# Two Terawatts average power output: the UK offshore wind resource

This paper presents a calculation of the offshore wind resource, and shows that the theoretical resource from offshore wind turbines in UK waters is approximately 2.2 TW  of  average (ie continuous output) of electricity. The figures are calculated across four ranges of sea depth: waters to 25 metres; 25-50 m, 50-100 m and deep waters to 700 m. The chart below shows the potential generation, with figures split out by sea depth and by distance from shore.

The available power is calculated by comparing power from an established wind farm, with the power available across the study area.

First, the calculations for an existing offshore wind farm in shallow waters; I’ll use that as a basis, to scale up the calculations for the resource in shallow, medium, deep and deepest waters.

### Kentish Flats

The wind power density (the mean density of the wind’s kinetic energy, over the year) at Kentish Flats is about 713 W per square metre of vertical plane, at 80 metres above sea level. The Vestas 3 MW turbines there have a net efficiency of approximately 30%, for that wind profile: i.e. 30% of the wind’s kinetic energy is converted to electricity; (so in this case the figure is efficiency, rather than capacity factor). The turbines have a swept diameter of 90 metres, and thus a swept area of 6360 m2 (45 m x 45 m x pi). So the power generated is:

713 W/m2 x 6360 m2 x 30% = 1.36 MWe per turbine.

The turbines are spaced in a 700 m x 700 m grid, so the turbine density is

1/(0.7×0.7) = 2.06 turbines per square km of seabed,

giving a density of harvested electricity of 2.8 MWe/km2, at 100% availability. A 90% availability gives around 2.5 MWe/km2.

And now to generalise to British waters, grouping the depths into shallow, medium and deep waters: 0-25 m, 25-50 m, and 50-700 m.

### 80 GW in shallow waters (0-25 m)

In the (territorial + EEZ) waters around Britain & Northern Ireland, there’s roughly 40 000 km2 of seabed shallower than 25 m, with a wind power density of 579 W/m2 (that’s per square metre in the vertical plane: the area that the turbine blades sweep through). So, for the depths of 0-25 m, scaling the Kentish Flats figure accordingly, the mean potential electricity density is:

2.5 MWe/km2 x 579/713 = 2 MWe/km2

40 000 km2 x 2 MWe/km2 = 80 GWe at 0-25m depths

### 270 GW in seas of depth 25-50m

For the 90 000 km2 of British waters in depths of 25-50 m, the wind power density is 868 W/m2(vertical). giving mean potential electricity of:

2.5 MWe/km2 x 868/713 = 3 MWe/km2

90 000 km2 x 3 MWe/km2 = 270 GWe at 25-50m depths

### 790GW in deeper waters 50-100m

And then there’s about 210 000 km2 at 50m-100m, with a wind power density of 1070 W/m2(vertical)

2.5 MWe/km2 x 1070/713 = 3.75 MWe/km2

210 000 km2 x 3.75 MWe/km2 = 790 GWe at 50-100m depths

### 1030GW from the deepest accessible waters 100m-700m

And then there’s about 220 000 km2 at 100m-700m, with a wind power density of 1340 W/m2(vertical)

2.5 MWe/km2 x 1340/713 = 4.7 MWe/km2

220 000 km2 x 4.7 MWe/km2 = 1030 GWe at 300-700m depths

### Total = 2.2 TWe

So, all together, that gives approximately 2,200 GWe, or 2.2 TWe.

For context, current UK energy demand (electricity + heat + transport) is about 0.25 TW, and current UK electricity demand is about 0.04 TWe.

The power available varies by depth, as seen above.  There is also some variation, depending on distance from shore.  The figures are as follows:

 MWe/km2 0-25m 25-50m 50-100m 100-700m <100km 1.9 2.9 3.7 4.7 100-200km 3.7 3.6 3.9 4.4 >200km 3.7 3.7 3.8 5.3

### Caveats and other thoughts

Now, that’s just a first-order estimate of the potential.

How much of that sea is needed for other uses – shipping or military exercises, is a matter of future negotiation. Fishing boats are happy to sail between turbines at their current spacing.

Current turbine spacing is on a grid of about 7 diameters by 7 diameters, as a precaution against damage from wake/turbulence: Kentish Flats is a square grid of 7.78×7.78 diameters.

