Question: AWS Grid emissions factors / renewable energy projects / Renewable Energy Credits

[Mike Chambers]

(Transferring this from issue thread on GitHub. Thanks.  https://github.com/cloud-carbon-footprint/cloud-carbon-coefficients/issues/13)

Hello. Great work here. Thanks.

To the best of my understanding, over the last few months Amazon has significantly increased it's portfolio of renewable energy projects. These can be seen here: https://sustainability.aboutamazon.com/about/around-the-globe?energyType=true

In addition to this, AWS report that:

AWS purchases and retires environmental attributes, like Renewable Energy Credits and Guarantees of Origin, to cover the non-renewable energy we use in these regions: - U.S. West (Oregon) - GovCloud (U.S. West) - Europe (Frankfurt) - Canada (Central) - Europe (Ireland)

Src: https://sustainability.aboutamazon.com/environment/the-cloud?energyType=true

Given this, Is there a method (or would you seek a method) to combine this data into /data/grid-emissions-factors-aws.csv ?

For example would there be an issue with setting the CO2e to zero for each of the regions listed above, and making reductions to other regions where Amazon has local renewable energy projects?

Thoughts?

>>> RESPONSE FROM David Mytton

Thanks for the question. Currently, the grid factors are used based on public data for the local region. RECs / REGOs have no impact on the carbon factor because they're market instruments, which is why the local emissions factors are used. Deploying new clean energy to the grid would reduce the overall grid emissions factor rather than zeroing the emissions for a specific cloud region (or DC).

[Mike Chambers]

So can you help me understand why RECs / REGOs can't be used to adjust our calculations?  This seems like an important issue for me to be able to understand.

Then around Amazon's renewable energy projects... Is the point here that these would be feeding the grid, therefore they would be included in the grid emissions calculations and so while a company may sponsor all kinds of initiatives they can't be attributed to one user of the grid, they need to be applied to all?  Is that about the size of it?

Thanks again.

[David Mytton]

That's right. Once power enters the grid it's not possible to trace where electrons go, so you can't say that electrons from clean sources go to DC A and electrons from fossil sources go to DC B. It's all part of the grid mix, which is what is reflected in the average emissions factors. This is itself a simplification because the grid mix changes on a minute to minute basis whereas the averages reflect the annual mix. This is what you see with work towards "100% clean energy 24/7" as opposed to annual matching (although that still has the same electron matching issue). Google's paper describes this: https://storage.googleapis.com/gweb-sustainability.appspot.com/pdf/24x7-carbon-free-energy-data-centers.pdf

RECs and their equivalents only say that clean energy was produced somewhere at some point, so aren't specific enough to be able to use in the way you are hoping. These are also complex instruments depending on whether they are attached to specific projects and whether those have additionality. Using these is an early step on a journey to netting off your emissions (buying clean energy that is equivalent to your usage), but is not sufficient.

I've written a few things on this topic if you want to read more: https://davidmytton.blog/how-can-data-centers-use-100-renewable-electricity/ is a good place to start and I wrote a report for Uptime Instute which is freely available if you submit the form at https://uptimeinstitute.com/resources/asset/renewable-energy-for-data-centers and the exec summary is at https://davidmytton.blog/recs-cannot-be-used-to-back-claims-of-100-renewable-energy/

[Mike Chambers]

Thanks for the detailed explanation, I do of course see your logic. It leaves me wondering how we incentivise cloud businesses to sponsor these kinds of initiatives and get some kudos/credit for doing so. 

I will read your articles, with interest and thanks.  

--
We're excited to hear from you and hope you're enjoying Cloud Carbon Footprint.
Please fill out our feedback form: https://forms.gle/nFzkRioryy4R1DGB6
Add your name to ADOPTERS.md: https://github.com/cloud-carbon-footprint/cloud-carbon-footprint/blob/trunk/ADOPTERS.md
Give us a star on the github if you're enjoying the tool!
---
You received this message because you are subscribed to a topic in the Google Groups "Cloud Carbon Footprint" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/cloud-carbon-footprint/_yjwrr3RU0I/unsubscribe.
To unsubscribe from this group and all its topics, send an email to cloud-carbon-foot...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/cloud-carbon-footprint/6d28bb51-c1d4-4e5a-a96e-8aac8e8628f4n%40googlegroups.com.