July 11, 2018 · 4 min read
Fossil fuels are the largest source of greenhouse gases. As they still count for 80% of our energy source, and 66% of our electricity source, it becomes urgent to get rid of them and transition to low-carbon electricity. Therefore, the best metric that describes our advancement towards a fossil-free electricity should be how much we can reduce CO2 emissions each time we install a renewable asset.
We should evaluate renewable installations on how much CO2 emissions they reduce
Although this may seem obvious, we are currently treating all renewable installations as equal. We measure the capacity added and not the emissions reduced by an installation with the hope that the renewable production will reduce emissions by displacing more carbon-intensive means of generation.
Luckily, we’ve already developed the algorithms to assess this: the marginal carbon intensity model. This model tells us what type of power plant reduces its production when renewables are producing instead. By associating each power plant type with its carbon emission factor, we can deduct the carbon impact associated: that’s the marginal carbon intensity of the electricity grid at a given time.
By knowing when a renewable produces its electricity, we can infer the carbon emissions avoided due to the displacement of an existing production source (in the best case scenario, this would be coal power plant).
Let’s investigate what would have happened if an additional kW of wind power capacity would have been available last year (2017) in Germany. By assuming that the new wind turbine produces electricity at the same time as others, and that they reach maximum production at the same time as others, a typical production profile (having a varying load factor) can be inferred:
Multiplying the marginal carbon intensity with the amount of grid electricity avoided by the additional wind production therefore gives us the CO2 avoided at each hour:
Two aspects come into play at each hour:
The CO2 reduction achieved depends on those two factors.
It is not sufficient to simply install more renewables: we also have to make sure to install them in places where the use of fossil fuels will be reduced.
Running the above analysis for all European countries for which we have data, we can construct a ranking:
Great Britain is the area where installing wind turbines yields the most CO2 reduction, as it displaces mostly fossil fuels (in this case gas). Note the interesting case of Eastern Denmark (DK-DK2) that displaces a surprisingly low amount of emissions. This is due to the fact that installing more wind production in Eastern Denmark has the effect of reducing hydro imports from Sweden to a larger extent than local coal and gas generation (see our previous article). As hydro generation already is low-carbon, the benefits of replacing it with wind power are very limited. The same is also true of Poland where a mix of local coal generation and Swedish hydro imports will be displaced, thus making it approximately the same level as Great Britain.
Furthermore, as building a wind turbine is not a carbon neutral investment (concrete is required to build them, ships to install them offshore..), we’ve added the additional CO2 emissions (from the IPCC) incurred as an indicator. Those emissions are calculated over the total life span of the asset, and converted to a yearly average. Using this indicator, installing a wind turbine in Norway — where it would only displace low-carbon hydro — would actually yield a slight CO2 increase, as the reductions would be insufficient to cover the initial emissions.
Currently, the calculations only take into account the short-term marginal emission factor, which means they are only valid when considering the installation of small assets. The installation of larger assets (such as a wind farm) requires taking into account long-term policy changes, capacity plannings or phase-out strategies.
Subscribe to get our latest content by email.
We'll send you updates on electricityMap, no more than once a month. You can unsubscribe anytime.