** DRAFT – DO NOT CITE ** California’s Carbon Challenge: Scenarios for Achieving 80% Emissions Reduction in 2050 Max Wei1, Jeff Greenblatt1 James H. Nelson2, Ana Mileva2, Josiah Johnston2, Daniel M. Kammen2* 1Environmental Energy Technologies Division, Lawrence Berkeley National Lab 2 Renewable and Appropriate Energy Laboratory, University of California, Berkeley • Class of 1935 Distinguished Professor of Energy Project leader: to whom correspondence should be addressed (Kammen@berkeley.edu) 1.
EXECUTIVE SUMMARY This study provides an updated analysis of long term energy system scenarios for California consistent with the State meeting its 2050 climate targets, including detailed analysis and assessment of electricity system build-­‐out, operation, and costs across the Western Electricity Coordinating Council (WECC) region. Four key elements are critical for the State to achieve its 2050 target of 80% greenhouse gas emissions (GHG) reduction from the 1990 level: aggressive energy efficiency, clean electricity (synonymous here with deep reduction if not zero-­‐carbon intensity); widespread electrification of passenger vehicles, building heating, and industry heating; and large scale production of low-­‐
carbon footprint biofuels which are needed to largely replace petroleum-­‐based liquid fuels. The approach taken here is similar to a back casting approach where technically achievable measures are taken and aggregated to meet the 80% reduction target. For the purposes of analysis, the energy and non-­‐energy sector are each assumed to have the objective of meeting an 80% reduction from their respective 1990 GHG levels, but of course, a different partitioning of energy and non-­‐energy sector GHG reductions is allowed if emission reductions in one sector are more economic or technically achievable than another. Figure ES.1 shows overall emissions for the key economy-­‐wide scenarios considered in this report. All cases are compliant or nearly compliant with the 2050 target suggesting that with aggressive policy support and continued technology development of largely existing technologies, multiple technical pathways exist to achieve the target. Recent downward revisions in the State’s long term population will abet the State’s efforts as will continued strong federal and state policy support for improved light duty vehicle efficiency standards. 1 ** DRAFT – DO NOT CITE ** Figure ES2 shows the progression of energy-­‐sector GHG reductions by key element (efficiency, clean electricity, electrification, etc.) starting from a business as usual (BAU)1 case to the Base Case, while Figure ES3 shows the percentage contribution to GHG savings by sector and key element. Transportation efficiency, clean electricity, and electrification are seen to be the largest levers with almost 50% of energy-­‐sector GHG reduction from the transportation sector. This work also considers detailed sensitivity analysis for the base compliant case (Figure ES4) providing quantification on the impact of partial or more aggressive implementation of efficiency, light duty electrification, building and industry electrification and other key factors. In addition to the large role for transportation efficiency, key hinge factors for the State include: the amount of imported low carbon biofuels, the degree to which the electricity system is decarbonized, and the amount of building and industry electrification. For example, if any one of these elements “falters” and does not achieve the GHG reduction amount assumed in the base case, energy system emissions can increase by more than 40% over the target level (or from 77Mt to about 110Mt). The critical role of developing a robust and cost-­‐effective clean electricity supply system is also highlighted. Clean electricity enables a decarbonization path in the building sector through electrified heating (e.g., heat pump-­‐based water and space heating) and through the electrification of building heating it also enables GHG-­‐compliance even with a reduced level of building shell and insulation measures (Reduced or 50% energy efficiency case in Figure ES1). This reduced efficiency case assumes both clean electricity and electrified building heating, and if either of these measures is not in place, emissions would increase by about 50% (Figure ES4). 1 BAU refers to the case where existing efficiency levels for end uses such as appliances and vehicles are frozen at current levels. 2 GHG [Mt CO2-­‐eq]
