Development of Bottom-up Representation of Industrial Energy Efficiency Technologies in Integrated Assessment Models for the Iron and Steel Sector

Adoption of efficient end-use technologies is one of the key measures for reducing greenhouse gas (GHG) emissions. How to effectively analyze and manage the costs associated with GHG reductions becomes extremely important for the industry and policy makers around the world.

Energy-climate (EC) models are often used for analyzing the costs of reducing GHG emissions for various emission-reduction measures, because an accurate estimation of these costs is critical for identifying and choosing optimal emission reduction measures, and for developing related policy options to accelerate market adoption and technology implementation. However, accuracies of assessing of GHG-emission reduction costs by taking into account the adoption of energy efficiency technologies will depend on how well these end-use technologies are represented in integrated assessment models (IAM) and other energy-climate models.

In this report, we first conduct a brief review of different representations of end-use technologies (mitigation measures) in various energy-climate models, followed by problem statements, and a description of the basic concepts of quantifying the cost of conserved energy including integrating no-regrets options. According to IPCC (2001), no-regrets opportunities for GHG emissions reduction are the options whose benefits such as reduced energy costs and reduced emissions of local or regional pollutants equal or exceed their costs to society, excluding the benefits of avoided climate change. In this report, a no-regrets option is defined as a GHG reduction option (i.e., via energy efficiency measure) that is cost effective over the lifetime of the technology compared with a given energy price, without considering benefits of avoided climate change. There are two types of treatments of no-regrets options: 1) options that include other benefits, e.g., reduced operational and maintenance costs and productivity benefits; and 2) options that exclude other benefits. Although existence of no-regret options is not acknowledged by some economists, a number of cost-effective measures were identified in the U.S. iron and steel sector, regardless whether or not other benefits are included. There are many factors, including market barriers and knowledge gap, which contribute to slower adoption of such measures in the markets.

Based upon reviews of literature and technologies, we develop information on costs of mitigation measures and technological change. These serve as the basis for collating the data on energy savings and costs for their future use in integrated assessment models. In addition to descriptions of the iron and steel making processes, and the mitigation measures identified in this study, the report includes tabulated databases on costs of measure implementation, energy savings, carbon-emission reduction, and lifetimes.

Through characterizing energy-efficiency technology costs and improvement potentials, we have developed and presented energy cost curves for energy efficiency measures applicable to the U.S. iron and steel industry for the years 1994 and 2002. The cost curves can change significantly under various scenarios: the baseline year, discount rate, energy intensity, production, industry structure (e.g., integrated versus secondary steel making and number of plants), efficiency (or mitigation) measures, share of iron and steel production to which the individual measures can be applied, and inclusion of other non-energy benefits. Inclusion of other non-energy benefits from implementing mitigation measures can reduce the costs of conserved energy significantly. In addition, costs of conserved energy (CCE) for individual mitigation measures increase with the increases in discount rates, resulting in a general increase in total cost of mitigation measures for implementation and operation with a higher discount rate. As all the cost data (U.S. dollars) are obtained and presented as the currency values for the respective reference years (i.e., 1994, 2002), a direct comparison of costs (U.S. dollars), when desired, can be made by converting the existing reference-year data (i.e., 1994, 2002 in this study) to a preferred reference year (e.g., 2007). The conversions may be accomplished by multiplying the existing cost numbers represented in a reference year by an inflation index based on Gross Domestic Product (GDP) for the preferred year (BEA 2009).

The cost curve data on mitigation measures are available over time, which allows an estimation of technological change over a decade-long historical period. In this study, we compared the same set of mitigation measures for both 1994 and 2002. No additional mitigation measures for year 2002 were included due to unavailability of such data. Based upon the available data and cost curves, the rate of change in the savings potential at a given cost can be evaluated and be used to estimate future rates of change that can be the input for energy-climate models.

