145个国家针对全球变暖、空气污染和能源不安全的低成本解决方案-斯坦福大学.pdf

This journal is © The Royal Society of Chemistry 2022 Energy Environ. Sci., 2022, 15, 3343–3359 | 3343 Cite this: Energy Environ. Sci., 2022, 15, 3343 Low-cost solutions to global warming, air pollution, and energy insecurity for 145 countries† Mark Z. Jacobson, * Anna-Katharina von Krauland, Stephen J. Coughlin, Emily Dukas, Alexander J. H. Nelson, Frances C. Palmer and Kylie R. Rasmussen Global warming, air pollution, and energy insecurity are three of the greatest problems facing humanity. Roadmaps are developed and grid analyses are performed here for 145 countries to address these problems. The roadmaps call for a 100% transition of all-purpose business-as-usual (BAU) energy to wind-water-solar (WWS) energy, efficiency, and storage, ideally by 2035, but by no later than 2050, with at least 80% by 2030. Grid stability analyses find that the countries, grouped into 24 regions, can exactly match demand with 100% WWS supply and storage, from 2050–2052. Worldwide, WWS reduces end- use energy by 56.4%, private annual energy costs by 62.7% (from $17.8 to $6.6 trillion per year), and social (private plus health plus climate) annual energy costs by 92.0% (from $83.2 to $6.6 trillion per year) at a present-value cost of B$61.5 trillion. The mean payback times of the capital cost due to energy- and social-cost savings are 5.5 and 0.8 years, respectively. WWS is estimated to create 28.4 million more long-term, full-time jobs than lost worldwide and may need only B0.17% and B0.36% of world land for new footprint and spacing, respectively. Thus, WWS requires less energy, costs less, and creates more jobs than BAU. Sensitivity test indicate the following. Increasing district heating and cooling may reduce costs by allowing flexible loads to replace inflexible loads, thereby replacing electricity storage and overgeneration with low-cost heat storage. A battery cost that is 50% higher than in the base case increases mean overall energy costs by only 3.2 (0.03–14.5)%. Almost all regions need fewer hours of load shifting than assumed in the base case, suggesting that actual load shifting may be easier than assumed. Increasing the use of electricity for hydrogen fuel-cell-electric vehicles instead of for battery-electric vehicles increases overall cost in most regions tested, due to the greater efficiency of battery-electric vehicles, but decreases overall cost in some regions by improving grid stability. Finally, shifting battery vehicle charging from day-night to mostly day charging reduces cost in the regions tested; shifting to mostly night charging increases cost. Ninety-five percent of the technologies needed to implement the plans proposed are already commercial. Broader context The world is undergoing a transition to clean, renewable energy to reduce air pollution, global warming, and energy insecurity. To minimize damage, all energy should ideally be transitioned by 2035. Whether this occurs will depend substantially on social and political factors. One concern is that a transition to intermittent wind and solar will cause blackouts. To analyze this issue, we examine the ability of 145 countries grouped into 24 regions to avoid blackouts under realistic weather conditions that affect both energy demand and supply, when energy for all purposes originates from 100% clean, renewable (zero air pollution and zero carbon) Wind-Water-Solar (WWS) and storage. Three-year (2050–52) grid stability analyses for all regions indicate that transitioning to WWS can keep the grid stable at low-cost, everywhere. Batteries are the main electricity storage option in most regions. No batteries with more than four hours of storage are needed. Instead, long-duration storage is obtained by concatenating batteries with 4 hour storage. The new land footprint and spacing areas required for WWS systems are small relative to the land covered by the fossil fuel industry. The transition may create millions more long-term, full-time jobs than lost and will eliminate carbon and air pollution from energy. 1. Introduction Global warming, air pollution, and energy insecurity remain three of the greatest problems facing the world. The Earth’s Dept. of Civil and Environmental Engineering, Stanford University, Stanford, California 94305-4020, USA. E-mail: jacobson@stanford.edu † Electronic supplementary information (ESI) available. See DOI: https://doi.org/ 10.1039/d2ee00722c Received 4th March 2022, Accepted 9th June 2022 DOI: 10.1039/d2ee00722c rsc.li/ees Energy reliance on centralized power plants and refineries; reliance on the need for a continuous supply of fuel that is subject to disruption arising from international war, civil war, embargos, bans, and labor disputes; and environmental damage due to continuous and widespread fuel mining and pollution. 