2017/81 k nKonw A A weldegdeg e ol n oNtoet e s eSrei r e ise s f ofro r p r&a c t hteh e nEenregryg y Etx itcrea c t i v e s G l o b a l P r a c t i c e The bottom line The Effects of Carbon Limits on Electricity Generation Integrated planning analysis can provide evidence to support and Coal Production: An Integrated Planning Approach decision making and consensus building about sensitive issues Applied to Poland in the energy sector. To support analysis, an optimization model Why is this issue important? can be quantified or modeled, systemwide estimates that show the imputes costs to a range of relative effects of closing different facilities make it easier for policy pertinent activities (capital and Essential efforts to reduce carbon emissions have makers to explain their decisions to citizens. operating costs of power plants, local consequences The necessary analysis must simultaneously consider wider operating and closing costs options in the electricity and mines sectors—including increased When 175 states signed on to the Paris Agreement on Climate for mines, emission control shares of gas and renewables, profitability, staffing, transportation Change in 2016, they set in motion a process that, for most, affects costs), thereby defining a set of fuel from mines and wells to power plants, imports of power and the way electricity is produced.1 For countries that purchase coal, of functions that can be solved coal from neighboring countries, and the installation of abatement oil, or other fuels for power generation, the effect may simply be a simultaneously to determine the equipment to meet emission standards. Such systems-level analyses decrease in import volumes. But for countries that export fossil fuels lowest system cost. are complex and data intensive, but they provide invaluable insights or use their own fossil-fuel resources for generation, reducing pro- duction may mean shutting down coal mines or oil wells. The natural on the impact of decisions. Debabrata question then becomes which facilities to shut down. The question Among the tools available for systemwide or multi-sectoral Chattopadhyay may be complex even when the power sector alone is considered; analysis are Times/Markal, Primes, Message, and Leap, but none is a senior energy produces results that are sufficiently granular to provide a full it becomes even more so when the extended effects of closures are specialist in the World understanding of specific choices. Bank’s Energy and taken into account. Extractives Global Practice (WB EEX). Mines and wells provide a livelihood for families; in some cases, whole communities are built around them. Haney and Shkaratan What is the key challenge? Jacek Filipowski is a consultant to the (2003) describe the social impact of mine closures in Romania, The lowest system cost must be determined and World Bank. Russia, and Ukraine; the effects include reduced employment, migration of labor, cuts in municipal and social services, and frayed presented in a transparent manner community ties. Perhaps the most effective way to illustrate the challenge, even Michael Stanley Because of these social effects, the question of mine closures at the technical level, is to break down the problem statement. is a lead mining specialist in the WB EEX. is often politically charged. In such contexts, where policy choices Consider the illustrative system in figure 1, where mines supply coal must be explained to citizens, evidenced-based decision making is to power plants, which, together with other plants fueled by gas or extremely important. Although not all of the implications of a closure renewables, must meet growing demand within specified emission Samuel Oguah targets. Several related factors must be considered: is an energy specialist in 1 http://newsroom.unfccc.int/paris-agreement/175-states-sign-paris-agreement/. the WB EEX. The Effects of Carbon Limits on Electricity Generation and Coal Production: Poland 2 T h e E f f e c t s o f C a r b o n L i m it s o n El e c t r i c it y G e n e r a ti o n a n d C o a l P r o d u c ti o n : P o l a n d • Which mines currently supply fuel to the most efficient fleet of costs of plants, operating and closing costs for mines, emission plants? With the cost of transportation taken into account, are control costs, and so on), it is possible to define a set of functions these necessarily the cheapest mines to operate? that can be solved simultaneously to determine the lowest system • Does the equation change when the number of employees at cost. The analysis of mining costs may even be extended to include each mine is taken into account? Does the fact that some mines the cost of various options—for example, acquiring new technology Where policy choices must supply fuel to plants that must be kept in