** DRAFT – DO NOT CITE ** 100
90
80
70
60
50
40
30
20
10
0
Non-­‐Energy Emissions
Energy Emissions
2050 CARB Target
Figure ES.1. Overall GHG emissions for scenarios in this report. All cases are compliant or nearly compliant with the 2050 target (the -­‐20% carbon cap electricity case is within 1Mt of overall 85Mt CO2-­‐eq target and others are at or below it). 2050 GHG Emissions [Mt CO2-­‐eq]
350
300
Efficiency
250
200
150
100
50
0
Clean Electricity
Electrification
Biofuels
Oil-­‐Gas Industry Replacement
Remaining GHG, 2050
Figure ES.2. Progression of energy-­‐sector GHG savings in 2050 from the BAU case to the base case. Total energy-­‐sector emissions in 2050 for the resultant base case is 75Mt CO2eq. 3 ** DRAFT – DO NOT CITE ** 50%
BAU Case [%]
Overall GHG Savings from
60%
Oil-­‐Gas Industry Replacement
40%
Biofuels
30%
Electrification
20%
10%
Clean Electricity
0%
Efficiency
Figure ES.3. Percentage of overall energy-­‐sector GHG savings by sector for the base case in 2050. Note that electrification savings are predicated by availability of clean electricity. In this compliant scenario, transportation sector contributes almost 50% of overall savings. 2050 GHG [Mt CO2-­‐eq]
40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190
No Energy Efficiency in Bldg, Industry, Transport
Light Duty Vehicle Electrification (100%, Base, 0%)
+151%
-­‐31%
50% EE in Bldg, Industry, Transport
+72%
Lower imported biofuels (0.9 Bgge instead of 7.5 Bgge)
Electricity Carbon Cap at -­‐40%, 14%, 50% of 1990
Building, Industry Electrification: 100%, Base, 0%
Growth -­‐/+20%
Biofuels LCA Factor 0%, 20%, 40%
PEV Adoption -­‐/+ 10 years
LDV Conservation -­‐/+20% VMT and +/-­‐20% Annual Sales
Aggressive Lightweighting, Light Duty Vehicles
50% EE in Bldg, Industry +77%
+54%
-­‐36%
+52%
-­‐37%
+48%
-­‐28%
+27%
-­‐23%
+23%
-­‐19%
+20%
-­‐19%
+19%
-­‐16%
+3%
Figure ES.4. Sensitivity plot for base case. Each of these sensitivities represents a scenario with base case conditions except for the measure or measures as noted. For example, “50% EE in Bldg, 4 ** DRAFT – DO NOT CITE ** Industry, Transport” has the same degree of decarbonized electricity and electrification adoption as the base case. Earlier studies indicate the importance of the non-­‐energy sector in future GHG emission. A more detailed treatment of this sector’s components, anticipated growth, and mitigation options is provided in this work. Non-­‐energy emissions alone are projected to exceed the total GHG target in the 2050 frozen case (Figure ES.1). A base case for non-­‐energy sector emissions achieving 80% GHG reduction from 1990 includes net zero GHG hydrogen production in industry, aggressive livestock manure management and soil N2O management, and methane capture and combustion for waste (Figure ES.5). High global warming potential gases (high GWP) are the key contributor in this sector while afforestation can be a potential path to sequester carbon. A scenario whereby the State does not replace high GWP gases with low GWP substitutes will increase the amount of GHG in 2050 by 40 Mt-­‐CO2eq. Similarly, if afforestation is not pursued, then GHG would increase by 17Mt, or conversely, if double the amount of carbon sequestration from afforestation is achieved (from 6.6% of California’s rangeland in the base case to approximately 13% of rangeland), GHG emissions can be further reduced by 17Mt from the base case. Considering total GHG emissions as a capped quantity in 2050, Figures ES4 and ES6 represent overall economy-­‐wide sensitivities to the overall target in the context of active carbon trading across energy and non-­‐energy sectors. These sensitivity plots also underscore the difficulty