and the “trend” scenario (about 48,800 kt). In this case BALCO2(BBT) <0.

Figure 8 shows that the construction of the BBT does not necessarily imply a reduction in CO2 emissions; indeed, the tunnel might lead to a considerable increase due to the traffic demand growth, as evidenced by the “trend” scenario, which yielded 4,400 kt CO2 more than the “minimum” one. Only if duly supported by a correct policy favouring railways the tunnel will help cut CO2 emissions (“consensus” scenario, −1,300 kt).

As far as “minimum” and “consensus” scenario are concerned, the overall CO2 emissions are cut only by about 3%. This value could be considered disappointing if taken by itself. It should rather be taken into account in connection with its relative traffic demand: in the period 2010–2035, a 29% emission increase in the “minimum” scenario (1,535–1,984 kt) is for instance accompanied from a 75% freight railway increase (51,660–85,033 trains).

As well, in “consensus” scenario the emissions are supposed to grow from 1,500 kt (in 2010) to 1,805 kt (in 2035, +20%), but the overall amount of freight trains circulating should treble (51,660 in 2010 and 164,345 in 2035) and is almost double than in “minimum” scenario. Similar considerations can be stated for the other traffic modes analysed.

Values and faults of the method described

The method described in this paper was conceived to relate carbon dioxide emissions to modifications in the transportation supply. It can be used either as an independent tool or within a Multi-Criteria Evaluation.

In both cases, further research is required on the interaction between the effects of the simplifications adopted and the accuracy of the model results. Particularly, as the impact of the emissions depends from a certain number of key variables (here identified with the estimation of overall road traffic, its modal shift, and the technological improvement of the vehicles circulating), the extension of the time frame at stake makes hard quantifying model uncertainty, thus denying an important information for decision-making.

If the construction of a new infrastructure is included in the process, then the considerable width of the time frame cannot be avoided and should be retained as a model feature. This obviously makes the estimation of the evolution of the key-variables not reliable on ex-post evaluations. As in any forecast, however, the containment of the referential period makes the control of the results easier.

The Brenner Base Tunnel was here reported as a case study. The main benefit from the construction of this infrastructure should be considered the time reduction for the connections between the cities along the stretch of the Munich-Verona high speed line at a lower average pollution rate. In mere terms of carbon dioxide, however, the results reveal that the effect of the tunnel may be strengthened if added to an adequate policy in favour of the railway and that the amounts of CO2 produced should not be considered by themselves but in connection with their relative travel demand.

Though it probably needs to be further refined and some of its limitations need to be better understood, the necessary changes having been made, the method described herein is considered to be fully applicable throughout transportation planning. In fact, it may cover all the cases of implementation of new systems, changes in existing systems, or construction of new infrastructures. Its hypotheses are sufficiently flexible, and the number of equations and assumptions used was kept as small as possible. The model only requires that input data be adequately detailed in terms of vehicle shares, fuel consumption per vehicle and distance travelled – a result that can be achieved from planners, decision-makers and stakeholders by allocating adequate resources for data collection.


  1. 1.

    The CO2 emission factors χh for final energy take into account the emissions related only to the energy quantified by the meter, whereas the emission factors χh for primary energy take into account also the share of emissions relating to the production, transportation of the energy source considered as well as the production, use and maintenance of the facilities used for its exploitation.

  2. 2.

    The CO2 emission factors φ k for primary energy (also referred to as “grey energy”) take into account the share of emissions generated by the entire process to produce the materials considered, spanning from the extraction of the raw materials to their disposal.

  3. 3.

    In detail, three main fields are considered among the socioeconomic tendencies: the division in different classes of age of the overall population, the gross domestic product and the import/export business. The social aspects involve mostly the life style of the families and its changes in the next years. The main differences among the hypothesis introduced are related to the transportation field: different market regulations, fiscal and infrastructural policies, prescriptions and bans.

  4. 4.

