UNDER PRESSURE...

​Somewhere back in your very first diving course there will have been reference to air being compressible because it’s a gas, as opposed to water being non-compressible. When we start thinking about how molecules interact, it’s helpful to go back to school science. In solids the molecules are packed tightly together, they can vibrate but they aren’t free to move around much. In liquids the molecules can move but stay interacting with each other and in gases the molecules fly around freely bouncing off the walls of whatever contains them.
Thinking about gas molecules as random whizzing around bumping into other molecules and their container, it’s easy to understand why if we push more molecules into the container the number of collisions with the container wall will increase and the pressure goes up. This kinetic model of gases assumes that the gas particles themselves are very small (they are) and there’s a lot of space between these gas particles. That leaves lots of extra space for us to jam more molecules in there and compress the gas together. Hence gases are compressible.  This nicely explains Boyle’s Law (remember that?) but makes a number of assumptions which create a concept of an ‘Ideal gas’. Sadly, Ideal gases don’t exist and we have to deal with real gases. But at low pressures Boyle’s Law is fairly useful.
Liquids don’t behave in the same way because the molecules are already close together and interacting to a limited extent. Each molecule forms temporary associations with the other liquid molecules. If you heat up a solid to melt it into a liquid, you can measure the temperature increasing. As you get to the point where the solid it melts, the temperature will stabilise (even though you are still heating it). This is the point where the molecules are absorbing heat energy to give them the kinetic (movement) energy to move around. Once all the solid molecules have absorbed enough energy to melt into liquid, then you can see the temperature start to rise again as you carry on heating it. During this phase the molecules will move faster, but they still interact with each other. Keep heating and give them enough energy and they will manage to escape the interactions and form a gas.
Water is a bit of a strange liquid, because it’s molecules interact more than other liquids, and this gives it some strange properties. Water molecules form hydrogen bonds to other water molecules, and then they break these bonds and reform them with other nearby water molecules. Although these bonds are quite weak, there are lots of them. If it weren’t for hydrogen bonds then water wouldn’t be a liquid at all. When the other elements in the same family of the periodic table, like Sulfur and Tellurium, bind to hydrogen they form gases not liquids.
Water is most dense (the molecules are packed tighter together) at 4 degrees. At this temperature the hydrogen bonds are quite structured and pull the molecules tighter. As the temperature rises the bonds start to make and break more frequently and allow the water molecules to move a little further away from each other, so water becomes less dense. At temperatures below 4 degrees, hydrogen bonds don’t form so well and so as water cools to become a solid, it also becomes less dense. This explains why ice floats on liquid water (good news if you’re a polar bear) and why cold water sinks into the ocean (that’s  thermoclines).
So, these little fairly weak hydrogen bonds have a lot of influence and it’s them we are fighting against when we try to compress water. They have already done the job of pulling the oxygen dihydride molecules far closer together than we’d expect from the other elements in their group. It’s because of them that water isn’t a gas at room temperature.  And once we’ve got as far as forming a liquid, there isn’t much compression left to achieve. At 4km down in the ocean, water has a measurable compressibility of just 1.8%. It’s not quite true to say water isn’t compressible, it’s just not very compressible and for the depths that we will visit we can probably ignore the marginally increased density.
Resistance to compression for any substance can be described by the bulk modulus value. This is a measure of how much pressure must be applied to reduce the volume by 1%. For solids, these values are predictably very high, eg diamond is 443,000,000,000 Pascals and steel is 160,000,000,000 Pa. For ice I would need 2,000,000,000 Pa to compress it by 1%.  For liquids we would expect the values are lower and generally they are. Water bucks the trend though and needs 2,200,000,000 Pa. So, I actually need more pressure to compress water than I do to compress ice. Blooming pesky things those hydrogen bonds!