of meeting the target with currently known technology since it may be difficult to achieve every measure shown to the full extent assumed in the base case. Many required elements must largely be in place by the 2030-­‐2040 time frame in order to either intercept retiring stocks of vehicles and/or appliances, or to provide adequate lead time for 2050 reductions to reach fruition e.g., afforestation, conservation. Moreover, each of the measures shown as “carbon negative” relative to the base case are quite challenging to achieve technically and at the scale required in this time frame (carbon negative electricity sector, carbon neutral biofuels, deep conservation in the passenger transportation sector, or full electrification of industry and residential heating for example). 5 ** DRAFT – DO NOT CITE ** 50
2050 BAU Non-­‐Energy Emissions
GHG Emissions [Mt CO2-­‐eq]
40
2050 Base Case Non-­‐Energy Emissions
30
20
10
0
-­‐10
-­‐20
Industrial Processes
Total Base Case Non-­‐
Energy Emissions in 2050
Figure ES.5. 2050 Non-­‐energy sector emissions for the frozen case and the base case. Total non-­‐
energy sector emissions for the Base case are 8.1 Mt CO2eq. High GWP (Non-­‐Ozone Depleting Substances)
Livestock
Agriculture and Forestry
Waste
6 ** DRAFT – DO NOT CITE ** 2050 Change in GHG from Base Case [Mt CO2-­‐eq]
-­‐20
-­‐10
0
10
20
30
40
50
No High GWP Substitutes
Afforestation (Double Amount vs No Afforestation)
No Manure Management
No Waste Capture
No Soil Treatment
No net zero hydrogen production
Figure ES.6. Sensitivity plot for non-­‐energy GHG emissions showing the increase or decrease in emissions from the compliant Base case. For example, not replacing high GWP gases with low GWP substitutes would increase emissions by 40 MMt-­‐CO2eq. in 2050. 7 ** DRAFT – DO NOT CITE ** 2.
ELECTRICITY EXECUTIVE SUMMARY [note – some of the values here will be updated with final results… the main expected difference is a few percent increase of non-­‐CCS natural gas in the generation mix by 2050] In this study we use a state-­‐of-­‐the-­‐art planning model for the electric power system – the SWITCH model – to investigate the evolution of the western North American power system (specifically WECC, the Western Electricity Coordinating Council) from present day to 2050 in the context of deep decarbonization of the energy system. As the cost of electricity is an important factor for the economic welfare of society, we employ a cost-­‐minimization framework, thereby simulating how policy goals, reliability requirements, and projected electricity demand might be met at the lowest possible cost. The power system is constrained to reach 14 % of 1990 CO2 emission levels by 2050 under a range of assumptions about future demand profiles, costs, policy mandates, technological availably, and electric system flexibility. The electricity system is of fundamental importance to the decarbonization of the entire energy system, as fuel-­‐switching away from oil and natural gas and towards electricity is a key decarbonization strategy. The scenarios presented here incorporate hourly electricity demand profiles resulting from the electrification of heating and vehicles, as well as from substantial energy efficiency. Even with aggressive energy efficiency measures, WECC-­‐wide electricity demand is found to increase by 75 % to 95 % between present day and 2050 (Figure 2-­‐1) due to population growth and additional demand from electric vehicles and electric heating. 240"
1.8"
1.6"
200"
1.4"
180"
and
Dem
160"
140"
r"co
Powe
120"
"in
rbon
st"(ca
"
d )"
clude
1"
"
r"cost
Powe cluded)"
n"ex
o
b
r
a
(c
100"
1.2"
80"
0.8"
0.6"
60"
0.4"
40"
0.2"
20"
0"
2010$
WECC$Electricity$Demand$(TWh/yr)$
Power$Cost$@$WECC$Average$($2013/MWh)$
220"
2020$
2030$
2040$
0"
2050$
