the science behind WGI depends partially on computer models and may have quite a few errors itself.

That may be true, but I think the sort of sloppy citation (grey lit, not lit), typo-type errors of fact and misrepresentation that have been getting press lately would be less common in WGI. That’s just my impression from trawling through the reviewer comments.

I don’t think there’s a way for the IPCC process to catch an error caused by a bug if the original group doing the implementation and model runs didn’t catch it, but the WGII stuff that’s been revealed is in the category of stuff the process can catch. That probably reveals a bit of a ’sampling bias’ (maybe that was your point?), since people can do simple fact checking to find these sorts of errors, but rerunning the models requires a significant ‘infrastructure’.

JStults,
WGI is about the physical science basis (how do we know climate is changing and how do we know humans are causing it). WGII is about the impact of climate change. I would expect more errors (and more speculation) in WGII, due to its nature, but the science behind WGI depends partially on computer models and may have quite a few errors itself.

I’ve never written, edited or reviewed an IPCC report, but I have done my share of reviewing papers. I don’t get paid by the journal, but it’s accepted at the institution where I work that this sort of thing is part of my job. I can hardly claim this work is unpaid. I imagine the same applies to most scientists’ involvement in the IPCC process.

Let’s face it, when you get paid for your hobby, you’re not in a position to complain too hard.