Turbine manufacturers advise that, to avoid power loss from shadowing, a grid of 5 x 3 is sufficient, which, if realised, would give potential power of more than triple the above figures. (because three grids of 5×3 diameters into less area than one grid of 7×7 diameters)

The mooring used on the Norwegian HyWind floating turbines are expected to be suitable out to depths of 700 metres, which is why I’ve chosen that depth as a cut-off for now, but I acknowledge that developments could, if necessary, open up deeper waters. But I think the figures show that deeper waters really shouldn’t be necessary!

### Units

For those of you who prefer units other than the SI standard ones, here are some alternatives. The UK offshore wind resource is:

• 2.2 TW of electricity
• 870 kWh per person per day for the UK’s 60 million residents
• 19 000 TWh per year
• 4.5 Mtoe (millions of tonnes of oil equivalent) per day
• 32 Mboe (millions of barrels of oil equivalent) per day; global oil production is about 80 million barrels per day.

### Sources

The Danish Wind turbine power calculator

Vattenfall datasheet on Kentish Flats offshore wind farm

DUKES 2009

Are these figures based on the maximum output of all the turbines or are they scaled to reflect the Danish experience which are only operating at 28% of the time over the year.?

Hi – the clue is in the title – average output – is is based on what the turbines would actually be producing on average, so it does allow max output to be scaled back due to the well known fact that it is not windy all the time.

Hello there Alistair,
yes, they do reflect capacity factors: they are not maximum output, they are based on the available wind power, as mapped at http://www.renewables-atlas.info/ .

Whereas wind turbines onshore in Britain average around 27% capacity factor, and near offshore tends to be more like 30-35%, the Round 3 sites are expected to be in the range 40-45%, and the very best sites can be even higher – the onshore site at Burradale in the Shetlands has seen annual capacity factors over 50%. This geographic variation is reflected in the calculations above.

So how is peak demand catered for please? Though this is an average figure – it could be from 3 months of zero production…

The TITAN 200, a Texas oil engineer’s solution for effectively and economically installing/operating large turbines offshore in deeper waters is the application of years of experience building and operating large offshore installations for oil and gas exploration. A self installing jack up platform, no ships/cranes required. My question is what is the expected productivity in terms of crude oil equivalent or electricity produces for a 5 MW turbine operating for 25 years in an offshore area of UK deemed good for wind in US \$’, or UK pounds ?
Thank you,
Colin Hardy,
Houston, Texas,
281-954-8985
832-771-6825cell
Offshore Wind Power Systems of Texas Titan Turbine Foundations

Hello there Colin,

Re the productivity of a 5MW turbine in an offshore area good for wind: well, we’d expect a capacity factor in the range 30-50%.

At the optimistic end, assuming a 25 year life and capacity factor 50% , the total energy would be:
50% x 5MW x 25 years x 8760 hours per year = 547,500 MWh

At the pessimistic end, taking a 30% capacity factor and a 20 year life, gives:
30% x 5MW x 20 years x 8760 hours per year = 262,800 MWh

In crude oil equivalent, that’s 155-322 kboe

Hi I did some calculations to work out how many turbines you would need. I use the 2.06 turbines per km2 of sea bed and round it to 2 for ease. Here goes

Sea Depth Area (km2) Num Turbines
Shallow 40000 80000
Low Deep 90000 18000
Med Deep 210000 420000
Very Deep 220000 440000

TOTAL 560000 1,120,000

So you suggest building over 1 million turbines around our coast. How much will this cost ?

Fred

2 Turbines per km2 seems far too low – do you have a reference for this?

Of what relevance is the cost figure on its own? You need to set that agains the value of power the turbines will generate over their lifetime.

I think you can work it out.

Offshore is about 3,000 uk per kW I think.

In these calculations I think there’s something you’re forgetting.You will agree that wind turbines are not 100% effective at capturing wind energy, yes? A very large proportion of the moving air mass interacting with the blades loses its energy without producing electricity. That energy has to go somewhere & it ends up as heat. Effectually, you’re doing the equivalent of hanging millions of 2kW electric heaters over the sea. Dumping a great deal of energy into the near-surface air-mass. What happens to it? If you heat that air, even slightly, it’s going to lose density & rise by convection. Do that over a large enough area & you change wind patterns. Your downwind turbines may not be physically shadowed by their upstream neighbours but the rising air column will have the same effect pushing the wind upwards. This phenomenon is common enough over landmasses & causes cloud formation & rain.
I’m not sure you can simply scale from small isolated wind generation facilities to something on this scale without consequent effects.

