The Counteracting Effects of Snowmelt Rate and Timing on Runoff

The declining mountain snowpack is expected to melt earlier and more slowly with climate warming. Previous work indicates that lower snowmelt rates are associated with decreased runoff. However, earlier snowmelt could increase runoff via lower vegetation water use in early spring. The relative importance of these factors with regard to runoff is linked to site-specific conditions such as plant available water storage (PAWS) and energy availability. To disentangle the effects of snowmelt rate and timing on runoff production, we conducted a hydrologic modeling experiment at sites in Colorado (NR1) and California (P301) that controlled for snowmelt rate and timing multicollinearity. We tested the sensitivity of snowmelt season potential runoff (R), changes in subsurface storage (Delta S), and other water budget components to snowmelt rate (sm(r)) and timing (sm(t)) using multiple linear regression and global sensitivity analysis (GSA). Regression results confirmed that R was governed by the competing influence of sm(r) and sm(t). At both sites, Delta S was more sensitive to sm(t) than sm(r) while R was more sensitive to sm(r) at P301 and to sm(t) at NR1, reflecting energy limitation at NR1. GSA analyses mirrored the regressions for R, confirming that sm(t) was more important at NR1 than P301. This work suggests that runoff increases from earlier snowmelt may counteract runoff losses due to slower snowmelt and that this process is mediated by PAWS and energy availability. These results suggest that R will be more susceptible to future changes in sm(r) and sm(t) at sites with greater PAWS and available energy. Key Points Potential runoff was most sensitive to snowmelt rate in California, USA, and snowmelt timing in Colorado, USA Energy availability and plant available water storage controlled intersite differences Decreased runoff due to slower snowmelt was counterbalanced by increases due to earlier snowmelt timing