Japan Tohoku Earthquake And Tsunami Warnings – Reading The Signals I
    By Bente Lilja Bye | March 21st 2011 10:43 AM | 10 comments | Print | E-mail | Track Comments
    About Bente Lilja

    Earth science expert and astrophysicist writes about Earth observation, geodesy, climate change, geohazards, water cycle and other science related...

    View Bente Lilja's Profile
    At the Top of the World.
    The massive 9.0 M earthquake at Tohoku, Japan, 11. March 2011 literally shook the entire planet. The signals were read even at the top of the world, close to the north pole. Read on to learn about earthquake and tsunami observations in general before you take a look at the unique earthquake recordings from the high-north.

    by Bente Lilja Bye and Ove Christian Dahl Omang

    Norwegain Mapping Authority's Geodetic Laboratory in Ny-Ålesund, Norway
    Ny-Ålesund Geodetic Laboratory, Svalbard, Norway. Credit: Ove Christian Dahl Omang

    After the 2004 Indian Ocean tsunami the international community stepped up its efforts on developing and building early warning systems. Japan is a leading nation with respect to expertise and implementation of both earthquake and tsunami early warning systems. The Tohoku earthquake and the subsequent tsunami that rushed towards and devastated the shores of Honshu island in Japan would have been a substantially larger disaster than it already is, were it not for the well functioning warning systems on the Japanese islands.

    Seismic networks in Japan
    In Japan there are a number of dense monitoring networks, like the national seismic networks that we see on this map. Credit: Japan Meteorological Agency.

    In Haiti Earthquake, science, early warning and mitigation a general description of warning systems is presented. An early warning system consist of 4 elements: Risk Knowledge; Warning Service; Dissemination and Response Capability. In order to be considered well-functioning all 4 elements of a warning system have to work and the Japanese early warning system worked very well.. We've seen a flood of heartbreaking images describing the consequences of the tsunami in particular. But what did and do scientists see through their instruments? We'll take a closer look at the Warning Service element that includes monitoring and forecasting impending events.

    The Tokohu-oki earthquake 11th March 2011 caused a massive tsunami that devastated the shores of Japan's Honshu island. IT will take years to rebuild this region. Credit: US Navy.

    Warning Service – reading the signals
    An effective warning system requires continuous monitoring of our planet; on local, regional and global scales. Several global infrastructures serves various nations needs depending on their particular natural hazards challenges. A multitude of instruments picks up a vast number of signals like crustal movements, seal level changes, seismic waves etc.

    Tsunami wave propagation
    Tsunami waves are modeled and included in tsunami early warning systems. Once the earthquake magnitude and location are known warnings of wave size and arrival times can be issued. Here we see the wave distribution of the Tohoku-oki tsunami. Credit: NOAA

    Our planet is a restless planet, sometimes more restless than others. The Earth rotates around it's own axis daily and circumvent the Sun once a year, that is pretty obvious to most of us. But less obvious are the more subtle movements like the tidal effects and continental drifts. Except of course, when sudden movements like earthquakes happens. Then it is not necessarily so subtle anymore. The science that describes the Earth's movements, shape, rotation and gravity field is called geodesy and a whole range of geodetic observing systems are included on the list of instruments reading the earthquake and tsunami signals. In addition we have the seismic signals that perhaps is more well-known in connection with earthquakes. The signal is basically sound-waves that travels around the globe almost instantly after an event.

    GNSS network
    The international GNSS (Global Navigation Satellite System) stations, which consist of GPS stations from countries all over the world, form the basis for monitoring crustal movements.

    It is equally important to monitor our planet before, during and also after events. Otherwise we will not be able to interpret the signals correctly and make the right decisions.

    Seismic, geodetic and oceanographic signals
    For a tsunami early warning system three different types of signals are monitored; seismic, geodetic and oceanographic.

