Papua New Guinea Earthquake: Did the M=7.9 shock trigger a distant M=5.6?

By Ross Stein, Temblor

See earthquakes around the world

Perhaps the most remarkable feature of yesterday’s Mw=7.9 mainshock was the occurrence of a M=5.6 earthquake 180 km (110 mi) to the southwest of the mainshock, just 15 minutes after the mainshock struck.

papua-new-guinea-earthquake-map
Relationship between the deep mainshock and candidate shallow remote aftershock. Due to its great depth, the M=7.9 mainshock caused widespread but only moderate shaking, leading to little reported damage.

 

Was the M=5.6 event an aftershock?

The seismic waves propagating from the 136-km-deep mainshock would have reached the site of the 10-km- deep M=5.6 event in about 100 seconds, or 2 min. So, if the M=5.6 was triggered, why the additional 13 minute wait? Perhaps small shocks began as soon as the waves struck, and one of these smaller (undetectable) shocks triggered the M=5.6. Nevertheless, the short time between these events hints at a connection. According to the USGS, one of the two possible fault planes on which the M=7.9 event struck was inclined to the southwest (USGS Fault Planes). If the rupture indeed struck on this plane, then the strongest seismic energy would have propagated parallel to the rupture plane, upwards and to the southwest, reaching the Earth’s surface at about 180 km away, tantalizingly close to the site of the M=5.6 quake. This, and the short time between them, make the strongest case for a causal connection between the two events, although it is far from demonstrated. The global rate of M≥4.5 earthquakes in the 24 hours since the mainshock is about the same as that in the preceding 24 hours, so I see no evidence of global aftershock triggering, which did occur after the much larger 2012 M=8.6 Indian Ocean earthquake.

 

What does this mean for urban aftershocks?

While the vast majority of damaging aftershocks struck close to the main fault rupture, there is growing evidence of remote ‘dynamic’ aftershock triggering as well. This means that after a large San Andreas, Hayward Fault, or Cascadia megathrust earthquake, for example, the public should be prepared for distant surprises in the minutes to days after the mainshock.

Coming back to the site of the mainshock, Bougainville Island hosts several active volcanoes. It will be fascinating to see if any of these awaken—either in small shocks or fumarole activity—in the next few days as a result of the mainshock shaking.

papua-new-guinea-earthquake-bougainville
Bagana volcano sits almost directly atop the M=7.9 event; will it reawaken?

 

 

Was the M=7.9 earthquake a surprise?

Not really. The plate tectonics in Papua New Guinea and the Solomon Islands region in the southwest Pacific looks more like the shards of broken plates. The ~100 mm/yr (4 inches per year) Pacific and Australian plate motion rates, combined with the profusion of faults, makes for one of the most seismically active places on Earth. Here is the setting of yesterday’s great quake:

papua-new-guinea-fault-map-earthquake-map
The map and C-D cross-section are annotated from Holm et al. (2016).

 

A very high rate of quakes during the past month is also evident in this region along most of the plate boundaries. A knot of seismicity exists in the region of the mainshocks, but those shocks lie 100-150 km (60-110 mi) west of the M=7.9 event, so it would be a stretch to call these foreshocks.

papua-new-guinea-earthquake-map-temblor
Temblor app screen shot of the region. The largest red dot is the M=7.9 mainshock.

 

One can also see Temblor’s Earthquake Forecast (colors) above, which is based on the GEAR model. A M~8 quake is expected at the site of yesterday’s quake about once a lifetime (put another way, with a 1% chance per year). However, the GEAR model is for quakes above 70 km depth, and the M=7.9 lies at 136 km depth.

 

Sources
US Geological Survey
Robert J. Holm, Gideon Rosenbaum, and Simon W. Richards (2016), Post 8 Ma reconstruction of Papua New Guinea and Solomon Islands: Microplate tectonics in a convergent plate boundary setting, Earth Science Reviews, 156, 66-81, doi: 10.1016/j.earscirev.2016.03.005

  • Ross Stein

    From Tom Parsons, USGS: “Given that there seem to be stray events scattered around during the period before, it would be hard for me to conclude that any one event is necessarily an aftershock. On the other hand, I think it’s also impossible to say that it isn’t given that there are approximately as many M>4 events during the ~2 days after the mainshock as there were in the preceding 20 days. Small numbers though.”

  • Ross Stein

    From Rob Holm, James Cook University, Australia, “Thanks a lot for your interest in my paper. It is certainly an interesting area and the earthquakes need a lot more attention! I have signed up for updates as well.”