This is seriously one of the easiest tank experiments I have ever run! And I have been completely overthinking it for the last couple of weeks.
Quick reminder: This is what we think hope will happen: On a slope, melt water from a dyed ice cube will sink, creating a Taylor column that will be driven down the slope by gravity and back up the slope by vorticity conservation, leading to a “westward” movement in a stretched, cyclonic trajectory.
We are using the DIYnamics setup: A LEGO-driven Lazy Susan. And as a tank, we are using a transparent plastic storage box that I have had for many years, and the sloping bottom is made out of two breakfast boards that happened to be a good size.
Water is filled to “just below the edge of the white clips when they are in the lower position” (forgot to take measurements, this is seriously what I wrote down in my notes. We didn’t really think this experiment would work…)
The tank is then rotated at the LEGO motor’s speed (one rotation approximately every 3 seconds) and spun into solid body rotation. We waited for approximately 10 minutes, although I think we had reached solid body rotation a lot faster. But we had a lot of surface waves that were induced by some rotation that we couldn’t track down and fix. But in the end they turned out to not matter.
To start the experiment, Torge released a blue ice cube in the eastern corner of the shallow end. As the ice cube started melting, the cold melt water sank down towards the ground, where it started flowing towards the bottom of the tank. That increased the water column’s positive relative vorticity, which drove it back up the slope.
This was super cool to watch, especially since the ice cube started spinning cyclonically itself, too, so was moving in the same direction and faster than the rotating tank.
You see this rotation quite well in the movie below (if you manage to watch without getting seasick. We have a co-rotating setup coming up, it’s just not ready yet…)
Very soon, these amazing meandering structures appear: Rossby waves! :-)
And over time it becomes clear that the eddies that are being shed from the column rotating with the ice cubes are constant throughout the whole water depth.
It is a little difficult to observe that the structure is really the same throughout the whole water column since the color in the eddies that were shed is very faint, especially compared to the ice cube and the melt water, but below you might be able to spot it for the big eddy on the left.
Or maybe here? (And note the surface waves that become visible in the reflection of the joint between the two breakfast boards that make up the sloping bottom. Why is there so much vibration in the system???)
You can definitely see the surface-to-bottom structures in the following movie if you don’t let yourself be distracted by a little #HamburgLove on the back of the breakfast boards. Watching this makes you feel really dizzy, and we’ve been starting at this for more than the 8 seconds of the clip below ;-)
After a while, the Taylor column with the ice cube floating on top starts visibly moving towards the west, too. See how it has almost reached the edge of the first breakfast board already?
And because this was so cool, we obviously had to repeat the experiment. New water, new ice cube.
But: This time with an audience of excited oceanographers :-)
This time round, we also added a second ice cube after the first one had moved almost all the way towards the west (btw, do you see how that one has this really cool eddy around it, whereas the one in the east is only just starting to rotate and create its own Taylor column?)
And last not least: Happy selfie because I realized that there are way too few pictures like this on my blog, where you see what things look like (in this case in the GEOMAR seminar room) and who I am playing with (left to right: Torge, Franzi, Joke, Jan) :-)
Hi, I’m Dan Wallace, a PhD student at the Institute of Sound and Vibration Research in Southampton, UK. I’m interested in lots of areas of acoustics — here I am working on a prototype 3D audio system in one of our anechoic chambers, as part of my Masters project.
Dan inside an anechoic chamber. Photo credit: Dan Wallace/University of Southampton
Today though, I’d like to talk about a machine I’ve designed, the CHocolate Oriented Machine-learning Processor (or CHOMP for short). CHOMP is a machine learning system designed to play a table-top game to super-human levels, with no silicon chips, no neural networks, not even any electricity… Introducing CHOMP!
Chomp’s hardware: A bunch of plastic boxes. Photo credit: Dan Wallace
This pile of plastic boxes has triumphed over PhD’s, employees of Google DeepMind, numerous seven-year olds and our University Vice-Chancellor – all it takes is a little training.
First, let’s introduce the game we’re playing. Like all the best games, our game is played with a big bar of chocolate, but for sustainability (of our waistlines), ours is 3D printed. By the way, NGCM stands for Next Generation Computational Modelling. NGCM is the Centre for Doctoral Training who are supporting me throughout my PhD with training, equipment and funding.
3D-printed chocolate bar. Photo credit: Dan Wallace
Players alternate taking bites out of the bar of chocolate from the bottom right corner, and the aim of the game is to avoid the poisoned square in the top left corner. Bites can be as big as you like, provided you follow one simple rule: Pick a square, remove it, then remove all squares below it and to the right.
