Carbon footprint of my household gas and electricity use

One of the single largest contributors to personal carbon dioxide emissions is energy usage at home.

In my case, this includes natural gas and electricity, but for others it may involve bottled gas or heating oil. Regardless of the specific form, the reason behind carbon dioxide emissions from energy usage is the burning of fuel. When we burn anything containing carbon in the air it generates carbon dioxide. For example, I burn carbon-containing natural gas to heat my house, heat my water, and cook my food. My electricity comes from the national grid which means I don't personally burn anything for it, but generators on the grid burn natural gas, coal, and biomass, like wood pellets, on my behalf to generate electricity.

In all these cases, carbon dioxide is emitted.

Calculating the amount of carbon dioxide emitted is theoretically straightforward, but obtaining accurate figures is challenging for various reasons.

The basic approach is to multiply the amount of energy used (measured in kilowatt-hours, kWh) by an emission factor that represents how much carbon dioxide is emitted per kilowatt-hour of energy used (measured in gCO_2,/kWh). This multiplication gives you the amount of carbon dioxide, in grams, which can easily be converted to kilograms or tonnes.

However, obtaining accurate values for energy usage and emission factors can be challenging. For now, I'll work through a first approximation using readily available information.

Working out personal usage

I'm the kind of person that keeps copies of documents like electricity bills and gas bills. I also take more meter readings than strictly necessary to see how my usage varies over time. For instance, I take readings before and after I leave the house for a trip away.

Unfortunately I soon realised that I didn't have enough meter readings at the right times. I wanted them as close as possible to January 1st, but in some years, the closest I had were in November or late January. I solved this by deciding on matched dates and accurate (i.e. real, not estimated) meter readings that roughly matched real years. For example, a "year" could run from November 15th one year to December 28th the next year. Finally, I corrected the numbers to account for the fact that we'd moved house several times and had meters changed more than once.

Although this means that some "years" are longer or shorter, it doesn't affect the data when considering it over a longer period. On average, my year length was 357±48 days, which is very close to the real year length, even though my shortest "year" was 256 days and my longest "year" was 496 days. By normalising for the number of days in each "year" and scaling back to a true year length, the numbers will be accurate and comparable.

After all this work, I can now plot gas and electricity usage between 2005 and 2022:

You can clearly see the impact of living in a smaller house: our gas consumption was much lower in 2009-2012 compared to recent usage. However, that house was only 52 m², much smaller and cheaper to heat than our current 110 m² house.

You can also see a consistent decrease in our gas usage since 2016, reflecting our ongoing efforts to improve insulation and upgrade the fabric of the house. I'll be exploring this journey in another article.

Converting personal usage to CO₂ emissions

As mentioned earlier, converting energy usage in kWh to carbon dioxide emissions in kilograms is straightforward by multiplying usage by the appropriate emission intensity factor. Although working out what the "proper" emission factors are is complex, for now I'm going to use the electricity emission factors provided by Andrew Crossland at mygridgb.co.uk¹. These factors represent national averages over the year and provide a high-level overview of my emissions. For natural gas I used an emission factor of 203 g CO2/kWh ².

While the emissions from burning natural gas remain stable over time, the emissions associated with electricity usage are dropping over time as more renewable generation is connected to the GB grid and with the replacement of coal by natural gas in power stations.

This improvement in emission factors over time looks like this:

This drop in emissions is remarkably quick for a large national grid, with a kilowatt-hour of electricity used in 2022 generating half as much carbon dioxide as a kilowatt-hour used in 2005.

By multiplying these emission factors with my usage, I can determine my personal household carbon dioxide emissions for gas and electricity:

What's clear from this is that burning natural gas is the single largest contributor to my household energy emissions, averaging at over 2 tonnes per year - equivalent to a flight from London, in the UK, to Santiago, in Chile.

Comparing consumption to national averages

It's useful to compare personal usage with average consumption.

Our electricity usage is remaining steady at about 2,500 kWh/year, or 6.9 kWh/day. Our gas usage is slowly declining but is currently about 10,000 kWh/year, or 27 kWh/day.

According to Ofgem³, typical "medium" usage for a household of 2-3 people is 2,900 kWh of electricity and 12,000 kWh of gas per year (7.9 kWh/day and 33 kWh/day, respectively). This means we use about 86% of typical electricity and 83% of typical gas, making our consumption fairly typical for a UK household.

Another perspective is to convert kWh/day into continuous power usage in Watts. In this case, our electricity usage is equivalent to a continuous power usage of 330 W and our gas usage is equivalent to a continuous consumption of 1.1 kW. It will be good to revisit these numbers as I investigate what fraction of our electricity is "base load", i.e. on 24/7 like fridges and Wi-Fi routers, and what fraction is "elective" i.e. things we switch on for specific purposes, like TVs and computers.

Overall, it's clear that there's still a lot of room for improvement in decreasing overall energy usage and associated emissions, especially by transitioning away from the gas-fired boiler.