Arrow of time, thermodynamics, and another transit to Yokohama

The second law of thermodynamics is the only physical law that is time-irreversible. All other laws, such as Newton's equation of motion or the Maxwell equations of electromagnetism, are insensitive to the direction of time (for example, the Earth is revolving around the Sun counterclockwise, and if you reverse the time, it would revolve clockwise, and this situation is perfectly fine). The second law of thermodynamics tells us that entropy (of a close system) can only increase with time, if you let the system alone. It's like your room getting messier and messier until you make your mind to clean it up. Why only the second law of thermodynamics has this arrow of time is still an unresolved problem in physics. Ludwig Boltzmann, the founder of statistical mechanics, once provided a microscopic justification, which was found to be rather incomplete, and he allegedly committed suicide because of this. Boltzmann's explanation is, however, what is commonly seen in most of textbooks on thermodynamics. Here I translate it to a more familiar example using a deck of cards. Suppose you start with a deck of 52 playing cards nicely ordered from the ace of spades to the king of clubs. You then shuffle it 100 times so that the cards are pretty much randomly ordered. Now you keep shuffling it and see if any order pops up. You may occasionally get lucky to see some partially ordered sequence (e.g., all aces in one pace), but your cards would stay looking randomly ordered for most of times, because there are so many combinations of card order that look random. How many different combinations of card order do we have? It's 52x51x50x….x1 = 52! ~ 8x10^67. If you're not familiar with the scientific notation, it's about 80000000000000000000000000000000000000000000000000000000000000000000 (67 zeros after 8) [try WolframAlpha for the exact result], and most of these combinations are pretty much randomly ordered. The initial order you started with is just one out of this huge number, so the probability of coming back to the initial order (the ace of spades to the king of clubs) by randomly shuffling is extremely small. It's not zero, but really, really small. Therefore, things are usually got more disordered with time (i.e., entropy increases), because a more random state is more likely to happen. But the probability of getting back to the initial state is not exactly zero, so some rigorous people are not satisfied with this sort of explanation based on thermodynamic improbability, and the debate over the origin of the arrow of time continues…

Now, why am I talking about this? Isn't this blog supposed to be about the Shatsky Rise cruise? OK, the reason is that we had another medical emergency, and we're currently heading to Yokohama again (this time, fortunately, no death is involved). Having two medical diversions in one cruise is pretty unusual. Will Sager, who has much longer sea-going experiences than I, told me that he has been on ~40 cruises over 33 years and had never had bad luck to this degree before. This cruise is only my 10th cruise, and this is actually my very first cruise as a chief scientist, and look what an experience I'm having! But if you think about the above thermodynamical improbability, the likelihood of having two medical diversions in one cruise may not be so small… Or I should think the other way around... Having similarly bad luck in future would be even less likely, so my future cruises may go entirely trouble-free!? Maybe it's time to start writing another sea-going proposal.

NSF has been very sympathetic to our situation, and because no further extension is possible this year (we need to get back to Honolulu by September 14th to give OBS to another cruise), they agreed to have another Shatsky Rise cruise sometime during late 2011 to early 2012 to finish any unaccomplished portion of our planned seismic survey. The science party is all grateful for this thoughtful decision.

There, in the water! A shark! A torpedo! A maggie?

Ah yes. The magnetometer or as we call it, Maggie. The magnetometer is a useful tool that has been around for over a century. A magnetometer is a scientific instrument used to measure the strength and/or direction of the magnetic field in the vicinity of the instrument. Magnetism varies from place to place and differences in Earth's magnetic field(the magnetosphere) can be caused by the differing nature of rocks and the interaction between charged particles from the Sun and the magnetosphere of a planet. Magnetometers are a frequent component instrument on spacecraft that explore planets or in our case a sea going vessel.

To keep it simplified: as the liquid metal core moves around the solid inner core, it creates a magnetic bipole or a North and South pole. The strength of this field would be continuous throughout the system if the Earth were a homogeneous structure. However, the Earth is a hetergeneous system and as such has effects on the magnetic field. The magnetometer allows us to detect the differences in the magnetic field created by changes within the structures of the Earth. Thus, a highly ferrous rock will have a greater effect on the surrounding magnetic field than a non-ferrous rock.

Magnetometers claim to fame: They discovered that parts of the seafloor were polarized one direction and parts were polarized in the other. Duayne has already covered seafloor spreading and magnetism so read his post for that. Without magnetometers the polar reversals would not have been discovered.

