Water samples from various sources were collected biweekly throughout
the year during October 2006 to October 2010. Samples of
precipitation, snowmelt, soil water, and stream water were collected
over the entire four-year period. Throughfall and snow pack sampling
was discontinued after two years, and groundwater sampling began in
the second year and continued through the end of the monitoring
period. Precipitation samples were collected in an existing rain gage
clearing at an elevation of 564 m. Precipitation collectors consisted
of 15 cm diameter by 50 cm long PVC pipes lined with polyethylene
collection bags. The collectors were mounted on vertical stakes 1.5
meters above the ground surface. When precipitation fell as rain,
mineral oil was added to the bags to minimize evaporation. To prevent
contamination by the mineral oil, samples were obtained by cutting a
hole in the bottom of the bag, allowing water to drain into the sample
vial. When precipitation fell as snow, samples were melted at room
temperature in a closed plastic bag after collection, and then
immediately poured into sample vials. Throughfall sample collection
was identical to precipitation, with the exception that a composite
sample was obtained from 6 collectors randomly located under the
forest canopy. The greater number of throughfall collectors was needed
to better homogenize the spatial variability caused by the forest
canopy.
Snowpack samples were collected by coring the entire snowpack with a
bevelled, PVC tube. Snow cores were placed in plastic bags, and as
with snow throughfall and precipitation, were melted at room
temperature before decanting into sample vials. Samples of water
draining from the bottom of the snowpack and soil were collected with
lysimeters (1.064 m2) installed near the
rain gage clearing. As described by Campbell et al. (2007), the
lysimeters consisted of heavy-duty (6 mm) PVC trays that drain by
gravity through a PVC pipe to an underground storage container. The
bottom of the lysimeter is impermeable and no roots bridged the soil
in the lysimeter and the surrounding soil, thus the lysimeter soil was
not directly hydrologically connected to the surrounding soil or the
groundwater. The water storage container was insulated to prevent the
drainage water from freezing. Soil lysimeters were installed in the
soil at a depth of 10 cm, whereas snow lysimeters were placed directly
on the surface of the forest floor. Both soil water and snowpack
meltwater samples consisted of a composite sample from three
lysimeters.
Groundwater was collected from two wells (Wells 1 and 27) in W3 that
are part of a network described by Detty and McGuire (2010). The wells
consist of 3cm diameter PVC pipe, with a 30 cm slotted screen at the
base, and were installed to a depth of about 10 cm into C horizon of
the soil (Well 1 is 88 cm and well 27 is 71 cm deep). Groundwater was
collected from each well and passed through a 0.45 μm nylon membrane
to remove sediment. Well 27 is located directly upslope from the
stream at an elevation of 565 meters whereas Well 1 is adjacent to the
stream at an elevation of 535 meters. Stream water samples were
collected at the W3 outlet, just above the weir, about 50m downstream
from Well 1.
All water samples were stored in 20 mL glass vials that were
completely filled with sample water and sealed with caps that
contained plastic conical inserts to remove headspace and prevent
evaporation. The caps of the vials were then dipped in paraffin wax
and placed in the dark at room temperature until analysis.
Isotopic results are reported in the standard δ notation in parts per
thousand (‰) relative to Vienna Standard Mean Ocean Water (VSMOW):
δ (‰) = (RSAMPLE -
RVSMOW)/ RVSMOW x 1000
where R is the
18O/16O or D/H
ratio of sample water or VSMOW. Oxygen isotopes of water were
initially determined using the
CO2-H2O equilibration
method (precision of 0.1‰) and hydrogen isotopes with the zinc
reduction method (precision of 0.4‰) using a mass spectrometer
following Coleman et al. (1982). After November 2008, the isotopic
composition of water samples was measured using cavity ring-down laser
spectroscopy as described by Lis et al. (2008) with an analytical
precision of 0.1‰ for oxygen isotopes and 0.8‰ for hydrogen isotopes.
A subset of 5 samples was run using both methods and showed a median
difference of -3% and 5% for D and 18O
(-1.9 and 0.7‰) respectively. The analytical uncertainty of
d, calculated as, d =
√u(D)2
+
u(18
O)2,
was 0.41‰ for the initial method and 0.81‰ for the cavity ring-down
method.
Campbell JL, Mitchell MJ, Mayer B, Groffman P, Christenson L. 2007.
Mobility of nitrogen-15-labeled nitrate and sulfur-34-labeled sulfate
during snowmelt. Soil Science Society of America
Journal 71: 1934–1944.
Coleman ML, Shepherd TJ, Durham JJ, Rouse JE, Moore GR. 1982.
Reduction of water with zinc for hydrogen isotope analysis.
Analytical Chemistry (Washington) 54: 993–995.
Detty JM, McGuire KJ. 2010. Topographic controls on shallow
groundwater dynamics: implications of hydrologic connectivity between
hillslopes and riparian zones in a till mantled catchment.
Hydrological Processes 24: 2222–2236.
Lis G, Wassenaar LI, Hendry MJ. 2008. High-precision laser
spectroscopy D/H and 18O/16O measurements of microliter natural water
samples. Analytical Chemistry 80: 287–293.
