Interesting observations indeed.  The 31st October GQN turned out another big disappointment participation-wise;  I'm certain that had stations been operating, I would have heard most of them, in the same manner that the usual activity on 3.677 MHz was 5x7.    I too heard ZL2BH at 2013, 5x2, then at 2019 heard another station 5x4 asking if the frequency (3.690 MHz) was in use, but he failed to identify, and like John, couldn't have heard  my CQs.  Repetitive CQs on 3.535 MHz CW also elicited no response. Suspect the D-layer may have still been ionized and was attenuating my high-angle NVIS signals.     The receive conditions at my portable site at Waitārere with an end-fed 1/2 wave seemed to be at least a good average, as evidenced by the 25 W CW-training transmission from VK1WI at northern Sydney suburb Dural, coming in 5x2  -  only slightly down on its normal level.  There was little to nothing in the way of typical 80 m QRM apart from occasional barely-detectable 10-second bursts of OTHR-type pulses.  So that left only one partial impediment to good signals - apart from the obvious lack of GQN participants - that one partial impediment being QRN, which averaged a tolerable S2 ~ S3 and only rarely peaked as high as S8.   Packed up at 22h05 and went home, despondent, without a single log entry.

Worst conditions yet - I believe.  S9+10 both at Conroys Dam Rec Reserve (gave it 15 mins before packing up) and at home. 

Wow!