When we first begin deploying seismic equipment, Maggie is put in the water first. It is towed behind the boat about 150m back and at about a depth of about 40m. Maggie is giving us updates constantly so we can track changes in the magnetic strength as we travel across the survey area. Using the data we collect we can build a very accurate magnetic profile of Shatsky Rise when we have completed the survey.

Although the Maggie is a humble looking piece of equipment, we are glad to have it and can thank its predecessors for helping solidify plate tectonics as an accepted theory of Earth's evolution.

Something about the air guns source

We are using air guns to implement OBS and MCS. Let's talk about something about the airguns source, mainly about its signature.

Source signature corresponds to the seismic wavelet in seismic survey. Seismic wavelet is the far-field response of energy of particle motion velocity (land seismic survey) or pressure (marine seismic survey) which propagates from the seismic source. A seismic wavelet should be carefully chosen in exploration seismology since the bandwidth, the length, and the shape of the seismic wavelet will affect the resolution of seismic survey. A good estimation of seismic wavelet is absolutely critical for seismic inversion.

A seismic wavelet can be defined with its amplitude spectrum, which shows its amplitude characteristics, and phase spectrum, which shows its phase characteristics and includes zero-phase, constant phase, minimum phase and mixed phase etc.. In exploration seismology, we can extract seismic wavelet from data by mainly three ways, pure deterministic way, pure statistical way, and through well-logging curve.

In our marine exploration seismology, air guns source has its deterministic signature, and we can extract the amplitude and phase information directly from the control terminal in the main lab after each shot. For each shot, we think they are almost consistent in amplitude and phase.

We then want to know which type of wavelet the air guns source signature belongs to, or it is unique and belongs to none of ideal wavelet. There are four types of seismic wavelet that are commonly used in seismic data processing software: Ricker wavelet, Ormsby wavelet, Klauder wavelet and Butterworth wavelet. See the following figures for their main characteristics.

(From up to bottom: Ricker, Ormsby, Klauder and Butterworth wavelets, a different choose of dominant frequency or bandpass frequencies and cutoff frequencies may give different bandwidths in frequency domain, and thus the temporal domain shapes are slightly different. )

And the following is the source signature for each air gun.

Although total source signature is not available, we can see from the shape of the single signature and conjecture from the time delay characteristics that the air gun source can be approximated by the minimum-phase Butterworth wavelet with certain parameters: both of them have vibrating evanescent tail, and the most important, minimum-phase Butterworth wavelet itself is physically realisable. Ricker wavelet is so ideal that although it is often used in wave-equation-based numerical modeling or inversion test, it is not realistic in practical cases. Another reason that we consider our air gun source signature is close to Butterworth wavelet is that the frequency range of our air guns source signature expands from low to high frequencies. We can see from MMO’s monitoring screen that after each shot, there will be a bright line in the frequency domain whose range is from very low (~ 0 Hz) to very high (~ 48kHz). Ormsby and Klauder wavelets are bandpass-type, and have nearly vertical cutoff edges at the boundary of frequency band. If we suppose them to be full-ranged, then their waveform will not be far away from practical cases. Butterworth wavelet however, will have soft edges at the boundary of its band. In fact it has a long tail in frequency domain.

To further explore the characteristics of our air guns source signature, a more careful mathematical treatment will be needed. Perhaps we can arrive at some different conclusions and find something new if we can extract the amplitude and phase information in some way, rather than merely use shape-based information, which can be inaccurate in some situations.

Another thing that should be noticed is that OBS and streamers receive different kinds of signals. As OBSs are located on the sea floor, the seismic wavelet they received are in the form of particle motions, while the streamers receive pressure as the signals, although the signals are both from air guns source. The reason is that when air guns source are triggered, energy is propagated to the sea floor and subsurface rocks through sea water in the form of pressure, since fluid can not transfer shear stress, and then only P-wave can propagate in the seawater. When energy arrives at the sea floor, it will convert to the kinematic energy of particle motions of subsurface rocks, and now there will be shear wave converted from P-wave when the P-wave meets subsurface reflectors. After some time, there will be reflections that go upwards to the sea floor from subsurface and at the contact of sea floor and seawater, all S-wave will vanish quickly, only P-wave propagates upwards to the streamers and received. As OBSs are in contact with sea floor, they will receive both S- and P-wave. And thus the wave signals received at the OBSs and streamers are different in nature.

Quiz: How many iPods does it take to store reflection seismic data?

Hi all, it's another day on the Langseth, and we are still recording multichannel seismic (MCS) data. This is reflection data collected by hydrophone's trailing behind the ship. Yesterday, Sam was kind enough to give a window into the actual deployment of the streamers (housing the hydrophone) and the depth control devices, otherwise known as birds. Today I want to give the reader a feel for the amount of data memory involved in the actual seismic survey. This is the juncture where geophysics meets computer science. I'll keep it brief.

