Today, Massachusetts residents get most of their electric power from natural gas generating plants. As we electrify vehicles and heating systems, electric power demand is expected to roughly double (see EoEEA, Energy Pathways to Deep Decarbonization (“pathways analysis”), Figure 22). If we are to reach “net zero” carbon emissions by 2050, we need to build a lot more green generating capacity.
Our state office of Energy and Environmental Affairs has developed alternative pathways for greening our power generation and reached the conclusion that the transition is feasible. EoEEA’s pathways analysis, released in December 2020, is a very thorough analysis of what it would take for Massachusetts — with its particular climate and resource profile — to get to a green grid.
The common element in all pathways is a significant expansion of off-shore wind farms in Massachusetts waters. In most scenarios, a minimum of 15 gigawatts of off-shore generating capacity is installed (pathways analysis at page 78). That amount of wind generating capacity would generate about as much power as the whole state currently uses each year.
Understanding wind facility size
The “capacity factor” is the ratio of average energy output across varying wind conditions to maximum “nameplate” capacity. The capacity factor depends on both wind patterns and turbine performance. The overall capacity factor for the 6 gigawatts (GW) of installed wind capacity in Denmark is 30%. Vineyard Wind is estimating a capacity factor of 45%. If the average capacity factor for Massachusetts’ projected 15GW of wind capacity were 40% over the 8760 hours in a year, it would generate 52.5 trillion watt-hours (Twh) of electricity. Massachusetts used approximately 52 Twh of electricty in 2019.
To reach 15 gigawatts of wind power by 2050 we will need to add roughly 500 megawatts of new capacity per year. Currently, we have two major wind projects moving through the permitting process in Massachusetts: Vineyard Wind and Mayflower Wind. If these two projects roll out as targeted by 2027, we will have added 1.6 gigawatts of capacity — 200 megawatts per year over 8 years. We will have to pick up the pace considerably. In March, section 91 of our climate roadmap legislation increased the 2027 goal from 1.6 to 4 gigawatts of capacity. That would put us on the necessary pace of development.
We will see soon whether the increased goal leads to the necessary new proposals — the Department of Energy Resources has just issued a request for proposals for additional wind construction with price constrained to a maximum of $77 per megawatt hour. This maximum price may suffice to attract proposals since as it is above recent project prices in the United Kingdom — see box below.
The number one challenge will be finding enough sites. While worldwide, most wind development is on-shore, the pathways analysis does not expect us to build much wind capacity onshore in Massachusetts. The expectation is that we will need to use floating as well as fixed off-shore farms. A related challenge is siting the transformer facilities for bringing the power into the power grid.
Example: Hornsea One — Largest Operating Wind Farm as of 2020
Hornsea One is a 1.2 Gigawatt wind farm 75 miles of the Yorkshire coast in the North Sea. It is comprised of 174 wind turbines each rated to produce 7 megawatts. It went online in 2020. The area occupied by the windfarm is 157 square miles. It is the first phase of a complex of wind farms to be built in that area. See project brochure site here. The project was built by Orsted, a Danish companty. Orsted retains 50% of the generating capacity. See Orsted financial details here. Orsted does publish audited production statistics for their offshore portfolio which includes the facility. They retain half (600MW) of the generating capacity at Hornsea 1 and their total owned offshore generating capacity is 4,379 as of March 31, 2021. Their financial results show power generation from that portfolio of 4.5TWH in the first quarter of 2021, which would translate to an annualized capacity factor of 47%.
The striking fact about the Hornsea One facility is the high cost. UK ratepayers will be paying approximately $227/Mwh for the wind power from Hornsea One. The price received by a renewable facility operator in the UK is set by a “Contract for Difference” where the Low Carbon Contracts Company (owned by the UK government) will pay the difference between an inflation adjusted “strike price” and the actual price of electricity. The strike price for Hornsea One is currently 164.96 pounds sterling per megawatt hour. (The pound is at $1.39 as of May 3.) By contrast market prices for electricity are between 20 and 40 pounds per megawatt hour, so the support for the Hornsea One plant is very deep. The subsidy is funded by a levy on non-renewable power producers. A second project in the same area, Hornsea Two, has come in at much lower level, $95/mwh. Some critics suggested that the lower price was too low to support construction, but Hornsea Two seems to moving forward full steam. More recent projects have even lower prices: Dogger Bank is at $65/Mwh.
