Aggregation
Based on data availability, comparability and sectoral variability, it is proposed that baselines be
constructed on the basis of energy intensity in the manufacture a ton of crude steel at the international
process routes level, i.e.:
1. Integrated Steel Plants (ISPs);
2. Scrap based Electric Arc Furnaces (scrap-EAF);
3. Direct Reduced Iron (gas-based) Electric Arc Furnaces (DRI (gas)-EAF); and
4. Direct Reduced Iron (coal-based) Electric Arc Furnaces (DRI (coal)-EAF).
No fuel separation is needed in categories 1 and 2 because in EAF the majority (more than 95%) of fuel
used is electricity and in Integrated Steel Plants it is almost all coal. However for EAFs, it is important to
examine the raw materials that used as iron-bearing materials (i.e. the proportion of scrap used) as these
have a significant impact on the energy use.
It is not possible to aggregate baselines across process routes as they differ significantly in the amount of
energy used to produce one ton of crude steel. Moreover, the different process routes use different
feedstocks, energy inputs and produce and different products. Each of the different routes are used in most
countries.
The DRI-EAF (coal-based) process route could be considered as an inefficient variation on the Integrated
Steel Plants concept, implying that GHG emission reduction projects from coal-based DRI-EAF should not
be credited at all. Therefore, this process route is excluded in the remainder of the report.
Since the technologies used for iron and steel production are internationally uniform, it is not necessary to
distinguish between geographical regions for technical reasons and standardised baselines for different
process routes may therefore be set up at an international (global) level. This level of disaggregation is also
in line with the recommendations of other studies on the suitability of multi-project baselines,
e.g. that
recommend using global, disaggregated benchmarks for commodities that are traded internationally and
products and feed stocks that are heterogeneous (Lazarus et. al., 1999),. However, the energy consumption
of the same technology in different places can differ substantially, reflecting the age of equipment, quality
of raw material inputs, the degree of integration of the process and the level of implementation of energy
efficient technologies.
It may be necessary to adjust for the specific characteristics of an individual plant or product mix within a
process route. For example, Phylipsen et al (1998), recommend correcting for the amount of hot-rolled and
cold-rolled steel produced.
In theory, standards could also be established at a higher degree of disaggregation. Thus, for example, it
would be possible to set standards for the different process steps within each process route, e.g. the energy
used in a blast furnace. However, such a disaggregation is not recommended as process integration is then
not rewarded as an option for emission reduction.
Setting baselines with still greater disaggregation (e.g. at the level of individual technology components
such as motors and pumps) would be extremely data intensive. Moreover, it would be difficult to assess the
COM/ENV/EPOC/IEA/SLT(2001)5
31
emission reductions from projects using this level of disaggregation, because it would require an
assumption regarding individual component use (e.g. how long. an individual motor runs7
.
Differences in SEC for the same technology operating in different places are generally caused by
differences in operation and maintenance, or the age of the technology. Other causes can be the quality of
coal and iron ore, which differs between countries.
3.3 Proposed baselines
As with baselines set in other sectors, ideal emissions baselines in the iron and steel sectors should meet
the following criteria (Ellis and Bosi, 1999):
• Be environmentally credible (to exclude projects that are not environmentally additional);
• Be transparent and verifiable by a third party;
• Be simple and inexpensive to draw up (low transaction costs/low costs for baseline development);
• Provide a reasonable level of crediting certainty for investors; and
• Have a potentially large volume of projects.
The variation in these criteria influences the number of projects that will be generated through JI and
CDM. For the suitability of the baselines that will be described in the following pages, the potential large
volume of projects is an important criterion.
Any baseline approach involves a trade-off between the criterion of environmental credibility and
generating a potential large volume of projects. The baselines proposed in the following sections will first
be described and then scored against these criteria.
