Table of Contents

• Coverage
• Dataset Description
  • Methods & Sampling
  • Data Processing Description
• Data Files
• Related Publications
• Related Datasets
• Parameters
• Instruments
• Project Information
• Funding

Nine coastal wetland soil cores were collected in June 2018 from Barataria Bay, Louisiana, a shallow open water basin located west of the Mississippi River Delta.  Soil cores were collected along three transects, roughly 1 meter apart, that consisted of three points:  the coastal fringe (0 m inland), 1 meter inland, and 2 meters inland. Soil cores were collected in polycarbonate tubes via the push core method to a depth of 150 cm, and field-extruded into 15 separate 10-cm intervals. Soils were stored  in polyethylene bags on ice and immediately transported back to the laboratory, where they were kept at 4 °C until sample analysis was complete. 

This dataset includes analyses of biogeochemical properties, organic matter fractionation, and stable isotope signatures. 

Moisture Content and Bulk Density: A subsample of soil was dried using a gravimetric oven at 70°C for 3 days or until a constant weight was achieved. Dried soils were ground using a SPEX Sample Prep 8000M Mixer/Mill (Metuchen, NJ). 

Total Carbon and Nitrogen content:  %C and %N were  determined using a Vario Micro Cube CHNS Analyzer on dried, ground subsamples.

Organic Matter Content: Dried, ground sub- samples were used to determine percent organic matter using the loss- on-ignition method, where soils were burned at 550°C in a muffle furnace for a total of 3 hours

Extractable Nitrate, Extractable Ammonium, Extractable Soluble Reactive Phosphorus:  2.5 grams of wet soil (both from the field and from the bottle incubation) were placed into 40 mL centrifuge tubes and 25 mL of 2 M KCl was added. Samples were then shaken continuously on an orbital shaker for 1 hour at 25 °C and 150 rpm, then centrifuged for 10 minutes at 10 °C and 5000 rpm.  Following the centrifuge step, samples were immediately filtered through Supor 0.45 μM filters and acidified with double distilled H2SO4 to a pH of < 2 for preservation. Samples of Extractable nutrients  were then analyzed using an AQ2 Automated Discrete Analyzer (Seal Analytical, Mequon, WI, EPA methods 231-A Rev.0, 210-A Rev.1, and 204-A Rev.0).

Carbon and Nitrogen isotopes: Stable isotope analysis was performed at the Stable Isotope Mass Spectroscopy Laboratory, Department of Geological Sciences at the University of Florida.  Dried, ground subsamples from only the 1 meter inland cores were initially combusted on a Carlo Erba NA1500 CNS elemental analyzer.  Following the removal of oxygen and water from the sample gas, the stream was passed through a 0.7 m GC column (120 °C), which separated the N2 gas from CO2.  Effluent then passed into a ConFlo II system, and into a Thermo Electron Delta V Advantage isotope ratio mass spectrometer, where sample gas was measured in relation to laboratory reference gases.  Carbon isotope results are expressed in relation to Vienna PDB, in standard delta notation.

Cellulose, Hemicellulose, and Refractory Carbon: Dried, ground subsamples were subjected to sequential extraction with H2SO4, following Rovira and Vallejo (2002) and Oades et al. (1970), with modifications.  The first fraction is referred to hereafter as Labile Pool 1 (LP1) and consists of either plant- or microbially-derived non-cellulosic polysaccharides, including hemicellulose.  Labile Pool 1 was extracted by adding 20 mL of 5 N H2SO4 into a 50 mL flask containing 0.5 grams of soil.  The solution was heated for 30 minutes at 105 °C and subsequently allowed to cool.  Samples were then filtered through Whatman #41 filters to separate particulates from the solution, and then diluted to a final volume of 50 mL.  Labile Pool 2 (LP2), which consists of cellulose, was determined by adding 2 mL of 26 N H2SO4 to 0.5 grams of dried, ground soil.  Samples were shaken at 100 rpm for 16 hours, then diluted to a final concentration of 2 N H2SO4 with deionized water.  Samples were heated for 3 hours at 105°C, then filtered in the same manner as LP1.  Determinations of LP1 and LP2 concentrations were conducted by use of a Shimadzu TOC-L (Shimadzu, Kyoto, Japan).  The refractory pool was calculated as total soil C minus the sum of the labile pools.  All pools were normalized to total soil C content.  