The other big challenge will be managing the variation in wind velocity. Wind power doesn’t fluctuate daily like solar power, but it does vary based on sustained weather patterns. If we are able to build enough wind capacity, EoEEA’s analysis contemplates exporting power to Quebec on good days, allowing Quebec hydro facilities to build up their reservoirs to release power when we have extra need because of sustained calm. This is how Denmark manages the variation — Denmark exchanges power with Norway which has hydro-power generating capacity that can be ramped up and down to offset wind variation.
Battery storage is another strategy for managing renewable energy variation, but it is not a central element or constraint in the scenarios modeled in the pathways analysis (see page 80). Instead, the analysis emphasizes overbuilding wind and solar and using electrolysis facilities to generate hydrogen fuel when wind and solar facilities are producing more power than we need. Another strategy that reduces demand for systemic battery storage is managing the timing of the new elements of electric load — charging of vehicles and generating building heat. Retaining some fossil fuel generating capacity to fire up during rare sustained wind power deficits saves $1000 per year per household over a scenario in which all fossil fuel generation capacity is decommissioned. Pathways analysis at note 53.
Our power grid — transmission lines and power transformer stations — will need to expand to handle the overall increase in load and also the new variable power flows from renewable generating facilities. The biggest challenge may be increasing the interconnection with Quebec. More generally, the report finds economic value in flexible sharing of generating capacity with other states. Regional cooperation will be important to our success.
While wind is the dominant power source envisioned, solar is also important. Most pathways contemplate approximately 25 gigawatts of installed solar capacity in 2050 (pathways analysis at page 55) — a 10 fold increase over today’s 2.5 installed gigawatts. The pathways analysis (p.84) finds that even if 1 in 3 rooftops in the state have solar installed, significant use of ground mounted solar will be needed to achieve the required solar capacity — approximately 66 thousand acres. In the maximal solar scenario, up to 3% of the state’s land area 158,000 acres would be covered with solar.
Understanding Solar Facility Size
As explained above for wind, the “capacity factor” is the ratio of average energy output across varying sun conditions to maximum “nameplate” capacity. For solar panels, year round capacity factors depend heavily on latitude and placement of the panels as well as weather and technology. Values range from 10 to 25% in typical installations. In my own solar installation — a good, unshaded location on a south facing roof — our capacity factor has been 15.1%. We installed 21 panels rated at 220 watts each to create a 4.63 kilowatt facility. Over a recent 5 year period, the array generated 30,648 kilowatt hours of power. If the facility had been in full sun 24/7 (8760 hours per year) over that period and each panel generated 220 watts continuously, it would have generated 202,356 kilowatt hours of power over 5 years. Divide 30,373 by 202,356 to get my installation’s capacity factor of 15.1%. This is apparently fairly average for Massachusetts installations. In Arizona and Utah, typical capacity factors average almost twice as high. The low solar capacity factor in Massachusetts as compared to wind explains why the pathways analysis emphasizes wind.
For ground mounted solar facilities, an important question is how much land area is required per megawatt of capacity. The pathways analysis uses a midpoint number of 4.06 acres per megawatt, with a range of 2.9 to 7.8. A solar developer of facilities in the 1MW to 10MW range said that his firm’s rule of thumb was 5 acres per megawatt plus a 20% allowance for connections, spaces, etc. That would suggest 6 acres per megawatt. Additionally, the developer pointed out that in many parcels that one might acquire, not all of the land area is usable — the parcel might include rocky areas or wetlands. It may be that the high side estimate of 7.8 acres per megawatt is closer to the truth.
The pathways analysis (at page 69) is encouraging as to the economic impact of decarbonization. Total energy expenditures are modestly higher than in the baseline fossil fuel case (10 to 20% depending on pathway), but the share of spending within the state goes up from roughly half to roughly 80%, as in-state construction projects replace out-of-state fossil-fuel purchases. Decarbonization also reduces the exposure of the future energy consumers to commodity price swings.
None of the identified pathways to net zero emissions depend on new technology, but they do depend on acceleration of our construction of wind, solar, and transmission systems. The pathways analysis (at page 89) acknowledges that the necessary pace of construction is “far in excess of the rates seen historically” for wind and solar. “Modeling . . . cannot make normative judgements about whether these build rates can be achieved and sustained.”