A game of chomp. Photo credit: Dan Wallace
Let’s look a little closer at CHOMP to see how the machine makes decisions. Each of the 33 boxes which make up CHOMP is labelled with a picture of the chocolate bar in a different state, and in every box are some coloured beads.
A glance inside the inner workings of Chomp: one of Chomp’s plastic boxes with colored beads inside. Photo credit: Dan Wallace
Each different coloured bead represents a different sized bite out of the chocolate bar – remember the rule: Pick a square, remove it, then remove all squares below it and to the right. When playing against people, we provide a handy map to show which coloured bead refers to which square. CHOMP lacks the dexterity to choose moves for itself, so we let players help by picking a bead at random from the correct box.
Which color bead corresponds to which move? Photo credit: Dan Wallace
As we play the game, we record the moves which are played by each player by placing the chosen bead on top of the box it came from. At the end of the game, we have two lists of moves, one of which ended in a loss (in this case, made by the human) and another which ended in a win (for CHOMP!)
Documenting a single game of Chomp. Displaying the moves that got chosen by the human and “the machine”. Photo credit: Dan Wallace
This information about good and bad moves enables us to train the machine.
Every losing move is removed from the game, decreasing the probability that CHOMP would choose that move, from that position, in the next game. Every winning move is returned to its box, with two extra beads, increasing the probability that CHOMP will make good moves. Through this process alone, called “Reinforcement Learning”, CHOMP learns the winning strategy.
Winning moves are reinforced by adding more beads the color of the winning move. Photo credit: Dan Wallace
And it works! We have recorded every game we’ve played with CHOMP since spring of 2018, marking games off in sets of seven. Human wins are marked in orange from the top of a set, and CHOMP wins are marked in green from the bottom. At the start of each tournament (the leftmost column in each chart) we reset CHOMP so that every bead in each box is equally likely to be chosen. The results show that while CHOMP might win a game by chance, six times out of seven, humans win. Over time though, CHOMP gets stronger and stronger, winning five or six games out of seven after around 70 games worth of training.
Documenting how the probability of winning against Chomp decreases the more Chomp has been trained. Photo credit: Dan Wallace
We usually give our opponents a Cadbury’s Chomp bar as a prize for beating the machine. If they lose, we still give them chocolate as a thank-you gift for helping to train up CHOMP. We are #notsponsored by Cadbury’s yet!
If you are lucky you win a Cadbury bar when winning against Chomp! #notsponsored by Cadbury (yet?). Photo credit: Dan Wallace
The beauty of CHOMP is that it is simple enough for a five-year-old to play, but powerful enough to surprise a professor. We’ve taken CHOMP to schools, science centres, university events and academic conferences, receiving some brilliant feedback from our defeated opponents. I’ll leave you with some quotes:
“So often, communication about AI and technology centres on the amazing tech, this game removes all that and shows how these machines operate beneath the algorithm!”
“Great presentation and concept. I loved seeing the “analogue” version of something so often thought of as digital!”
“Very interesting, it tears apart the fear of AI, because it’s just plastic boxes!”
Chomp on a winner’s podium. Photo credit: Dan Wallace
A pink swirl going across a styrofoam block underneath a layer of yellow water? What’s going on here?
The picture was taken in a water tank, simulating the circulation of water masses in a fjord. A fjord is a long and narrow bay, usually with a sill that is separating the bay from the open ocean. And those sills play an important role in on the one hand preventing water exchange between the fjord and the open ocean (because everything below sill depth has a really hard time getting across the sill) and on the other hand mixing water masses inside and outside of the fjord (which we see visualized with the pink dye).
And here is why the sill is so important: Every time the tide goes in or out of the fjord (so pretty much all the time), the sill acts as an obstacle to the water that wants to go in or out. And flow across a ridge tends to create mixing downstream of the ridge.
In the picture below, we see a sketch of the situation in an outgoing tide, which is what we also see represented in the photo above: Water wants to push out of the fjord and has to accelerate to get through the much smaller cross section where the sill is located. This leads to strong currents and strong mixing “downstream” of the obstacle.
Except that “downstream” is on the other side of the sill only a couple of hours later, when the tide is pushing water into the fjord, but is again hindered by the sill.
