Friday, 24 November 2017

The Carbon Intensity of UK Grid Electricity

What it Means for Low Carbon Buildings

Take a look at this chart. It's nothing short of astonishing. Up to 2012 the amount of carbon dioxide emissions associated with the delivery of one unit (kilowatt hour, or kWh) of electricity in the UK was hovering around 500gCO2/kWh. Since then, the amount of carbon dioxide that is emitted for each unit of electricity has plummeted. In 2016 the average was 269gCO2/kWh, a fall of nearly half in only four years. This change has far-reaching implications for regulators, not least those involved in ensuring the low carbon transition of the UK building stock, both newly constructed buildings and the improvement of the existing stock.

So what's behind the fall?

The first factor is the retreat of coal-fired power stations. In 2012, the government's Digest of UK Energy Statistics (DUKES) has coal fired power stations producing 44% of our electricity nuclear plants were suffering from outages and gas prices had risen, so coal use was at a high. By 2016 the corresponding figure for coal was only 9%. In the same period, gas fired power stations rose from 24% to 42% of UK power generation. This matters for two reasons. First of all, because coal is made up of long-chain hydrocarbons, with a higher ratio of carbon atoms to hydrogen atoms it produces about 60% more carbon dioxide than natural gas for each unit of heat energy produced in burning. Second, gas is more often burnt in a Combined Cycle Gas Turbine (CCGT) power plants with conversion efficiencies of up to 60%, compared to 40% for conventional steam turbines.

The second factor is the increasing contribution from renewables in the electricity supply. Enormous amounts of wind energy, biofuel fired generation and solar energy have come online. In 2012 renewables and 'other' represented 11% of UK electricity supply. In 2016, this had risen to 27.8%.

As a result the average carbon intensity of electricity in 2016 at 269 gCO2/kWh was only just higher than that for gas (216 gCO2/kWh). When you add in an efficiency for a gas boiler at (say) 80%, the gap disappears.

This is huge.

For years electricity has been the bad boy in low carbon building design. People fretted as a series of reports from the Energy Savings Trust showed that heat pump installations in the UK were operating nowhere near their advertised efficiencies and were consequently underperforming gas boilers for carbon emissions. Simple resistive electrical heating by panel heaters or immersion heaters for hot water were to be avoided at all costs.

Four short years later and all this is is turned on its head.

And we're only just getting started with renewables. In September, Dong Energy announced that it would move forward with the world's largest offshore wind farm, Hornsea 2 off the Yorkshire coast, with development costs that had fallen by half compared to previous offshore farms. A couple of week later, and not to be outdone, the UK's first subsidy-free solar farm was announced. It's still a bit of an outlier combining solar with battery energy storage and using pre-existing grid connections from with an earlier development, but it's a clear sign of the direction of travel. The carbon intensity of grid electricity is heading only in one direction.

But there's another wrinkle to consider. The carbon intensity of the grid is not a static value. It varies constantly as the mix of generators fluctuate to meet different levels of electricity demand and in response to changes in wind and sunlight. On 11th June this year, it was windy and sunny at the same time. Records tumbled. The carbon intensity of grid electricity in the middle of the day on was below 80gCO2/kWh.

So now the moment when you choose to take power from the grid is a strong determinant of the actual instantaneous carbon emissions your electricity use is creating.

Some uses of electricity - for example for preparing domestic hot water, or to some extent space heating buildings could be relatively time independent.  If I'd known ahead of time that carbon emissions would be so low on 11th July, I'd have been able to set a timer for my immersion heater to heat water for me at midday and got my tank of hot water at fully one third of the carbon emissions of using gas heating.

And the technology to do this is just around the corner.  This awesome new grid carbon intensity forcasting service has been recently launched by the National Grid the Met Office and WWF, with an API that software developers could use to do just this kind of thing.


So where does this leave low carbon building?

The current building regulations in England and Wales were last reviewed in 2012 and set minimum carbon emissions rates that developers must design to. The carbon intensity of electricity in the approved calculation (the Standard Assessment Procedure or SAP) is currently 519gCO2/kWh, which was accurate at the time. Now it is woefully behind the curve.

Buildings are normally intended to be long-lasting. If we allow ourselves to imagine a future where digital technologies, the smart distribution of electricity, demand response, energy storage and renewables combine in a so-called 'Smart Grid' then a number of significant observations about low carbon building emerge:

  • Even based on the current carbon intensity, never mind the future direction of travel over the life of a building, it is utterly beyond me that any new build or significant refurbishment should include gas heating.

  • The current enthusiasm among UK policy makers and local authorities for district heating (for example this recent consultation by Scottish government) could also be a troubling dead end. District heating itself is neither intrinsically clean nor green - it all depends what heat source you put at the other end of the pipes you're going to dig up all the streets to install. Gas fired combined heat and power may be seen as low carbon at the moment, but how long will it look so appealing if electricity continues on its current path?

  • Building codes are currently focused on regulating carbon emissions. In a world of low carbon electricity you can meet a carbon target with a draughty garden shed full of electric fan heaters. It's time to move to energy targets (kWh/m2) to create buildings that sip energy and liberate power for the demands created by the electrification of transportation.