Hi Dave,
Surprised & reassured that my comment has occasioned such interest.
Sorry to say that so many wind farm advocates usually seem oblivious of
practical considerations.

Since my post, another aspect occurred that I see no indication of being
considered. A gigawatt of electricity generated is regarded as a
gigawatt of electricity available. This is far from true. To get the
power from where it’s generated to where it’s consumed it has to be
transmitted & the cable it runs through is not a perfect conductor.
All cable has resistance. What is fed in at one end is not what comes
out the other. 200 km of cable is a lot of cable & a lot of
resistance so there’s going to be a big discrepancy between what’s
generated & what’s landed. In addition, it’s likely to be landed a
long way from where it’s consumed. Largely in the SE of England. The
National Grid isn’t designed to do this sort of thing. It’s primarily
for load balancing as currently most power is generated not far from
where it’s consumed. If you were costing this project you’d need to add
in supplying a new Grid. More importantly, it makes a difference to the
energy available at the demand end because you need to factor in not
only the losses between the wind farm & point of landing but also
between there & the wall sockets. The trouble is, there’s a positive
feedback happening here. The more power you feed in at one end of the
system, the more gets lost in the system itself so the more needs to be
fed in. More wind turbines. As far as I’m aware it’s what makes PV’s in
the Sahara an impractical method of supplying Europe’s electricity.
Nothing comes out of the wire in Germany.
In common with my original post, there’s also an environmental aspect.
The resistance in the cables generates heat & that heat will be
dumped into the sea. It’s going into deep water & it’s never been
there before. Any idea what it will do to the currents or the water
column?
Sorry if I’m raining on your parade but my experience in the practical
world teaches that the rake handle of unintended consequences does tend
to slap one between the eyes with monotonous regularity.

Pete Lewis
Fuengirola, Spain.

Pete,

If yours were the comments about windfarms significantly altering weather etc then your concerns / claims were addressed / shot down in flames in recent responses.

Your comments (the ones in this latest posting) will possibly attract further interest – to point out they too don’t stand much scrutiny – when considering the scale and effects of the losses.

Actually the HV Grid losses are relatively minor (in terms of MWh delivered / MWh lost per km) certainly in comparison to the losses in the shorter distances from the step-down transformers to the LV (33kV / 11 kV/ 240 V) electricity networks to homes, offices, etc. In 2009 the electricity transmission losses were 7.1 % of the 380 TWh/y generated or 27 TWh/y. Of this, about 22 % was lost from the HV Grid, or about 6 TWh/y (see DUKES 2010 figures below) and most of the remainder about 20 TWh/y from the LV wires. So the HV transmission losses were about 1.5 % of electricity generated.

I have not seen much data on the estimated losses for cables from windfarms to shore but I doubt they would add significantly to the the current system’s 6 TWh/y losses in delivering unit output generated to the UK’s step-down transformers. Indeed, future massed windfarms on Dogger Bank / southern North Sea / Irish Sea may not be much further away in geographical distance to demand centres (particularly in the south of England) than current power stations (much capacity in Northern England).

Its the LV side that needs more investigation in my view – in terms of the supposed significant move to electric heating and possibly electric vehicle recharging. The litmus test for any future UK energy system is to deliver about 6+ TWh of ENERGY to the UK building stock/industry per DAY for a severe cold snap in a windless week+ winter anti-cycone. The recent clod snap demand was 1.2 TWh per DAY electricity and 5 TWh per DAY gas ( ie to gas CH boilers, etc, ex CCGTs). Anyone advocating a move to 6+ TWh per day by LV wire needs to consider this. Even with domestic hot water storage and widely-deployed ground-source heat pumps in urban areas (practicality ?) taking the edge off the peak heating needs an ‘all electric heating’ system would still be carrying significantly more LV electricity than is currently the case. So the future of the gas network is an interesting question.

