Hot Water Drilling In South Pole Ice: IceCube

That only leaves the enormous engineering problem of drilling a hole two kilometers deep in the incredibly cold (-50C) ice at the south pole. It is done with a hot water drill. Essentially a hose two kilometers long is slowly lowered into the ice, as hot water is continually being pumped out the nozzle. Run a pump from the surface of the hole to recapture the now cooled water, for reheating. A cylindrical hole two kilometers deep is formed. The hose is then pulled back up, still pumping out hot water until it reaches the surface to keep the hole from freezing in on itself. A string of PMTs is then lowered into the new hole and allowed to freeze into the ice. Sounds simple!

My job was to create software which modelled this process for two reasons. First it takes a lot of fuel to heat of the enormous volume of water needed to melt a two kilometer hole, so we would like to optimize the process as much as possible. Secondly, we do not want to use too little fuel or we will not have put enough heat into the ice. Then the hole might either freeze around the expensive hose or the hole may freeze before the PMTs have been deployed to the bottom of the hole.

Two models were developed, DrillGUI and the block profile model.

1. DrillGUI: A graphical Matlab interface over a c engine. It's job is to actually solve the heat equation with time dependent boundary conditions to watch the movement of the hole wall. Its output is a "drill strategy." A drill strategy tells the drill operator what speed to lower the drill as a function of depth. There are two criteria for an optimal hole.
   
   
   1. The hole never freezes around the hose.
   2. At the time the PMTs are fully lowered into the hole, the hole should be no bigger than the size of a PMT
   
   
   After a drill strategy is completed the GUI allows the user to graph any number of quantities. DrillGUI also allows the user suggest a strategy and plots the results.


2. Block Profile Model: This Matlab model bypasses the solving of the heat equation by using "appropriate" asymptotic solutions to the heat equation to model the radial temperature profile in the ice. It models the movement of the hole wall with conservation of energy arguments and watching the flux of heat through the wall governed by the temperature gradients.

Also an argument for the maximum drill speed attainable by conservation of energy.