Interpreting the Mixed Layer

(Today's post was written by scientist and co-Principal Investigator Eric Kunze)

As pointed out in a previous blog, in order to follow post-storm restratification of the mixed-layer, we need to simultaneously find a deep mixed-layer, horizontal density gradients which imply vertical shear (through the thermal-wind balance) and storms to homogenize the mixed-layer. At the same time, we need to avoid too much horizontal shear which would act to disperse our swarm of freely drifting profiling floats. This is a tough combination but historical measurements were favorable. We have had no trouble finding density fronts but have been wary of deploying the floats in them because of our concern of finding horizontal shear.  At our first site, stratification extended to the surface with only hints of a previous mixed layer. Furthermore, passing storms from the west tended to veer north of our latitude before reaching our longitude. Moving north, our second site had a mixed-layer thickness of 20 m (when it was supposed to be 100 m) and the storms continued to veer north. We have moved further north, though limited by the strength of the oncoming storms, to find a mixed layer 70-80 m thick at 35N, 139W. 23 floats are now deployed across a narrow density front, all but 2 that we brought along.

To get a better sense of whether the mixed layer was behaving as expected, time-series of various variables have been averaged over the mixed layer for each float depth profile (roughly once an hour). The easiest to explain are the top and bottom panels which show mixed-layer density σθ (top), east velocity u (red) and north velocity v (blue) bottom panel). Density lightens midway through 24 MAR, then undergoes weak diurnal oscillations, lightening during daylight and getting denser at night. The spread of 0.02 is a signature of horizontal variability that is needed for restratification. The velocities show that there is not a single but multiple frequencies of a day or so. The north velocity cresting before east velocity is a signature of low-frequency (1-day period) internal waves. 

Figure showing various characteristics of the mixed layer, including density, velocity, stratification, and shear. (Figure credit: Eric Kunze)

The second panel from the top shows mixed-layer stratification N (green) and vertical shear |Vz| (red).  Before 28 MAR, stratification and shear undergo daily oscillations, strengthening during the day and weakening with occasional very weak N at night. The daylight restratification could be due to the the advection of lighter water over heavier water that we are looking for or warming by the sun. After mid 27 MAR, the stratification increases by a factor of 2 and no longer exhibits nocturnal weakening, a signature of strong restratification. perhaps associated with the larger density contrasts. 

In the third panel from the top, vertical shear is repeated (red) along with horizontal density gradients normalized to mimic the thermal-wind balance that dominates much low-frequency variability in the ocean. Thermal wind balance in essence is a balance between horizontal density gradients and vertical shear.  Strictly speaking, both should be smoothed over more than a day for the balance to hold. The signals are comparable but sometimes horizontal gradients are larger than shear, sometimes small. This comparison can be confounded by wind-driven mixed-layer shear and near-inertial wave vertical shear, both of which can act with or against thermal wind shear. 

So we don't have the answer yet. 

Years of analysis will be needed to combine results from the EM floats shown here with towyo measurements from SWIMS and the air-sea flux buoy.  But we do see a signal of restratification and have enough horizontal coverage to evaluate contributions from advection vs. air-sea fluxes.

Looking at Data, Episode 4 (round and round we go!)

We have all likely had the experience in a classroom where we were expected to take something on faith; someone else discovered it and provided evidence, and that had to be good enough. In science, many of these ideas challenge our innate perception of the world, and it becomes hard to fully integrate them into how we perceive the world without coming up with the evidence ourselves. Newton’s Laws of Motion are a popular target for rote memorization in physical science classes, something so many can repeat and yet so few fundamentally embrace. Aristotle may not have grasped all the intricacies of physics, but he certainly had a solid handle on how humans see things.

The path through which the buoy has drifted over the last several days provides indirect evidence of inertial motions in the mixed layer of the ocean. It does not move in a perfectly repeatable circle because there are always other effects super-imposed on top, such as a mean current.

