Sampling Methods
Sediment traps (ST) were constructed by hollowing the caps of 1L
HDPE bottles to accommodate rigid 30-cm long tubes of 4.75-cm
diameter. The tubes were secured silicone glue and the caps were
screwed onto acid-washed 1L HDPE bottles. Traps were immobilized
using zip ties on a rope that was attached to an anchor at the
bottom and floating buoy at the top. Upon retrieval the sediment
traps caps were replaced while underwater and bottles were
immediately brought into an anoxic N2-filled
portable glove bag on the boat. The trap contents transferred into
1L acid-washed glass serum bottles that were sealed with a butyl
rubber stopper. The sediment trap samples were stored in the dark at
4C. Inside a 100% N2 glovebox (Vacuum
Atmospheres), the sediment trap material was transferred into
acid-washed and stoppered glass serum bottles that were centrifuged
at 1700 rpm for 10 minutes to concentrate solids. The solids were
then washed three times with anoxic ultrapure water to remove
soluble salts and dried under a vacuum in the glovebox.
A 60-cm frozen core was collected from Brownie Lake on January 12,
2018, at a water depth of 13 m using a freeze corer (Shapiro 1958)
rented from the National Lacustrine Core Facility at the University
of Minnesota. The core was transported on dry ice to the lab, where
the surface of the slab was smoothed using a wood planer and cut
into 2.5 cm-wide vertical strips on a band saw while cooled by dry
ice (Harrison et al. 2015). The strips were shipped to Iowa State
University on dry ice and stored at -80C freezer until further
processing.
Gravity cores (GC) were collected using a National Lakes Assessment
gravity corer (Aquatic Research Instruments). The core was extruded
on the same day as collection, either in the field or after
transport to the lab after sealing with a rubber stopper for
transport. In August 2018, the core was extruded in air. In October
2019 and February 2021, the core was extruded inside a disposable
N2-filled glovebag. Sediment samples were
placed into acid-washed vessels for porewater extraction inside a
disposable N2-filled glovebag using Rhizons
of 2.5 mm diameter and pore size of 0.15 µm (Rhizosphere Research
Products). Porewaters were collected into 20 mL syringes, with
suction created by wedging open the syringe connected to the Rhizon.
In August 2018 porewaters were analyzed immediately or preserved for
later analysis. In February 2021 the wet sediments were stored at a
-80C until extraction inside a 100% N2
glovebox. The remaining sediments were washed three times with
ultrapure water to remove soluble salts and dried under a vacuum in
the glovebox.
Analytical Methods
Lake water and porewaters of intact gravity cores were analyzed
using Hg-plated gold microelectrodes and a DLK-70 potentiostat (AIS;
Analytical Instrument Systems). One core was collected from beneath
the epilimnion and the other from beneath the monimolimnion. Lake
water samples were pumped to a boat from target depths using a
groundwater pump (Lambrecht et al. 2018) and into a homemade
flow-through cell with no headspace that was constant flushed at a
low flow rate during analysis. Porewater profiles were collected
using an AIS micromanipulator.
Scans were collected optimized for analytes in the water samples: 1)
with oxygen (O2), cathodic scan at 200 mV/S, 2) without hydrogen
sulfide, cathodic scan at 200 mV/s, conditioning at -0.1 V for 10 s;
3) with hydrogen sulfide, cathodic scan at 200 mV/s, conditioning at
-0.9 V for 10 s conditioning at -0.1 V for 10 s; 4) abundant
hydrogen sulfide anodic square wave (ASW) from -1.5 to -0.1 V and a
step height of 50 mV with no conditioning.
Porewaters from the August 2018 core were chemically fixed or
analyzed immediately after extraction (Xiong et al. 2019). pH was
measured with a calibrated Mettler Toledo AG 8603. Precision is 0.01
(QC1). The ferrozine reagent was used bind and quantify dissolved
ferrous iron (Fe2+) (Stookey 1970).
Dissolved sulfide was determined using the Cline assay after fixing
with 10 mM zinc acetate (Cline 1969). Dissolved phosphate was
measured with the molybdate blue method (Murphy and Riley 1962).
Porewater samples for DIC and δ13C-DIC
were injected into evacuated, He-flushed exetainers containing 1 mL
of concentrated phosphoric acid. Samples were analyzed at the
University of California-Davis Stable Isotope Facility and are
reported as δ13C in permil (‰) relative
to Vienna PeeDee Belemnite (VPDB) with a long-term standard
deviation of 0.1 ‰ (QC1).
Porewater measurements of the February 2021 gravity core were made
inside a 100% N2 glovebox. pH was measured
within 30 minutes of extraction using a Mettler Toledo SevenCompact
pH meter calibrated with standards of pH 4, 7, and 10. Precision is
0.01 (QC1).
