near IR, heat flux direction, CDRs
I think this will be a superb addition to the web pages.
At present the impression is given that insolation is irrelevant to the skin effect, since the point is made that visible absorption is negligible in the skin layer. But something like 10% (from memory) of the SSI is in near IR wavelengths where absorption length scales are comparable to 1 mm. Thus, when driven with a radiation model that includes the near IR bands, the likes of the Fairall skin model show insolation-related variability in the skin effect of about a third of the mean effect. I think distinguishing near and thermal IR would be useful.
Although GHRSST has a global focus because of its origin, I also wonder about the emphasis in the discussions on the usual case of the heat flux being out of the ocean. In coastal zones and inland waters the opposite flux is less unusual. Generalising the discussion to deal with either case would not greatly complicate the text, I think.
I'm not comfortable with "foundation temperature is defined and suggested for use in climate applications". It reads as if such a choice solves the need to account for skin and diurnal variability when generating CDRs. But the problem in estimating SST-f is the same as if we choose a shallower depth: to infer a sub-surface SST (whether foundation or depth) from skin-sensitive observations at some arbitrary time of day. Compatibility of (at least some versions of) satellite CDRs with historical in situ CDRs is a strong requirement of climate users, and SST-depth is a natural choice to pursue this.
Other than that, good stuff, thanks for drafting it.
(Please correct the use of ms^-1 as a unit of speed -- it is a unit of frequency in SI.)