LTER Synthesis working group -- The Flux Gradient Project: Understanding the methane sink-source capacity of natural ecosystems

PIs:

• Sparkle L. Malone, Assistant Professor, Yale University
• Jackie H. Matthes, Senior Scientist, Harvard University

Project Summary

• https://lternet.edu/working-groups/the-flux-gradient-project/

Repository Structure

lterwg-flux-gradient-eval/
├── functions/               # Core analysis functions
│   └── calc.*.R             # Calculation functions
│   
├── workflows/               # Complete analysis workflows
│   └── flow.*.R             # Analysis workflows
│
├── exploratory/             # Preliminary analyses and development
│   └── flow.evaluation.*.R
│
└── aop/             # Preparation of AOP, canopy diversity, and caopy information data
    ├── flow.NEONAOP.EVI.Download.R
    ├── flow_AOP_FormatLayers.R
    ├── flow.AOP.viz.R
    └── flow.StructuralDiversity.R 

Data Frame Organization

• Column name should use snake case include units, last "_" proceeds units (i.e. var_molperm3)
• no "- or /" in column names

Getting Started

1. Clone this repository

   git clone https://github.com/lter/lterwg-flux-gradient.git
   cd lterwg-flux-gradient

2. Install required R packages

   # Core packages
   install.packages(c("tidyverse", "neonUtilities", "rhdf5", "googledrive", 
                     "foreach", "doParallel", "lubridate", "ggplot2"))
   
   # Additional packages
   install.packages(c("gslnls", "terra", "sf", "ggh4x"))
   
   # If using BiocManager
   if (!requireNamespace("BiocManager", quietly = TRUE))
       install.packages("BiocManager")
   BiocManager::install("rhdf5")

Evaluation Workflow

1. flow.evaluation.batch → Creates the data needed to evaluate gradient fluxes using the products of lterwg-flux-gradient/workflows/flow.evaluation.dataframe.R. First data is filtered (flow.evaluation.filter.R) to produce: FilteredData_ALLSites.Rdata and FilterReport_ALLSites.Rdata. Next, the One2One analysis (flow.evaluation.One2One.R) is done on filtered data to produce: One2One_ALLSites.Rdata and FilteredData_ALLSites_BH.Rdata. BH stands for the best height, which is determined by the height levels with the highest R2. Next the diurnal analysis (flow.evaluation.diurnal.R) produces: DiurnalSummary_ALLSites_BH.Rdata. Finally, we fit carbon exchange parameters (flow.evaluation.cparms.R) to produce: CarbonParms.Rdata.

2. flow.evaluation.figures → produces the figures and tables that evaluate the gradient flux. The products of flow.evaluation.batch are used.

   SITE_Evaluation.RData → flow.evaluation.batch
                         ├─ flow.evaluation.filter.R → FilteredData_ALLSites.Rdata, FilterReport_ALLSites.Rdata
                         ├─ flow.evaluation.One2One.R → One2One_ALLSites.Rdata, FilteredData_ALLSites_BH.Rdata
                         ├─ flow.evaluation.diurnal.R → DiurnalSummary_ALLSites_BH.Rdata
                         └─ flow.evaluation.cparms.R → CarbonParms.Rdata
   
   FilteredData_ALLSites.Rdata, etc. → flow.evaluation.figures → Figures and tables
   

AOP Workflow

1. flow.NEONAOP.Download.R → Downloads and mosaics NEON AOP Data. Data is stored locally on the Malone Lab server.

2. flow.neon.site.squarebuffers.R → Creates square buffers for LTER-NEON co-located sites using this file: Ameriflux_NEON field-sites.csv to produce: NEONLTERsiteBuffers.Rdata.

3. flow.neon.site.simplefeatures.R → This script uses the NEONLTERsiteBuffers.Rdata and breaks each buffer into wedges to produce: FG_Site_Wdges.RDATA.

4. flow_AOP_FormatLayers → Extracts AOP information for each wedge to produce a summary file.

   NEON AOP Data → flow.NEONAOP.Download.R → Downloaded AOP data 
   
   Ameriflux_NEON field-sites.csv → flow.neon.site.squarebuffers.R → NEONLTERsiteBuffers.Rdata
   
   NEONLTERsiteBuffers.Rdata → flow.neon.site.simplefeatures.R → FG_Site_Wdges.RDATA
   
   FG_Site_Wdges.RDATA + Downloaded AOP data → flow_AOP_FormatLayers → Summary files
   

Function Folder

• Use hierarchical naming with the active verb first, i.e. "flag.iqr.R"
• This is where all functions called by flow. scripts in Workflow Folder are stored

Exploratory Folder

• Wild West, this is where preliminary functions and workflows are stored

Related Repositories

• lterwg-flux-gradient: Main repo, contains code for filtering
• lterwg-flux-gradient-methane: Code for the methane paper

Supplementary Resources

LTER Scientific Computing Team website & NCEAS' Resources for Working Groups