Andrews’ inquiry into the glaciation of Baffin Island extends over five decades, covering most of eastern Baffin Island and adjacent shelves. Although the original paradigm was simple deglaciation from a Last Glacial Maximum (LGM) ice advance to the shelf break, strong vertical gradients in bedrock weathering demanded differential glacial erosion or long periods without glaciation, as did thick sedimentary fills spanning the entire Quaternary, resulting in a model of restricted ice at the LGM in the Eastern Canadian Arctic (ECA). One of Andrews' legacies was to embrace new technologies, including early forays into enriched and AMS 14C, U-Series, Amino Acid Racemization, and more recently Cosmogenic Exposure (CE) dating.

The past 30 years of Baffin Island research reveal important insights on ice-sheet dynamics and rates of change. The contrast in rates of glacial processes operating in fiords compared to the adjacent terrestrial realm is striking. In fiords and narrow valleys basal sheer stresses and surface gradients were low, resulting in low-gradient, and probably fast-moving outlet glaciers, separated by relatively slow-moving, polar ice on the intervening crystalline uplands. Many outlet glaciers reached the shelf edge at the LGM. However, some coastal uplands remained above the limit of glacial erosion; lakes in these regions contain undisturbed sediments extending back to the last interglacial, in support of the controversial proposal of the late 1960s for limited LGM ice in the ECA. Recent CE dating shows that non-erosive cold-based ice covered many coastal sediment fills at the LGM without leaving any trace beyond a few scattered erratics.

Andrews' work on Labrador Sea sediments showed that many Heinrich events originated as partial collapses of the Laurentide Ice Sheet through the Hudson Strait and Cumberland Sound conduits. These events were unexpectedly rapid in both their advance and retreat cycles, often lasting only a few decades to centuries. Some advances flowed independent of topography, breaching 1000-m-high barriers with little imprint on regional ice flow patterns, which can only be explained by a near decoupling of the ice from its bed. Perhaps Andrews’ most enduring legacy has been to populate academia with PhDs who started their careers studying the glacial history of Baffin Island.