service (often these or combining operations of adjacent mines. be explained to citizens, are plants that provide heat as well as electricity) change the And this brings us to the second challenge: data requirements. equation? Data requirements for the optimization problem just posed can evidenced-based decision • What is the best way to meet emission targets? Is it more be extensive, especially since both the power and mining sectors are making is extremely cost-effective to install emission-control systems on existing implicated. In countries with good data-management systems, this important. Although not coal-fired plants or to replace coal with less-polluting but type of information is routinely aggregated and readily available. But all of the implications of a more-expensive technologies (such as gas and renewable in other countries, a great deal of effort may be required to collect energy)? What is the optimal mix of the two approaches? and collate it. The challenge is often not with the existence of the closure can be quantified • How does the imposition of emission targets change the ranking data per se but rather with the format (hard copy versus electronic, or modeled, systemwide of mines? or incompatible electronic files) in which data are available and, on estimates that show the occasion, their quality. Even—or especially!—in such cases, sys- relative effects of closing This optimization problem can be formulated as a cost-reduction tems-level analysis can advance the cause of building databases that function. By imputing costs to all activities (capital and operating can be beneficial to the sector. different facilities make it easier for policy makers to explain their decisions to Figure 1. Framework for analysis of a coal-supplied generating system in the presence of emission limits citizens. Coal mine 1 Coal mine 2 Coal mine N Make decisions on production levels and mine closures Decide on new capacity requirements Coal plant 1 Coal plant 2 Gas/renewables and generation technologies Meet generation requirement Emission controls/CCS Emission controls/CCS CCS Decide controls to be installed Meet emission targets CCS = carbon capture and storage 3 T h e E f f e c t s o f C a r b o n L i m it s o n El e c t r i c it y G e n e r a ti o n a n d C o a l P r o d u c ti o n : P o l a n d Figure 2. Poland’s generation mix, 1972–2014 180 Geothermal/solar/wind 160 Biofuels/waste Hydro Is it more cost-effective Natural gas to install emission-control 140 Oil Coal systems on existing coal- 120 fired plants or to replace coal with less-polluting TWh 100 but more-expensive 80 technologies (such as gas and renewable energy)? 60 What is the optimal mix of the two approaches? 40 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014 Source: IEA. How has the World Bank addressed the challenge generation. Gas generation, which has been on the rise, accounts for 2 percent of emissions. Emissions of local pollutants are also of integrated planning analysis? significant; they include 3.28 million tons of particulate matter and Poland provides an illustration more than a million tons of SO2. Reducing these emissions will entail cutting down on coal pro- Poland produces electricity and heat from aging plants (half are more duction (closing the most unproductive mines), developing alternative than 30 years old). Although coal production has been on the decline sources of generation (gas and renewables), grappling with the over the past 20 years, Poland’s 31 active mines still produce more unemployment caused by mine closures, and investing in equipment than 100 million metric tons of coal each year to generate electricity to control pollution and, potentially, to capture and store carbon. and heat (IEA 2011). These mines employ more than 80,000 people, Asked to help the government of Poland analyze options for and coal accounts for more than 80 percent of generation. reducing emissions, a World Bank team developed an integrated Although this state of affairs would seem to present an oppor- planning model to address the following questions related to coal, tunity to introduce newer, cleaner technologies, serious issues of heat, and mines through 2030: investment requirements, energy security, and mine employment • What are the present generation mix and associated coal must be dealt with (IEA 2011). These problems have long been recog- requirements for electricity and heat production (referred to nized, but to date only limited steps have been taken to modernize below as “business as usual”)? and decarbonize the sector through increases in wind, biomass, and • How would a carbon constraint affect the cost of supplying gas generation (figure 2). electricity and heat, the generation mix, coal production, and The country’s production of electricity and heat emits 166 million mine employment? tons of CO2 each year, 97 percent of which comes from coal-based • What would be the best ways to meet the imposed carbon limit? 