Kilokat7 posted this waterfall he took of 1629 kHz, showing carriers from Australian MW stations at 1108z on 16 Sep 2015:
I see four carriers (he thinks maybe there is a fifth hiding on the strongest at the left), at about: 1628.925, 1629.000, 1629.015, 1628.04 kHz
I recorded I/Q data this morning centered at 1600 kHz, 250 kHz wide. I modified my software to handle 9 kHz spacing, and found I did indeed have carriers on many of those MW channels.
For example, here is 1521 kHz, used by powerhouse Saudi Arabia, which can be heard with audio here on the east coast:
Time goes down the waterfall. You can see the signal faded out. I don't display the time axis, but knowing the update rate and checking where it faded out on the image, it was about 1025z, 6:25 AM local time. Sunrise here is about half an hour later. I'm surprised that Saudi Arabia's carrier would be present well after their local sunrise, but perhaps this was them? Anyone else it could be?
I then decided to check 1629 kHz, looking for Australia. Here's the recording:
You can see five carriers faded in at around 1030z and faded around around 1100z. The frequencies are about 1628.92, 1628.98, 1628.99, 1628.01, 1628.04 kHz. My netSDR is running without the GPS reference right now, so the frequencies could be off by a few Hz.
Kilokat7 saw these carriers: 1628.925, 1629.000, 1629.015, 1628.04 kHz. Pretty close?