Data processing:
All statistical analysis was performed in R (R Institute for Statistical Computing, Vienna, Austria) using RStudio (RStudio Inc., Boston, MA, USA). The Shapiro-Wilk test was used to verify assumptions of normality, and a logarithmic transformation was performed on all datasets.  Levene’s test was used to determine homogeneity of variance. A linear mixed-effect model (package ‘lmer’) was used to test the following predictor variables: depth, distance inland, and the interaction of depth and distance inland on the samples collected from the marsh. Transect was included as a random effect. Distance inland was found to be a non-significant predictor variable for all parameters except for extractable NH4+. As such, distance inland and the interaction of depth and distance was removed as a predictor variable from models except extractable NH4+. Isotopic determinations and quantitative PCR analysis was performed exclusively on the three replicate cores taken 1 meter inland, and thus depth was the only predictor variable tested for those parameters. Following determination of significance within one of the predictor variables, the package ‘lsmeans’ was used for post-hoc pairwise comparisons using the Tukey method.

BCO-DMO processing:
- Added a conventional header with dataset name, PI names, version date
- Adjusted parameter names to comply with database requirements
- Units removed and added to Parameter Description metadata section
- Added Latitude and Longitude columns, converted DMS to DD
- Added column for Date of collection
- Entries with 'Analysis Not Conducted' were replaced with 'nd' (BCO-DMO's default missing data identifier) 

Description from NSF award abstract:
Coastal Louisiana is currently experiencing net sea level rise at rates higher than most of the world's coastlines and within the global range predicted to occur in the next 65 - 85 years, making Louisiana an ideal site to study potential future impacts of rising sea level on coastal systems. This project will use field collection and controlled tank experiments to study the changing organic carbon cycle resulting from erosion of marsh soils along with its impact on associated biogeochemical processes. The hypothesis tested in this study is that the majority of eroded soil organic carbon is converted to carbon dioxide (CO2) and released to the atmosphere, representing an addition to the anthropogenic input of CO2. This process has not been quantified and could be an important missing component in predictive models of atmospheric CO2 changes. While this process may be of only regional importance today in comparison to other sources of CO2, this study of the Louisiana coast will greatly enhance our full understanding of the potential impacts on the global carbon cycle that may result from coastal erosion as global sea level continues to rise.

The project will train graduate and undergraduate students in interdisciplinary research involving marine and wetland biogeochemistry, microbiology, and ecological modeling. It will also fund development of an interactive, educational display on the loss of coastal wetlands for the Louisiana Sea Grant's annual Ocean Commotion educational event attended by area middle and high school students, teachers, and parents. Results from this study may also inform community planners both regionally and worldwide as they prepare for sea level rise in coastal communities.

Eustatic sea level rise and regional subsidence have created a much greater rate of coastline loss in Louisiana than is being experienced in most of the world's coastal regions, reaching global rates that are predicted to occur worldwide in 65 - 85 years. This provides a unique potential to extrapolate data from Louisiana's changing coastal carbon cycle to both regional and global models of the future impact of sea level rise and coastal erosion. By quantifying and modeling the importance of CO2 emissions resulting directly from mineralized soil organic matter from eroding coastlines, a missing element can be added to climate change models. The PIs here plan to investigate the fate of the coastal wetland carbon pool as it erodes using field sampling, laboratory analysis, mesocosm manipulations, and the creation of a coupled physical-biogeochemical model for the basin being studied. Beyond quantifying increased CO2 emission, the PIs will also address the potential for increased eutrophication due to input of nutrients from eroded soils, as well as the potential for future contribution to existing hypoxic zones in the northern Gulf of Mexico that result from excessive nutrient input from the Mississippi River watershed.