Oscar

Sounds good! The only thing that could be improved is the warning message that pandas gives - it's pretty opaque! Though appreciate this is not your department. Might be useful...

Thanks for the quick and comprehensive response. I'll fork, and try to find time to work on this. Got quite a lot on with other projects at the moment though,...

Following on from #1196: The docs currently have '0.7.0' written at the top of the page and in the page title, implying that they relate to the latest release. Until...

Modified carbon_fns with fsolve calls so uncertainties no longer cause errors. HOWEVER: Error propagation in fsolve functions is invalid. Not yet implemented in uncertainties package - lebigot/uncertainties#69.

Might be able to implement it with existing uncertainties.wrap function, now that zero finders are no longer vectorised (#2)?

Good thoughts @tompc35. Briefly: 1. TB: You can also specify BT at S=35 manually, and it scales it with salinity. Falls back on Lee value *IF* not value is provided....

Hi @markusritschel, thanks for getting in touch! The short answer to your specific question is 'no'. It's not currently possible to select alternative parameter sets as an option. I've only...

Probably not in practice - but it's no extra work, so might as well add it for completeness? Just pushed an updated cbsyst.py to the co2sys_compare branch (https://github.com/oscarbranson/cbsyst/commit/9110fd31994b8be43cd3c438722139a67c2c25ba), which contains...

I *think* I've completed this. Just need to test. Best way of testing probably against GLODAPv2 data, which have both in-situ ('phtsinsitutp') and 25 C 0 Bar ('pht25p0') variables. Preliminary...

Thanks @tompc35 - that would be really useful. I've actually made a start on this, but would welcome help. I was thinking of comparing cbsyst to GLODAP/WOA data, after [Orr...