Surface and porewater physicochemistry
Water physicochemical properties were monitored monthly from both
surface water (n = 24) and porewater (n = 24) at 15-cm below surface
sediment. Salinity and temperature were measured with a YSI Model 600
XL (Xylem, Inc., Yellow Springs, OH, USA). Water samples were
collected using a 60-mL syringe and filtered samples were run through
a 0.7-μm glass fiber filters. Water samples were stored at -20°C until
analysis at the Southeast Environmental Research Center Nutrient
Analysis Laboratory at Florida International University. Filtered
samples from porewater and surface water were analyzed for dissolved
organic carbon (DOC), nitrite (NO2), nitrate (NO3), ammonium (NH4),
and, soluble reactive P (SRP). Unfiltered water samples were also
analyzed for total organic carbon (TOC), total nitrogen (TN), and
total P (TP). Inorganic and organic nutrients were analyzed with an
Alpkem RFA 300 auto-analyzer (OI Analytical, College Station, Texas,
USA). Organic carbon was analyzed with a Shimadzu 5000 TOC Analyzer
(Shimadzu Scientific Instruments, Columbia, Maryland, USA).
Organic matter and elemental stoichiometry
We determined aboveground biomass from randomly selected fifteen (or
the number of available) sawgrass culms in each plot for total six
times. We used a non-destructive allometric method derived from
density, the length of the tallest leaf on each culm, and culm
diameter at base (Daoust and Childers 1998).
Live belowground biomass was determined by taking one core (2.4 cm^2 9
30 cm depth) from each monolith at the end of the experiment,
separating the core into 10-cm sections, and storing at 4°C until
processing (within two weeks). The live roots, those that floated in
water and were not visibly dark and dead, were separated by washing
over a 1-mm sieve, dried at 60°C, and weighed.
We measured decomposition rates of sawgrass leaf litter on the surface
sediment beginning 01 April 2017 and incubated for 155 and 317 d (n =
24). We filled a 1 mm-mesh bags with air-dried sawgrass leaf litter
(~2 g) and incubated on surface sediments. We collected samples and
returned them on ice. In the laboratory, samples were rinsed with
deionized water, oven-dried at 60°C until mass stabilized, and
subsequently weighted for individual mass remaining. The decay
constant, k, was calculated over sampling intervals with an
exponential decay equation: y = e^-kt, where y is the fraction of
initial mass remaining at time t. We further assessed C, N, and P
concentrations in sawgrass leaf litter. We ground dried litter using
an 8000-D ball mill (Spex SamplePrep, Metuchen, New Jersey, USA), and
measured C and N (Zimmerman et al. 1997) using a CHN Analyzer (Carlo
Erba 1500, Milan, Italy) and P using the ash/acid extraction method
followed by the ascorbic acid method for spectrophotometric analysis
(Solórzano and Sharp 1980).
Upon the termination of mesocosm experiment (June 2018), samples for
C, N, and P composition were collected from soil cores (n = 24), live
sawgrass leaf (n = 24) and root (n = 24). Soil cores (8 x 7 x 25 cm)
were oven-dried until mass stabilized to determine soil bulk density,
which was calculated as the dry weight divided by core volume. Dried
samples of sawgrass leaf, root, and soil were ground and subsampled
for C, N, and P analysis as described above.
Ecosystem CO2 metabolism
Measurements of ecosystem CO2 exchange were conducted every other
month from December 2015 to January 2017 (eight sampling events) on a
subset (n = 4 per treatment) of monoliths. Prior to each measurement,
a polycarbonate collar (67 cm long X 49 cm wide X 58 cm high) was
inserted between the container holding the monolith and the box
containing the monolith and surrounding water. The collar was fitted
with a foam platform on which the chamber could sit while enclosing
both the plants and soil. The clear polycarbonate chamber (53 cm long
9 38 cm wide 9 150 cm high) was placed onto the foam platform and
sealed using bungee cords to ensure an airtight seal during
measurements. A pump sent air from the chamber to an infrared gas
analyzer (LI-840, LI-COR) and back to the chamber. The chamber was
allowed to equilibrate for 2 min, then CO2 concentration was measured
every second for 3 min in both full light and in the dark (Wilson et
al. 2018a). The flux was calculated as the linear slope of CO2
concentration over time. Net ecosystem productivity (NEP) was measured
in full light, whereas ecosystem respiration of CO2 (ERCO2) was
measured in the dark immediately after light measurements by covering
the chamber with a dark cloth that blocked out all sunlight. Gross
ecosystem productivity (GEP) was calculated from NEP and ERCO2 as GEP
= -NEP- ERCO2 where NEP is instantaneous CO2 flux in light and ERCO2
is the CO2 flux in the dark. Ecosystem flux measurements were not
taken from February to November 2015 because the experimental setup
had not yet been equipped to handle these kinds of measurements.