As for electricity losses causing adverse-consequence heating in North Sea or seas generally – the effects would be tiny (say roughly 2 % maximum sub-sea, amounting to 10 TWh/y even in a massive 500 TWh/y marine renewable deployment). If you wish to concern yourself about the adverse effects of power generation then consider (apart from global warming) the thermal losses (reject heat) from current coal gas and nuclear power stations dotted along the UK’s coasts, and the more enclosed estuaries and rivers. Such heating currently amounts to at least 500 TWh/y (compared to maybe 10 TWh/y at most from massed marine renewables). Indeed, the overall energy production-releated heating of the seas around the UK will FALL dramatically as more offshore renewables are deployed and future thermal power stations (hopefully mainly CHP with CCS) play more of a back-up roll. I would be about a billion times more concerned about the adverse effects and unintended consequences of global warming – especially of climate scientists like Hansen are right (advocating 350 ppm as safe-ish).

The HVDC links from wind and CSP schemes (which could be thermal or PV) in north Africa to northern Europe would lose about 10 % of the power exported, not the 100 % as you surmised.

Unintended consequences should always be looked into – and I think this group has generally been good at that. Indeed, I hope this and other posting have addressed your current concerns about large-scale renewable energy deployments – if possibly at the expense of introducing you to the far more serious concerns about some of the adverse consequences of current fossil and nuclear energy generation.

Neil

——————
Extract from UK Energy Digest (DUKES 2010) page 117 :

5.13 Losses as a proportion of electricity demand in 2009, at 7.1 per cent, were slightly higher than in 2008 (6.9 per cent). The losses item has three components:

• transmission losses from the high voltage transmission system, which represented about 22 per cent of the figure in 2009,
• distribution losses, which occur between the gateways to the public supply system’s network and the customers’ meters, and accounted for about 74 per cent of losses,
• theft or meter fraud (around 4 per cent)

——————

On 25 Mar 2011, at 08:49, dave andrews wrote:

On Friday, 25 March 2011, wrote:

Hi Dave,
Surprised & reassured that my comment has occasioned such interest.
Sorry to say that so many wind farm advocates usually seem oblivious of
practical considerations.

Since my post, another aspect occurred that I see no indication of being
considered. A gigawatt of electricity generated is regarded as a
gigawatt of electricity available. This is far from true. To get the
power from where it’s generated to where it’s consumed it has to be
transmitted & the cable it runs through is not a perfect conductor.
All cable has resistance. What is fed in at one end is not what comes
out the other. 200 km of cable is a lot of cable & a lot of
resistance so there’s going to be a big discrepancy between what’s
generated & what’s landed. In addition, it’s likely to be landed a
long way from where it’s consumed. Largely in the SE of England. The
National Grid isn’t designed to do this sort of thing. It’s primarily
for load balancing as currently most power is generated not far from
where it’s consumed. If you were costing this project you’d need to add
in supplying a new Grid. More importantly, it makes a difference to the
energy available at the demand end because you need to factor in not
only the losses between the wind farm & point of landing but also
between there & the wall sockets. The trouble is, there’s a positive
feedback happening here. The more power you feed in at one end of the
system, the more gets lost in the system itself so the more needs to be
fed in. More wind turbines. As far as I’m aware it’s what makes PV’s in
the Sahara an impractical method of supplying Europe’s electricity.
Nothing comes out of the wire in Germany.
In common with my original post, there’s also an environmental aspect.
The resistance in the cables generates heat & that heat will be
dumped into the sea. It’s going into deep water & it’s never been
there before. Any idea what it will do to the currents or the water
column?
Sorry if I’m raining on your parade but my experience in the practical
world teaches that the rake handle of unintended consequences does tend
to slap one between the eyes with monotonous regularity.

Pete Lewis
Fuengirola, Spain.

No doubt well intentioned but Peter displays supremely confident
ignorance. He only has to look on Wikipedia National Grid to see that
the cost of doubling the UK National Grid is fractions of a p/kWh.

And it has long been recognised by Claverton that csp from the Sahara is very expensive – but that is not due to the losses en route which are about 2% per 1000km.

Note that all the energy in the wind ends up as heat anyway, with or without turbines. There’s no escaping entropy. The turbines just affect where and when it turns up as heat, that’s all.