The study of fluids (like the ocean) is different from the study of individual particles (you, me, a car, a tennis ball) in many ways. The basic laws still apply: applying forces affects the motion of the water, mass can’t appear or disappear, and all the energy that enters and leaves the system must be accounted for. However, the standard fare of tennis balls and ramps doesn’t work as well in a continuous fluid. True, fluids will flow downhill under the force of gravity, but while they do so they can be influenced by the topography of their path and interactions with the rest of the environment (heating/cooling, evaporation, etc). When we look at the ocean, we are not only dealing with a fluid, but also have to consider the fact that the Earth is rotating. Oftentimes, there are so many simultaneous processes going on in the ocean that it is hard to pick out the ones you have been taught to expect theoretically in the classroom.

The core of Newton’s laws is the concept of inertia: unbalanced forces are required to change the motion of an object. It makes sense that you have to push on something to make it speed up, but it is often challenging to grasp that things are only slowing down because forces are acting on them, not because they have some inherent property that makes them want to slow down and be at rest. Without friction, you could roll something across the floor and it would continue indefinitely. Weird.

A "progressive vector diagram" showing the expected path of a parcel of water if it had the observed velocity. This is more direct evidence of inertial motions, since it is based on direct measurements of the velocity. (Figure credit: John Mickett)

In a fluid ocean, when you factor in the rotation of the Earth, it gets even weirder. Imagine you have a strong gust of wind that pushes on the surface of the ocean and makes it start moving. In a non-rotating model, if we were to ignore friction, that chunk of water would keep moving along in a straight line. On and on and on. However, because the ocean is on a curved and rotating surface, this is not what happens. Once a particle is in motion, if that motion persists long enough, it should follow a curved path, clockwise in the northern hemisphere and counterclockwise in the southern hemisphere (assuming the same perspective). This behavior is attributed to the Coriolis force, although in reality this is a way to explain the motion of a particle on a rotating platform from a stationary perspective. Ultimately, the water is merely conserving its momentum, traveling the path that maintains a constant momentum at all times in the absence of forces. This is perfectly analogous to the tennis ball rolling across the frictionless floor in a straight line indefinitely, once we incorporate rotation into the picture.

Although this figure can be a bit difficult to decipher, it shows that the "u" (east-west) and "v" (north-south) velocities are each switching direction twice a day, as one would expect for a particle moving in a circle. This is indicated by the shift from bluer shades to yellower shades that occurs regularly in the top 60 meters. This figure is based directly on ADCP current data. (Figure credit: John Mickett)

In real life, this can be hard to observe. The optimal conditions for producing these “inertial motions” are a strong wind-gust followed by a period of calm (persistent wind will continuously force the water to change its motion). In addition, the platform you use to observe the motions must travel with the mixed layer, which we sometimes treat as a “slab” or uniform chunk for approximations, since the surface ocean is what is being directly forced by the wind. You need to know the position of the object regularly enough to trace out a reasonable path. Finally, you must be able to observe over a considerable length of time, because these motions take close to a day to complete a rotation at our latitude.

In this case, we don’t just have to believe it because the textbooks claim it happens. Both our buoy (which we have following the mixed layer right now with the positioning of the drag elements) and the floats appear to be following largely inertial motions, collectively following very similar curved paths roughly once a day. The signature also shows up in our ADCP current data that we are measuring directly from the ship. Without any additional forcing, these motions would die out due to friction, but for the past few days, we have been able to observe a set of wonderfully predictable oscillations. Physics works!

Close-up view of the sea surface temperature in our region from the infrared satellite image. As you can see from the range of colors, we are right in the middle of a region with substantial variations in temperature. Although the image may look grainy, the resolution is actually quite impressive for satellite data, and is incredibly valuable in getting an overview of the region.

Searching the Sea

Finding anything at sea is a tricky business. The antenna on our EM-APEX floats only sticks up about a foot above the water, and often sinks below the surface when a wave passes by. Without accurate GPS positions, finding the floats would be impossible, but even when we know exactly where they are it can take some time to see them in the waves.

We brought out a total of 25 EM-APEX and have been deploying them in tightly-packed clusters to sample the ocean's small-scale horizontal variability and how it evolves in time. But after we are finished in one location we need to round them all up. Then the hunt begins.

