I could well be missing something here about what is really happening. Near as I can tell the only difference in the two situations was the temperature as described in the OP.
Battery SOC (near full) was taken out of play AFAIK by using loads at higher amps than the panel can ever do then seeing how much of the load the panel took (as seen on Trimetric) and also on the controller's own ammeter reading.
Panel is tilted up with a breeze behind it so no trapped heat, just the sun on the front, temp taken at back on the white part.
Panel wattage taken from controller's readout, which appears to use its ammeter amps x battery voltage. One thing I see is that the panel wattage stays about the same while you can vary the battery voltage with loads. As the batt voltage changes, that makes the amps change where the wattage is steady. In the Now test as seen in the OP, panel wattage was only 181-4ish when in better times it is closer to 200. (of 230)
It is a little cooler today in a sea breeze but with same good sunshine blue sky so I will get another set of numbers for comparison with yesterday's if panel temp is lower.
EDIT- another thought--the panel to controller wiring is strung out along the ground from the contraption to the trailer, so that wire is getting hotter in the sun too. I haven't measured comparable voltage drops Spring/Now, but does wire resistance go up if the wire is hotter?