4,5 It is postulated here that a transition entirely to a clean, renewable wind-water-solar (WWS) electricity, heat, storage, transmission, and equipment system (Fig. S1, ESI†) will sub- stantially reduce or eliminate these three problems and at low cost. Given their severity and their rapid growth, these problems must be addressed quickly. Ideally, 80% of the problems will be solved by 2030 and 100%, by 2035–2050 (Section 3.1). Given the goals of addressing air pollution and energy insecurity simultaneously with global warming, the transition must also avoid emissions of air pollutants and improve energy security. For these reasons, we do not include carbon capture (CC), direct air capture (DAC), bioenergy (B), nuclear power (N), or blue hydrogen (BH). Such technologies either increase or hold constant fuel mining (CC, DAC, BH), increase or hold constant air pollution (CC, DAC, B, BH), reduce little CO 2 while locking in combustion pollution (CC, DAC, B, BH), are costly (CC, DAC, N), have long time lags between planning and operation (N), or carry meltdown, weapons proliferation, waste, and mining risks (N). 4 Given that eliminating 80% of all emissions by 2030 and 100% by 2035–2050 with WWS, without these technologies, avoids 1.5 1C warming (Section 3.1), such non-WWS technologies are also not needed. Many research groups have examined 100% renewable energy (RE) systems in one or all energy sectors and have found that RE systems keep the grid stable at low cost. 6–39 Most closely related to this study, are studies to transition 139 countries 21 and 143 countries 25,35,36 to 100% WWS across all energy sectors while keeping the grid stable. All energy sectors include electricity, transportation, building heating/cooling, industry, agriculture-forestry-fishing, and the military. This study, which examines a transition of 145 countries, improves upon the previous studies in several respects. First, two additional countries (Lao, PDR and Equatorial Guinea) are included beyond the 143-country studies. For grid stability analysis purposes, the 145 countries are grouped into 24 regions (Table 1), as in the 143-country studies. Second, raw end-use energy consumption data for each sector in each country originate here from 2018 (the latest update) 40 rather than from 2016 25,35,36 or 2012. 21 Similarly, new cost data for electricity generation, storage, and installed name- plate capacities are used. The new costs, in particular, are lower than were the previous costs for several WWS technologies. Third, a significant unique feature of this study is the calculation and use of building heating and cooling loads worldwide every 30 seconds for a full three years, 2050–2052. The loads are calculated consistently with wind and solar generation in each country using a weather prediction/climate model. In the previous base studies, 25,40 such loads were estimated from daily heating and cooling degree day data. Fourth, four-hour batteries are concatenated here to provide both long-duration electricity storage and substantial instantaneous peaking power. Because battery costs have dropped dramatically and because four-hour batteries are now readily available, it is now justifiable to include a larger penetration of batteries than in the previous studies. Fifth, five new sensitivity tests are performed. In one, the fraction of district heating and cooling is increased in the most expensive regions, which are mostly small countries and islands, to examine the impact of increasing district heating and cooling on the cost of keeping the grid stable. In the second, the percent increases in the levelized and annual costs of energy are estimated when battery costs are 50% higher than those assumed in the base case. This sensitivity test is important because future battery costs are expected to drop but are uncertain, and a large share of electricity storage here is battery storage. In the third test, the maximum number of hours needed to shift a flexible load forward in time is reduced from a baseline value of eight hours to see how many hours of load shifting are actually needed in each region. If the maximum time needed is less than eight hours, then implementing demand response should be easier than proposed here. In the fourth test, the cost of increasing the penetration of electrolytic hydrogen fuel-cell-electric vehicles at the expense of battery- electric vehicles is examined. Finally, the cost of constant day and night versus mostly day versus mostly night battery-electric vehicle charging is examined. 