GeoNet Rapid - Why is it different?

GeoNet Rapid has been up and running for a few months now so i thought i would go over a few things, the first is:

Quake history:

SeisComP3 is fast as it locates the earthquakes automatically (no humans needed).  As soon as it has enough data it is able to post a first location on GeoNet rapid, if you look at the pic above you can see the first location was posted 42seconds after the actual earthquake. 

As more data arrives, from more distant stations, the system refines the location (up to 4 minutes in the pic above)  and finally finishes when the last data has arrived (10min on the pic). This delay is caused by the time it takes from the earthquake P wave to travel to the stations - eg if there was an earthquake at Stewart Island, it would take around seven minutes for the earthquake wave to travel and be picked up by our sensors at Cape Reinga (top of the North Island). The very last change comes later on when one of our duty seismologists looks at the event.

Status/ quality:

As the event changes you will notice both the 'status' and 'quality' change along with it.  The status lets you know if the solution has been located 'automatically' or 'reviewed' by a human   - see figure B in the pic below.  the Quality changes as more information comes into the system - with few stations the solution has 'caution', many stations has 'good' and finally 'best' when a human has reviewed it - see figure A in the pic below.

Descriptions found by hovering over the ? on the website.

We have also added a keyword / colour coding system to the earthquake intensity so you can easily tell if an earthquake was damaging or unnoticeable - see figure C in the pic above.

So why is it different to www.geonet.org.nz?

 

Yes the earthquake locations on GeoNet Rapid will be slightly different to those on www.geonet.org.nz The main reason for this is that SeisComP3 uses a three dimensional velocity model, whereas www.geonet.org.nz uses a one dimensional model.

The three dimensional data model uses a lot of data collected by the GeoNet project as well as many other research projects in NZ. By using this model we are really improving the way we located earthquakes.

 Next....

GeoNet Rapid will be in full production by late 2012, and will bring with it a whole new look to the GeoNet website and information, some exciting things to come!

PS: The official GeoNet app for your smartphones can be downloaded here for  Android  and the iPhone/Pad app will be available next month!

Have an iPhone or iPad?

Well the GeoNet app is nearly ready, and here is a sneak peak.




The app can be in your hot hands next month!

Myth Busting!

Over my time here at GeoNet there have been various hazards come and go, and with them come the various myths/legends and some weird theories.

So I've decided to do a 'Monday Myth Buster' series on our  Facebook and Twitter pages, to clarify some of these up, and you might get a laugh or two out of them!

Today we start with 'Earthquake Weather'. Now this belief dates back to the 4th Century where the famous Greek Aristotle said that earthquakes were caused by winds trapped in underground caves. Small quakes were thought to have been caused by air pushing on the cave roofs, and large ones by the air breaking the surface. This led to the belief in earthquake weather, if  a large amount of air was trapped underground, the weather would be hot and calm before an earthquake

As cool as this theory sounds, its just isn't true.  Earthquakes begin below the grounds surface, where the weather is too far away to have any effect. And statistics have shown that earthquakes occur in about the same numbers in wet / dry / cold and hot conditions.

Out in the field - Christchurch

Christchurch cityscape

On Thursday i traveled down to Christchurch to help one of our Scientists Lara put out some new seismic test sites in Canterbury,  as she was the 'boss' of the trip i took on the role of navigator and hole digger!

We put out three test sites in total, they will stay in place for a month and then we come back and pack them up. Back at GeoNet HQ the team then look at the all of the data the sites collected, if  they test well (the signals are clear with no noise eg. farm activity) we will potentially build permanent stations there to help us better locate the earthquakes in the Canterbury area. 

All three sites were in the Rakaia district and all on dairy farms so we saw lots of cows (our poor rental ute was a bit messy too!)  The first  test site was in a nice dry grass paddock - so digging the hole for the sensor was 'fun'. It takes about 45min to put out a test site, as long as everything goes to plan - with this site we had a slight issue with the case that houses the data-logger above ground - being good kiwis however, it was nothing that we couldn't fix with a bit of duct-tape!

Rakaia 1

Rakaia 2

We have to dig a small hole at each test site to house to seismometer. The second site  had the added bonus of stony soil - even more fun for digging!  The photo below shows the finished product with the equipment wrapped up in a tarp to protect from the weather, and the electric fence to keep the nosy cows out.
 





 The photo below is of the final site of the day, although it looks like she is in the middle of cake baking, here Lara is mixing up the ready-set concrete to go into the hole. A concrete paver is carefully placed (and leveled) onto the concrete and its on top of this that the seismometer sits.

Rakaia 3

 On the second day we traveled out to Mount Pleasant here  as we would like to install a new strong motion sensor. These measure the very strong shaking associated with damaging earthquakes and are located in towns or near faults and are usually installed inside buildings.








We found a suitable area down a small terrace and a homeowner kindly let us have a look at his garage which would be a suitable place (they generally need to be on rock or concrete and have a power source). Although there were about eight houses down this particular terrace, one has been demolished and only two are now inhabited since the earthquakes - and the damage is quite obvious as you can see with the shed above.