Ecosystem CO2 metabolism was measured in 16 experimental plots
approximately every four weeks during midday (~10:00 to ~15:00) from
May 2017 to June 2018, representing the 2018 water year. We used
transparent polycarbonate chambers consisting of a bottom collar, the
main chamber (0.2 m^2 x 120 cm), and a top lid (0.2 m^2 x 30 cm). We
inserted the collar into a container holding a monolith and water. We
put the main chamber on the top of the collar where a window foam
helped create airtight sealing between the parts. We used a pair of
brushless fans to homogenize air inside the chamber. We used a micro
diaphragm pump (Parker Hannifin, Hollis, New Hampshire) circulating
gas samples between the chamber and a CO2/H2O analyzer (Li-840A,
LI-COR Biosciences, Lincoln, Nebraska, USA). We closed the lid and
waited until CO2 concentration stabilized (<1 min) before we
measured CO2 concentration every 1s for 3min. We measured net
ecosystem productivity (NEP) under three light levels (100, 57, 30% of
natural photosynthetically available irradiance; PAR) by using a layer
of neutral mesh-screens (57 and 30%). We also measured ecosystem
respiration (ER) under the dark condition by using a double layer of
black plastic bags. Simultaneously, we measured temperature and
incident PAR (LI-190 quantum sensor, LI-1400 data-logger, LI-COR
Biosciences, Lincoln, NE, USA) at the beginning and end of each
measurement. A CO2 flux was calculated as the slope of linear
regression throughout incubation (~3 min) that resulted in the median
R^2 of 0.99 for all sampling events. Gross ecosystem productivity was
estimated by summing NEP and ER.
Soil CO2 and CH4 fluxes
One 10-cm diameter PVC collar was installed 5 cm into the soil of each
peat monolith for soil C efflux measurements. Soil CO2 efflux (n = 6
per treatment) was measured monthly near noon between February 2015
and September 2016 (20 measurements total) on all 24 monoliths using a
portable infrared gas analyzer (LI8100, LI-COR, Lincoln, Nebraska,
USA) equipped with a 10-cm diameter chamber. Each flux measurement was
taken for 120 s. The flux was calculated as the linear slope of CO2
concentration over time.
Soil CH4 efflux was measured monthly between February and November
2015 (seven sampling events) from a subset (n = 4 per treatment) of
monoliths using the LI-8100 modified to collect a subset of air for
trace gas sampling. The chamber was sealed and, immediately following
closure, 25 mL of gas was withdrawn using a 60-mL syringe from a port
in-line with the instrument. After 15 min, another gas sample was
collected. The gas was sealed in a 20-mL evacuated glass vial and
transported back to the lab for analysis. Samples were run within 2 d
of collection on a gas chromatograph (Shimadzu Scientific Instruments
GC 8A, Columbia, Maryland, USA) fitted with a flame ionization
detector (FID). Methane flux was calculated as the slope of CH4
concentration over time. No soil gas flux measurements were taken in
March, April, and December 2015, or January 2016 because of equipment
failure.
Enzyme
We measured the fluorometric activities of extracellular acid
phosphatase, arylsulfatase, β-1,4-glucosidase, and
β-1,4-cellobiosidase from soil sub-samples using the4-MUF-phosphate,
4-MUF-β-D cellobiosidase, 4-MUF-β-D-glucopyranoside, and
2-MUF-sulfate, respectively. Soil microbial enzyme activities were
assayed using previously described methods (Saiya-Cork et al., 2002).
Soil sub-samples were collected (approximately 1 g) from each
sawgrass-peat monolith, homogenized in 60 mL of 50mM sodium acetate
buffer, and loaded onto a 96-well plate with the appropriate
substrate. Fluorescence was read at 365 nm excitation and 450 nm
emission using a Synergy H1 microplate reader (BioTek, Winooski,
Vermont, USA). We incorporated blanks and controls within each
microplate to account for auto fluorescence and quenching. We also
calculated enzyme activities (phosphatase, arylsulfatase,
β-1,4-glucosidase, and β-1,4-cellobiosidase) associated with
decomposing root litter using the same method for soil enzyme
activities.