So what is happening is this: The tidal current goes in and out, and mixing occurs on one or the other side of the sill. So the situation looks like this:
This is what that looks like in our tank (the “tidal waves” are generated by lifting the right end of the tank and then just slushing back and forth):
Of course, in reality we don’t see pink swirls, and the surface layer isn’t a different color from the deep layer, either. But that’s why tank experiments are so cool: They show us what’s going on deep below the waves, that we can otherwise only deduce from complicated measurements of temperatures, salinities or mixing rates, which require highly specialized equipment, a research ship, and lots of technical know how to process and analyse and display. Which, of course, is also being done, but this demonstration gives a quick and easy visual representation of the processes at play at sills all around the world.
P.S.: The photos in this blog post were taken when I ran the fjord circulation experiment with Steffi and Ailin at GFI earlier this year. I am posting about this again now because I wanted to use the picture for other purposes and realized that I never actually wrote about this feature in as much detail as it deserves!
On publishing in a journal peer-reviewed by kids, and suggesting it as a first journal new PhD students should be asked to write for
You guys might remember my favourite experiment with the ice cubes melting in freshwater and saltwater. This experiment can be used for almost any teaching purpose (Introduction to experimenting? Check! Thermohaline circulation? Check! Lab safety? Check! Scientific process? Check! And the list goes on and on…) and for any audience (necessary observation skills start a taking the time it takes ice cubes to melt in the easiest case, to observing the finest details of the melt). In short, I love this experiment!
A different format of science communication
After using it in all kinds of settings for years, I wrote up the experiment for Frontiers Young Minds, a journal which is written for, and peer-reviewed by, kids (link to my article). I love the idea of not only tailoring your science communication to the audience of young readers, but making sure that it actually works well for them by including them in the process. Additionally, the peer-reviewers get a great insight into how a publishing process (and thus an important step in science) works, too.
The whole peer-review and publication process was a really positive experience. Speciality chief editor for “Earth and its resources“, Mark Brandon, and the whole team were super responsive and helpful all the way from initial article idea until publication.
Writing for and being peer-reviewed by young readers
Having my writing peer-reviewed by the “young readers” was super interesting. For example, on one of my articles, they commented on how, as kids growing up in the US, they were not familiar with metric units and could I please give them units they could actually relate to? This is an issue I should probably have been aware of, but I totally wasn’t.
Another example from the other article: a different young reader commented that English was their second language, and could I replace difficult words like “puddle” and “dye” with easier words. As a non-native English speaker myself, this feedback was super helpful — I thought that I was writing in an easy language already, but clearly my perception of “easy language” has drifted into specialized vocabulary — super valuable feedback!
And then both teams reviewing both my articles had a science mentor helping them, and also commenting him/herself on the article and how the review process with the kids went and suggesting further edits, that would make it easier for kids to work with the article.
Illustration by Jessie Miller for Frontiers Young Minds, used with permission
And then, of course, there are Jessie Miller‘s super cute illustrations! After seeing what she did for my first article, I couldn’t wait to see what would happen for this one, and I am super excited about another illustration that makes me feel completely understood and seen.
Writing your first ever article for FYM?
So all in all, publishing with FYM is something I would totally recommend to anyone. And I would even go so far as to recommend it as the first article that PhD students should be asked to write. Why?
Articles for FYM can be written on “core concepts”, which can mean basically writing a literature review on the topic you are about to write a PhD thesis on, and one that is broken down so far that you will really have to have understood things. There is this saying attributed to basically all science educators in one form or another, that only if you can explain your topic to a child, do you actually understand it yourself. So explaining to children is actually a super helpful step in the process of getting into a topic yourself.
Writing something that is designed to be understood by a wide variety of audiences is really useful for another reason, too: to give to all your family and friends as an easy insight into what it is you are spending all your time on.
The feedback you get on how you talk about your topic will be helpful for all future communications about it; Practicing scicomm as early as possible is always a good idea :-)
Having a really positive publishing experience is a great start into a PhD, because surely other kinds of experiences will follow sooner or later. The submission through the uploads and forms and stuff works the same way for FYM as for all other journals (including the “oh crap, they want the images in a different format than I prepared them in! Let’s google how to convert them”, “Really? They need an abstract? Maybe I should have read the instructions more carefully…”, or “They are really counting the words on the submission! So now I need to cut an extra paragraph that I thought I could get away with…” surprises that are typical for the “Let me quickly submit this article and go for lunch! Oh wait, half a day later and I am still nowhere near the end of the process” experience that is so common when submitting articles. At the same time, the stakes feel a little lower for this kind of article, since as an early PhD student, you are writing about other people’s work, not yet your own (at least when writing a core concept article, there is also the “cutting edge research” article type, in which you are writing about some newly published article of yours). And then, as I described above, the whole process is really positive and friendly and supportive throughout, even though all the steps are the same as for any other journal (Waiting for the editor to send the article out to the reviewers. Seeing that stuff is waiting on a desk somewhere and compulsively checking every day whether it has been moved on and the email notification just didn’t make it through. Replying to a reviewer. That kind of things). So I believe that it’s a really good way to be introduced to the publishing process without being pushed into super cold water right away, building up confidence for later submissions of your own work.