This acknowledgement of the unprecedented pace of construction is the principal red flag in the pathways analysis. When I see numbers representing historically slow production of wind capacity, I see the angry local forest advocates who thronged the state house to protest the installation of a mountain pass wind farm and I recall the successful state and national campaign by wealthy beach front owners to kill the wind project in Nantucket sound. I can foresee passionate local opposition to covering meadow lands with thousand of acres of solar panels.
The Role of Nuclear Power (Added May 2)
Thank you to commenters who pushed me to acknowledge the role of nuclear.
Having recognized the potential construction challenges for wind and solar, I should have branched to a discussion of the nuclear option. Nuclear power is not an economically viable option at this time, but it may be later and we should keep it as part of our conversation. For more on the short-run non-viability of nuclear, see this further post.
Before achieving a zero emissions grid, we may well run into limits on acceptance of additional wind and solar installations. Those limits could relate to Nimby-ism, to ecological concerns that compete legitimately with carbon reduction, or to inability to manage the power variation of wind and solar. When that looks like it is soon going to happen, we will have to re examine the viability of nuclear power as part of our clean energy solution.
For now, we are putting a lot of emphasis on off-shore wind which compares favorably to nuclear on a cost and risk basis. The favorable agreements that we have entered into for wind power supply appear to be cheaper than nuclear. For example, including sales of renewable energy credits, power from the Vineyard Wind project is expected to cost rate payers $79/Mwh ($79 per megawatt hour) over a 20 year life. This price is just a bit higher than the National Renewable Energy Laboratory’s high range estimate for Levelized Cost of Electricity(LCOE) for offshore wind coming online in 2025: $77/Mwh (downloaded May 2). The same source (NREL) estimates mid-range price (no low-high range offered) of $74/Mwh for nuclear coming online in 2025, but Lazard Frere’s 2020 LCOE analysis puts offshore wind at $86/Mwh and nuclear much higher at $129/Mwh to $198/Mwh.
These favorable wind prices may not be real. It remains to be seen whether the projects will be built at those favorable prices. For the Hornsea One project which was actually built, the price was $227/Mwh. Almost three times the Vineyard Wind bid.
But one thing is clearer than comparative costs: even large offshore wind investments are more scalable than nuclear and therefore less risky. We can make more incremental bets. A range of project sizes for the Vineyard lease from 200MW to 800MW were considered. Even at full size, the estimated project size is much less than a nuclear plant: $2.8 billion (early estimate) vs. $14 billion (pre-escalation cost of Vogtle). Unlike a nuclear plant, if the project were halted before completion, some of the installed turbines could likely be used: The 800MW project will come online in two 400MW phases.
A final critical distinction between nuclear and wind is that nuclear cannot be built without a rate payer commitment to cover construction costs (see this post). We are committing to pay for power delivered from the wind projects at a fixed price, but we are not bearing the risks for the projects failing or coming in at an unexpectedly high construction or operating cost (see, the draft 83C PPA’s).
Except for residents of Everett and Chelsea who reside next to the oil and gas terminals, most Massachusetts residents experience energy production and delivery as aseptic activities. We don’t see the smoke when we turn on the light switch. Mining and fracking don’t exist in Massachusetts. Bringing energy generation home to Massachusetts in the form of wind and solar farms will require all of us to accept more of the environmental costs of our energy consumption. While we may have to see and hear more of what it takes to produce energy, the good news is that as vehicles and buildings go electric, our air will be less polluted locally and we will be doing our part to reduce global carbon emissions.
Wind and solar lifecycle carbon emissions
An important question to ask is how zero emission power generation technologies like wind and solar look on a lifecycle basis. The good news is that while there are carbon costs in manufacture and construction of wind and solar facilities, the lifecycle emissions are vastly lower than fossil generation technologies. See World Energy Council (2004); nuclear power industry sponsored literature review (2011); for additional reviews of wind facility life cycle emissions, see Life cycle costs and carbon emissions of wind power: Executive Summary (2015) or Life cycle greenhouse gas emission from wind farms in reference to turbine sizes and capacity factors (2020); for deeper review for solar, see Life Cycle Greenhouse Gas Emissions from Solar Photovoltaics (2012)