The pulse of a montane ecosystem: coupling between daily cycles in solar flux, snowmelt, transpiration, groundwater, and streamflow at Sagehen Creek and Independence Creek, Sierra Nevada, USA

Abstract

Abstract. Water levels in streams and aquifers often exhibit daily cycles during rainless periods, reflecting daytime extraction of shallow groundwater by evapotranspiration (ET) and, during snowmelt, daytime additions of meltwater. These cycles can aid in understanding the mechanisms that couple solar forcing of ET and snowmelt to changes in streamflow. Here we analyze 3 years of 30 min solar flux, sap flow, stream stage, and groundwater level measurements at Sagehen Creek and Independence Creek, two snow-dominated headwater catchments in California's Sierra Nevada mountains. Despite their sharply contrasting geological settings (most of the Independence basin is glacially scoured granodiorite, whereas Sagehen is underlain by hundreds of meters of volcanic and volcaniclastic deposits that host an extensive groundwater aquifer), both streams respond similarly to snowmelt and ET forcing. During snow-free summer periods, daily cycles in solar flux are tightly correlated with variations in sap flow, and with the rates of water level rise and fall in streams and riparian aquifers. During these periods, stream stages and riparian groundwater levels decline during the day and rebound at night. These cycles are reversed during snowmelt, with stream stages and riparian groundwater levels rising during the day in response to snowmelt inputs and falling at night as the riparian aquifer drains. Streamflow and groundwater maxima and minima (during snowmelt- and ET-dominated periods, respectively) lag the midday peak in solar flux by several hours. A simple conceptual model explains this lag: streamflows depend on riparian aquifer water levels, which integrate snowmelt inputs and ET losses over time, and thus will be phase-shifted relative to the peaks in snowmelt and evapotranspiration rates. Thus, although the lag between solar forcing and water level cycles is often interpreted as a travel-time lag, our analysis shows that it is mostly a dynamical phase lag, at least in small catchments. Furthermore, although daily cycles in streamflow have often been used to estimate ET fluxes, our simple conceptual model demonstrates that this is infeasible unless the response time of the riparian aquifer can be determined. As the snowmelt season progresses, snowmelt forcing of groundwater and streamflow weakens and evapotranspiration forcing strengthens. The relative dominance of snowmelt vs. ET can be quantified by the diel cycle index, which measures the correlation between the solar flux and the rate of rise or fall in streamflow or groundwater. When the snowpack melts out at an individual location, the local groundwater shifts abruptly from snowmelt-dominated cycles to ET-dominated cycles. Melt-out and the corresponding shift in the diel cycle index occur earlier at lower altitudes and on south-facing slopes, and streamflow integrates these transitions over the drainage network. Thus the diel cycle index in streamflow shifts gradually, beginning when the snowpack melts out near the gauging station and ending, months later, when the snowpack melts out at the top of the basin and the entire drainage network becomes dominated by ET cycles. During this long transition, snowmelt signals generated in the upper basin are gradually overprinted by ET signals generated lower down in the basin. The gradual springtime transition in the diel cycle index is mirrored in sequences of Landsat images showing the springtime retreat of the snowpack to higher elevations and the corresponding advance of photosynthetic activity across the basin. Trends in the catchment-averaged MODIS enhanced vegetation index (EVI2) also correlate closely with the late springtime shift from snowmelt to ET cycles and with the autumn shift back toward snowmelt cycles. Seasonal changes in streamflow cycles therefore reflect catchment-scale shifts in snowpack and vegetation activity that can be seen from Earth orbit. The data and analyses presented here illustrate how streams can act as mirrors of the landscape, integrating physical and ecohydrological signals across their contributing drainage networks.

Diel streamflow cycles suggest more sensitive snowmelt-driven streamflow to climate change than land surface modeling

“…We tried several correlation cutoffs (r>0.5, 0.6, 0.7, 0.8 and 0.9; see Figure 1 for r>0.6) to assess their effects on the detection algorithm (Figure A2). The preliminary lag window of 6 to 18 hours was used to avoid confounding snowmelt signals with evapotranspiration (ET)-induced streamflow diel responses (Kirchner et al, 2020;Mutzner et al, 2015;Woelber et al, 2018). ET-induced streamflow diel response can positively correlate with solar radiation with lags below 6 hours due to the previous day's ET, and above 18 hours due to the next day's ET diurnal signal (Kirchner et al, 2020).…”

“…Few simple, low-cost observational tools exist to separate rainfall-driven from snowmelt-driven contributions to streamflow or to separate this year's melt from previous years' melt and storage. One method that can be straightforwardly applied to existing long-term observations is based on coupled diel cycles in solar radiation, snowmelt, and streamflow (Kirchner et al, 2020;Lundquist and Cayan, 2002). Coupled diel cycles have been used to study kinematic wave celerity (Kirchner et al, 2020), the impact of snowpack variability on streamflow timing (Lundquist and Dettinger, 2005), groundwater fluctuations (Loheide and Lundquist, 2009), and transitions from snowmelt to evapotranspiration-dominated streamflow fluctuations (Kirchner et al, 2020;Mutzner et al, 2015;Woelber et al, 2018).…”

“…One method that can be straightforwardly applied to existing long-term observations is based on coupled diel cycles in solar radiation, snowmelt, and streamflow (Kirchner et al, 2020;Lundquist and Cayan, 2002). Coupled diel cycles have been used to study kinematic wave celerity (Kirchner et al, 2020), the impact of snowpack variability on streamflow timing (Lundquist and Dettinger, 2005), groundwater fluctuations (Loheide and Lundquist, 2009), and transitions from snowmelt to evapotranspiration-dominated streamflow fluctuations (Kirchner et al, 2020;Mutzner et al, 2015;Woelber et al, 2018). More recently, Kirchner et al (2020) combined local observations and remote sensing to show that streamflow diel response was tightly controlled by the timing of snowpack disappearance.…”

“…Coupled diel cycles have been used to study kinematic wave celerity (Kirchner et al, 2020), the impact of snowpack variability on streamflow timing (Lundquist and Dettinger, 2005), groundwater fluctuations (Loheide and Lundquist, 2009), and transitions from snowmelt to evapotranspiration-dominated streamflow fluctuations (Kirchner et al, 2020;Mutzner et al, 2015;Woelber et al, 2018). More recently, Kirchner et al (2020) combined local observations and remote sensing to show that streamflow diel response was tightly controlled by the timing of snowpack disappearance. Here, we extend the 'diel cycle index' approach of Kirchner et al (2020) using diel streamflow observations to detect the occurrence of days when streamflow is coupled to snowmelt inputs, and investigate their contributions to historical variability in streamflow amount and timing.…”

“…More recently, Kirchner et al (2020) combined local observations and remote sensing to show that streamflow diel response was tightly controlled by the timing of snowpack disappearance. Here, we extend the 'diel cycle index' approach of Kirchner et al (2020) using diel streamflow observations to detect the occurrence of days when streamflow is coupled to snowmelt inputs, and investigate their contributions to historical variability in streamflow amount and timing. We compare STS end-of-century predictions under an RCP8.…”