Subtitle: Comments too good to get lost.
This is a page where very interesting comments from the crowd are stored.
They might go into a post later but as long as those are questions which are frequently asked, it is nice to have a place to store them.
Judith explained the letters used in http://www.avcan.org/?m=Noticias&a=noticia&N=931 to Diana
Are these the answers tgo the letters you may be looking at?
First part…. Types of names of code deanery of band for seismometers… those who have no corner period are assumed to be greater than 10 sec code band = band, sample rate (hz) and corner type period (sec)
E = period extremely short, greater than or equal 80 and less than 250 and less than 10 sec S period = period short, greater than or equal to 10 and less than 80 and period of less than 10 sec
H = High broadband, greater than or equal 80 and less than 250 and greater than or equal to 10 sec
B period = broad band greater than or equal to 10 and less than 80 and period greater than or equal to 10 sec
M = period average, more that one and less than 10 L = long, close to 1
V = very long period, close to 1
U = very short period close to 0.01
Intrumentos
H= High gaing sismometer
L= low gaing sismometer
G= gravimeter
M = mass position seismometer
N= acelerómetro
Orientación
Z N E = tradicional (Vertical, norte –sur, este-oeste respectivamente)
A B C = Triaxial (along the edges of a cube turned up on a corner)
T R = For formed beams (transversal , Radial)
A Follow Up comment to Henri le revenants comment which was used in this post: Who stole my Earthqucak or why does BBC hate EL Hierro
Following along….
We were all quite happy when IGN/Pevloca started putting out 3 axis GPS data. It allowed a clean view of what the surface terrain was doing without having to go through some very convoluted brain twisting to back out the N-S and E-W vectors.
Many of you devote hour upon hour gazing at the tremor and quake plots and marvel at the spectrograph meanderings of each and every little tick… hoping to be the one to spot the first vestiges of something happening.
Theres nothing wrong with that. I do it too.
But before we (as in the Internet community in general) start to question what the technicians, analysts are doing at IGN, maybe we should ask ourselves… “Can I do any better?” If you think so, then you are either well trained, or an idiot.
Seismograph data, Helicorder outputs, and most of the data that is available for general public, is not really intended for detailed scrutiny. It’s mostly geared towards the way that we use it… to see if something is going on. If you had the nuts and bolts data, the raw waveforms… you might be able to do some second guessing that could stand up to other opinions.
Here is a sample of what you would work with. This is from Routine Data Processing in Earthquake Seismology… Figure 2.21.
http://i45.tinypic.com/2127jw5.png
The top three traces are the Z axis (vertical motion) the N (S) axis, and the E (W) axis. N is most sensitive to motion along the 0° and 180° alignment, and natch, E is most sensitive to motion along the 90° to 270° bearings.
P waves tend to show up the best on the Z axis… the reason is that they tend to lift or lower the sensor as as this body wave arrives… being a body wave, it took as close to a straight path to the seismo as it could manage. (though in truth, it isn’t very straight at all) S waves show up best on the two horizontal axises, N and E. Why? They are side to side motion… “transverse” unlike the compressional P wave (longitudinal).
If you notice in that graphic, some of the points have labels. These are pick points.
Pn, Pg, Sg, and Sn. .. but what happened to the P and S thing? Well, thats them. They get labeled by the vagaries of whatever path they took to get to the seismo.
Here are a few. (there are a shitload more)
Which particular wiggle on a seismo has to do with where it is in relation the quake and what the sesmologist knows about the makeup of the crust between the monies and the quake. Generally, the refer to a propagation model to determine the delay times. The current one is called AK135 which is in improvement on the IASP91 model.
All this is pretty heavy for us out here in the general public. Can imagine the freaking nightmare of trying to get an automated system to correctly pick these phases and do the analysis.
I’ve worked with automated detection and tracking systems… though not in this field. It’s not easy to get your detection thresholds correct without getting under or over reporting…. and that was just for data with only 4 or 5 detection parameters.
Yeah… I admire the work of the people who diligently work to get the picture and data right. And I’m not gonna second guess what they do.. or that they intentionally drop data. I, for one, am not qualified to question their expertise.
Schteve here my script for capuring the webcams to timelapse:
Pre-requistes:
Linux OS
perl is installed on your OS
RPMFusion is installed on your OS (this is a repository that allows yum install of ffmpeg)
I have created a /root/tolbachik on my OS and in there i locate 2 scripts:
photoGrab.pl
convert.sh
photoGrab.pl is used to pull jpegs from the webcam sites and save it into a sub-directory on you system. I create a subfolder for each volcanoe i want to watch..