    In order to ease the comprehension of the modal choice, the values of the costs in the scenarios analyzed in this article are here provided. In “trend” scenario, the values are as follows: for the goods, the distance costs are 34.10 € for every freight train and kilometre, reduced to 31.90 € from 2015, and 0.57 € for every HGV; time costs are, respectively, 68.20 and 34.10 €. This value considers a type-train of 420 t up to 2015 and of 500 t train from 2015; HGVs are divided in 10 classes, considering the different goods transported; the capacity is supposed stable up to 2035. For the passengers, the distance costs are currently 0.085 € for every train passenger and 0.10 € for every car passenger; time costs are 9.15 € both for rail and road. In “consensus” scenario, the values are as follows: for the goods, the distance costs are the same as in “trend” scenario for rail; only the capacity of the type freight train is supposed to change, being 425 t up to 2015, 550 t in 2015 and 605 t in 2025. For road freight transportation, costs increase to 0.68 € for every HGV from 2015 to 0.75 € from 2025. The capacity as well as the time costs are the same as in the “trend” scenario.

  5. 5.

    In 2021, CO2 emissions are supposed to be 1,696 kt for the “trend” and 1,542 for the “consensus” scenario – it makes a difference of 144 kt. In 2035 this difference rises up to 371 kt–2,176 kt in “trend” and 1,805 kt in “consensus” scenarios.


  1. 1.

    Aiscat (2009) Aiscat informazioni. Quarterly Report Online at: [01-02-2011]

  2. 2.

    Barbier-Saint-Hilaire F et al (1999) TRIBUT – a Bicriterion approach for equilibrium assignment. PTV AG, Karlsruhe

    Google Scholar 

  3. 3.

    BBT SE (2008a) Technische Projektaufbereitung - Elaborazione tecnica del progetto. Online at: [26.05.2011]

  4. 4.

    BBT SE (2008b) Potenziamento asse ferroviario Monaco – Verona. Galleria di Base del Brennero. Progetto definitivo. Relazione tecnica [29-03-2008]

  5. 5.

    BBT SE (2009) Il sistema della Galleria di Base del Brennero. Online at: [02-03-2011]

  6. 6.

    Black WR (2010) Sustainable transportation: issues and solutions. The Guilford Press, New York and London

    Google Scholar 

  7. 7.

    Bollen J, van der Zwaan B, Eerens HC, Brink C (2009) Local air pollution and global climate change: a combined cost-benefit analysis. Resour Energy Econom 31:161–181

    Article  Google Scholar 

  8. 8.

    Booz Allen Hamilton Ltd (2007) Estimated carbon impact of a new north-south line. Research report. Online at: [10-10-2011]

  9. 9.

    Börjesson PII (1996) Emissions of CO2 from biomass production and transportation in agriculture and forestry. Energ Convers Manag 37(6–8):1235–1240

    Article  Google Scholar 

  10. 10.

    Bundesministerium für Verkehr, Innovation und Technologie (BMVIT) (2008) Lageplan Gesamtstrecke der Achse München-Verona

  11. 11.

    Campos Gouvea VB, Ramos RAR, de Miranda D, Correia S (2009) Multi criteria analysis procedure for sustainable mobility evaluation in urban areas. J Adv Transp 43(4):371–390

    Google Scholar 

  12. 12.

    Contaldi R, Ilacqua M (2003) Analisi dei fattori di emissione di CO2 dal settore dei trasporti, APAT report 28/2003

  13. 13.

    de Carvalho Macedoa I (1998) Greenhouse gas emissions and energy balances in bio-ethanol production and utilization in Brazil (1996). Biomass Bioenergy 14(1):77–81

    Article  Google Scholar 

  14. 14.

    European Environment Agency (EEA) (2010) The European environment state and outlook 2010 synthesis. Copenaghen, 2010. Online at: [13-10-2011]

  15. 15.

    EMEP/EEA (2009) Air pollutant emission inventory guidebook. Online at: [01-03-2011]

  16. 16.

    Forschungsstelle für Energiewirtschaft (FFE) (2009) Online at: [01-12-2009]

  17. 17.

