Hydrological and Meteorological Data
Four (2015) and two (2016) field stints of up to three weeks occurred
during each open-channel season (05/18 – 09/17), with hydrological
sampling balanced among other field priorities. Hydrological and
meteorological stations were setup for continuous monitoring at 30 or 60
minute intervals (Figures 1 and 3). At the meteorological stations (854,
1425, and 1750 m asl), data loggers recorded air temperature, ground
temperature (2 cm and 30 cm depth), and rainfall. The meteorological
station at 854 m asl also recorded solar radiation and barometric
pressure. Meteorological instrumentation is described in Broadman et al.
(2019). At the hydrological stations, we used In-Situ TROLL 9500
(“TROLL”) instruments to measure water-pressure and turbidity, and
Hobo Onset U20 (“Hobo”) instruments to measure water pressure.
Additionally, time-lapse cameras photographed the hydrological stations
hourly throughout the open-channel seasons. In Carnivore Creek, and
during the 2015 season in Chamberlin Creek, the TROLL and Hobo
instruments were secured in thalwegs of relatively stable reaches close
to Lake Peters, upstream from where the main channel anabranches (CAR
and CHBa on Figure 1). During the 2016 season in
Chamberlin Creek, the instruments were secured farther upstream at the
apex of the alluvial fan (CHBb on Figure 1). Turbidity
data were used after removing erroneous data due to fouling or
instrument dislodgement, resulting in discontinuous time-series.
Unrecorded or erroneous stage data was infilled using regressions or
photographic reconstructions where available (Broadman et al., 2019;
Thurston, 2017). To generate statistics, unrecorded temperature data was
also infilled using regressions (Appendix I), but only the original
meteorological records were used for sediment modeling.
SSCs in Carnivore and Chamberlin Creeks were sampled manually at the
hydrological stations using a depth-integrated handheld sampler (US
DH-48). The liquid volume of each sample was recorded, prior to
filtering with GN-4 Metricel membrane disc filters (0.8 μm). Filters
were later dried in a laboratory oven and weighed (± 0.01 mg). For
discharge, we used a Hach FH950 portable velocity meter to measure
current velocities and cross-sections near the hydrological stations. In
the 2016 field season the velocity meter was inoperable, so only
cross-section areas were used for calculating discharge (Broadman et
al., 2019). The discharge data was used to construct stage-discharge
rating curves and establish continuous discharge records (Broadman et
al., 2019; Thurston, 2017).