Well, convection still isn't over the center, but it is closer to the center and of slightly better "quality" than yesterday. Also, as the NHC noted, the radius of maximum winds has contracted since yesterday, another sign of better organization. Thus, given the flat thermal profiles that recon found -- which aren't all that far off from what the models had shown, probably because of the Jacksonville and Charleston upper air obs and their trajectories -- we've got a subtropical storm.

In my opinion, it's still a borderline case, but given the proximity to land it gets the upgrade. If this were out in the central Atlantic, I'm not so sure it gets the upgrade regardless of recon status. As for impacts from this one, think Tammy from 2005, only on the opposite trajectory (southwest then south rather than north then northwest). Tammy was another storm with dubious structure that eventually brought an inch or so of rain to parts of the Southeast US.

NHC forecast looks good and reasonable; this is one where you really don't need modeling, at least for the intensity portion. SSTs aren't warming -- and in fact cool from here on out -- and the upper level temperatures at the tropopause aren't cooling, so it's like a TC over cooler waters in the East Pacific. A slow decay should be expected over the next couple of days, made slower by the fact that it still has a signature aloft.

It's been an interesting week for the storm system that spawned Andrea, starting off in Texas with two consecutive nights of MCS formation (including one night with a derecho) and progressing eastward along the periphery of the subtropical ridge that had been parked near Florida. The last TC that I can recall that had any tie to an MCS (large thunderstorm complex over land) was Danny in 1997, which formed in the Gulf after one such complex moved out over water and did not dissipate like most of them do.