Stickers on a property

Its strange how you can travel around some parts of Canterbury and even Christchurch and its hard to tell the earthquakes ever happened, and then you can go to certain areas and its very clear!  As we were traveling out to the airport we drove past buildings in various states of demolition, and empty lots where this has already occurred, at the same time there are little signs of hope and resilience - the church below had a cute picture painted on it with 'I'll kiss it better"

A church being demolished in the cbd.

Out in the Field - Problem solving (and a helicopter)

Here at GeoNet we have a fancy system called 'BigBrother' that keeps an eye on all of the equipment we have out and about in NZ, when something goes wrong it alerts our technicians so they can get it fixed ASAP.

Occasionally this involves a wee trip in a helicopter....  

Before this (fun part) however, the technicians problem solve to figure out exactly what is/could be wrong with the sites. Once they have that figured out they load up with tools and various spare parts and head out to the often remote parts of NZ - that's why we need the helicopter.

OHOK site, looking for a spot to land

Today our first stop was the Okoha (OHOK) cGPS site.  Here Sam and Andrew replaced the radio as it was old and a bit temperamental.

Boys playing nicely together.

Testing the comms.

Flying over DURV

Next stop was D'Urville Island here  to our cGPS site DURV.  which was also having some communication problems, this was fixed by replacing a broken freewave radio.

Sam fixing!

cGPS and our chopper for the morning.

The boys also did some general site maintenance and upgrading whilst we were there. Both sites are now back up and running.

A cute little island and lighthouse!

And to finish, a couple of scenic shots - if only every day was like this!




thanks to JK @ Helipro for flying us today!  

Makara WindFarm (West of Wellington)

You asked for a faster GeoNet......

 Well next week, you get it!

 On Tuesday 6th of March our Beta website "GeoNet Rapid" goes live.  The team here at GeoNet have been working on this new system for the past year and are excited to see what you think!

The basics:
- Speed : You will be seeing details of earthquakes from around 2 minutes of their occurrence, as a computer system locates the earthquake as soon as the data arrives (rather than a person waiting for all of the data).
- Numbers : More events will be posted to the rapid site, not just the larger/more widely felt events that we put on the 'recent quakes page' (we located over 19,000 earthquakes last year, so there are plenty!)
- Regions :  Its easier to view earthquakes in your area, with options to see various regions in NZ as well as 'all' or 'felt' events.  You can also view statistics on the earthquakes in the various regions in the last week/month and year.
- History : As more data comes in, the details of an earthquake can change (magnitude/depth etc.) You can see all of these changes right up to the final location as all of the data has been received from our instruments out in the field. The first location can be made in under a minute, as the first data comes in!
- Issues:  This is a beta testing site, so things can go wrong!  Don't panic, thats why we have the test site, to discover all issues and fix them!

- Remember: geonet.org.nz is still the official source of EQ information, and our duty officers will still be locating the quakes here.

Screenshot of the 'regions' page

Links to the site will be available on the GeoNet website  on Tuesday (and on facebook / twitter etc) as well as a comprehensive news story with information from the history of the new system, to what exactly it does!

I will be giving more blog updates on how to use the website and as more features are added, and for those more g33k types out there - you may like to check out the GeoNet Development blog here.

A great example of how fast the new system will be, is shown in the diagram below comparing automatic location times and manual 'human' times to events located on the 23rd of December.

On 23 December 2011 (UTC) there were 106 earthquakes over magnitude 3 in the Canterbury region.  On average SeisComP3 (SC3) had a first automatic location two minutes after the earthquake occurred and a final automatic location after four minutes.  Compare this to the fifteen to twenty minutes it typically takes to make a manual location

In January i gave a sneak peak at the GeoNet App that we have been working on, here. We plan to have this available on the Android marketplace the following week! < and iPhone later on in the year>

Drums - not the musical kind!

Now i did a blog last year on how to read the earthquake drums on our website (you can read it here) but we have had quite a few queries recently about the scale and things showing up on various drums - so i thought I'd have another look!

Some key points:
- The drums have scales (top right)
- The scales automatically change depending on the current activity (we don't change them to 'hide activity', if they didn't change you wouldn't be able to see anything!
- The drums often show noise -wind, stock, traffic etc. as the instruments are very sensitive (although we do try and put them in quiet areas).
- If you see an event on the drum and its not on the 'recent quakes' page, it probably wasn't felt.
- There are several different kinds of seismometer in use, so you cannot directly compare them.

And we don't use these drum images to locate the earthquakes, they are just there for you to look at!

The drums on the left, are in the 'Canterbury Quakes' tab here, they will have different scales and you can't directly compare quakes.

 Under our earthquake tab you can find Quake Drums, these show instruments from all over NZ. The picture on the right is from December 23 2011, as you can see it clearly shows how widely events were picked up across the country.

Edgecumbe 1987 event on  Hicks Bay drum

Here is an example of what the earthquakes looked like on the old drums. This picture (left) is off a drum in Hicks Bay and shows the March 1987 Edgecumbe earthquake and aftershocks, as you can see - its a mess!