As the days grow shorter, the sunsets get pinker and the air temperature drops, the short Antarctic autumn kicks in. From the hills behind the base we can look north and see heavy grey clouds rolling over the blue black ocean as water evaporates and cools the warm ocean, stirring up the colder air. Ice forms in off-white patches but wind and waves move it away, or mix the cooled surface with the layers just below, still warmer than freezing from the remains of the heat absorbed over the summer. Until all this heat has been dissipated any ice that forms is only temporary, and certainly not suitable for us to stand on.

Closer to the base thicker ice is left over from previous years, some having been around for over twenty years, held fast by the coastline and wedged in by islands or the floating ends of glaciers. Parts of this receive heavy snowfall, enough to push the top of the floating ice below sea level. This eventually floods with sea water and freezes. As the years pass, more snow falls, pushing the ice lower and forming more snow ice on top. Each summer the bottom gets melted away by the warm ocean, while in winter the thick ice and snow limits the amount that can freeze onto the bottom, so over time the whole depth of the ice is made up of this frozen sea-snow mixture.

Right now, at the site we're using as a platform to measure the ocean, there's about two meters of this snow-ice, a few centimetres of slushy sea and snow, and eighty centimetres of snow. That's thinner than you'd expect from ice that's been left alone for so long, probably because strong currents in the area enhance melting of the bottom during the summer. It's also being squashed up against Ross Island by the slow advance of the McMurdo ice shelf, bending it into a series of buckles. These are annoying because they combine with the surface flooding and freezing to cause significant variation in the total ice thickness. This means that we have to be careful when measuring the thickness, as otherwise we'd not know if a measurement this week was lower because some ice melted off the bottom, or just because we'd drilled a hole in a bit of that was thinner anyway. We can sort of see these in the snow on the surface, but they're quite subtle, and snow gets sculpted by the wind, so we can't really know what's happening to the ice by just looking at the snow. The only solution is to dig a lot of holes in the snow, then drill through the ice at the bottom to profile these ridges and troughs. Measuring the ice thickness is pretty easy and only takes a minute or two, but getting there in the first place takes a bit of an effort.

Digging holes is something of an art. A small hole is pretty easy, you just whack the snow with a shovel a couple of times and you'll quickly have a hole down to your knees. Once there, though, things get more tricky. Not only have you got to break up the snow, you've got to get it out of the hole (and probably back in again later). You have to be sure to make the hole big enough to fit yourself and your shovel or you'll just end up bending down from the surface, and that never ends well. Here the correct choice of shovel is very important. Snow likes to stick together, and it's best moved about in big blocks. A good shovel will pop out smooth blocks without breaking them apart, but still have enough of a curve to scoop out the few loose bits left over from the levering out the block. Too much like a spade and the soft stuff slides off, too much like a shovel and the blocks break up and need too many swings to empty out the hole. You can adopt a double shovel technique here, taking one blade-like spade to cut the blocks, and a huge bucket of a shovel to lift them from the hole, but this would be a little excessive. The best way to dig a lot of holes, though, is to get someone to help you. If nothing else, moving snow about is just as good as going to the gym, gets you outside, and provides a much better view.

Having dug close to 40 holes as deep as our chests we've got a good handle on the shape of the ice at our current site and slightly larger muscles in our arms. For now any ice work we do has to happen at this snowy site as the new ice hasn't formed, so we expect to do a bit more digging before we're done...