Figure 2-­‐1: WECC average power cost and WECC average electricity demand by investment period in the Base Scenario. The power cost with ‘carbon included’ differs from that with ‘carbon excluded’ by the cost of carbon permits. The error bars represent the range of costs found in scenarios other than the Base Scenario. 8 ** DRAFT – DO NOT CITE ** We find drastic emission reductions in the power system by 2050 to be feasible under a wide range of possible futures. The WECC-­‐wide average cost of power in 2050 is found to range between $145/MWh and $232/MWh across scenarios, a 17 to 89 % increase relative to a business-­‐as-­‐usual scenario in which emissions stay flat after 2020, and a 34 to 114 % increase (in real terms) relative to the present day cost of power. As this study assumes little technological progress by default in many parts of the electricity system, these cost estimates are likely an upper bound. We demonstrate that breakthroughs in the cost of solar energy or the deployment of demand response could contribute greatly to containing the cost of electricity decarbonization. Between present day and 2030, the evolution of the WECC power system is dominated by the implementation of aggressive energy efficiency measures, installation of renewable energy facilities due to renewable portfolio standard (RPS) targets, and the replacement of coal-­‐fired generation with gas-­‐fired generation (Figure 2-­‐2). In the 2030 timeframe, the flexibility provided by the existing transmission network, existing hydroelectric facilities, the expansion of balancing areas, and a large fleet of gas-­‐fired generation units is largely sufficient to integrate 45 -­‐ 86 GW of wind and solar power capacity in WECC, representing 12 -­‐ 21 % of total electricity produced. Consequently, deployment of new storage or long-­‐distance, high voltage transmission capacity is not necessary at multi-­‐GW scale in this timeframe unless efficiency implementation is relatively slow. Transmission capacity into California, made available in part by the retirement of out-­‐of-­‐state coal generation, is dominated by renewable power in the form of bundled Renewable Energy Certificates (RECs) in the 2030 timeframe. The cost of power stays almost constant until 2030 due to moderate gas prices, the expiration of existing generator sunk costs, and the development of high quality renewable resources. Intermediate renewable energy policy targets – either a 12 GW distributed generation mandate in California by 2020 or a California 50 % RPS by 2030 – can help to deploy renewable generation in-­‐
state and on an accelerated schedule. However, these policy targets have less effect on the generation mix in the 2040 to 2050 timeframe as the cap on carbon emissions dominates the deployment of renewable energy. 9 ** DRAFT – DO NOT CITE ** 180"
600"
California"Genera4on,"Storage,""
and"Transmission"Capacity"(GW)"
Rest"of"WECC"Genera6on,"Storage,""
and"Transmission"Capacity"(GW)"
Transmission)Import/Export)
160"
Storage)
140"
Solar)
Wind)
120"
Biopower)
Biopower)CCS)
100"
Gas)
80"
Gas)CCS)
Nuclear)
60"
Geothermal)
Coal)CCS)
40"
Coal)
20"
0"
Hydro)
2013"
2020"
2030"
2040"
550"
Transmission)Import/Export)
500"
Storage)
Solar)
450"
Wind)
400"
Biopower)
350"
Biopower)CCS)
300"
Gas)
Gas)CCS)
250"
Nuclear)
200"
Geothermal)
150"
Coal)CCS)
Coal)
100"
Hydro)
50"
0"
2050"
2013"
2020"
2030"
2040"
2050"
Figure 2-2: Base Scenario generation, storage, and transmission import/export capacity as a
function of investment period in California and the rest of WECC.
80"
80"
70"
70"
60"
60"
Exports: Renewable!
50"
50"
Exports: Non-Renewable!
40"
40"
30"
30"
20"
20"
10"
10"
Imports: Renewable!
Solar!
Biopower!
Biopower CCS!
Gas!
V40%"Carbon"Cap"/"BioCCS"
V20%"Carbon"Cap"/"BioCCS"
New"Nuclear"
No"CCS"
California"50%"RPS"
12GW"Distributed"PV"
Low"Gas"Price"
Sunshot"Solar"
Demand"Response"
Expensive"Transmission"
Limited"Hydro"
Small"Balancing"Areas"
Aggressive"Electrifica@on"
Reduced"Efficiency"Implementa@on"
Gas CCS!
Base"Case"
0"
Imports: Non-Renewable!
Wind!
Business"As"Usual"
2050 Average California Generation (GW)!
0"
Nuclear!
Geothermal!
Coal CCS!
Coal!
Hydro!
Demand!
Figure 2-3: California average generation mix by fuel, average imports and exports by renewable
average electricity demand in 2050 for all scenarios. A similar figure for the rest of WECC can
be found in the main text.