Much of the following analysis mainly focuses on existing plants. New plants usually operate at a Specific
Energy Consumption that is approximately equal to the world wide best practice and the SEC of
worldwide best practice is thus considered the most suitable standardised baseline for new plants. This is
particularly true as a number of plants in developing countries have been built and operated at a SEC that is
at the world-wide best practice. However, exceptions do exist: for example, in the United States where it is
not considered economically viable to build new plants with the lowest specific energy consumption due to
low energy prices, lower efficiency plants are still being constructed. Thus, a baseline that incorporated
these new facilities into the average might yield higher crediting levels.
Corex© plants too operate at approximately equal efficiencies world-wide, with most plants that are
installed operate at a specific energy consumption near the best practice value. Thus, the best practice value
of 19 GJ/ton crude steel is proposed to be a suitable baseline, both for new and refurbishment Corex plants.
Benchmarking energy use
Emission baselines could also be developed based on standardised (or “benchmarked”) energy
performance of a plant. In such cases, country specific GHG emission factors would need to be applied to
7 Note that motors and lighting from such industries as iron and steel must be excluded from a general “motor
project” if there also is a iron and steel project to avoid double counting.
COM/ENV/EPOC/IEA/SLT(2001)5
32
the standardised energy value to calculate the accompanying GHG emissions reduction8
. Separating the
energy and GHG intensity of the process avoids the requirement to set default GHG intensity standards for
electricity generation, significantly simplifying the baseline setting process.
Several possible baselines based on the concept of benchmarking can be generated, such as average,
median or “better than average”. Average and “better than average” benchmarks can also be thought of in
terms of the stringency of a baseline. A standardised baseline based on the energy intensity of different
steel production routes would be very stringent if based on world-wide best practice values, but much more
lax if based on world-wide average performance.
A baseline based on the “national average” performance
could be more or less stringent than a world-wide average, depending on the country.
Benchmarking based on recent technology additions is another option and indeed, one that is
recommended in other sectors. However, for new plants, this option seems to be essentially the same as
benchmarking on the world-wide best practice level (see e.g. the latest new iron and steel plants in South
Korea) and so is not considered separately.
Five different possibilities for a standardised energy baseline for each of the production routes are
discussed below:
1) A baseline based on the best practice energy performance world-wide.
Assuming that the plant operating with the best practice energy performance world-wide indicates the level
above which environmental additionality is guaranteed, a baseline can be set at this level. In practice,
adjustments to this level will need to be made in order to account for the amount of steel that is hot-rolled
and cold-rolled (see Table 12) and for the quality of the coal and ore used. In such a case, only projects
using advanced technology would generate emission credits.
2) A baseline based on the world-wide average SEC per process route.
Using this value would lead to a relatively lax baseline as plants that are responsible for approximately half
of total iron and steel production world-wide would qualify as CDM/JI projects if the baseline “test” were
the only additionality criteria.
3) A baseline based on “better than average” performance
Such a baseline would be between the values of the world average and best practice performance baselines
outlined above. It would therefore also result in an intermediate level of stringency and an intermediate
level of project numbers. It could be drawn up at an arbitrary level, such as 90% of the “average” level.
4)
A baseline based on benchmarking on the country average SEC per production route
This option would result in baselines differentiated by country, as well as by production route. This type of
baseline is not consistent with aggregation purely at a production route level, but it would have the
advantage of being able to take into account widely differing national circumstances. As for baselines
based on a worldwide average, this type of baseline could be drawn up at different levels of stringency
(e.g. country average, 10% better than average).
5) A “graduated crediting” baseline
This alternative allows a percentage of the reductions to be counted - with the amount varied according to
the stringency of the benchmark used (see Figure 9). The values could be set to take into account both
8 The determination of the country specific emission factors is dependent on the developments in the power sector of
a country. The Electricity Case Study suggests how such baselines could be established.
COM/ENV/EPOC/IEA/SLT(2001)5
33
national performance and BAT performance. Such an approach is more stringent than one based on
national average figures (because not all emission reductions would be credited at that level), but would
allow a larger number of projects than a standard set at the BAT energy performance level.