Barographs detect shock waves


In my last blog I introduced you to this mystery: 

A yacht skipper at anchor in a river on the NE end of Baie de Prony (at SE end of New Caledonia) last Wednesday morning noticed his barometer jump up 3 hPa in an hour and then down in the next hour, and once again. These jumps were superimposed upon the normal semi-diurnal atmospheric tide that the barometer was recording as a series of bumps that look line Sine waves. The STRANGE thing is that nothing happened to the wind or cloud in the area (usually a change of this magnitude is associated with an incoming gale or a nearby squall).   Nothing.

Later he found out that another boat anchored 10 miles further south observed much the same thing.

Here is the weather at the time, Black lines are isobars. Shading is the rain, and green lines are swell height. Wind barbs are coloured code for speed. Baie de Prony is in the trade wind belt.


Here is a screen shot of the barometer readings – Diurnal variations show up as the gentle bumps,  and I asked you all for an explanation of the strange up-down barometer behavior.



I received two replies :


1) Rich West said  if there are power supply inconsistencies or electrical noise, it can cause this. The green line I’ve drawn might be the normal diurnal variation. Noise would cause part of it to not be recorded at the correct level, thus the spikes. Another clue to this is at the blue arrow. Here, there seems to have been a power interruption. If power is interrupted to the unit, when it’s repowered, it picks up at a new level, thus the inconsistency at the blue arrow.

It could be a bad connection to the barometer or it might be RF noise from an HF radio that causes this. I sure see this when I’m messing with my electrical system and interrupt power.


This is worthy of consideration, but in this case we can probably rule it out because another vessel  10 miles away measured  the same signal .

The other reply was from Nicolas Remy who observed much the same trace  from his anchorage in Noumea:


2) We are moored in Nouméa and following your message I checked our barometer recording, which has the exact same hickups as those mentioned. I include a screenshot of the reading and the raw data (times are UTC, I excluded the 1hr tendency on the graph and the yellow line is temperature, but the sensor is inside the boat and is clearly not normalised).

Screen Shot 2014-09-29 at 12.41.20

Around 8 local time, there is a first bump, then returning to normal, and a second peak around 10 local time, with slopes of up to 1 hPa in 10 minutes ! In fact it even has got a dip later on, late in the evening, with pressure dropping 1 hPa in 5 minutes at some point, and then shooting right up again. I notice the picture of your correspondent’s barometer has the same dip as well.

Some kind of small mobile ridge ? A shock wave due to something ? The barometer was clearly more bouncy during these 24 hours than before and after…


What made me think about a shock wave was the fact that it took some time for the atmosphere to settle back into a normal behavior, but I’m also unable to account for some other aspects. I thought about a volcanic eruption*, but since there was none known close by, it would need to be quite big, so would be heard of, and furthermore other stations would have recorded the shock wave in the whole of the South Pacific.

Any data from Norfolk Island, or from Vanuatu available ? Has anything been recorded in NZ or Australia ? Now we could also think about a smaller scale event.

I found that the meteorite blast over Russia a year ago or so circled the planet twice and that some waves had periods up to 100 minutes.

What do you think about the idea that a very much smaller meteorite than that could have exploded somewhere over the ocean where no one would have witnessed it ? It could account for the period and amplitude of the wave. However, one thing that does not change : it would have to be recorded elsewhere as well. So readings in the region would be most interesting (Fiji ?)


*  The event above  occurred on Wednesday 24 Sep so has nothing to do with the Volcanic eruption of  Mount Ontake in Japan  last weekend.

I like the explanation of it possibly being the shock wave of an (unreported) meteor.  So I have searched for some nearby barographs:

For example Noumea Airport data for last week is  plotted by Wunderground  at


This trace is based on METAR reports that round the pressure DOWN to the nearest hPa, and thus tends to give a jagged step pattern to the barograph.  However I think you’ll agree we can see the same pattern in the barograph between Tuesday and Wednesday ( dates are UTC ,  we need to add 11 hours to UTC to get Noumea local time , so late Tuesday UTC= Wednesday morning Noumea time).


Below are the observations from Fiji, with only a slight possible  trace of the blip, closer to “Wednesday” so this suggests the phenomenon was best observed  to the New Caledonia area.


There was no detectable signal in data from Vila,  Brisbane or northern NZ. Below are the observations from Norfolk Island ,  and yes, they did record a  weak signal :


(By the way: One thing I like about this  Norfolk Island trace is how the barograph portrays a passing anticyclone, with a slow rise and fall, and also the semi-diurnal atmospheric tide) .

SO   data around the region tends to confirm that the shock wave seemed to have a period of around 100 minutes and, around Noumea,  an amplitude of around 2 to 3 hPa.  This sort of low period atmospheric “wave” is consistent with the shock wave given by an exploding meteor, but in this case we haven’t received any reports of the meteor itself.

My compliments to the observers who spotted this.  It reminds us all to watch our barographs with always a quizzical brain,  like  Sherlock Holmes.

Bob McDavitt for MetBob