"As we move to a more electric world we must realise that the "when?" of energy use is as important as the "how much?" for accurate carbon emissions accounting. Once we understand that we will make better decisions and decarbonise faster." Barny Evans
As we aim to decarbonise WSP (and the entire UK economy) it is crucial that we get the calculation of these emissions correct. As part of the next step in our leadership in this area we have been considering how we could more accurately calculate CO2 emissions.
With most fuels this calculation can be a simple static calculation of the CO2 emissions emitted; with electricity however, it is more complicated. Electricity in the UK is generated by a mixture of nuclear power, gas and coal-fired powered stations, solar, wind and hydro among others. The mix of these fuels gives an overall CO2 emission factor for our mains electricity. Typically, a static emissions factor has been used to capture this mix (for Building Regulations Part L this is currently 0.519g CO2 per kWh). These static factors are important as they are used in corporate reporting and in defining “zero carbon” homes and cities.
However, these factors don’t capture the variability of CO2 intensity of the grid and rapid changes in the way electricity is now being generated. We are now able to take real data from the National Grid every hour to get live CO2 emissions factors for electricity (considering the mix of technologies generating at each point in time). The graph below shows the grid electricity emissions factors for March to December 2018, comparing the Part L and Defra static emissions factors to the live grid intensity.
Figure 1 - Comparison of CO2 Emission Factors over time

You can see the difference is significant; our current Building Regulations require a flat rate of 0.519, whereas the “actual” emission factor on some occasions is <0.100, but varies nearly fourfold, even during a day. The Defra emission factor we use for company GHG emissions reporting also just has a flat rate. When applying this methodology to the energy use at our Newcastle office over this same period, we discovered that the static emissions factors gave a result that was more than double that when we used the live grid data (70 tonnes of CO2 versus 30 tonnes of CO2). WSP regularly provide advice on carbon reduction so it is important to get this right.
Electrical supply from renewables varies both seasonally and depending on the time of day. Solar energy generation, for example, is most prominent in the summer daytime, which allows us to turn-off higher carbon power stations, such as coal. The result of this variation in generation potential and demand is that during the summer, and during the middle of the day, or middle of the night, the carbon intensity of electricity is much lower than in a winter evening. This means that a kWh used at one time can be >4 times as CO2 intensive as another.
This should be guiding decision making about when we use power to help increase use of electricity when it is low carbon and reduce it when it is high. With static emission factors though we don’t recognise this at all and at WSP we think it is likely that we will have to change that soon. For example, by updating the methodology to take account of the variability make the business case of deploying energy storage which is a crucial part of our low carbon future.
At WSP we are exploring how we can improve CO2 emissions calculations and therefore decision making. We have made a start by suggesting that we all use the hourly emission factor to see more accurately what our emissions are, and how we can best reduce them. We can report these in parallel with standard reporting until these are updated. There are more issues around the long term and marginal emission factors we are investigating.
We are keen to hear from stakeholders in this area and see how we can work together to drive this forward.
By Sabbir Sidat, Principal Consultant, Sustainable Places and Energy Management
Barny Evans, Technical Director - Energy, Waste & Sustainable Places, WSP