Turbidity currents in submarine canyons transport large volumes of sediment and carbon to the deep sea and are known to present a major risk to submarine infrastructure. Understanding the origin, the triggers, the recurrence, and the timing of these events is important for predicting future events and mitigating their impact. Depending on the morphological and latitudinal setting of submarine canyons, different external controls will govern the recurrence of turbidity currents. Here, we assess the recurrence of turbidity currents in shelf-incising submarine canyons off eastern Canada in order to examine the effects of external forcings such as glacier retreat and sea level on the deep-water sedimentary record. We used multibeam bathymetry, sub-bottom profiles, and the analysis of turbidites in sediment cores to infer the triggers of turbidity currents over time and propose a conceptual model for the activity of turbidity currents during glacial retreat. The chronostratigraphy of turbidites shows that turbidity current activity in the glaciated The Gully submarine canyon (eastern Canada) was highest between 24 ka cal BP (LGM) and 17 ka cal BP, with > 100 turbidites per 1,000 yr, when the ice sheet was directly delivering sediment to submarine canyons. As the ice margin retreated, the dominant sediment supply switched to glaciofluvial and then to longshore drift, while RSL remained low. The recurrence of turbidity currents nonetheless decreased drastically to < 10 per 1000 yr during that time, pre-dating the rise in RSL. This timing suggests that the reduction of turbidity-current activity is closely linked to retreating glaciers rather than to sea-level rise, which occurred later. Following the retreat of the ice sheet, sea level rose progressively to drown the shallow banks on the continental shelf, and turbidity currents ceased being active after 13 ka cal BP. In the late Holocene, landslide and concomitant turbidity-current recurrence increased to 1 per 1,000 yrs, with at least four new events recorded in deep water. This study shows that glacial sediment supply and sea level controlled the type of sediment supply to the continental slope, which in turn controlled the triggers of turbidity currents over time and the flushing of sediment to the deep water. By comparing with other glaciated margins, we propose a conceptual model explaining the recurrence of turbidity currents, taking into account RSL change and the position of the ice margin relative to the shelf edge. This conceptual model can help predict turbidity-current activity and offshore geohazards on other ancient and modern glaciated continental margins.