Science “Connections” in 10 Steps
Ice Shelves, Materials Density, Sea Levels, and More
As news of “a Delaware sized piece of the antarctic ice shelf breaking off” continues to circulate, with various levels of rhetoric and misunderstanding, I thought it might be a good time for a simplified overview of some basic science**
1) It’s important to first understand what an “Ice Shelf” is, and why it is NOT the same as a “Glacier”:
1a) Glaciers form on, and are supported by, land. They are made entirely from accumulated, packed precipitation over geologically long periods. They are massive “fresh water stores” that have been gradually taken out of the climate cycle, thus lowering the sea level over time. When glaciers melt off (or chunks break off and fall in), 100% of their stored water goes (sooner or later) back into the water cycle and ocean, and *can* cause the ocean levels to rise.
1b) On the other hand Ice Shelves may look like landmasses, but they form on, and are actually directly floating on, the sea’s surface, with no land supporting them. This happens due to a combination of seasonal freezing of water on the ocean’s surface from extreme cold temps, from some new accumulated snowpack (once there is a solid top on which to accumulate), and the slow flow of ice from thicker “inland” sources of ice flowing glacially outward.
1c) Bonus word: “iceberg” — it just means free-floating sea ice, and makes no distinction as to whether the ice came from a detached piece of an ice shelf, formed on the shore during the winter and floated off, formed in extreme temperatures in the open ocean, or broke off from a “permanent” glacier directly into the sea, and floated away.
2) Pieces of ice shelves break off regularly, due to things like:
- long term stresses from ocean waves and tides (like bending a credit card back and forth until it breaks).
- seasonal melt as surface and ocean temperatures rise.
- long term climate trends
Sure, the piece that just broke off is larger than any which have broken off in recent years, but there have been larger ones in the past. This regular “breaking off” has been going on since prehistory, and throughout recorded history, and is by no means an unprecedented or alarming event on its own.
3) “But this break must be mostly because of the march of global warming, right?” Not necessarily. Antarctic sea ice was actually at near record growth, trending up solidly for years, until August 2016 when summer hit the southern hemisphere and the ice started rapidly shrinking, bringing it to its least extent since 1997. The shelf was only able to grow so large because of colder-than-normal temperatures across multiple years, and there was even less ice than now as recently as 20 years ago. Which is to say: it’s warm there now, but its been unusually cold for a long time. It not out of band for historically periodic fluctuations.
4) “When pieces of the shelf break off, float away, and melt, won’t the thousands of square miles of ice that are above sea level going to contribute to sea levels rising?” No, because floating ice (which the shelf is, and has always been since it formed) doesn’t increase water volume as it melts. This is thanks to the science of water and ice density, which is going to need some explaining…
5) “Density” is the ratio of a thing’s “weight per volume”. Everything that “floats” at a stable position in any fluid (i.e. is not actively sinking or rising, but is vertically stable) only does so because it is LESS dense than the fluid below it, and MORE dense than the fluid above it. Fluids (of which air is one), and even some loose solids, tend to settle into layers of different densities, heaviest on the bottom.
- If you have a 1 inch cube of wood and another of steel, the steel cube is heavier, right? That means steel is more dense than wood. (I’m pretty sure the facts check out on this).
- If you weighed a 1 inch cube of air, the steel and wood cubes would both be heavier than it. That means they are both more dense than air, and both will sink in air, rather than float (pretty sure this also checks out).
- Comparing our solid cubes to a 1 inch cube of water now, the wooden cube is actually lighter than the water cube, and the steel cube heavier (feel free to check my math on that). Therefore we can expect the steel cube to sink through both air and water, but the wooden cube will sink only through the less-dense air, and then FLOAT on the more-dense water, settling stably between the interface of the water and air (at least given a sufficient volume of water so that it can displace a little and be seen to actually float, which I’ll come back to).
5b) If you find the above cubes to be too abstract, consider this alternative example: A beachball gets “perfectly filled” with the same air and pressure that’s outside of it. Its total weight by volume necessarily includes the 1mm vinyl skin of the ball as well as the air inside of it. Since air is very light, that vinyl skin is actually a significant part of the total weight of the air-filled ball, and because of its heavy skin the total weight pre volume will be heavier than the same volume of air. If you filled the ball with water, I believe most would rightly expect it to weight more than the air-filled ball (and with water being so heavy, the skin becomes a negligible part of the total weight). So an air-filled beachball weighs more than the same volume of air, but less than the same volume of water. That translates to the beach ball being denser than air and less dense than water. That’s why it FLOATS between the two.
5c) You can even see density layers in loose solids: Take a well-mixed jar of sand and “styrofoam snow”, and agitate it side-to-side. Or if you can, find one of those great old “magic window” toys from the 70’s and look at a how a behaves [there’s a video link to a newfangled one at the end].
