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

Abstract

Abstract. Climate warming may cause mountain snowpacks to melt earlier, reducing summer streamflow and threatening water supplies and ecosystems. Few observations allow separating rain and snowmelt contributions to streamflow, so physically based models are needed for hydrological predictions and analyses. We develop an observational technique for detecting streamflow responses to snowmelt using incoming solar radiation and diel (daily) cycles of streamflow. We measure the 20th percentile of snowmelt days (DOS20), across 31 watersheds in the western US, as a proxy for the beginning of snowmelt-initiated streamflow. Historic DOS20 varies from mid-January to late May, with warmer sites having earlier and more intermittent snowmelt-mediated streamflow. Mean annual DOS20 strongly correlates with the dates of 25 % and 50 % annual streamflow volume (DOQ25 and DOQ50, both R2 = 0.85), suggesting that a one-day earlier DOS20 corresponds with a one-day earlier DOQ25 and 0.7-day earlier DOQ50. Empirical projections of future DOS20 (RCP8.5, late 21st century), using space-for-time substitution, show that DOS20 will occur 11 ± 4 days earlier per 1 °C of warming, and that colder places (mean November–February air temperature, TNDJF <−8 °C) are 70 % more sensitive to climate change on average than warmer places (TNDJF > 0 °C). Moreover, empirical space-for-time based projections of DOQ25 and DOQ50 are about four and two times more sensitive to earlier streamflow than those from NoahMP-WRF. Given the importance of changing streamflow timing for headwater resources, snowmelt detection methods such as DOS20 based on diel streamflow cycles may constrain hydrological models and improve hydrological predictions.

Pulses within pulses: Concentration‐discharge relationships across temporal scales in a snowmelt‐dominated Rocky Mountain catchment

“…These seasonal pulses represent the dominant hydrologic forcing over interannual time scales. However, closer inspection often reveals they are not singular pulses, but composed of numerous individual pulses (Kirchner et al, 2020; Krogh et al, 2021; Pellerin et al, 2012). These ‘pulses within the pulse’ originate from a combination of diel variation in rates of snowmelt (i.e., greater daytime melting) and also rain events which contribute additional water.…”

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

“…A steep stage-discharge relationship, in which small changes in discharge are associated with large changes in stage, is ideal to observe small diel streamflow changes with sufficient precision. The second limitation originates from the assumption that the majority of snowmelt is correlated with solar radiation, which is supported by the dominant role of solar radiation in process-based studies of mar- itime and continental snowpacks (Cline, 1997;Jepsen et al, 2012;Marks and Dozier, 1992). Because our method allows the lag time between solar radiation and streamflow to vary within a predefined window, we expect it to capture the effects of other important energy fluxes, such as sensible heat, that often lag the diel patterns of solar radiation by several hours (Ohmura, 2001).…”

“…Precipitation phase (rainfall versus snowfall) is mediated by basin elevation and hypsometry (Jennings et al, 2018;Wayand et al, 2015), which also influences precipitation amounts (Houze, 2012), with higher elevations and steeper watersheds typically having higher precipitation and snowfall. Solar radiation is the primary energy source for snowmelt in snow-dominated montane watersheds (Cline, 1997;Marks and Dozier, 1992). Conversely, cloudiness lowers solar radiation and melt rates (Sumargo and Cayan, 2018).…”

Near-field variability of evaposublimation in a montane conifer forest

“…Montane forests are a dominant land cover type in the Sierra Nevada range of the western United States (Barbour et al, 1991). Processes that govern snow accumulation and ablation in montane forests are therefore critical for understanding the water cycle, monitoring forest health and forecasting spring runoff (Heckmann et al, 2008;Kirchner et al, 2020;Krogh et al, 2022). Evaposublimation-the combination of evaporation and sublimation-is an important ablation process in a montane environment, where wintertime air temperatures vacillate around freezing.…”