If I was building my own Grand Design right now, my future-proof forever home based on these observations here's what I'd go for:

  • High levels of insulation and air tightness to drive down space heating demand to a practical minimum

  • Eliminate the wet heating system - I'd go underfloor electric coupled to a high thermal mass floor to allow price and carbon responsive electricity purchase to heat the slab at times of excess renewable generation

  • Direct electric hot water cylinder - again allowing price-responsive purchase of electricity as well as diversion of excess generation from...

  • the inevitable....beautiful solar panels on the roof - as many as possible!

Could this be the future direction energy efficient buildings? What do you think?


  1. Yup, yup and yup.
    I would just add clever storage heaters - since the problems are all existing homes not future grand designs.
    The internal walls in old brick-built houses could be useful too: with "underfloor" heating elements applied as wall-paper.

  2. Very interesting, thought provoking piece. I never knew the underlying assumptions in building regulations were so out of date. With renewable capacity ever increasing (both supported and subsidy free) the downward trend will continue further. Does anyone know when/if the assumptions will be updated?

    Ultimately if it is easier to meet building regulations solely with electricity, gas will likely rapidly disappear from new builds. For developers not only would building regs be easier to achieve, they would also avoid needing to pay network connection costs and they can avoid the costs associated with installing the equipment.

    Over time this will lead to a "utility death spiral" for the gas network. As there would be no new connections the cost of maintaining the gas network would fall to an ever shrinking number of existing users, as some buildings were demolished/refurbished and existing users use less gas through efficiency. At the same time the increased demand would mean the costs of maintaining the electricity network would fall on a per unit basis, unless if course there was rapid widespread distributed generation. Eventually the costs of gas for existing users would become so great that it would pay to switch to electricity, even when including retrofitting/upgrade costs. The challenge would be ensuring support is provided to the vulnerable who potentially could not afford the up-front costs of switching from gas to electricity.

    PS - The big offshore wind farm is called Hornsea, not Horningsea.

  3. Great piece as ever. But wy go for electric UFH when a heat pump can have and efficiency of 300% if installed right? New smart switches let you ramp heat pumps up and down according to the amount of excess PV you're generating and in future will do the same to react to the carbon intensity of the grid.

    1. Hi Leah

      its the time dependent element that got me thinking about direct electric. Look how low the carbon intensity fell on 11 June for just a few hours.

      Think also about a sunny/cloudy day where the excess solar is only there for short periods on a home with solar.

      Heat pumps prefer long stable running periods and tend to 'trickle' heat into buildings slowly over time. This is one of the principal reasons that it's been such a struggle to get people to love the technology in the UK where we're used to gas boilers that can have your house warm an hour after you boost them.

      What we'd need from a responsive heating technology that can respond to market signals about low carbon/cost energy is that it can operate at a high power for short periods to take best advantage of availability. That's what has turned my thinking towards direct electric coupled to high thermal mass.


  4. wonderful share about the carbon intensity of UK grid electricity. this installation will really help consumers saving energy as they are so energy efficient. I think these housing energy efficiency buildings are the future.

  5. Hello,

    I like your blog.

    I want to say that the analysis of the carbon value of the grid could be different.

    From your blog:
    "As a result the average carbon intensity of electricity in 2016 at 269 gCO2/kWh was only just higher than that for gas (216 gCO2/kWh). When you add in an efficiency for a gas boiler at (say) 80%, the gap disappears."

    The UK electricity supply grid and the UK heat supply grid are not the same. What I mean to say is that the heat demand of most UK properties is met by gas, and so is not included in the UK electricity grid supply numbers. Yes, the electricity grid includes some home heating, but this is a minor fraction, instead most homes are independent of the electricity grid for heating. To do the analysis you want to do about the carbon worth of electric versus gas heating, you would need to look at the carbon intensity of the entire UK energy supply for heat and electric, not just look at the carbon value of the electricity grid. If you look at the relative size of these two supply networks you will find that the electricity grid couldn't supply heat to homes at the carbon value you are supporting(currently anyway - maybe in future).

    From your blog;
    "So now the moment when you choose to take power from the grid is a strong determinant of the actual instantaneous carbon emissions your electricity use is creating."

    The average carbon grid value here is misleading. The 'average carbon value' is not the 'marginal carbon value' of increasing or decreasing the use of electricity at any time(e.g., the carbon cost of using one more or one less unit of energy). Except for a few special days (like 11th June), in the UK there is nearly always some gas fired electricity being produced. As this is the electricity source that will flexible respond to demand, if we were to decrease the electricity demand by 1GW, then we stop using 1GW's worth of gas. Likewise, if we increase demand by 1GW then we use 1GW's worth more of gas. In short, if you turn on the kettle, then a gas fired plant will respond to that demand, so the carbon value of boiling that kettle will be at the carbon cost of gas. Similarly, it doesn't matter when you boil the kettle as the demand will be met by a gas fired power plant day or night. So the marginal carbon cost of another unit of electricity is pretty stable. Only, when we have excess electricity power on the grid (and the gas fired power plants have turned off completely), will we be at the point where it matters when you turn on your appliances.

    Hope this makes sense.

  6. Carbon intensity is not a joke and people should keep their eyes on it!