highspeed passenger trains, intercity passenger
trains, regional passenger trains, urban passenger
trains, and goods trains. Early models were based
on empirical correlations between emissions, train
type, operating speed, and other simple variables
such as the distance between stops, and average
emission factors for train types [88, 89]. More
recent models allow the user to specify an operational matrix for a journey, from which energy
consumption and emissions can be calculated
[90, 91]. This approach is a broadly analogous to
that used in modal models for road transport. As
with road vehicles, passenger trains consume
energy for loads other than propulsion. These
include the heating and cooling loads, and electrical power to operate lights, instruments, and other
electrical accessories. However, the energy consumption for auxiliary power is fairly low compared with that required for propulsion. The
effects of regenerative braking on energy consumption may also be important.
Shipping Models
For shipping, both simple and detailed methods
have been developed for calculating emissions
[33, 92, 93]. The simple methods are essentially
similar to that for road transport (see Eq. 1), and
combine fuel use (in tonnes) with fuel-specific
emission factors. Detailed calculation approaches
firstly involve obtaining ship movement data,
including sailing routes and distances between
ports. Different ship operating modes may also
be considered, such as cruising, in-port operation
(hotelling, loading and unloading), and maneuvering. Each vessel also needs to be characterized.
National and international shipping covers a wide
range of vessels, from recreational craft to large
oceangoing cargo ships that are driven primarily
by large, low-, and medium-speed diesel engines.
In addition, marine engines can be further classified according to their principal fuel: bunker fuel
oil, marine diesel oil, or marine gas oil. Again,
auxiliary engines are used to provide power and
services when berthed [33]. In Europe, a widely
used set of emission factors was determined in a
study for the European Commission [92].
The fuel consumption and emissions of vessels
operating on inland waterways were initially
modelled by analogy to marine shipping. However, the operational characteristics of inland
ships are rather different from those of their seagoing counterparts. Firstly, they sail at lower
speeds and in restricted waterways. The lower
speeds reduce energy consumption and emissions, whereas shallower depths increase sailing
resistance and hence fuel consumption and emissions. In order to make the most of the limited
space on inland waterways, vessels tend to be
squarer in shape than seagoing ships. This also
causes the sailing resistance to differ from that of
seagoing ships [94]. Recent models are therefore
more specific to inland vessels and their operating
conditions. For example, in the ARTEMIS project, a model was developed and validated for nine
types of vessel based on size and type of waterway
[94]. The model calculates the energy consumption and emissions under different service conditions (speed, waterway dimensions, load, etc.).
Activity Data
For all transport modes the accurate characterization of traffic activity (such as numbers vehicles,
trip distances, and modes of operation) and the
fleet composition is vital to the estimation of
emissions. Although models and emission factors
are continually improving, poor activity data
remains one of the main sources of uncertainty
in the calculation of emissions [95]. For many
years, the direct measurement of road traffic has
been the primary source of activity data. The
collection of accurate information on important
parameters such as traffic speed, road gradient,
or vehicle load can often be difficult and timeconsuming, rendering it impractical for large geographic areas. However, the quality and availability of activity data has increased substantially in
recent years as a result of improved data collection
technologies (e.g., tag-and-beacon systems, cameras, in-vehicle data collection, and satellite technology) and online dissemination. Furthermore,
there have been significant advances in microsimulation traffic models, the outputs from
which are well suited to integration with instantaneous emission data. Some workers have linked
microscale traffic and emission models for practical applications [96].
Air Quality, Surface Transportation Impacts on
61
trains, regional passenger trains, urban passenger
trains, and goods trains. Early models were based
on empirical correlations between emissions, train
type, operating speed, and other simple variables
such as the distance between stops, and average
emission factors for train types [88, 89]. More
recent models allow the user to specify an operational matrix for a journey, from which energy
consumption and emissions can be calculated
[90, 91]. This approach is a broadly analogous to
that used in modal models for road transport. As
with road vehicles, passenger trains consume
energy for loads other than propulsion. These
include the heating and cooling loads, and electrical power to operate lights, instruments, and other
electrical accessories. However, the energy consumption for auxiliary power is fairly low compared with that required for propulsion. The
effects of regenerative braking on energy consumption may also be important.
Shipping Models
For shipping, both simple and detailed methods
have been developed for calculating emissions
[33, 92, 93]. The simple methods are essentially
similar to that for road transport (see Eq. 1), and
combine fuel use (in tonnes) with fuel-specific
emission factors. Detailed calculation approaches
firstly involve obtaining ship movement data,
including sailing routes and distances between
ports. Different ship operating modes may also
be considered, such as cruising, in-port operation
(hotelling, loading and unloading), and maneuvering. Each vessel also needs to be characterized.
National and international shipping covers a wide
range of vessels, from recreational craft to large
oceangoing cargo ships that are driven primarily
by large, low-, and medium-speed diesel engines.
In addition, marine engines can be further classified according to their principal fuel: bunker fuel
oil, marine diesel oil, or marine gas oil. Again,
auxiliary engines are used to provide power and
services when berthed [33]. In Europe, a widely
used set of emission factors was determined in a
study for the European Commission [92].
The fuel consumption and emissions of vessels
operating on inland waterways were initially
modelled by analogy to marine shipping. However, the operational characteristics of inland
ships are rather different from those of their seagoing counterparts. Firstly, they sail at lower
speeds and in restricted waterways. The lower
speeds reduce energy consumption and emissions, whereas shallower depths increase sailing
resistance and hence fuel consumption and emissions. In order to make the most of the limited
space on inland waterways, vessels tend to be
squarer in shape than seagoing ships. This also
causes the sailing resistance to differ from that of
seagoing ships [94]. Recent models are therefore
more specific to inland vessels and their operating
conditions. For example, in the ARTEMIS project, a model was developed and validated for nine
types of vessel based on size and type of waterway
[94]. The model calculates the energy consumption and emissions under different service conditions (speed, waterway dimensions, load, etc.).
Activity Data
For all transport modes the accurate characterization of traffic activity (such as numbers vehicles,
trip distances, and modes of operation) and the
fleet composition is vital to the estimation of
emissions. Although models and emission factors
are continually improving, poor activity data
remains one of the main sources of uncertainty
in the calculation of emissions [95]. For many
years, the direct measurement of road traffic has
been the primary source of activity data. The
collection of accurate information on important
parameters such as traffic speed, road gradient,
or vehicle load can often be difficult and timeconsuming, rendering it impractical for large geographic areas. However, the quality and availability of activity data has increased substantially in
recent years as a result of improved data collection
technologies (e.g., tag-and-beacon systems, cameras, in-vehicle data collection, and satellite technology) and online dissemination. Furthermore,
there have been significant advances in microsimulation traffic models, the outputs from
which are well suited to integration with instantaneous emission data. Some workers have linked
microscale traffic and emission models for practical applications [96].
Air Quality, Surface Transportation Impacts on
61