convert.sh is used to pull all the jpegs in a sub-directory into an mp4 in the same subdirectory. After i do this i normall mv the jpegs into an archive sub-folder
an example of execution for tolbachik every 60 seconds:
perl photoGrab.pl tolbachik 60
then after a while you can run the conver.sh:
./convert tolbachik
Here are the 2 scripts (between the tags:
photoGrab.pl:
#! /bin/perl -w
my $jpg;
print “$ARGV[0]\n”;
#argument 1 is the name of the volcano to watch
#argument 2 is the time to sleep in seconds between webcam updates
#you need a sub directory of the same name in which to save the jpegs too
#THERE IS NO ERROR CHECKING
#to capture tolbachik tun: perl photoGrab.pl tolbachik 60
my $volc = $ARGV[0];
while(1)
{
for(my $i=0; $i<360; $i++)
{
@timeData = localtime(time);
$month = $timeData[4] +1;
;
$year = $timeData[5] +1900;
if($volc eq “stromboli”)
{
system(“wgethttp://www.ct.ingv.it//sezioniesterne/webcam/WebcamEolie.php"
#open the file and search for the string match
open (TEST_DATA, “WebcamEolie.php”);
while ()
{
if($_ =~ m/^Ultimo aggiornamento/)
{
#change Sqt to pic some other webcam view
#Sqt = close up thermal
#Sqv = close up visual
#there is also Spt, Spv, Vvt and Vlv
m/(Sqt\/\d{8}\/\d{4}\/Sqt\d{4}.jpg)/;
print (“$1\n”);
$jpg = $1;
}
}
close(TEST_DATA);
unlink (“WebcamEolie.php”);
system(“wget http://www.ct.ingv.it/dati/webcams/$jpg“);
}
if($volc eq “colimo”)
{
#I think this was colimo, but it was offline at the time i added it and never tested it
#system(“wget http://www.opentopia.com/images/data/cams/9360/big.jpg“);
}
if($volc eq “tolbachik”)
{
system(“wget http://www.emsd.ru/video/Tolbachik/img_1.jpg“);
#TODO add timestamp to jpeg image, note the timestamp is local time, it should
really be changed to the timezone you are capturing the image from
#system(“./addTimeStamp.sh img_1.jpg $timeData[2]_$timeData[1]_$timeData[0]_$timeData[3]_$month_$year”)
}
if($volc eq “whiteIsland”)
{
system(“wgethttp://www.geonet.org.nz/p/volcano/cameras/latest/whiteislandcraterfloor.jpg“);
}
if($volc eq “tongariro”)
{
system(“wgethttp://www.geonet.org.nz/p/volcano/cameras/latest/tongariro.jpg“);
}
#move the image into sub_folder and add timestamp to filename.
system(“mv *.jpg ./$volc/$year$month$timeData[3]_$timeData[2]_$timeData[1]_$timeData[0].jpg”);
#time in seconds to sleep before grabbing the next picture
#for tolbachick every 60 seconds, whiteIsland and tongario every 900 seconds
sleep $ARGV[0];
}
}
convert.sh (note the copyright, I have modified this script a slight bit from where i copied it from)
#!/bin/bash
# Simple Camera Image Archiver by Stephen Cropp of skcservices.com
# Copyright (C) 2011 by Stephen Cropp
# This script is released under the terms of the BSD license.
# You can find a summary and text of the license terms at the following website
# http://creativecommons.org/licenses/BSD/
# THIS SOFTWARE IS PROVIDED BY STEPHEN CROPP “AS IS” AND ANY
EXPRESS OR IMPLIED
# WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
# MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO
# EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT,
# INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
# (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
# SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
# CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
# OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
# OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
#ldate=$(date +%Y%m%d -d “1 hour ago”)
#lhour=$(date +%H -d “1 hour ago”)
#archive=/root/tolbachik/archive
#recordings=/root/tolbachik/recordings
volcano=$1
x=1;
for f in $(ls -r -t $volcano/*jpg); do
counter=$(printf %04d $x);
cp “$f” /root/tolbachik/tmp/cam1_”$counter”.jpg;
x=$(($x+1));
done
ffmpeg -y -f image2 -r 4 -i /root/tolbachik/tmp/cam1_%04d.jpg $volcano/timelapse.mp4
rm /root/tolbachik/tmp/cam1_*.jpg
………………………………………………………………………………………
This was in response to floodwarn, but a technical difficulty (unrelated to forum operations) wound up having me post at the end of the comments. In essence, I had a crash about halfway through the post and had to pick up from where I left off.
It is possible… but its a pretty rare set of circumstances that could cause it.
In order for a tsunami to form, a large quantity of water has to be rapidly displaced. Subduction zones have a plate being pushed under another, from time to time (usually always) these two plates that are pushing towards each other will stick. That does not stop the overall motion of the plates, and one can develop a sizable amount of energy stored in a flexed plate… one that slowly bends to accommodate the movement. When the sticking part gives way, it allows the flexed region to relax. This adds to the energy release. “Reverse Faults” are the usual focal mechanism, depending on size, they can lift the ocean floor a few meters in a very short period of time. This is a megathrust quake, that can make events like the Japan or Sunda Plate tsunamis.
The atlantic does not have a lot of places where this can happen. Down by the lesser and Greater Antilles, there are subduction zones, but they are pretty small.
Another configuration that can generate a tsunami, are areas where an undersea mountain is rapidly shoved to one side. In this case, the water is displaced in a similar method to a using a wedge in golf. This would be possible along transforms (normally seen in fracture zones), but for it to be a hazard, you would have to be down range of the direction that the undersea mountain is shoved towards.
In all cases, the undersea topology works to focus or defocus the energy from these events.
A third potential source would be a continental shelf landslide. These tend to be more localized though, they are essentially point source events and the energy dissipates quite rapidly the further you are from the source. (1929 Grand Banks earthquake)
Tsunamis are wave structures. Not that they are “waves” in what you see down at the beach. They are longitudinal waves. Compression events. But… by being an energy “wave,” some of the concepts of wave mechanics are at play. Specifically Near Field, and Far Field. In the Near field, an EM wave has not gained coherence and the energy is subject to local reflections and interference. The wave front looses energy as an ever increasing bubble. Once the wave transitions into “far field,” it travels as a coherent planar wave. At this point, energy loss is much less since the energy effectively travels as a beam. The issue with megathrust quakes, is that the tsunamis that the form, come into being as planar wave fronts since a large section of the sea-floor is lifted or moved as one unit. (or large units). Landslide or shelf collapse generated tsunamis are formed as point source events. They loose energy with distance until the wave transitions to “far field”
What those distances are for tsunamis, I don’t know. For EM waves (radio, radar etc…) it’s about ten wavelengths.
I think that the largest tsunami that England has encountered recently (think holocene) was from the Storegga Slide events.