Total alkalinity as CaCO3 was determined e
using Hach TNT Total Alkalinity test kits and a Hach DR1900
spectrophotometer within 1 hour of extraction. The detection range
was 25 to 400 mg/L CaCO3 (QC1 and QC2). If
samples were above the detection range they were and measured again,
with the result from the diluted sample reported.
Particle size analysis of gravity cored sediments was determined
with a Malvern Master Sizer 3000 coupled with a Hydro MV dispersion
unit for controlled dispersion of the sediment grains. Samples of
~0.045 g were optimal for the required 3-5% obscuration level on the
instrument. Grain sizes were determined using Gradistat version 8.0.
Grain size QC1 indicate the minimum size in microns of particles in
the size bin. The notes indicate the classification of the sample.
Elemental analysis was performed on sediments fumigated with
hydrochloric acid using a Costech Elemental Analyzer at the Large
Lakes Observatory, University of Minnesota, Duluth using a BBOT
standard.
Elemental abundances were determined by X-ray Fluorescence (XRF) at
Minnesota State University, Mankato on a Rigaku Supermini 200 under
100% He with Pd anode X-rays at 50 kV and 4.0 mA. Samples were
freeze dried and homogenized and powdered by mortar and pestle.
Powdered samples of 0.5-3 g and placed in a polyethylene cup with a
transparent Etnom film on the bottom. Elemental abundances produced
from XRF were normalized as anhydrous oxides. Precision was
calculated by the long-term standard deviation of replicate analyses
of the standards and duplicates, and accuracy was determined by
comparison of the standard values with United States Geological
Survey (USGS) values for the same standards. Accuracy is reported as
QC1 and used as the detection limit. Precision is reported as QC2.
Acid volatile sulfur (AVS) and chromium reducible sulfur (CRS) were
sequentially extracted from the February 2021 gravity core using a
modified method from (Canfield et al. 1986; Fossing and Jørgensen
1989). An average 4 g of homogenized samples at 2 cm resolution from
the top 30 cm of the sediment gravity core collected in February
2021 were used for the analysis. The samples were dried in the
vacuum chamber of an anoxic glovebox (100%
N2) at room temperature. AVS in the samples
were converted to H2S gas by reacting the
samples with 10 ml ethanol and 25 ml of 6M HCl. The evolved
H2S gas was carried via
N2 to an
AgNO3-NH4OH trap
solution, where it was quantitatively converted to silver sulfide
(Ag2S). The residual samples were reacted
with 25 ml of chromium chloride (1 M
CrCl3.6H20 in 0.5 M
HCl) solution for CRS extraction and were also converted to
Ag2S using the same steps, using a fresh
batch of AgNO3-NH4OH trap solution. The
Ag2S precipitates were recovered on a 0.45 μm
polycarbonate membrane filter using vacuum filtration before being
dried at 60°C and weighed to determine the AVS and CRS weight
percent using the stoichiometries of FeS and
FeS2, respectively.
δ34SAVS and
δ34SCRS values
were measured on the AVS- and CRS-extracted material from the
February 2021 gravity core using a Thermo Delta Plus IRMS connected
to an Isolink EA via a Conflo IV interface in the Earth System
Evolution Lab, Iowa State University. The
δ34S values are reported with reference
to Vienna-Canyon Diablo Troilite (VCDT). Precision of
δ34S measurements was 0.5‰ and is
reported as QC1.
Bulk X-ray absorption near-edge structure (XANES) spectroscopy at
the S, Mn, and Fe k-edges was performed at beamline 9-BM at the
Advanced Photon Source (APS) in a helium-filled sample chamber.
Analyses were performed in July and October 2020 on sediment trap
material and gravity core material from October 2019. Samples were
analyzed using a Vortex four-element silicon drift detector. For
each sample, 15 to 20 spectra were collected on the S k-edge, 3-4
scans on the Fe k-edge, and 5-10 scans on the Mn k-edge.
Samples of the January 2018 freeze core, October 2019 gravity core,
and 2019 sediment traps were mapped using XRF at beamline 13-IDE at
2500 eV, 4500 eV and 7200 eV with a frequency of 1-2 MHz and a 50 µm
aluminum filter. Analyses were performed in August 2019 and October
2020. Samples were stored, transported, and handled in a nitrogen
atmosphere and analyzed in a He-purged bag. Multi-energy maps of
<1x1 mm were collected at energies 2469, 2470, 2471 and 2472 eV
and used to visualize the distribution of sulfur species. Spots of
overlap between Fe and S were chosen for microscale XANES in
fluorescence mode at the S, Mn, and Fe K-edges using a Canberra
SXD-7 7-element silicon drift detector.
