Besides easier maintenance, a piston motor may have better energy efficiency. The big two-stroke Diesel engines, like those used on ships, have energy efficien

Besides easier maintenance, a piston motor may have better energy efficiency. The big two-stroke Diesel engines, like those used on ships, have energy efficiencies over 50%, e.g. around 55%. I do not know if there have been some recent improvements in gas turbines, but at least a few years ago the efficiency of gas turbines was lower, sometimes as low as 40%. The reason is that the exhaust gases of a gas turbine are very hot, retaining inside a significant fraction of energy from the consumed fuel. Electrical turbogenerators are preferred for mobile or military applications due to much smaller sizes for a given power and higher reliability, not because of good efficiency. In order to have good efficiency when using a gas turbine, one must recover the heat from the exhaust gases, like in all combined-cycle power plants, where high efficiencies, e.g. over 65%, are attained by using a cascade of steam turbines, typically 3 steam turbines, for heat recovery. This increases a lot the size of the power plant and it requires a lot of water, which in Antarctica could be obtained only by melting ice, reducing the efficiency gains and complicating a lot the power plant (a special construction would be necessary to prevent water freezing in undesirable places and to use the water in closed cycle, otherwise the energy efficiency of a combined cycle starting from ice may be less than of a simple gas turbine). A more realistic solution for using a gas turbine in Antarctica would be to recover the heat from the gas turbine that burns fuel with a closed-cycle gas turbine using supercritical carbon dioxide. While this is a promising method for replacing the entire cascade of steam turbines (with steam turbines, you need a cascade because heat is exchanged at fixed temperatures, where water is boiling; while with a supercritical fluid heat can be exchanged at a continuously decreasing temperature of the exhaust gases, recovering most of the waste heat in a single heat exchanger), it was mostly experimental and I do not know if there are any suitable commercial products at this time.