FYM announces new articles on their social media (with lovely tweets!), which have a fairly wide reach, well above what most of us have, and that’s a great opportunity to be seen as authority on a topic by a large number of potentially interested people. Great opportunity to expand your network!
And, as I said before, I just love the illustrations and I would imagine that having something like this when you start working on a new topic would be super exciting and motivating :-)
What do you think? Will you suggest writing a FYM article to all your new PhD students now?
Seeing this illustration (and, of course, having the article published) was a super nice surprise during the busy run-up to my big event, which is actually happening right now (good thing I know how to schedule blog posts ;-)). The illustration makes me suuuuper happy because to me it really captures what the article is about and, more importantly, what my goal in writing the article was. And I feel seen and understood in a profound way, and reminded of who I am. Never underestimate the power of #scicart! Thank you, Jessie!
Glessmer, M. S. (2019) How Does Ice Form in the Sea? Front. Young Minds 7:79. doi: 10.3389/frym.2019.00079
Showing double-diffusive mixing in tank experiments is a pain if you try to do it the proper way with carefully measured temperatures and salinities. It is, however, super simple, if you go for the quick and dirty route: Cream in tea! Even easier than the “forget the salt, just add food dye” salt fingering experiment I’ve been recommending until now.
The result of double-diffusive mixing of cream in tea is probably familiar to most (see above), but have you ever looked closely at the process?
Below, we pour cold cream into hot tea. The cream initially sinks to the bottom of the tea cup, but then quickly heats up and fingers start raising to the surface of the cup. They are visible as fingers because while the heat has quickly diffused into the cream, the actual mixing of substances takes longer and the opaque milk stays visible in the clear tea. Only when the fingers have risen to the surface the substances begin to mix due to shear and diffusion of substances. Hence the name “double diffusion”: First diffusion of heat, then of particles afterwards.
Pretty cool, isn’t it?
If you happened to stir the tea before pouring the cream, it looks even more awesome. Home-made galaxies :-)
And isn’t it fascinating how the blob of cream in the middle of the cup stays intact for quite some time?
So now you know the only reason why I am drinking black tea: So I can do salt fingering experiments with it! :-)
My friend Alice Langhans runs a super cool science communication Instagram (@edu_al_ice), where she posts about her experiences as PhD student in physics education research. And there is a lot more going on on that Instagram than just pretty (but oh so pretty!) pictures. I make sure to read all her posts, because there are always interesting, motivating, inspiring thoughts hidden behind that “read more” button. And now she’s even started a new series of physics experiments on #experimentalfriday, and I am super excited that she wrote this guest post for me!
But now look at the picture below, and then read about some magic! :-)
Magic! One of the arrows changes its direction and here is why:
Click for large picture. Picture by Alice Langhans.
First, the arrows are unchanged and visible through the glass.
Click for large picture. Picture by Alice Langhans.
Adding water to the glass, the image of the arrow gets bigger and appears mirrored!
Click for large picture. Picture by Alice Langhans.
With even more water even the second arrow appears bigger and mirrored.
Click for large picture. Picture by Alice Langhans.
The waterglass I used is round and the refraction of light in water is different than in air, which makes the water glass act like a positive (converging) lens. This is why the image of the arrow appears bigger and mirrored.
Think of the arrow as many points, each of which is the source of a divergent bundle of light. The light coming from the point that is the arrowhead on the right, is refracted through the waterglass and reaches our eye to the left. The light from the left end of the arrow refracts in such a way that it now enters our eye on the right side.
Notice, how you can also see how the upper arrow appears even bigger? The glass is more wide at that height, magnifying properties of the water glass lens are therefore increased.
Isn’t that a super nice demo? I love it! Thank you for writing this guest post, Alice! :-)
P.S.: Alice has just been interviewed for a podcast. Curious what she’s talking about on there? Me too, but that’s why I follow her Instagram (@edu_al_ice) — to never miss out on all the cool stuff she’s up to! :-)