Keaton Snyder, Avery Snyder, and First Mate Eric Piper on the lookout for EM-APEX floats. (photo credit: James Girton) 

There's a float in this photo. Can you find it? (photo credit: JG)

Once spotted, the challenge is to maneuver the ship carefully up to the tiny float in the water and bring it into range of the hooking poles near the back of the ship without running it over. This is especially tricky when the float carries fragile instruments on top. In addition to the antenna and CTD, which are relatively sturdy, about half of our EM-APEX floats carry temperature microstructure sensors for measuring turbulent mixing. The tiny glass thermistor bead on the top of these can be broken with even a light brush of the hand, let alone being bashed against the side of the ship. Needless to say, we brought spares, but as of this posting (about 3/4 of the way through the cruise) we haven't broken any! Our success is likely a combination of skill (in ship driving and hook-wielding), experience (knowing what methods have been successful in the past, as well as easing cautiously into any new procedures), and, of course, luck (in both the favorable weather on our recovery days, as well as in the way the waves and ship happened to line up during that final approach and grab). You never know what the ocean will throw your way, but it pays to be prepared.

Captain Diego Mello and First Mate Eric Piper steer the ship for an EM-APEX pickup. (photo credit: JG)

Field engineers Avery Snyder and Eric Boget prepare to snag the EM-APEX. (photo credit: JG)

Caught! With the snap hook around the
lifting loops, the EM-APEX is ready to be
brought on board. (photo credit: JG)

Safe on board (including the microstructure
probes--tiny glass beads on top of metal stalks
on the gray cylinder at the top end of the float).
(photo credit: JG)

On our first big float recovery day, we even had calm enough wind and wave conditions that we were able to put in one of the small boats that the Sikuliaq carries. This makes recovery much easier, since two people can lift the float out of the water by hand. We had planned to alternate pickups, but the boat was able to speed ahead of the Sikuliaq and pick up all 7 remaining floats before the ship could even reach one!

The workboat with a load of
EM-APEX. (photo credit: JG)

When conditions allow, the Sikuliaq's workboat
provides a quicker way to pick up multiple floats.
(photo credit: JG)

Looking at Data, Episode 3 (also: Suprises!)

One of the joys of science is the opportunity to be surprised. We are approaching the world with curiosity and questions, and looking for ways to make sense of what we observe. As much as we may understand about the universe from centuries of careful study, every year scientists collectively discover unexpected and exciting new things, and this kind of discovery is what motivates us to spend a month in an anything-but-stationary floating lab. It also requires a degree of humility and open-mindedness that can be difficult at times, yet is essential to the successful execution of an experiment.

During our second deployment, we eventually reached the point where we had the buoy with its chain of dangling instruments, 15 EM-APEX floats, and the ship towing SWIMS all surveying simultaneously, in addition to periodic CTD casts from the ship and a continuously running ADCP mounted on the ship. This may seem like an extraordinary amount of equipment to be directed towards the same small area, but because each instrument has its own particular set of advantages and limitations, the way we learn the most about what is happening is by utilizing them all simultaneously.

 

The final step in deploying the buoy is lifting the surface structure with the A-frame while all the subsurface instruments hang in the water below. Definitely a team effort. (Photo credit: RE)

Both the buoy and the floats are characterized as “Lagrangian”, meaning that they are intended to drift with a particular water mass. With the buoy, the X-wings on the chain are intended to facilitate this, and therefore the expectation is that the buoy will nominally follow the water mass at the depth of the X-wings. The floats are a little more difficult to predict, because they are continuously moving up and down, passing through shear (different velocities at different depths), and so they may not be tracking one specific water parcel. (This is one of the things we also need to keep in mind when looking at the data: differences in the density measured by a float may be due to temporal changes, spatial changes, or simply the fact that the float could be entering and exiting different chunks of water). In any event, observing how the movement of these two platforms evolves is always a matter of great curiosity.

Bluer values of salinity represent fresher water, while redder
values represent saltier water.

