Add TARGWP20, TARGWP100 and TARGWP500 from third assessment report

[mikapfl]

We actually found a usage of global warming potentials from the third assessment report in the wild; therefore, we need them. Source is linked. I transferred it from pdf by hand (a mind-numbing exercise), so there might be errors. I'm willing to re-check tomorrow, but today it wouldn't make much sense.

[znicholls]

I'll take a look now @mikapfl

[znicholls]

Nice. A few typos and questions I think. Was indeed hard going through this with a pdf.

Is there a reason we're adding extra gases like "HGalden1040x"? Or is it just cause we've read them now, so may as well keep them?

[mikapfl]

Note: I didn't include -(CF2)4CH(OH)- because the - are significant and we can't deal with them easily and it's just not worth the headache to figure something out given this is likely not used anywhere we care about.

[rgieseke]

Note: I didn't include -(CF2)4CH(OH)- because the - are significant and we can't deal with them easily and it's just not worth the headache to figure something out given this is likely not used anywhere we care about.

Where is this not working? Should be a valid JS object or Python dictionary key, isn't it?

[mikapfl]

Where is this not working? Should be a valid JS object or Python dictionary key, isn't it?

It won't work as a normal unit in pint, i.e. in openscm-units, I think.

[mikapfl]

Thanks for checking everything, Zeb! Very much appreciated. To be extra sure, I now did an alternative extraction using https://smallpdf.com/pdf-to-excel and targeted copy+pasting, which means I didn't have the chance to introduce typos using this way. I found two missing values for HFE356pcc3 (I was apparently also confused, it is super confusing because the gases are not really ordered alphabetically or something).

Now I'm very confident that the values included here reflect what's in the official pdf.

[mikapfl]

Is there a reason we're adding extra gases like "HGalden1040x"? Or is it just cause we've read them now, so may as well keep them?

I just figured, if it is the pdf, let's add it now while we deal with this pdf and then never have to think about it again, hopefully.

[rgieseke]

Where is this not working? Should be a valid JS object or Python dictionary key, isn't it?

It won't work as a normal unit in pint, i.e. in openscm-units, I think.

Maybe openscm-units should then just ignore it (it's very likely not handled by any SCM I guess). Not sure what level of completeness we have for other reports. If we include 'HGalden1040x' then why not '-(CF2)4CH(OH)-'?

[mikapfl]

Maybe openscm-units should then just ignore it (it's very likely not handled by any SCM I guess).

Ok, I added it and will deal with the fallout in openscm-units.

I did the same amount of checking for it (extract via two different routes) so am confident in the values as well.

[znicholls]

Now I'm very confident that the values included here reflect what's in the official pdf.

Nice!

Ok, I added it and will deal with the fallout in openscm-units

Hmph, but fair I guess

[znicholls]

Super, thank you!

[rgieseke]

Released as 0.10.0, thanks @mikapfl and @znicholls!

[rgieseke]

0.10.1 - forgot to re-run make.

[mikapfl]

0.10.1 - forgot to re-run make.

Something which could be included in pre-commit or other CI 😉