If you are fortunate enough to have access to a port-side win-dow when flying into McMurdo Sound, Antarctica, you’ll see a long, slender glacier that spills off the south-western flank of Mount Erebus and then floats out into the waters of the Sound. This is the Erebus Glacier tongue, as big as any glacier in New Zealand, but tiny in Antarctic terms. It is also uncommonly narrow relative to its length, and edged by substantial undulations. It has been the focus of research ever since Scott’s last expedition, with geologist Griffith Taylor documenting its structure (Taylor 1922). Remarkably, given the decades-long interval between occurrences, the last few kilometres of the glacier broke off during their time there (Stevens et al. 2013). The abnormal structure, history of research, and ease of access mean it has been scientifically pored over for decades. It was also the starting point for the development of some Antarctic infrastructure that has been a mainstay of New Zealand sea ice research for more than thirty years. Ernest Rutherford is supposed to have said ‘we don’t have much money so we have to think’ (da C. Andrade 1964). In the era of ~10% success rates for funding applications, this could be augmented to additionally say ‘and be flexible and be prepared to take one’s time’. Big science is about big missions, large teams focused on a fundable goal, with its end-points well defined. Certainly, within this envelope, research takes its meanders and, science being science, advances are often found in the side-meanders. What about a different kind of mission? What if the mission is built upon meanders? What if we could go from listening to a glacier through to predicting the impacts of a changing climate?