    Old Jpanese tide gaugeIn the case of an event (earthquake) the seismic signals are immediately recorded and used to make the first estimates of where (location) the earthquake takes place and it's size (magnitude). However, in cases where the earthquakes are so big as the Tokohu earthquake (9.0), the seismic techniques are not quick enough to produce correct estimates of the magnitudes which in turn provides the basis for tsunami warnings with times of arrival, sea level and inundation maps. Recent development of geodetic techniques shorten the time for magnitude estimates. (Will be described in a later article.) Geodetic techniques are also an integral element of the oceanographic network. Tide gauges, ocean buoys and submarine pressure gauges are all part of the oceanographic observing system that are included in a tsunami early warning system. Often the     oceanographic instruments are co-located with geodetic            Old Japanese tide gauge.                   instruments like GPS and various instruments measuring gravity.

    Gravity measurements of the Tohoku earthquake close to the North Pole

    While there are dedicated instrumental networks for tsunami early warnings, signals from other networks are also contributing to both the before and after analysis that goes into the warning systems.

    The abrupt movement of the tectonic plates create sound waves that the global seismic network sees. As the movement implies a displacement of huge masses, it results also in a change in the gravity field. Geodetic experts keep track of our planet's gravity field with the help a suite of instruments like GRACE and GOCE from space and various gravimeters on land, airborne and on ocean (both airborne and oceanographic gravity measurements are taken on a campaign-basis. )

    The Norwegian Mapping Authority's geodetic observatory at Ny-Ålesund, Svalbard, which, by the way, is the world's northern most society at 79,9 degrees north, operates several different types of geodetic instruments, such as GPS, VLBI and a Superconducting Gravimeter.The fine tuned superconducting gravimeter instrument was originally commissioned and operated by the National Astronomical Observatory of Japan, in close cooperation with Norwegian authorities who took over the operations in it's entirety in 2010. It has made continuous observations (every 1 second!) of the Earth's gravity since September 1999.

    We say that the superconducting gravimeters are so sensitive to changes in gravity that if you choose to take a leak close by, the instrument will reveal your actions (your fluid adds weight to the local environment). Weather this is true or not has no bearing on the observations of the Tohoku earthquake. The signals were so powerful it knocked out the instrument after about 20 min (see graph below). After a short brake it was fully operational and was recording again. The gravitational signal of the earthquake reached Ny-Ålesund and the superconducting gravimeter approximately 10 min the event in japan.

    Seen above the very instrument, super conducting gravimeter in Ny-Ålesund, Svalbard, Norway, that measured the Tohoku-oki earthquake in Japan. Credit: Ove Christian Dahl Omang.

    The thin line before the horizontal black wall, are real measurements of gravity, and no error or irrelevant markings as one might be tempted to think. The signal extends way beyond the normal data range. Credit: Ove Christian Dahl Omang.

    Tohoku-oki gravity signal in Ny-Ålesund
    Here we see the signal with greater resolution. It shows a rather immediate and dramatic change in amplitude. Also we can see how the data acquisition unit stopped recording. Note that the signal is measured in volts rather than gal, the unit for gravity, due to the superconducting/electric instrumentation. 
    Credit: Ove Christian Dahl Omang.

    These recordings will surely be integrated in the numerous analysis of this massive earthquake.

    The co-author, Dr. Ove Christian Dahl Omang, is responsible for and operates the superconducting gravimeter in Ny-Ålesund. He uses the instrument in connection with his research on post glacial rebound in the polar and high-north regions.

    More readings and links

    Japan early warning related links - observing networks and information

    Historic seismicity outside Honshu, Japan

    Japanese seismic network.

    Geospatial Information Authority,Japan (national GPS network etc)

    Japanese tide gauges

    List of tide gauge stations in Japan.

    The tsunami as it looked at the ocean buoy outside Japan. GPS measurements.

    Some of the international observations systems that are being used:

    International GNSS Service - IGS

    IRIS - global seismic network.

    GEO - supersite for Tohoku-oki

    Japan Sendai Earthquake portal (at Harvard)

    International Charter on Space and Major Disasters.

    UN-SPIDER web site.

