Eye watering numbers for offshore wind!
A recent public utility filing for Revolution Wind 2 in Rhode Island gives us some insights into the eye watering costs of offshore wind.
I hope you are all having a great summer so far. After a diesel powered road trip in the top of the south with the kids, lots of swimming and a spot of hunting it’s time to get back into the blog.
Today let’s have a look into the economics of offshore wind.
Offshore Wind Economics…
The US offshore wind market is reasonably well established. We can gain a lot of insights as to what will unfold in New Zealand from the substantial “headwinds” that the US industry is facing. This is despite the massive state and federal subsidies being poured in via various schemes including the ironically named “inflation reduction act”.
Hat tip to Meghan Lapp for this analysis which was shared on David Blackmon’s excellent substack. I’ll pull out a few key passages for comment with a New Zealand perspective but the whole article is well worth a read and can be found here Required Reading From Meghan Lapp.
Let’s start with the costs. Specifically capital costs to build the wind farm (CAPEX) and operating costs after construction (OPEX).
Note that this 804MW wind farm is located offshore from Martha’s vineyard in about 35-45m water depth. This is a similar water depth and size to the farms proposed for South Taranaki. There are four farms currently proposed for south Taranaki ranging in size from 800 - 1000MW (0.8 - 1GW).
In a recent filing of Revolution Wind 2 with the RI Public Utilities Commission, the developer’s own information showed that compared to a combined cycle gas plant capital cost of $1,115 per kW, the corresponding offshore wind cost of its project would be $6,249 per kW (courtesy of Allen Brooks at Energy Musings. The corresponding fixed O&M costs for natural gas were listed $64 per kW/year, and for offshore wind $129 per kW/year. Have fun with that, ratepayers.
Let’s apply these numbers to the New Zealand context. As I have previously written about in NZ pension gone with the wind, the NZ super fund has partnered with Copenhagen Infrastructure Partners to build a 1GW wind farm off the coast of South Taranaki in ~30m of water. This makes this a directly comparable development, with the caveat that there is no offshore wind industry in NZ, so our costs can be expected to be higher.
Using a currency conversion of USD to NZD of 1.61 this gives us the following figures for the same project in New Zealand:
Construction = NZD $10.05 Billion.
Annual opex = NZD $207.7 Million.
Now Meghan makes a few other very interesting observations regarding the capacity factor. For those unfamiliar with this term capacity factor it is the actual power generated, represented as a % of the installed capacity.
The U.S. EIA itself estimates the total system LCOE for combined cycle natural gas plants entering service in 2027 at $39.94 per MWhr, and offshore wind at $136.51 per MWhr, except it makes this assumption for offshore wind based on a 44% capacity factor. Since Dominion said a 42% capacity factor performance guarantee is “untenable” and threatened to walk away from its project if that was required, the EIA estimate is probably wrong. And think about adding intermittency costs to all that- not knowing when extra electricity will have to be purchased, how much extra, and at what price.
For clarity. LCOE is levelized cost of energy. Which is a measure of the average net present cost of electricity generation for a generator over its lifetime. Be wary of the LCOE metric generally speaking as it is easy manipulated. It is manipulated by changing where in the project the analysis starts (pre or post construction for example). However, this does give us an order of magnitude comparison.
What the capacity factor of 42% (being generous) means is that on average the wind farm with produce 42% of the installed capacity. In the case of our South Taranaki 1GW farm this is 420MW which is not a lot for a $10 billion investment.
The problem, which not costed here, is that at times it will generate close to 1GW and at other times nothing. The grid (transmission network, other generators and consumers) needs to be able to absorb and mitigate these wild and rapid swings in generation output. Building a grid that can cope with this will also be a hugely expensive undertaking, which is not captured in the LCOE metric, but does need to be funded through lines charges and other fixed fee structures.
Megan continues….
Another fun fact is that the bigger the wind farms get, and the more offshore wind farms are built, the less wind there is and the worse they perform. It’s called the wind wake effect. Any sailor is well familiar with this- a sailboat in front of you will steal your wind, and make your sailboat go slower if you stay behind it. It’s not rocket science.
But somehow, this escaped wind farm productivity analysis for years. In 2018, Harvard University flagged this and found that the average power density of a wind farm was up to 100 times lower than estimates by some leading energy experts because previous estimates failed to account for the wind wake effect.
When wind developer Orsted finally had to admit in 2019 that the wind wake effect reduced its estimated project outputs, its stock crashed over 7% immediately. A more recent 2022 report by ArcVera Renewables entitled “Estimating Long-Range External Wake Losses in Energy Yield and Operational Performance Assessments Using the WRF Wind Farm Parameterization” specifically analyzed the potential for large project to project wake impacts for offshore wind leases off NY and NJ, resulting in simulations depicting wind speed deficits of 7% up to 100 km away from the wind facility, with a 28.9% loss of wind at the wind farm itself.
Now let’s consider that there are currently four companies touting windfarms off of the South Taranaki coast. Are they considering the wake effect in their modelling?
As a side note having four wind farms of 800 - 1000MW all in the same general location and generating based on the same wind patterns would be chaos. At least when wind farms are geographically distributed, they won’t all be stopping and starting at exactly the same time, making their impact on the grid still chaotic but somewhat less so.
So, let’s do some very crude back of the fag packet economic calcs for the Super Fund’s proposed wind farm.
Investment = $10 billion (shared with Copenhagen Infrastructure Partners).
Expected return on investment somewhere around 8% pa = $800M
Annual OPEX = $207M
Annual revenue required = $1.07B
Based on an average (highly optimistic) output estimate of 420MW of capacity factor corrected average generation we get ~3.7M MW/hrs annually.
This is around $270 / MWhr or $0.27 / KWhr which is already higher than my current retail power bill rate.
Note this does not include any of the following that would need to be factored in:
ABEX (abandonment / decommissioning expenditure) which would need to be recovered over a relatively short 20 year period and could be greater than the installation costs.
Transmission network upgrading that would undoubtedly be necessary to deal with this amount of intermittent power generation. This will easily be several hundred million.
Also not discussed is curtailment costs. Which is the money that has to be paid to other generation companies not to produce when the wind is blowing but to remain available for when the wind drops off. We haven’t seen this in NZ yet but it will appear in some form as renewables dominate the market.
As you can see from this very crude example, with generation rates already higher than current retail rates, without including any of the extras. It is not unreasonable to expect your power bill to double or even triple if offshore wind generation capacity increases in the New Zealand market.
As we have noted in this blog before low density disperse energy = high resource intensity = high cost. All of which is a function of a low productivity energy system.