2. Methodology WWS electricity-generating technologies include onshore and offshore wind turbines (Wind); tidal and wave devices, geothermal electric power plants, and hydroelectric power plants (Water); and Paper Energy (2) Estimate the 2050 reduction in demand due to electrifying or providing direct heat for each fuel type in each sector in each country and providing the electricity and heat with WWS (Note S2, ESI†); (3) Perform resource analyses, then estimate mixes of wind- water-solar (WWS) electricity and heat generators required to meet the total demand in each country in the annual average (Note S2, ESI†); (4) Use a prognostic global weather-climate-air pollution computer model (GATOR-GCMOM), which accounts for com- petition among wind turbines for available kinetic energy, to estimate wind and solar radiation fields and building heat and cold loads every 30 seconds for three years in each country (Note S3, ESI†); (5) Group the 145 countries into 24 world regions and use a computer model (LOADMATCH) to match variable energy demand with variable energy supply, storage, and demand response (DR) in each region every 30 seconds, from 2050 to 2052 (Notes S4–S6, ESI†); (6) Evaluate energy, health, and climate costs of WWS vs. BAU (Note S7, ESI†); (7) Calculate land area requirements due to WWS energy generators (Note S8, ESI†); (8) Calculate changes in WWS versus BAU employment numbers (Note S9, ESI†); and. (9) Discuss and evaluate uncertainties (Main text). In summary, three types of models are used: a spreadsheet model (Steps 1–3, Note S2, ESI†), a 3-D global weather-climate- air pollution model (GATOR-GCMOM) (Step 4, Note S3, ESI†), and a grid integration model (LOADMATCH) (Steps 5–8, Notes S4–S6, ESI†). With regard to the spreadsheet calculations, 2018 end-use BAU energy consumption from IEA 40 is first projected to 2050 for each country. End-use energy differs from primary energy (Note S2, ESI†). IEA provides end-use energy data for each of seven fuel types (oil, natural gas, coal, electricity, heat for sale, solar and geothermal heat, and wood and waste heat) in each of six energy sectors (residential, commercial, transportation, industrial, agriculture-forestry-fishing, and military-other) in each of 145 countries. These countries represent over 99.7% Table 1 The 24 world regions comprised of 145 countries treated in this study Region Country(ies) within each region Africa Algeria, Angola, Benin, Botswana, Cameroon, Congo, Democratic Republic of the Congo, Coˆte d’Ivoire, Egypt, Equatorial Guinea, Eritrea, Ethiopia, Gabon, Ghana, Kenya, Libya, Morocco, Mozambique, Namibia, Niger, Nigeria, Senegal, South Africa, South Sudan, Sudan, Tanzania, Togo, Tunisia, Zambia, Zimbabwe Australia Australia Canada Canada Central America Costa Rica, El Salvador, Guatemala, Honduras, Mexico, Nicaragua, Panama Central Asia Kazakhstan, Kyrgyz Republic, Pakistan, Tajikistan, Turkmenistan, Uzbekistan China China, Hong Kong, Democratic People’s Republic of Korea, Mongolia Cuba Cuba Europe Albania, Austria, Belarus, Belgium, Bosnia-Herzegovina, Bulgaria, Croatia, Cyprus, Czech Republic, Denmark, Estonia, Finland, France, Germany, Gibraltar, Greece, Hungary, Ireland, Italy, Kosovo, Latvia, Lithuania, Luxembourg, Macedonia, Malta, Mol- dova Republic, Montenegro, Netherlands, Norway, Poland, Portugal, Romania, Serbia, Slovakia, Slovenia, Spain, Sweden, Switzerland, Ukraine, United Kingdom Haiti Dominican Republic, Haiti Iceland Iceland India Bangladesh, India, Nepal, Sri Lanka Israel Israel Jamaica Jamaica Japan Japan Mauritius Mauritius Mideast Armenia, Azerbaijan, Bahrain, Iran, Iraq, Jordan, Kuwait, Lebanon, Oman, Qatar, Saudi Arabia, Syrian Arab Republic, Turkey, United Arab Emirates, Yemen New Zealand New Zealand Philippines Philippines Russia Georgia, Russia South America Argentina, Bolivia, Brazil, Chile, Colombia, Curacao, Ecuador, Paraguay, Peru, Suriname, Trinidad and Tobago, Uruguay, Venezuela Southeast Asia Brunei Darussalam, Cambodia, Indonesia, Lao PDR, Malaysia, Myanmar, Singapore, Thailand, Vietnam South Korea Republic of Korea Taiwan Taiwan United States United States Energy direct heat from solar thermal; and building heat and cold loads. Heat and cold loads are derived from modeled outdoor air temperatures, a specified indoor tempera- ture, and assumptions about building areas and U-values 35 (Note S3, ESI†). From the wind data, time-dependent wave power output is also derived. The time-dependent data from the file are then input into LOADMATCH 16,21,25,35,36 (Notes S4–S6, ESI†), which simulates the matching of energy demand with supply and storage over time. LOADMATCH is a trial-and-error simulation model. It works by running multiple simulations for each grid region, one at a time. Each simulation advances forward one timestep at a time, just as the real world does, for any number of years that sufficient input data are available for. The main constraint is that the sum of the electricity, heat, cold, and hydrogen loads plus losses, adjusted by demand response, must equal energy supply and storage during every timestep of the simulation. If load is not met during any timestep, the simulation stops. Inputs (either the nameplate capacity of one or more genera- tors; the peak charge rate, peak discharge rate, or peak storage capacity; or characteristics of demand response) are then adjusted one at a time based on an examination of what caused the load mismatch (hence the description ‘‘trial-and-error’’ model). Another simulation is then run from the beginning. New simulations (usually less than 10) are run until load is met