With warming, spring streamflow peaks are more coupled with vegetation green-up than snowmelt in the northeastern United States

In forested watersheds, climate change may have substantial implications for both ecological and hydrological processes by altering the snow and vegetation seasonal dynamics. However, to date, there have been few studies which systematically analysed their combined effects on seasonal streamflow behaviour at the watershed scale. Using long-term remote-sensed vegetation data and snowpack measurements, we characterized the long-term patterns in the timings of green-up/senescence and subsequent growing season length, as well as snowpack development/disappearance and subsequent snowpack duration. We further investigated the correlations of phenological and snowpack seasonal variations with the cumulative percentiles of daily streamflow over two transition periods: vernal (between snowmelt and green-up) and autumnal (between senescence and snowpack development) windows. We found growing season significantly lengthened by similar to 30 days over the last four decades at the study watershed, driven by both earlier green-up and later senescence. The long-term snowfall significantly decreased by similar to 2.4 cm, and the snowpack duration has been shortened by similar to 25 days, mostly driven by earlier snow disappearance. While there were no significant changes in the lengths of vernal and autumnal windows, the normalized streamflow distributions over the vernal window have widened and the low-frequency peak flow over the autumnal window has shifted earlier over the study period. Interestingly, although the snow disappearance timing has advanced by similar to 18 days, there was no evidence of changes in the timing of spring peak discharge. Our results suggest that with long-term decreases in winter snowfall and snowpack duration, the impacts of snowpack dynamics on seasonal streamflow regimes over the vernal and autumnal windows have weakened. Meanwhile, the seasonal streamflow dynamics are more closely mediated by vegetation green-up with time. This study emphasizes the importance of understanding the phenological responses to climate change for prediction of future seasonal flow regimes in forested watersheds.