Metadata

Description
In each plot, within a 1 × 1m sub-plot we counted the number of live and dead plants of each species, and the height and diameter of n = 15 random sawgrass culms in order to estimate aboveground biomass using previously derived allometric relationships between culm height, diameter and biomass.
From each plot, sawgrass plants (including roots) were collected before and after fire at different times; all samples were conserved at 4°C until processing. The biomass was divided into shoots (aboveground), and coarse and fine (< 2mm) roots (belowground) and oven-dried at 40°C for 72 h. Afterwards, sub-samples were ground using an 8000D ball mill (SPEX SamplePrep, New Jersey, United States) and analyzed for C, N and P concentrations. For CN samples were ran through a CE Flash 1112 elemental analyzer (CE Elantech, New Jersey, USA) using standard procedures. For P samples underwent a dry-oxidation acid hydrolysis extraction followed by a colorimetric analysis of phosphate using a UV-2450 spectrophotometer (Shimadzu, Kyoto, Japan).
Every three months, using a digital water quality meter probe (YSI, Ohio, USA), we measured in each plot water temperature, pH, conductivity and dissolved oxygen; water depth was recorded each time as well.
From each plot, two (filtered and unfiltered) surface water samples were collected before and after fire at different times. All samples were collected into high‐density polyethylene (HDPE) bottles, and one sample at each plot was filtered through a 0.45‐μm nylon membrane filter (Whatman, Inc.). Unfiltered samples were conserved at 4 °C until analysis, while the filtered samples were frozen. The unfiltered samples were analyzed for total organic C (TOC), and total N (TN) and P (TP), while the filtered samples were analyzed for dissolved organic C (DOC), total inorganic N (TIN) and soluble reactive phosphorus (SRP). Total/dissolved organic C was determined by introducing the sample into a combustion tube and oxidizing it to form CO2, which was detected by a Non Dispersive InfraRed (NDIR) detector. TN was determined through conversion to nitric oxide, reaction with ozone, and detection of the chemiluminescent emission by a photomultiplier tube. TP was determined by oxidizing and hydrolyzing all of the phosphorus-containing compounds to SRP. Analysis for inorganic filtered nutrients, ammonia/ammonium as N (NH3/NH4-N), nitrite as N (NO2-N), nitrate and nitrite as N (N+N), and soluble reactive phosphorus as P (SRP), were simultaneously performed by wet chemical analysis using a four-channel Rapid Flow Analyzer based on standard EPA procedures.
From each plot, one periphyton and one floc (i.e., flocculent detrital organic matter) samples were collected before and after fire at different times. Periphyton was grabbed floating from the water surface or, when the system was dry, from the soil surface. Floc was collected in the water near the soil surface using a baster. Periphyton samples were frozen until analysis, while floc samples were conserved at 4°C. Both periphyton (after cleaning it from dead plant material and rocks) and floc were oven dried at 40°C for several days until their high moisture content completely evaporated. The samples were then ground and analyzed for total C, N and P concentrations.
In each plot we measured soil depth by sticking a metal rod down in the soil until it reached the underlying limestone.
From each plot, two soil cores (54-mm diameter) were collected before and after fire at different times; all samples were conserved at 4°C until analysis. At the time of sampling, soil cores depth was recorded in order to derive the volume needed for bulk density calculation. After collection, the fresh weight of soil sub-samples was obtained, then the sub-samples were over-dried at 40°C for 120 h before dry weight measurement. Soil bulk density was calculated as the ratio between the estimated dry weight of the sample and its volume. The dried sub-samples were then ground and analyzed for C, N, and P concentrations.
Air-dried sawgrass root and leaf material collected at the LP and HP sites was used to study decomposition rates in the soil. The material, divided into litter (leaf), coarse roots and fine roots was placed into 1-mm nylon mesh bags; each bag contained ~3 (leaf litter), ~5 (coarse roots), or ~1 (fine roots) g of dry matter. Before deployment, sub-samples were placed in the oven at 60°C for 72 h to estimate residual moisture content. Root decomposition bags were placed in the soil, at a depth between surface and 20cm, inside the holes dug during soil cores collection. Leaf litter bags were placed on the soil surface, homogenized with the rest of the litter. Each bag was tied with fishing line to a marker stick and to a weight that prevented it from floating. All decomposition bags were incubated in the soil for 1 year before the burn treatment (January 2019 to December 2020), then new bags were placed after the burn for another year (March 2020 to March 2021). After each 1-year period of incubation, the bags were retrieved from the soil, and the material, after being rinsed over a sieve, was oven-dried at 60°C for 72 h before weighing. The difference in mass between deployment date and retrieval date allowed us to calculate decomposition rates. Decomposition rates, expressed as k, were estimated using a linear regression of the ln‐transformed fraction of dry mass remaining versus time. The model is shown in the following equation:
$$M_{t} = M_{0} \times e^{-kt}$$
Where M0 is the initial mass, Mt is the mass on a given sampling day, and t is the time of incubation in days.