    Gkatzoflias D, Ntziachristos L, Samaras Z (2007) Computer programme to calculate emissions from road transport - users manual. European topic centre on air and climate change. Online at: [01-03-2011]

  18. 18.

    Hansen J, Sato M, Kharecha P, Beerling D, Masson-Delmotte V, Pagani M, Raymo Maureen, Royer DL, Zachos JC (2008) Target atmospheric CO2: where should humanity aim? Open Atmos Sci J 2:pp 217–231. Online at: [13-10-2011]

  19. 19.

    Hidalgo I, Szabo L, Ciscar JC, Soria A (2005) Technological prospects and CO2 emission trading analyses in the iron and steel industry: a global model. Energy 30(5):583–610

    Article  Google Scholar 

  20. 20.

    INFRAS (2010) The Handbook Emission Factors for Road Transport (HBEFA). [15-01-2010]

  21. 21.

    Intergovernmental Panel on Climate Change (IPCC) (2006) 2006 IPCC guidelines for national greenhouse gas inventories, prepared by the national greenhouse gas inventories programme. In: Eggleston HS, Buendia L, Miwa K, et al. (eds) Published: IGES, Japan

  22. 22.

    Intergovernmental Panel on Climate Change (IPCC) (2007) Contribution of working groups I, II and III to the fourth assessment report of the intergovernmental panel on climate change. Climate Change 2007: Synthesis Report. Geneva

  23. 23.

    Italferr (2010) Rapporto di sintesi sulla emissione (rimozione) della CO2. Progetto Definitivo “Potenziamento linea Rho-Arona: tratta Rho – Gallarate”. Online at: [01-02-2011]

  24. 24.

    Keller M (2010) Handbook emission factors for road transport 3.1. Quick reference. Infras, Bern

  25. 25.

    Kiese R, Butterbach-Bahl K (2002) N2O and CO2 emissions from three different tropical forest sites in the wet tropics of Queensland, Australia. Soil Biol Biochem 34(7):975–987

    Article  Google Scholar 

  26. 26.

    Köne AC, Büke T (2010) Forecasting of CO2 emissions from fuel combustion using trend analysis. Renew Sustain Energ Rev 14(9):2906–2915

    Article  Google Scholar 

  27. 27.

    Konidari P, Mavrakis D (2007) A multi-criteria evaluation method for climate change mitigation policy instruments. Energy Policy 35–12:6235–6257

    Article  Google Scholar 

  28. 28.

    Kouloumpis V, Kouikoglou VS, Phillis YA (2008) Sustainability assessment of nations and related decision making using fuzzy logic. IEEE Syst J 2(2):224–236

    Article  Google Scholar 

  29. 29.

    Land Tirol (2009) Verkehrsentwicklung in Tirol 08/09. Online at: [01-02-2011]

  30. 30.

    McNally MG (2008) The four step model. Center for activity systems analysis. Institute of Transportation Studies, UC Irvine

  31. 31.

    Network rail (2009) Comparing environmental impact of conventional and high speed rail. Online at: [01-02-2011]

  32. 32.

    Nocera S, Cavallaro F (2011) Policy effectiveness for containing CO2 emissions in transportation. Proced Soc Behav Sci 20:703–713

    Article  Google Scholar 

  33. 33.

    Nocera S, Maino F, Wagner M (2010) Calculation of the energy consumption for the construction of a railway tunnel. In: Martin U. et al. (eds) Network for Mobility 2010 (Proceedings of the 5th International Symposium). Stuttgart: FOVUS, 2010-09-30

  34. 34.

    Nordhaus W (2007) The challenge of global warming: economic models and environmental policy. Online at: [13-10-2011]

  35. 35.

    Öko-Institute e V (2011) GEMIS - global emission model for integrated systems. Online: [01-03-2011]

  36. 36.

    Phillis YA, Grigoroudis E, Kouikoglou VS (2011) Sustainability ranking and improvement of countries. Ecol Econ 70(3):542–553

    Article  Google Scholar 

  37. 37.