10 ** DRAFT – DO NOT CITE ** WECC$Electricity$Dispatch$in$2050$(GW)$
Post-­‐2030, the electricity system undergoes a radical transformation in order to eliminate almost all carbon emissions from the generation mix. In the 2040 timeframe, deployment of wind, solar, and geothermal power reduce power system emissions by displacing gas-­‐fired generation. In the 2050 timeframe this deployment trend continues for wind and solar, but is accompanied by large amounts of new storage and long distance, high-­‐voltage transmission capacity. Electricity storage is used primarily to move solar power from the daytime into the night in order to charge electric vehicles and meet demand from electrified heating (Figure 2-­‐4). Low cost solar power (the Sunshot Solar Scenario) could increase the demand for electricity storage. If demand response is deployed in large scale in this timeframe, it substitutes for the functionality of storage, thereby strongly incentivizing the deployment of solar generation, especially in California. 300$
$Jan$$$$$Feb$$$$$$Mar$$$$Apr$$$$$May$$$$$Jun$$$$$$Jul$$$$$$Aug$$$$$Sep$$$$$Oct$$$$$$Nov$$$$Dec$
250$
200$
150$
100$
50$
0$
!50$
19$ 3$ 11$18$ 2$ 10$19$ 3$ 11$18$ 2$ 10$19$ 3$ 11$18$ 2$ 10$17$ 1$ 9$ 18$ 2$ 10$17$ 1$ 9$ 18$ 2$ 10$18$ 2$ 10$17$ 1$ 9$ 18$ 2$ 10$17$ 1$ 9$ 18$ 2$ 10$17$ 1$ 9$ 17$ 1$ 9$ 18$ 2$ 10$18$ 2$ 10$17$ 1$ 9$ 19$ 3$ 11$18$ 2$ 10$18$ 2$ 10$19$ 3$ 11$
Hour$of$Day$(PST)$
Nuclear$
Geothermal$
Biopower$
Coal$
Coal$CCS$
Gas$(baseload)$
Gas$CCS$
Gas$(intermediate)$
Gas$(peaker)$
Storage$(discharging)$
Hydro$(non!pumped)$
Solar$
Wind$
Storage$(charging)$
Demand$
Figure 2-4: Base Scenario hourly power system dispatch across WECC in 2050. Each month
includes two sample days. Total generation exceeds demand due to distribution, transmission,
and storage losses, as well as variable renewable energy curtailment.
Through 2050, transmission lines that exist today between Pacific coast states are found to be mostly sufficient to move power between coastal states. New transmission capacity is built primarily to move power from the inside of the continent towards demand centers on the coast. Transmission capacity over the California border is increased by 40 -­‐ 220 %, implying that transmission siting and permitting will become increasingly important over time. California remains a net electricity importer in all scenarios investigated. The percent of electricity imported 11 ** DRAFT – DO NOT CITE ** into California ranges from 22 % to 60 %, with most values occurring at ~40 %. Scenarios that implement demand response programs or that have expensive costs for the building of new transmission capacity are found to require the least import/export capacity into California. Scenarios that allow the deployment of out-­‐of-­‐state nuclear power or exclude carbon capture and sequestration (CCS) from the generation fleet are found to require the most California transmission import/export capacity. 2050$
Electricity)Transmi/ed)
(Average)GW))
0)C)0.5)GW)
0.5)C)1)GW)
1)C)3)))))GW)
3)C)5)))))GW)
5)C)10)))GW)
10)C)20)GW)
Electricity)Generated)
(Average)GW))
2)
5)
10)
20)
Hydro)
Wind)
Gas)
Geothermal)
Nuclear)
Solar)
Coal)
Biopower)
Gas)CCS)
Figure 2-5: Base Scenario average generation mix by fuel within each SWITCH load area, and
average transmission flow between load areas in 2050.