In 1994, integrated steel mills in the U.S. produced 55.4 Mt steel and secondary steel mills produced 35.9 Mt steel, for a total of 91.3Mt steel production in the United States (IISI 1994). Primary energy use for integrated steel making was 1,444 petajoules (PJ), over three times the energy use in secondary steel making, which was 426 PJ. The total carbon emissions from steel making related to energy use in 1994 were 34.3 MtC, with 78% of these emissions from integrated steel making (26.9 MtC) and the rest (22%) from the secondary steel making. In 2002, integrated steel mills in the U.S. produced 50.1 Mt steel and secondary steel mills produced 50.8 Mt steel, for a total of 100.9 Mt annual steel production. Primary energy use for integrated steel making was 1115 PJ, about twice of the energy use in secondary steel making, which was 519 PJ. The total carbon emissions from steel making related to energy use in 2002 was 30.6 MtC, with 71% of these emissions from integrated steel making (21.9 MtC) and the rest (29%) from the secondary steel making. We calculated that from 1994 to 2002 the steel production energy intensity has decreased by 15% and 14% for integrated steel and secondary steel, respectively indicating efficiency technology uptakes for both sectors over the period of time. In addition, the production shift from integrated steel to much less energy intensive secondary steel, in combination with the observed technology uptakes, resulted in an overall reduction in energy intensity by 21% for the U.S. iron and steel industry from 1994 to 2002.

We estimated that the potential savings of final energy use resulting from applicable mitigations measures was 397 PJ in 1994 (287 PJ for integrated steel making, and 110 PJ for secondary steel making), and 304 PJ in 2002 (223 PJ for integrated steel making, and 81 PJ for secondary steel making). The potential annual energy savings corresponded to 25% and 24% of total annual final energy use in the U.S. iron and steel sector in 1994 and 2002, respectively.

We have identified a number of cost-effective mitigation measures in this study. Furthermore, inclusion of other benefits from implementing mitigation measures can reduce the costs of conserved energy significantly, making more measures cost-effective. Using the final energy price of US$2.59/GJ in 1994 and US$3.49/GJ in 2002, a number of measures are identified to be cost-effective in this study when including non-energy benefits. We estimated that the potential savings of final energy use resulting from the cost-effective mitigations measures was 251 PJ in 1994 (186 PJ for integrated steel making, and 65 PJ for secondary steel making), and 217 PJ in 2002 (144 PJ for integrated steel making, and 73 PJ for secondary steel making). Overall, implementing applicable cost-effective mitigation measures could result in potential final energy savings by 16% and 17% of the total annual final energy use in the U.S. iron and steel sector in 1994 and 2002, respectively.

We also estimated overall potentials in carbon-emission reductions due to mitigation measures for both years (1994 and 2002), respectively. In this study, we have developed and defined the concept of cost curves for carbon reduction associated with the mitigation measures. The potential reduction of carbon emissions resulting from the applicable mitigation measures was 6.1 million ton of carbon (MtC) in 1994 (3.9 MtC from integrated steel making, and 2.2 MtC from secondary steel making), and 5.7 MtC in 2002 (3.7 MtC from integrated steel making, and 2.0 MtC from secondary steel making), corresponding to 18% and 19% of annual energy-related carbon emissions in 1994 and 2002, respectively. Applying cost-effective measures would reduce carbon emissions by 4.7 MtC in 1994 (3.4 MtC from integrated steel making, and 1.3 MtC from secondary steel making), and 4.4 MtC in 2002 (2.7 MtC from integrated steel making, and 1.7 MtC from secondary steel making), corresponding to approximately 14% of annual energy-related carbon emissions in each year.

We have also concluded that based upon the cost curves derived from available information on mitigation measures for both years, the rate of change in the energy-savings or carbon-reduction potential at a given cost can be evaluated and be used to estimate future rates of change for input in energy-climate models. Accuracies of such estimation of the rate change may be improved as more comprehensive information on characterizing the mitigation measures becomes available. Implementing existing cost effective measures can result in significant energy savings and carbon-emission reduction for both years relative to their technical potential in energy savings and carbon-emission reduction. In addition, total costs of conserved energy increase with the increases in discount rates. The outcomes from this research provide information on initial technology database that can be accessible to integrated assessment modeling groups seeking to enhance their empirical descriptions of technologies.

While many energy efficiency technologies have become cost-effective to mitigate long-term climate change, it is important and necessary to continue to incorporate new information on technology characteristics, and their evolution and response to energy and carbon price into various integrated assessment models to enhance empirical descriptions of the technologies, e.g., econometric models, service demand models, discrete choice models, or computational general equilibrium (CGE) models.

There appears to be a need to develop and refine sectoral algorithms and produce databases that can be used to match the needs of different integrated assessment modeling of climate policies. New algorithms should allow transformation of information on behavioral responses, technology costs, energy savings, other benefits, and policy costs into meaningful and functional data forms. Developing such algorithms may require customization and automation of database functions that would account for many variables. Furthermore, the desired data-model linking effort will require close interfaces between modelers and the developers of the cost-curve databases on energy efficiency measures. Future efforts should also include additional business sectors.