About two weeks ago one of reddit comments mentioned "...acoustic measurements from the Comprehensive Nuclear Test Ban Treaty (http://en.wikipedia.org/wiki/Comprehensive_Nuclear-Test-Ban_Treaty) acoustic station at Cape Leeuwin (SW tip of Western Australia), and acoustic logger from near Rottnest Island (from CMST - http://cmst.curtin.edu.au/) was used to determine bearing to an acoustic event, possibly a sea surface impact.
If you took the bearing of this event and intersected it with the '7.5th' ping arc you end up with the search region that the Chinese ship and Ocean Shield were searching in.
My source is a colleague at CMST."
Rather than simply "taking a bearing", one may produce a ring for the "acoustic measurements", centred at Cape Leeuwin, and intersecting with the 8:11 (00:11 UTC) or 8:19 (00:19 UTC) Inmarsat ring at the current search area (the one where Ocean Shield has detected 33 kHz signals). This point is approximately at (20.80S, 103.75E).
The snapshot from Google Earth below shows two rings: Inmarsat 8:11 ring and the Cape Leeuwin's "acoustic" ring.
By design, the current search area defines the intersection and the "acoustic" ring. Interestingly, the previous search area is also very close to the other intersection of the two rings (pointed to by the red dotted lines). This other point is approximately at (32.60S, 95.35E). Both points are shown here approximately, just to illustrate the idea. More precise intersections would need both Inmarsat data for 8:19 (00:19 UTC) ping ring and Cape Leeuwin acoustic measurements.
In any case, this analysis may explain why the search efforts continue to be focussed on southerly routes. The reasons why acoustic measurements are not made public are not obvious, but may be related to sensitivities around the Comprehensive Nuclear Test Ban Treaty.
A Google Earth KML file with two rings can be found below. It is based on one of the KML file produced by Duncan Steel.