Carbon accounts 2021: looking back and looking ahead

Click the image for a larger version. Representation of the reduction in domestic carbon dioxide emissions from gas and electricity over the last four years. Also shown are the steps I have taken to achieve these reductions.

[Article Summary: it is actually pretty difficult to estimate carbon dioxide emissions – but it is important to try.]

Friends, it’s New Year’s Day.

And at this still point in the ever-rolling cycle of the years, it seemed like a good time to summarise progress on my project to reduce carbon dioxide emissions from the house.

It may seem like a good time, but actually this is not a good time at all to summarise annual emissions. The best time to do this is 6 months away in the summer. This is because household emissions peak in the winter and fall to practically zero in the summer.

You can see this on the graphs below which show cumulative electricity and gas consumption through the last few years.

Gas

To estimate the emissions from gas I have estimated the meter reading which I would have had in the summer of 2018 and used this as a baseline.

I then subtract weekly meter readings from this baseline and convert them to kWh of thermal energy, and then multiply the number of kWh of gas by 200 gCO2 per kWh.

Burning methane gas in a boiler releases around 183 gCO2 per kWh of gas – as documented in this official spreadsheet. (Look up the ‘Fuels’ tab and use cell F42) But some additional CO2 emissions are associated with delivering the gas to my home: compressors drive the gas along pipelines and ships deliver gas across the oceans.

The actual value of these ‘upstream’ emissions is difficult to know precisely, but actual experts suggest it amounts to roughly 24 gCO2 per kWh of gas delivered. So in principle the best estimate of CO2 emissions from gas delivered would be 183 + 24 = 207 gCO2 per kWh. This figure is 4% larger than the figure I used.

However, it is likely that direct methane leaks at wells and in handling plants are underestimated (example). Evaluated over a decade after leakage, methane is 84 times more powerful as a greenhouse gas than CO2. So if even 0.1% of methane leaked on its way to my home, the CO2 equivalent emission would be increased by 17 gCO2/kWh. However some people estimate that actual leakage is more than 1%. If that were so that would practically double the climate impact of using gas to nearly 400 gCO2 per kWh. The unknown magnitude of leaks is just one more reason to stop using methane gas.

Given these uncertainties I have used a figure of 200 gCO2 per kWh as a likely underestimate of true emissions which is not obviously wrong, but which is a convenient round number.

Click the image for a larger version. The graph shows the cumulative emissions of carbon dioxide emissions from domestic gas use over the last three years.

The graph above is based on weekly gas meter readings.

The data form a series of ‘steps’ and it is clear that measuring from one step level to the next gives a better estimate of the yearly emissions than choosing an arbitrary point on the ‘riser’ of the staircase. This implies measuring from summer to summer

The reason is that if the winter is mild before the New Year but cold after New Year, the emissions fall in different years even though they arise from the same winter.

But however one analyses the data, it is clear that there has been no step this winter of 2021/22. We now use gas only for cooking and I hope shortly to stop even this use and make that curve go entirely flat. For ever!

Electricity

To estimate the emissions from electricity I have multiplied meter readings in kWh by 230 gCO2 per kWh.

As with gas, it is not obvious how much carbon dioxide is emitted for each kWh of electricity consumed from the grid. Depending on the generation source, the so-called carbon intensity of the electricity can vary significantly. For example, as I write – with low demand and high winds – the carbon intensity of the electricity is just 111 gCO2 per kWh.

The MyGridGB web site maintains a live monitor of carbon intensity, and shows an annual summary of average carbon intensity through the year.

Over the last 3 years the average carbon intensity in the UK has been 245, 222, and 235 gCO2 per kWh. Since these figures are within a few percent of each other I have used a rounded value of 230 gCO2 per kWh for the entire range of the measurements.

Click the image for a larger version. The graph shows the average carbon intensity for each year. The values for eth last three years have been 245, 222, and 235 gCO2 per kWh. The red line at 100 gCO2 per kWh is the target carbon intensity for the year 2030.

