So first off my opinion
Carbon Capture and Storage is a strange one.
My first reaction was:
Why don’t we just stop producing the CO2?
Which is probably the obvious reaction.
But the more I’ve learned about energy and heavy industry, the more I’ve realised that “just electrify it” isn’t always an immediate answer.
Some industrial processes inherently produce carbon dioxide.
Some existing infrastructure isn’t disappearing tomorrow.
And some sectors are considerably harder to decarbonise than others.
So the question becomes:
If we’re going to produce CO2 anyway, can we catch it before it reaches the atmosphere?
That’s Carbon Capture and Storage.
CCS.
So how does it work?
At a very high level there are three parts.
Capture it.
Move it.
Store it.
The CO2 is separated from industrial exhaust gases or other processes.
It’s then compressed and transported, potentially using pipelines or ships.
Finally it’s injected underground into suitable geological formations.
Potentially kilometres below us.
And that’s the bit that originally made me raise an eyebrow.
We’re going to pump enormous quantities of CO2 underground and leave it there?
Apparently, yes.
But obviously the geology matters rather a lot.
Storage isn’t just finding a hole
This is where my original mental picture was wrong.
We’re not digging a giant cave and filling it with carbon dioxide.
The CO2 can be injected into porous rock formations with impermeable rock above them helping keep it contained.
Old oil and gas reservoirs can potentially be useful because we already know something about their geology.
Saline aquifers are another possibility.
So suddenly CCS becomes partly an energy problem and partly a geology problem.
Which means yet another collection of data.
The North Sea makes this particularly interesting
For the UK this feels especially relevant.
We’ve spent decades developing offshore oil and gas expertise.
We have pipelines.
We have offshore engineering knowledge.
We have geological data.
We have depleted reservoirs.
So some of the infrastructure and knowledge built for extracting carbon from beneath the North Sea might potentially have another life putting carbon back underground.
There is something rather circular about that.
But capture isn’t free
This is the bit I think sometimes gets lost.
Capturing CO2 requires energy.
Compressing it requires energy.
Moving it requires infrastructure.
Injecting it requires infrastructure.
Monitoring the storage site requires work.
So CCS isn’t:
Build this box and emissions disappear.
There is an efficiency and financial cost.
Which means whether CCS makes sense depends heavily on where it’s being used.
If we can cheaply avoid producing the carbon in the first place, that’s obviously attractive.
But if the alternative is an industrial process that is extremely difficult to decarbonise, CCS starts looking rather different.
And then comes the monitoring
This is the part that interests me from a data point of view.
If we’re going to put millions of tonnes of CO2 underground and claim it’s safely stored, presumably we need evidence.
How much was captured?
How much entered the pipeline?
How much arrived?
How much was injected?
At what pressure?
Into which well?
What’s happening underground?
Is there evidence of migration?
Are the wells healthy?
Is the amount stored consistent with the amount we think we’ve injected?
That’s a serious chain of data.
And presumably that data needs to remain trustworthy for a very long time.
Data lineage suddenly becomes physical
In normal data engineering we talk about lineage.
Where did this number come from?
What transformed it?
Where did it go?
CCS potentially gives that concept a very physical meaning.
A tonne of CO2 has effectively got a lifecycle.
Created here.
Captured here.
Transported through this infrastructure.
Injected here.
Monitored afterwards.
If governments, businesses and the public are going to trust claims about captured carbon, the data behind those claims has to be pretty solid.
That’s something I find particularly interesting.
So where is my head at?
I don’t see CCS as an excuse to keep emitting carbon forever.
But I also don’t think the choice is necessarily:
CCS or renewables.
There are industrial processes and existing assets where the real choice may be much messier.
My current view is that CCS probably makes most sense where eliminating the original emission is genuinely difficult.
And if we’re going to do it, the monitoring and data side needs to be treated as part of the infrastructure rather than something bolted on afterwards.
Because burying CO2 is one thing.
Proving where it went and what happened to it over the following decades is another.
That’s where my head is at the moment.
Gareth Winterman