Let's start with the set up of the reflection experiment. Behind the ship we are currently towing a streamer cable that is approximately 6 km long. On this streamer cable there are 468 channels each recording data every 2 milliseconds. It might not be immediately obvious the enormous data requirements involved in the data acquisition, but a simple order of magnitude calculation provides an intuition for how much memory is required. In an earlier blog post by the chief scientist, Jun Korenaga, we learnt that the Shatsky is about the size of California. We will be making several trips across the Shatsky rise, and in total we will collect 3, 500 km of reflection data, which enable us to "see" into the crust beneath the ocean floor.

So what does 3, 500 km of MCS data translate to? Quick math: at 20 shots per km (because we shoot at every ~50 m), 3, 500 km of seismic survey will require 70, 000 shots. Each of these shots is recorded by the 468 channels, so we have 468 x 70, 000 = 32,760,000 shot traces. Now, as I mentioned, each channel samples data every 2 milliseconds and for each shot we have sampling continuing for 16 seconds (so 8,000 samples per trace). Four byte per sample will then gives the total memory required as: ( 32,760,000 * 8,000 * 4) bytes = 1,048,320,000,000 bytes. Large? Maybe not so much. This is actually 976 Gb (close to 1 tera bytes). I have a 6 Gb iPod touch, and if I were to store all the data on this model, I would require ~ 163 iPods. Yes, that's a lot of iPods. An interesting story is the history of the growth of memory capability for seismic acquisition, but that's for another day. Right now, I need to get back to my watch.

Streamer out----let's get the multi-channel reflection started

Cheers! Streamer 3 is being pulled out from the deck since this morning. The streamer has 6 km long, with 480 maximum channels in 12.5 m interval (but we just activate 468 channels for work this time). In addition, we need to install birds(streamer depth contrllers) on it in certain interval to control the depth (picture: people are installing the red bird onto the yellow streamer, and then the streamer will go out into the ocean with the help of the hanging and transmittion mechine on the left hand side of the picture). Although it will take hours to complete the whole streamer deployment, that means our multi-channel reflection seismic survey is about to start around the sunset today, hopefully. Let's do it!

P.S. my name is Jinchang Zhang (people also call me Sam). I am a PhD student of oceanography at Texas A&M University, working with Dr Will Sager on the research topic of Shatsky Rise formation. We are trying to use kinds of marine geophysical data to interpret the geological nature of Shatsky Rise, like bathymetry, seismic, magnetic, gravity, etc. So the seismic data collected by this cruise means a lot to my entire PhD study. And I am going to employ ProMAX to process the reflection seismic data of this survey, which I will talk more about later on. See you later!

Did Somebody Say XBT Party?

I did!! What is an XBT, you ask? Well, XBT stands for Expendable Bathythermograph Probe and it is a disposable temperature probe that the science team launches from the boat once every twenty-four hours. Our XBT brand of choice is made by Lockheed Martin Sippican Proprietary, and comes in two different flavors, T-5(max depth of 1830m, max vessel speed of 6 knots) and T-7(max depth of 760m, max vessel speed of 15 knots). But what does an XBT do and what can it tell us? Well, an XBT actually contains a wire link which transmits data to our main lab's computer, which in turn stores information on depth versus water temperature and depth versus sound velocity. This data can then be pooled together from several different research vessels to compile weather and climate forecasting as well as climate research. But why would we care about sound velocity in the water? Once we can calculate accurate estimates of the speed of sound through water, scientists can then use this information to create more precise bathymetric maps. To demonstrate how these depth vs. temperature and depth vs. sound velocity profiles can vastly change across an ocean basin I present some of our recent findings: the first set of profiles, in red, were collected at the beginning of the trip in close proximity to Oahu. The next set, in blue, were collected off the coast of Japan in transit back to our study area.

An update

Around the noon time on July 30th, there was a medical emergency, which warranted the immediate stoppage of all science operations. Unfortunately, the person who had the emergency was officially pronounced as deceased at ~2PM, and soon after Langseth got underway for Yokohama, Japan. The vessel arrived in Yokohama four days later (August 3rd, or August 4th in Japan Standard Time), and after offloading the remains of our deceased crew member, we started our return to Shatsky Rise. We're currently still in transit. NSF very generously granted us a 7-day extension to minimize the loss of science days caused by this Yokohama transit. With the current speed (10-11 knots), we expect to return to where we stopped, by the afternoon of August 7th.

Our sympathies go to the family of the deceased crew member, and we are grateful for those who were involved in the cruise rescheduling.