Multiplying the number of kWh used by 230 gCO2 per kWh tells me the emissions associated with my use of grid electricity. This is shown below on the same vertical scale as on the gas graph above.

Click the image for a larger version. The graph shows the cumulative emissions of carbon dioxide emissions from domestic electricity use over the last three years. The scale is the same as in the previous graph showing carbon dioxide emissions from gas use.

The graph above is based on weekly electricity meter readings.

During 2019 and 2020 carbon emissions occurred at a regular rate with no seasonal steps. Even, looking closely, I cannot detect the point in November 2020 when solar panels were installed.

But in March 2021 when our Powerwall battery was installed, the curve goes flat as the combination of solar panels and battery was sufficient to take us practically off-grid for the summer.

In September 2021, as solar generation weakened, we began to draw electricity from the grid again, and also began heating with electricity using our air-source heat pump. Currently we are using 20 to 25 kWh/day – more than twice the previous rate. This will probably continue until March

It is hard to estimate precisely, but I think – with colder months ahead – the summer-to-summer emissions will be similar or slightly less than last year.

Gas and Electricity

To estimate the emissions from both gas and electricity use, I have added the data from the two previous graphs together.

Click the image for a larger version. The graph shows the cumulative emissions of carbon dioxide emissions from domestic gas and electricity use over the last three years. The scale is the same as in the previous graphs.

Anticipating data from the spring of 2022, it looks like emissions will have fallen from about 3.6 tonnes in 2018/2019 to (hopefully) only 0.7 tonnes in 2021/22. This is an 80% reduction.

But even though this has already been a tedious article, this is not quite the end of the story.

Embodied Carbon

To achieve that 80% cut in annual emissions, I had to buy things which involved the emission of carbon dioxide – so called embodied carbon.

It is difficult to estimate the amount of embodied carbon in a particular object, but after quite some effort I have come up with the following estimates.

Intervention	Embodied tonnes of CO2
EWI PU Boards	1.6
EWI Mortar	1
Argon Triple Glazing	1.9
Solar Panels	1.6
Battery	1.4
Heat Pump	1.5
Air Conditioning	1.5
Total	10.5

They amount to 10.5 tonnes of embodied carbon. To find out when this embodied carbon has been ‘paid for’ I need to compare the CO2 emissions described above with the so-called counter-factual: the emissions which would have occurred if I had done nothing.

If I had done nothing then my guess is that emissions over the last 4 years would be simply 4 x 3.6 tonnes of CO2 – or 14.4 tonnes.

Actually, CO2 emissions over the last 4 years have been 3.6 + 2.9 + 2.0 + 0.7 = 9.2 tonnes.

So my ‘investment’ of 10.5 tonnes of embodied carbon will have saved 5.2 tonnes of emissions by summer 2022, and should continue to save (3.6 – 0.7) = 2.9 tonnes per year for several years to come. So I should ‘break even’ during the year 2024. Everything beyond that will be pure emissions savings.

Looking ahead

Click the image for a larger version. The graph shows the estimated household emissions from 2018 to 2040. The red line shows the emissions which would have occurred if I had done nothing. The green line shows the emissions according to the current plan. The dotted line shows emissions if the money I pay to Climeworks is not a scam.

Looking further ahead, the tonnes of carbon ‘debt’ I have incurred seems less significant. And carbon dioxide emissions avoided by 2040 amount to 60 tonnes.

Additionally since March 2021, I have been paying Climeworks £40/month to permanently remove 50 kg/month (0.6 tonnes/year) of carbon dioxide. If they are actually doing this – and I have no real way of knowing! – then our household is very nearly carbon neutral.

However none of what I have discussed accounts for emissions arising from consumption, or travel, or from my pension investments – all of which are likely to be quite significant.

So there is still lots to do in the new year.