Over the course of approximately 7 days several things happened which surprised us all. By about halfway through, the floats had moved in a vaguely semi-circular pattern and we started joking that it would be funny if they turned again and ended up back where they started. Well, they did. Almost exactly. In fact, the full pattern (as seen in the images) describes a shape that looks somewhat like a carabiner used in rock climbing. The figures included here show a couple different properties measured by the floats, all for a depth near 20 meters: temperature, salinity, and horizontal velocity. Additional work will be necessary to refine these data, but even in a rough form changes in velocity (indicated by the length of the arrows), temperature, and salinity are visible, as well as the fact that they came back to the start!

Notice the warm-ish water in the upper left corner and
the much colder water to the right. 

Even more surprising, the buoy also followed almost the same track. At one point, it continued to head south on its own and we almost started taking bets about whether it would actually keep going that way and have to be rescued or turn and follow the floats. (It’s hard not to anthropomorphize in these moments. Will the buoy say, “Hey guys, wait for me!” and start going north?) To our surprise, it took an abrupt left turn and started following the floats north again. What must be happening with the velocity field for this to happen? (Having the buoy so closely track the position of the floats is incredibly informative, since it has both subsurface ADCPs that measure currents but also has a suite of meteorological instruments mounted on top, which allow very localized measurements of rain, wind, and heat fluxes.)

 

In looking at the velocity, it almost looks as though it goes in phases: alternating faster and slower periods. 

In the grand scheme of things, these are fairly small surprises, but they point to the truth that unexpected things can happen any time in science. Hopefully, we will find even more interesting features in the data as we dive deeper.

The weather, and other considerations

Doing research from a ship poses a number of logistical challenges that must be taken into consideration along with the science goals of the mission. We are out here to look for and study a particular phenomenon, but this means not only finding it in the first place (as discussed in earlier posts), but also having acceptable conditions in which to operate. In the case of this project, since we are hoping to look at restratification processes (several days of de-mixing) following a strong mixing event (i.e. a storm) at a front (a fairly narrow physical feature), we need to be in the right place at the right time without putting the ship in danger. Ideally, we’d like to time it so that the equipment is already in the water when the storm hits, because if things get too exciting, we either can’t deploy our instruments or we need to vacate the region, neither of which is good for the science. Fortunately, we have a very experienced ship crew on the Sikuliaq that is both very helpful in making sure the science happens but also conscientious about making sure that no unnecessary risks are taken.

Consulting about the optimal measurement spot on the bridge. From left: Chief Scientist James Girton, Captain of the Sikuliaq Diego Mello, Co-PI Eric Kunze, Co-PI John Mickett, and Second Mate John Hamill. (Photo credit: RE)

We are currently surveying the region of our final deployment with SWIMS to determine where exactly we want to place the buoy and the floats for the next round of observations. The first step in selecting this location was looking at the most recent data sent back from Argo[1] floats in the region to see if we could find places with a deeper mixed layer that would also better coincide with the storm track in the northern Pacific Ocean. We also looked at satellite altimetry (ocean height) and sea-surface temperature measurements to confirm that the deeper mixed layers could be found near a front. This allowed us to define a general box of interest, with the intent of doing on-site verification of the required features as we approached.

Ryan Newell, Eric Kunze, and Chris Siani work on reconnecting the winch cable to SWIMS after repairing the electrical connection, which is essential for transmitting data back to the ship as the instrument profiles. (Photo credit: RE)

Because the ship has to go so much slower when we are towing SWIMS, once we arrived at the beginning of our large spatial box, we used a different instrument to get a rough picture of the mixed layer, known as an “expendable bathythermograph” (XBT). This instrument is standard fare in the world of oceanography, and is frequently used on ships of opportunity because they are so simple (and inexpensive) to use. The XBT is a small torpedo shaped instrument with a lead weight at one end and a thermistor that measures temperature. Depth is calculated from a simple speed-of-fall estimation, and the data is transmitted to the ship in real time as a thin copper wire unspools rapidly. All three deployments, spaced roughly one degree of latitude apart, showed a mixed layer closer to 70 or 80 meters, rather than the 30 meters we observed at our last site, which is great news.