Do bear in mind that this is not a proposal to install two terawatts of wind turbines. It’s an examination of what the laws of physics say is our potential offshore wind resource. Just 9% of this resource would provide 200GW, which is the UK’s total energy demand (electricity, heating, transport)!

Hi. I was directed here from http://cooltheearth.wordpress.com/2011/03/29/why-george-monbiot-is-wrong-on-nuclear-power/ and my first thought was what George Monbiot’s thoughts on this would be. (I respect Mr Monbiot a great deal; I believe that he is owed a vote of thanks for his efforts in attempting to counter the AGW misinformation campaign – although his latest position regarding nuclear power is a concern).

My second thought was OMG. If these numbers are right (and I don’t know enough to doubt them) then why on Earth are we in the UK not pursuing this more actively?

Many thanks,
Colin

Basically Mackay got it wrong.

From time to time power outages happen. Sometimes they last for hours, sometimes days, and in extreme cases—sometimes even months. So you have a choice: you could either sit there and cross your fingers in the dark, hoping that the electricity comes back on quick. Or you can be prepared with a powerful generator that will automatically switch on and at once restore power to your home or business.

Basically Mackay got it wrong.

I’m looking at MacKay’s chapter on Wind and comparing it with your figures. MacKay seems more generous in his figures for output per square metre[1] but MacKay assumes that only 1/3 of the total area is available for wind gerneration – and quotes even more pessimistic estimates from the DTI[2]

You also include generation – the lion’s share of your 2TWe – from waters over twice as deep as those in which even the two turbines in the Beatrice field demonstration project are operating! On what do you base your assumption that we could even operate generators in waters of this depth?

[1] MacKay: “I’ll assume that a power per unit area of 3 W/m^2 … is an appropriate ?gure for offshore wind farms around the UK.”
You: ” So, for the depths of 0-25 m, scaling the Kentish Flats figure accordingly, the mean potential electricity density is:
2.5 MWe/km^2 x 579/713 = 2 MWe/km^2″

[2]

The Department of Trade and Industry’s (2002) document “Future Offshore” gives a detailed breakdown of areas that are useful for offshore wind power. …The DTI’s estimated power contribution, if these areas were entirely ?lled with windmills, is 146 kWh/d per person (consisting of 52 kWh/d/p from the shallow and 94 kWh/d/p from the deep). But the DTI’s estimate
of the potential offshore wind generation resource is just 4.6 kWh per day per person. It might be interesting to describe how they get down from this potential resource of 146 kWh/d per person to 4.6 kWh/d per person. Why a ?nal ?gure so much lower than ours? First, they imposed these limits: the water must be within 30 km of the shore and less than 40 m deep; the sea bed must not have gradient greater than 5? ; shipping lanes, military zones,
pipelines, ?shing grounds, and wildlife reserves are excluded. Second, they assumed that only 5% of potential sites will be developed (as a result of seabed composition or planning constraints); they reduced the capacity by 50% for all sites less than 10 miles from shore, for reasons of public acceptability; they further reduced the capacity of sites with wind speed over
9 m/s by 95% to account for “development barriers presented by the hostile environment;” and other sites with average wind speed 8–9 m/s had their capacities reduced by 5%.

@John:
The basis of his assumption is given in the original article:

“The mooring used on the Norwegian HyWind floating turbines are expected to be suitable out to depths of 700 metres, which is why I’ve chosen that depth as a cut-off for now, but I acknowledge that developments could, if necessary, open up deeper waters. But I think the figures show that deeper waters really shouldn’t be necessary!”

Where did you get your areas of depth ranges from? ie 40000 km2 depth less than 25m

The depth data was from the SRTM30 dataset. The calculation was done using GIS layers from:

1. the BERR Renewables Atlas;
2. the SRTM30; and
3. the UK Continental Shelf boundary.

Hi,

I am currently carrying out a masters degree project on the potential of offshore wind in the UK.
I have tried accessing the files as you recommended but can not seem to open them in the correct format. I am looking for a break down of total areas offshore available in the UK at distances from shore: 0-10km, 10-20km, 20-30km, 30-40km, 40-50km and 50km, But any other breakdown like depth as you have used would also help.
This is for a calculation similar to your however taking into account further developments in the field.

I would be thankful for any help

Regards
Stuart