Figure 1: Digital elevation model, location of study
watersheds and snow measurement sites. Elevation was calculated at 0.5 m
resolution from a LIDAR flight in October,
2014.
Figure 2: Baseflow separation example for GOLD100 (a) and
NONM100 (b). Baseflow (red) is separated from total streamflow (black)
using Eq. 2. Direct runoff (blue) is calculated by subtracting baseflow
from total
streamflow.
Figure 3: Quickflow hydrograph separation example for
LIBB400 during the 2018 snowmelt period. Quickflow discharge (b) is
subtracted from total streamflow discharge (a) based on natural
occurring snowmelt-induced diurnal cycles. Inflection points (a; blue
circles) used to extract quickflow were identified with a computer code
when instantaneous streamflow curvature was at a daily maximum.
Figure 4: Three-component hydrograph separation for LIBB400 during 2018.
Green represents baseflow separated from the total hydrograph using a
mass-balance method. Blue and red represent two components of direct
runoff: throughflow and quickflow, respectively.
Figure 5: Normalized discharge (black) and specific conductance (blue)
for GOLD100 (a) and NONM100 (b) in 2017. Specific conductance vs.
streamflow for GOLD100 (c) and NONM100 (d) in 2017. Color denotes the
day of the water year.
Figure 6: Time series of streamflow (blue) and specific conductance
(red) for NFLL200 during peak flow conditions (a). Dashed lines show the
timing of daily peak streamflow and how they relate to the timing of
specific conductance. The solid lines show how specific conductance is
typically lower on the falling limb of diurnal streamflow cycles
compared to the same recorded streamflow value on the rising limb. Daily
normalized SC – Q plots during the rising limb (b) and falling limb (c)
along with the daily hysteresis index (HI25). Colors denote hour past
noon for respective dates with arrows denoting the direction of
hysteresis.