Marine Tech Bern explains how the XBT works to volunteer Ethan Brush. The tip of the actual instrument is just visible inside the black plastic cylinder it is housed in until deployment. (Photo credit: RE)

As mentioned earlier, however, picking a site is not as simple as finding a place in the ocean that has the right properties and going to work. Since we are now closer to the storm track, it is also important that we are not putting the ship at risk when doing the science. The box we had initially outlined for this portion of the cruise included some regions that were considerably farther north. After consultation between the chief scientists, the captain, and the mates, it was determined that if we could achieve the same scientific goals without entering that region of the ocean, both the ship and our instruments would probably be better off. As a result, we are about to embark on round three of buoy and float deployment for the home stretch of the project.

---

[1] The Argo program is a worldwide consortium of researchers using floats very similar to our EM-APEX floats that do a 2000 meter temperature-salinity-depth profile once every 10 days, often for 5 or more years. (Some floats are now equipped with additional sensors, such as dissolved oxygen and nitrogen sensors). One huge asset to this program is that the floats are distributed throughout the world ocean and profile year-round, often in places that are not readily accessible to ships all the time. You can find out more about the program and the data at argo.ucsd.edu.

Rolling around

We’re currently in the middle of a 48-hour transit to what will likely be our final measurement site, which means we’re cruising along at close to 10 knots rather than the 3 or 4 knots we go when we’re towing SWIMS behind us. One of the consequences of this is that the ship is rolling a lot more than it does when we’re going slowly. This makes a number of ordinary tasks quite challenging: drinking coffee, walking across the room, sleeping. Using the treadmill is as much of an arm workout as it is a leg workout if you want to make sure you don’t pitch off the edge. The extra (and often unexpected) rolling also means that it is even more important to make sure all the valuables are strapped down tightly. Our computers and electronic equipment are more or less bolted in place, using a combination of non-slip mats, bungee cords, hooks, and screws. The other day we rolled to a 25-degree angle, which quite effectively woke up anyone who was sleeping and had everyone else grabbing for the nearest handle as their chairs slid across the room. The Sikuliaq tends to roll more than many of the other research vessels because the bottom is rounded to allow it to work effectively in the icy conditions of the Arctic.

If you’re curious what this is like and wish to experience it vicariously, you have several options. Watching the opening scenes to Mary Poppins where the ladies of the house are holding onto furniture and china is a remarkably accurate representation of what it’s like in the lab when we hit a particularly big roll. The dancing scene at the end of Grease, where Sandy and Danny are walking around on the teeter-totter-esque platform is also pretty good (except that scene should really last 48 hours to do justice the ship’s motion). However, if neither of these quite do it for you, you can always abruptly lift a table at one end and see what falls off.

Plastic in the water around us

A couple of days ago an engine strainer – a big metal strainer used to filter the water taken in by the ship’s engine to cool itself – was taken off for a weekly cleanup, and our engineers, knowing that we have an ecologist on board, kindly suggested that I take a look. 

It was strewn with technicolor plastic bits, as well as tiny living beings that had got caught in the strainer during the week of filtering through the ocean water at 6-meter depth.

I should say, this unexpected strainer – frankly, I had no idea than an engine even had one - gave me a much clearer vision of what is really happening in the water here than any of the samples I'd tried to look at before that. 

Well, this part of the ocean appears to be sadly full of small-sized plastic, even though you won’t see it when looking at the gorgeously, impeccably blue water surface. It’s beautiful, by the way, how this principle seems to hold for everything in life: looking just at the surface, more often than not you are in for some very wrong conclusions.

These plastic bits are submerged in the water column, and establishing the layer, or the depth of their predominant concentration is very easy – one has to simply filter through the chosen depth for a while - but not technically available to me on this cruise.

What I will have a chance to do, though, is take a good look at the strainer a couple more times to see how much it accumulates every week while we’re steaming away northwards from the center of the Gyre.