Stand-replacing windthrow in the boreal forests of eastern Quebec

The frequency of stand-replacing windthrow has rarely been evaluated over large spatio-temporal scales in boreal forests. This study examines stand-replacing windthrow that occurred within a forested area of 3 061 000 ha, both at the regional scale and at the scale of individual windthrow events. We specifically examined windthrow that covered at least 5 ha and for which mortality affected >75% of pre-windthrow canopy cover. At the regional scale, we found that windthrow affected, on average, 0.0255% of the area per year between 1971 and 2000. Logistic regressions indicate that stand age was the most important predictor variable at this scale, with severe windthrow less likely to occur in relatively young stands (<75 years). At the scale of individual windthrow events, slope direction was the most important predictor variable, with slopes facing the wind affected more. However, this factor was not significant at the regional scale probably because wind direction varied from one windthrow event to another. Overall, our results suggest that stand-replacing windthrow is difficult to predict and integrate into regional-scale forest management scenarios because it is primarily associated with the random occurrence of exceptional windstorms, while other biotic and abiotic variables play a relatively marginal role.