6) Water and ice (meaning “ice made from water” specifically) are an interesting case because they are both made of water. So why does ice float? Shouldn’t it be the same density and just bob around “wherever” in the water? No. Remember that water expands some as it freezes. Which is to say that water gets volumetrically larger as it becomes ice. Since its weight remains the same, but the volume larger, it becomes less dense overall (same weight for a now-higher volume). Since ice is lower density than water, it will float.
7) Back to “displacement”: One of the really cool rules of “floating” is that everything which floats on something else (no matter what it’s made of, or what fluid it’s floating on) will displace only as much weight (not volume) in the fluid below as the floating object itself weighs (aka how much force it presses down with into the fluid). With objects of densities that are closely tuned to the fluid they are floated on, it can mean an object that’s just barely peeking out of the surface of the fluid, or one sitting on top as if the fluid is a solid surface.
7a) Note that very heavy ships, made of very heavy steel, manage to float (at least under ideal circumstances). This is because of their shape: If all the spaces inside of a ship, which could potentially be filled with and hold water, are actually just filled with air, the total density of the shape is very low. It’s much like a beach ball with an oddly shaped metal skin: All that air inside the shape reduces the overall density to the point that it’s less dense than water, and will float. Science doesn’t much care what the materials are; it only cares about the weights per volume, and how much of the fluid the floating object needs to displace in order to match its own weight.
7b) So if a ship with 30 foot walls weighs 20,000 tons on land, it will displace 20,000 tons of water. While that sounds like a lot, water is pretty heavy and the ship may only be enough displace enough water for it to sink 18 feet into the water, leaving 12 feet of high and dry walls above the surface. Now start replacing the air inside the ship with water, making it gradually heavier. With each bucket of water you pour inside (simultaneously displacing 1 bucket of air upwards in the process) the ship will displace another bucket of water below, making the vessel sink lower and lower. Eventually the total steel+water weight is more than the total water that the ship’s shape could possibly displace before sinking all the way.
8) As noted above, water and its ice have different densities. An important note is that their densities are fairly constant (at least for all practical “science 101” purposes, ignoring small variability due to impurities, temperature, etc). The constancy of their relative densities mean there is also a constant ratio of how far ice will sink into water. It doesn’t matter if it’s a glass of water with ice cubes floating in it, or an iceberg preparing to sink the Titanic: approximately 90% of the ice’s volume will be under water, and 10% will rest above the surface. That’s just the math based on how much water expands when it freezes, and how much the density therefore drops. Incidentally, this constant relationship is also where the phrase “what you see is just the tip of the iceberg” originates.
9) Thought experiment: If you float some ice cubes in a glass of water that’s filled right up to the rim, some of the ice will be sticking up ABOVE the rim of the glass (in fact, we know it’s about 10% of the ice). So… will the glass overflow as the ice melts into it? The answer is ‘no’. Remember that ice takes up more space than the water it was made from, and it will again take up less space as it melts.
9a) “OK, so it gets smaller. But will it be small enough to not overflow?” Yes! In fact, it will be EXACTLY the right size. We already know that the ice is “displacing a volume of water equal to its own total weight”. We also know that whether frozen or melted, the ice’s weight is constant. And since it happens to be melting into the same kind of liquid that it was made from, its liquid volume must be equal to the volume that it had been displacing. The melt will perfectly fill the “void” left as the ice melted, leaving the water level constant exactly at the rim. [there’s a video link of this at the end]
10) So when an ice shelf, made mostly from frozen ocean water, whose volume floats 10% above the surface, finally melts back into the sea (or breaks off, floats away, and melts) does the sea level rise? No. Just like the glass of water, there’s no appreciable net change. The ice shelf’s water was never “stored long-term” (geologically speaking), having only recently been frozen out of the sea, or temporarily prevented from precipitating directly into the sea as part of the normal water cycle. Yes, there are surely some small differences in water volume because of some older inland ice in the mix, and because of impurities, and other niggling reasons, but it’s all pretty insignificant compared to what happens with melting glacial ice, which is a 100% “new” contribution to the sea’s volume, at least as far as human history is concerned.
- 2013 Story on Record Sea Ice
- (current story) NPR — Why is the Ice Shelf Cracking
- How Density is Measured in the Lab
- Melting Ice in a Full Glass of Water [video]
- Solid Density Layers / Magic Window [video]
** I carefully stated “simplified” and “basic” in the introduction. Sure, I could have cut fewer descriptive corners, used technically more-correct terms, taken the time to explain those, pointed out exceptions and caveats, and explained the reasons for them, too. I absolutely could have been far more scientifically rigorous and pedantic about all the details. But that’s not the point here. This material is meant to be “intro level” - to get the scientifically curious to better understand some basics of science, in general, and as it applies to current events. My hope is that it sparks some people into learning more about a subject that they perhaps never spent much time on before.