    You can also see for yourself how the warning system worked in practice in Japan. Maybe even more interesting now that you know a few things about what goes into such a warning system.

    Richard M Allen, UC Berkely, collection of info about the warning in Japan.


    Bente:  a most informative article.  Thank you for sharing.
    Thank you Patrick - for this and that :-) Hopefully more students will be able to benefit from this article. :-)
    Bente Lilja Bye is the author of Lilja - A bouquet of stories about the Earth
    Oliver Knevitt
    Thanks, Bente&Ove; incredible that an event so far away can affect a gravimeter in that way! Looking forwards to the next installment.
    Thanks Oliver,

    Yeah, those graphs Ove produced are really illustrating the drama of the event - from a scientist's perspective. The real life drama is unfortunately not so easy to handle, to put it mildly.

    We've got more interesting stuff from this geoscientifically violent event. Hopefully within a few days. :-)

    Bente Lilja Bye is the author of Lilja - A bouquet of stories about the Earth
    To anonymous.

    Your loss saddens me.

    I can assure you that Bente means no disrespect to anyone in writing about the science of disaster prediction and response.   By discussing these tragedies, we all hope to contribute to a better understanding so that our emergency responses can be ever more effective.
    The gravitational signal of the earthquake reached Ny-Ålesund and the superconducting gravimeter approximately 10 min the event in japan.
    This all seems to claim that the gravimeter can extract the gravitational signal (e.g. tidal forces) from the vibrational acceleration background, and that after extraction the pure gravity change signal is indeed that large. It seems you confuse (not in your mind maybe, but as it is written in your article) gravimeters and seismometers, or to say it differently, your gravimeter just worked as a sensitive seismometer during the quake, while gravity changes were totally swamped.
    Or are you on the general relativity level where gravity and acceleration are equivalent anyways? ;-)
    Hi Sascha,

    You are correct in that the superconducting gravimeter (SG) functions more like a seismograph in these particular cases. However, SG closer to the epicenter do observe permanent changes in gravity after quakes. The Japanese have of course SG in Japan also, not only close to the North Pole, and you can read more about their observations and analysis of other Japanese earthquakes in Imanishi et al, 2004 Science, 306, p 476-478.

    Ove and Bente
    Bente Lilja Bye is the author of Lilja - A bouquet of stories about the Earth
    I am very interested in Dr.Ove Christian Dahl Omang's work on glacial rebound Lilja .Is there a website where i can follow his progress?

    I was sent by colleagues in Japan a series of GPS readings from Tokaimura (which is on Ibaraki prefecture coast and was hit by the earthquake and tsunami, though not as hard as in Miyagi ) The readings extend from April,2010 up to March 11, 2011, the date of the Tohoku earthquake. The horizontal coordinates are fairly stable, but the altitude shows a kind of breathing with a maximum in August, 2010 and then a slow decline by about three meters untill the end of February, 2011. And then a suddent steepening of the slope and another three meter drop but this time in the 10 days preceding the earthquake. It is quite impressive and it is hard to believe that, based on such sudden and violent changes, there is no way to give much earlier warnings and save many more lives. The break in the slope is quite visible after the first few days, say on the third of february though the resolution of my plot is not very good. In any case a five days advance warning would have been possible if this had been interpreted as a precursor. Do you have informations about the use of GPS in early warning systems ?

    There are multiple ways of using GPS in early warning systems, both before, during and after an event.

    Real-time or near-real-time GPS techniques have been developed for use in early warning systems and my next article in the ...Reading the signals...series will describe a pilot project undertaken by NASA's JPL and University of Nevada. Describing and understanding the physics of the rupture of a submarine earthquake are of out-most importance for tsunami warnings as the parameters go into the wave models that give the inundation (wave height etc) and arrival times of the tsunami.

    Are you planning on publishing your analysis of the GPS data from Japan? You could also add them to the GEO supersite for the Tohoku-oki event.
    Bente Lilja Bye is the author of Lilja - A bouquet of stories about the Earth