Dataset:
PeriphytonCores_Chemistry
Ten mL periphyton mat subsamples were collected bimonthly, starting in
April 2015, from each mesocosm using a plastic syringe. In the lab,
the mat samples were placed in a 500-mL beaker and homogenized with
deionized water (DI) water to a periphyton slurry with a minimum
volume of 200-mL. The beaker was placed on a stir plate to facilitate
continuous mixing while the slurry was subsampled for biomass,
chl-
a
, total carbon (TC), total nitrogen (TN),
and total phosphorus (TP) analyses. The biomass and
chl-
subsamples were analyzed using the same
methods described for the settlement plate samples. The TP subsample
was dried at 80 degrees C and pulverized with a mortar and pestle. We
processed a known amount of each subsample and used colorimetric
analysis to estimate TP concentration, expressed as micrograms gram
AFDM, following the methods of Solorzano and Sharp (1980). TC and TN
subsamples were ground down to a fine powder by mortar and pestle,
dried at 60 degrees C, and analyzed for total C and N content using a
Flash 1112 elemental analyzer (CE Elantech, Lakewood, New Jersey, USA)
following standard procedures.
PeriphytonPlates_AccumulationRates
Clear, acrylic settlement plates (approximately 104 square
centimeters) were deployed in each mesocosm to measure periphyton
biomass and chlorophyll-a accumulation rates. The settlement plates
were incubated on the soil surface of the mesocosm for two months to
allow periphyton mats in the mesocosms to colonize the plates. At the
end of the two-month incubation period, settlement plates were
retrieved and transported to the laboratory, where they were scrapped
clean using a razor blade and deionized water. Settlement plates were
sterilized using a bleach solution and redeployed the following month.
This was done quarterly from March 2015 to April 2016 for a total of 4
settlement plate samples per mesocosm over the duration of the
experiment. The material scraped from the settlement plates was
collected in a beaker and homogenized with deionized water (DI) water
to a minimum volume of 200-mL. The beaker was placed on a stir plate
to facilitate continuous mixing while the slurry was subsampled for
biomass, chlorophyll-
, and diatom analysis. The
biomass subsample was placed in a drying oven at 80 degrees C for at
least 48 h to obtain a dry mass (g) measurement and then in a furnace
where it was combusted at 500 degrees C for 1 h to obtain the ash
(mineral) mass (g). We calculated ash-free dry mass (AFDM) by the
loss-on-ignition method as the difference between the ash mass and the
total dry mass. The chl-
subsample was filtered
onto Whatman 25-mm diameter glass fiber filter (GFF) paper, frozen,
and later subjected to 48 h 90 percent cold acetone extraction at 20
degrees C. Chl-
concentrations were determined
using a Gildford FLUORO IV fluorometer (Gilford Instrument, Oberlin,
Ohio, USA; excitation 435 nm; emission 667 nm). AFDM and
chl-
accumulation rates
(mgPerMeterSquaredPerDay) were calculated by standardizing AFDM and
chl-
values by the area of the settlement plates
and the number of days incubated.
PeriphytonPlates_DiatomRelAbund
A subsample of the periphyton material scrapped from the settlement
plates was used to analyze diatom species composition. Diatom
subsamples were cleaned of mineral debris and organic matter using the
sulfuric acid oxidation methods prescribed by Hasle and Fryxell
(1970). We pipetted a known volume of cleaned diatom sample onto a
glass coverslip, permanently mounted it on a microscope slide with
Naphrax (PhycoTech Inc., St. Joseph, Michigan, USA) mounting medium,
and viewed the diatoms under a compound light microscope (Axioskop 2;
Zeiss, Thornwood, New York) equipped with differential interference
contrast and a digital camera (Leica DFC425, Wetzlar, Germany). We
counted at least 500 diatom valves along random transects at 600×
magnification under oil immersion and identified them to the species
level. Raw diatom counts were converted to relative abundance by
standardizing the number of valves counted for each species by the
total number of valves counted. We chose to analyze periphyton diatom
species composition from the settlement plates rather than samples
taken from the mesocosm soil surface because, as diatoms are counted
and identified from cleaned (oxidized) samples, testing treatment
effects on the diatom assemblage using established mats would have
been confounded by the presence of dead diatom frustules; using
settlement plates allowed us to better pick up diatom community
response to treatments over time.