    ProgTrans AG (2007) Aggiornamento della previsione sul traffico merci e passeggeri per il Brennero al 2015 e 2025. Research Report, Zürich, Switzerland. Online at: [13.10.2011]

  38. 38.

    Ruffini FV, Maino F, Nocera S (2010) Sostenibilità del tunnel di Base del Brennero in rapporto alle emissioni di CO2. Research Report, Bolzano, Italy

  39. 39.

    Sinha KC, Labi S (2007) Transportation decision making - principles of project evaluation and programming. Wiley, New York

    Google Scholar 

  40. 40.

    Smit R, Ntziachristo L, Boulter P (2010) Validation of road vehicle and traffic emission models. A review and meta-analysis. Atmos Environ pp 2943–2953

  41. 41.

    Song C (2006) Global challenges and strategies for control, conversion and utilization of CO2 for sustainable development involving energy, catalysis, adsorption and chemical processing. Catal Today 115:2–32

    Article  Google Scholar 

  42. 42.

    Stern N (2006) Stern review on the economics of climate change, UK Treasury. Online at: [22-06-2011]

  43. 43.

    Swiss Center for Life Cycle Inventories (2011) Ecoinvent database. Online at: [01-03-2011]

  44. 44.

    Terna (2009) Dati statistici sull’energia elettrica in Italia. Confronti internazionali. Online: [01-03-2011]

  45. 45.

    Thornes JE (1992) The impact of weather and climate on transport in the U.K. progress. Phys Geogr 16:187–208

    Article  Google Scholar 

  46. 46.

    Transport & Mobility Leuven (2007) TREMOVE model description. Online at: [21.06.2011]

  47. 47.

    Transport & Mobility Leuven (2010) TREMOVE – a policy assessment model to study the effects of different transport and environment policies on the transport sector for all European countries. [07-01-2009]

  48. 48.

    Transportation Research Board (TRB) (2008) Potential impacts of climate change on U.S. Transportation. Special Report 290. National Research Council, Transportation Research Board, Washington, DC

  49. 49.

    Tuchschmid M (2009) Carbon footprint of high-speeds railway infrastructure (Pre-study). Methodology and application of high speed railway operation of European railways. Online at: [01-02-2011]

  50. 50.

    Turkenburg W (1997) Sustainable development, climate change, and carbon dioxide removal. Energ Convers Manag 38(1):S3–S12

    Article  Google Scholar 

  51. 51.

    UN, United Nations (1998) Kyoto protocol to the United Nations framework convention on climate change. Online at: [22-06-2011]

  52. 52.

    Von Rozycki C, Koeser H, Schwarz H (2003) Ecology profile of the German high-speed rail passenger transport system, ICE. Int J LCA 8(2):83–91

    Article  Google Scholar 

  53. 53.

    World Bank (2010) World development report 2010: development and climate change. The international bank for reconstruction and development. The World Bank, Washington DC

    Google Scholar 

  54. 54.

    Worrell E, Price L, Martin N, Hendriks C, Ozawa Meida L (2001) Carbon dioxide emissions from the global cement industry. Annu Rev Energ Environ 26:303–329

    Google Scholar 

Download references


This research was developed at the European Academy in Bolzano and partly funded under grant No. D0363 by both the European Academy of Bolzano and BBT public limited company. A noticeable amount of data and values used in the article was provided by BBT-SE, whose support is here gratefully acknowledged.

Author information



Corresponding author

Correspondence to Silvio Nocera.

Rights and permissions

Open Access This article is distributed under the terms of the Creative Commons Attribution 2.0 International License (, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Reprints and Permissions

About this article

Cite this article

Nocera, S., Maino, F. & Cavallaro, F. A heuristic method for determining CO2 efficiency in transportation planning. Eur. Transp. Res. Rev. 4, 91–106 (2012).

Download citation


  • CO2 Emissions
  • Transportation
  • Heuristic
  • Brenner railway tunnel