Palmer's initial topic of study was
soil mechanics, particularly at low temperature; he would later investigate
ice flow and the mechanical properties of
ice, which would remain a recurrent, long-term interest of his. He would deploy
dimensional analysis, which he described as 'a magical way of finding useful results with almost no effort,' as well as physical models of systems that, while simple, nevertheless captured a relevant aspect of the problem and allowed for experimentation and optimisation cheaply, which was especially important before
digital computers were powerful enough to simulate complex systems. Sometimes, the models were not so small: Palmer realised that the 1:20 scale modelling of storm hazards near
Western Australia were insufficient, and instead was involved in building a large 1:6
flow cell. His contribution to understanding how pipelines
buckle form the basis of modern pipelines are designed to avoid this hazard, and he introduced a new way of laying pipelines in deep water partly-filled with seawater (previously, they had been laid empty), so that the walls did not need to be as thick (significantly reducing costs) to stop the pipelines buckling under the pressure. == Other work ==