4 T h e E f f e c t s o f C a r b o n L i m it s o n El e c t r i c it y G e n e r a ti o n a n d C o a l P r o d u c ti o n : P o l a n d Figure 3. Generation mix through 2030 under business as usual 160,000 140,000 120,000 The country’s production Municipal wastes Generation (GWh) Biogas of electricity and heat 100,000 Oil Biomass emits 166 million tons of 80,000 Power imports CO2 each year.... Reducing Gas 60,000 Coking coal these emissions will entail Imported coal 40,000 cutting down on coal Lignite Domestic coal production, developing 20,000 alternative sources of 0 generation, grappling with 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 the unemployment caused by mine closures, and investing in equipment to An optimization model was custom-built for this purpose by the in light of the resulting production level, the generation level of the control pollution. Bank Power System Planning Group. It draws on an integrated coal- downstream power station should be adjusted. power model that one of the authors developed for the Indian power Under business as usual, with demand stable, Poland’s gener- sector in the 1990s.2 The integrated coal-power model developed ation mix changes only gradually through 2030. The most notable for Poland determines (i) the dispatch of individual power stations change is an increase in gas generation, as coal plants are retired over the period, (ii) the optimal selection of new projects to meet under the least-cost plan and replaced by gas (figure 3). CO2 emis- demand and emissions constraints, and (iii) upstream production sions decrease slightly, from 166 million tons in 2017 to 144 million and closure decisions for individual mines. The upstream linkage tons in 2030. from power plants to mines is very useful in deciding how emissions Lowering annual emissions to 107 million metric tons annually of carbon and other pollutants can be effectively mitigated through by 2030—a 26 percent reduction from business as usual and a 40 a mix of fuel switching (e.g., coal to gas or coal to renewables) and percent reduction from the 2005 emission level (as per the Poland’s through associated decisions on mines. The model can optimize declared target3)—would increase the cost of the system by $2.4 coal production by capturing the productivity of individual mines billion, a 2.1 percent annual increase over 2016–30. For the years (expressed as the number of employees per ton of coal produced), 2017–20, the reduction is based on a Polish government report the minimum technical production level that must be maintained (KASHUE-KOBIZE 2010), and, for the rest of the period, on the linear to supply coal-fired plants, and the mine’s costs. It performs the trend needed to reach a level of emissions that is 40 percent lower optimization by taking into consideration the lowest economically than the 2005 level. feasible employment and production levels and determining how, 2 See Parikh and Chattopadhyay (1996). A variant of this model including emissions control 3 “Intended nationally determined contributions” for reductions in greenhouse gas emis- decisions was deployed for China Light and Power in Hong Kong in 2001/02 to decide on type sions were declared in advance of the 2015 United Nations Climate Change Conference in Paris and timing of installation of controls. CLP continues to use the model for fuel budgeting and in December 2015. dispatch analysis. 5 T h e E f f e c t s o f C a r b o n L i m it s o n El e c t r i c it y G e n e r a ti o n a n d C o a l P r o d u c ti o n : P o l a n d Figure 4. Generation mix if CO2 limits are observed over 2017–30 160,000 140,000 Municipal wastes 120,000 Although generation from Biogas Generation (GWh) Oil lignite is presently the 100,000 Biomass cheapest way for Poland 80,000 Power imports Gas to produce power, there is Coking coal 60,000 no way to avoid reducing it Imported coal Lignite 40,000 if near-term CO2 emission Domestic coal limits are to be met. 20,000 0 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 Because the agreed CO2 limit is not binding until 2022, Polish mines continue their production at full capacity, five close down by decision makers may recoil from the cost of meeting it ahead of time. 2030. In any event, if existing carbon commitments are scrupulously Mine employment stood at about 80,000 in 2016. With business observed, gas-fired units will have to play a much greater role in as usual, that level declines slowly to about 67,000 by 2030. Under the generation mix after 2023, replacing old lignite power plants the carbon-constrained scenario, however, it drops to 54,000—a drop (figure 4). Although generation from lignite is presently the cheapest of 20 percent of the workforce. way for Poland to produce power, there is no way to avoid reducing In summary, the model applied to Poland reveals that under a it if near-term CO2 emission limits are to be met. Under the car- business-as-usual scenario, with no emissions constraint, coal’s bon-constrained scenario, gas generation increases in the model share in generation and heating would likely fall from about 80 to almost 50 TWh in 2030 compared with 24 TWh under business percent in 2017 to 63 percent in 2030. Gas and renewables would as usual. Generation from renewables is unaffected, because the play a growing role, coming to account for about 30 percent of the economic case is the same in both scenarios, although biomass generation mix in 2030. generation increases slightly over the period from 4.5 TWh in the The cost of meeting the specified carbon-reduction target by business-as-usual case to 6.3 TWh with the carbon constraint. 2030 is relatively small: a 2.1 percent annual increase in overall Respecting carbon limits has a significant effect on coal mines. system costs. Under this scenario, coal’s share in generation would Under business as usual, annual coal production drops from 100 drop to 48 percent by 2030 and that of gas and renewables would million tons currently to 90 million tons by 2030 (figure 5a). Under rise to 43 percent. Lignite’s share would fall steeply from 28 percent the carbon-constrained scenario, however, the binding CO2 limit under business as usual case to below 10 percent. But these gains after 2022 drives down lignite production sharply, and overall coal would mean closing five coal mines and laying off about 20 percent production falls to 48 million tons in 2030 (figure 5b). While some of Poland’s remaining coal miners. 6 T h e E f f e c t s o f C a r b o n L i m it s o n El e c t r i c it y G e n e r a ti o n a n d C o a l P r o d u c ti o n : P o l a n d Figure 5. Coal production under two scenarios Make further a. Business-as-usual b. CO2 limit connections 120 120 Electricity Live Wire 2014/5. Heating 100 100 “Understanding CO2 Emissions from the Global Energy 80 80 Millions tons Millions tons Sector,” by Vivien Foster and Daron Bedrosyan. 60 60 40 40 Live Wire 2014/18. “Exploiting Market-Based Mechanisms 20 20 to Meet Utilities’ Energy Efficiency Obligations,” by 0 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 0 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 Jonathan Sinton and Joeri de Wit. Live Wire 2014/24. “Capturing and Storing Carbon: The World What is the message for our clients? References Bank’s Role,” by Nataliya Integrated planning analysis supports good Haney, Michael, and Maria Shkaratan. 2003. “Mine closure and its Kulichenko, Richard H. impact on the community: Five years after mine closure in Zechter, and Asad Ali Ahmed. decisions—and vital intersectoral coordination Romania, Russia, and Ukraine.” Social Development Department, Live Wire 2015/43. “Integrating Applying integrated planning analysis to a real case reveals its ability World Bank, Washington, DC. Climate Model Data into to support evidence-based decision making, which can be useful IEA (International Energy Agency). 2011. Poland—2011 Review. Power System Planning,” by in dialogues with stakeholders. Yet the analysis is just that: a tool Energy Policy for IEA Countries Series, Paris. Debabrata Chattopadhyay and for decision making; and it must be complemented by programs to KASHUE-KOBIZE. 2010. “Metodyka wraz z przykładowym obliczeniem Rhonda L. Jordan. spread awareness and to cope with the dislocations caused by the ‘limitu’ krajowej emisji gazów cieplarnianych dla Polski na lata decisions it underpins. 2013-2020.” KASHUE, Warsaw. Live Wire 2016/55. “Designing While the data requirements of this type of analysis can be heavy, Parikh, J., and D. Chattopadhyay. 1996. “A multi-area linear program- Effective National Programs the process of assembling those data fosters discussion among ming approach for analysis of economic operation of the Indian to Improve Industrial Energy sectors that should focus their efforts in pursuit of common national power system. IEEE Transactions on Power Systems 11(1): 52–58. Efficiency Programs,” by Feng goals. Collaboration in data gathering also improves the chances that http://ieeexplore.ieee.org/document/485985/?denied Lieu and Robert Tromop. the results of the ultimate analysis will be embraced by a wide range of stakeholders. We gratefully acknowledge the contributions to this work of Leszek Pawel Live Wire 2017/78. “Minerals Kasek and Ryszard Malarski in the World Bank’s office in Poland. and Metals to Meet the Needs of a Low Carbon Economy,” by Kirsten Lori Hund, Daniele La Porta, and John Drexhage.