Wind and solar power are key elements in the decarbonization of the power system, providing 37 – 56 % and 17 – 32 % of energy generated respectively across WECC in 2050 if no new nuclear capacity is built. At these penetration levels of variable renewable energy, the least cost strategy for meeting policy, reliability, and demand targets includes the curtailment of wind, and to a lesser extent solar facilities at hours of high renewable output and/or low electricity demand (Figure 2-­‐4). As we approach 2050, determining how the cost of variable renewable curtailment is to be compensated will become increasingly important. In an effort to integrate wind and solar resources into the power system, the amount of installed gas capacity remains relatively constant between present day and 2050 (Figure 2-­‐2), though CCS is installed on some gas plants by 2050. The fleet-­‐wide average capacity factor of non-­‐CCS gas generation drops steeply between 2030 and 2050, reaching only ~X-­‐X (to be calculated from final results, likely 2-­‐6 or so)% in 2050, indicating that gas plants are only operated for a handful of 12 ** DRAFT – DO NOT CITE ** hours each year but are of extremely high value during those few hours. This result suggests the difficulty of supporting gas generation through energy and ancillary service market revenues, and implies the need for other revenue streams such as a capacity market. As there is little space in the carbon cap for fossil fuel emissions by 2050, sub-­‐hourly reserves – spinning and quickstart reserves – are almost exclusively provided by hydroelectric and storage facilities. Sub-­‐hourly reserves are dominated by gas-­‐fired generation in the present day, so this represents a drastic shift that will need to be accompanied by ancillary service markets that are able to efficiently procure reserves from zero or low carbon sources. Both gas-­‐fired CCS and nuclear power are found to be economical in the context of deep emission reductions, but neither is found to be essential to meeting 2050 emission targets. Both technologies are subject to large political and/or technical uncertainty and therefore economics may not be the driving force for installation. The deployment of moderate amounts of flexible gas CCS to balance variable renewable generation is found to be one of the most effective ways to contain the costs of reducing carbon emissions, especially in California. Gas CCS is not found to be economical to run in baseload mode due to the prevalence of inexpensive wind and solar power, as well as incomplete emissions capture by the CCS system. Coal-­‐fired CCS is not deployed at scale in any scenario investigated due to unfavorable economics and incomplete emissions capture. The finding that baseload fossil fuel CCS is not economical at deep carbon reduction levels is counter to much of the prevailing thinking about CCS and follows directly from using a detailed modeling platform such as SWITCH. Biomass CCS can be effective at reducing power sector emissions far below zero by 2050, and can therefore be thought of as a hedge against incomplete decarbonization of other sectors (notably the transportation sector). The cost to make the power system net carbon negative is moderate if biomass is made available to the electric power system instead of to the production of biofuels. PROJECT PUBLICATIONS (AVAILABLE ONLINE AT http://rael.berkeley.edu) Mileva, A., Nelson, J. H., Johnston, J., and Kammen, D. M. (2013) “SunShot Solar Power Reduces Costs and Uncertainty in Future Low-­‐Carbon Electricity Systems,” Environmental Science & Technology, dx.doi.org/10.1021/es401898f Tidwell, V. C., L. Dale, G. Franco, K. Averyt, M. Wei, D. M. Kammen, and J. H. Nelson. 2013. “Energy: Supply, Demand, and Impacts.” In Assessment of Climate Change in the Southwest United States: A Report Prepared for the National Climate Assessment, edited by G. Garfin, A. Jardine, R. Merideth, M. Black, and S. LeRoy, 240–266. A report by the Southwest Climate Alliance. Washington, DC: Island Press. 13 ** DRAFT – DO NOT CITE ** Wei, M., Nelson, J. H., Greenblatt, J. B., Mileva, A., Johnston, J., Ting, M., Yang, C., Jones, C., McMahon, J. E. and Kammen, D. M. (2013) “Deep carbon reductions in California require electrification and integration across economic sectors”, Environmental Research Letters, 8, doi:10.1088/1748-­‐
9326/8/1/014038 Nelson, J. H., Johnston, J., , A., Fripp, M., Hoffman, I., Petros-­‐Good, A., Blanco, C., and Kammen, D. M. (2012) “High-­‐resolution modeling of the western North American power system demonstrates low-­‐cost and low-­‐carbon futures”, Energy Policy, 43, 436–447. 14