👋 The workshop was successfully held in August 2025. https://fluxnet.org/bridging-the-gap-flux-data-meets-land-surface-models-a-successful-workshop/

Authors: Yujie Liu, Roel Ruzol, Sreenath Paleri, Md Shamsuzzaman, Bailey Murphy

This repo contains tutorials for flux data gap-filling and partitioning (REddyProc), FLUXNET data product exploration, and comparison of observed tower data with ELM outputs.

🔗 What is Binder?

Binder is an open-source service that makes GitHub repositories interactive. With just one click, users can launch a virtual compute environment with all dependencies installed. It is especially useful for teaching, code demonstrations, and sharing reproducible research.

Please click on the badge below:

Then you will see a virtual computing environment, as shown in the images below. Typically, it takes > 5 mins to launch. If it is still loading after 3 mins, please refresh the browser page.

🧪 Binder launches a temporary session (in a Docker container) based on GitHub repo, but:

• 📝 Any edits you make inside Binder (e.g., to code or files) exist only in that temporary environment
• ⏳ When the session ends, all changes are lost

💾 Luckily, you can save files with your edits manually from the Binder file browser:

1. 📁 In the Files pane (usually lower-right corner), navigate to the file or folder you want
2. ☑️ Check the box next to the file
3. ⚙️ Click the More button (gear icon), then select Export...
4. ⬇️ Your browser will download the selected file

💡 Please note: We will use Binder for Tutorial 1, as it takes > 20 mins to install REddyProc. For Tutorials 2 and 3, you may run the scripts either using Binder or locally on your own machine.

---

📚 Three tutorials are included here:

1. 📦 REddyProc workflow (day 1)
2. 🌐 FLUXNET data application (day 1)
3. 🔍 Flux model comparison (day 3)

📦 Tutorial 1: REddyProc workflow

🔗 What is REddyProc?

REddyProc is a R package about standard and extensible Eddy-Covariance data post-processing (Wutzler et al., 2018) includes u* filtering, gap-filling, and flux-partitioning.
🌿 The Eddy-Covariance (EC) micrometeorological technique quantifies continuous exchange fluxes of gases, energy, and momentum between an ecosystem and the atmosphere.
⚙️ It is important for understanding ecosystem dynamics and upscaling exchange fluxes (Aubinet et al., 2012).

This package includes functions for post-processing half-hourly flux data:

• 1️⃣ A quality-check and filtering is performed based on the relationship between measured flux and friction velocity (uStar) to discard biased data (Papale et al., 2006).
• 2️⃣ Gaps in the data are filled based on information from environmental conditions (Reichstein et al., 2005).
• 3️⃣ The net flux of carbon dioxide is partitioned into its gross fluxes in and out of the ecosystem by night-time based and day-time based approaches (Lasslop et al., 2010).

Where to download the package?

• 📦 CRAN release of REddyProc
• 🚧 Development version on GitHub
• 🌐 REddyProcWeb online tool

🎯 Tutorial outcomes

• Comprehension: flux postprocessing, including u* filtering, gap-filling and partitioning;
• Application: working on one typical workflow using REddyProc;

📘 Tutorial instructions

• Script: 01_REddyProc.rmd;
• Data: AmeriFlux BASE data for US-Syv;
• We use data from US-Syv for demonstration. Please do not change to other sites if you are using REddyProc for the first time;
• If you extra time, please explore 'Bonus training' suggested in the R script;
• If you want to use REddyProc for your own study sites, we have a help session after the turtorial.

🌐 Tutorial 2: FLUXNET data application

🔗 What is FLUXNET?

• 📊 The FLUXNET2015 dataset includes data collected at sites from multiple regional flux networks.
• 🚀 The next generation of global, open, and accessible FLUXNET data will be available soon.

🎯 Tutorial outcomes

• Comprehension: Understand FLUXNET data products (data structure, key variables, uncertainty, qualify flags, etc);
• Application: Work in groups to utilize FLUXNET data to understand your study site;
• Analysis: Explore temporal trends in meteorological and flux data for your study site;

📘 Tutorial instructions

• Script: 02_FLUXNET.rmd;
• Working in group for different study sites, and data for different study sites can be downloaded here here;
• We provide example figures using FLUXNET data for US-Syv. Feel free to use these as a starting point, discuss them with your group, and create additional figures to deepen your understanding of your own study site;
• Required task (4 tasks): Explore temporal trends (annual sums, monthly average and qualify flags) for meteorological and flux data;
• Bonus task (3 tasks): ecosystem water budget, light use efficiency, and energy balance closure. Please explore 1–3 of them within your group.
• Please keep records of your figures and discussion notes in group slides.

🔍 Tutorial 3: Comparing FLUXNET data and ELM output

🎯 Tutorial outcomes

• Comprehension: Time-series comparison of observed vs model data
• Application: Work in groups to utilize FLUXNET data and ELM outputs (default and adjusted runs)
• Analysis: Determine the level of agreement between flux tower data and ELM outputs across years with available data

📘 Tutorial instructions

• Script: 03_flux-model-comparison.rmd;
• Working in group for different study sites, and data for different study sites can be found in the data folder here
• In this tutorial, we will compare the FLUXNET data with the ELM outputs. There are 10 overlapping variables that we can compare between the two data sets. Check the crosswalk table for these variables, units, and conversion factors.
• There are 4 parts to this tutorial. Parts 1 to 3 will create time-series graphs for comparing environmental variables, energy exchange variables, and carbon flux variables. In Part 4, we use a simple tool to discern the level of agreement between flux tower data and ELM output.
• Please take time to interpret the figures with your group and drop them in your group slides.

---

Other learning resources

⏳ Working with half-hourly data:

• 📺 Youtube: FLUXNET-ECN Workshop – Thomas Wutzler
• ❓ FAQ for REddyProc
• 📚 EGU 2019 Short Course Materials (REddyProc)
• 🐾 Gapfilling flux data using a machine learning model: XGBoost
• 🌫️ Correcting RH-dependent water flux underestimation under high RH conditions

⚡ Working with high-frequency data:

• 🔬 EcoFlux Lab: High-Frequency Data Processing Documentation

References

• 🔸 Wutzler et al., 2018: https://doi.org/10.5194/bg-15-5015-2018
• 🔸 Aubinet et al., 2012: https://doi.org/10.1007/978-94-007-2351-1
• 🔸 Papale et al., 2006: https://doi.org/10.5194/bg-3-571-2006
• 🔸 Reichstein et al., 2005: https://doi.org/10.1111/j.1365-2486.2005.001002.x
• 🔸 Lasslop et al., 2010: https://doi.org/10.1111/j.1365-2486.2009.02041.x
• 🔸 Liu et al., 2025: https://doi.org/10.1016/j.agrformet.2025.110438

Acknowledgement

• Andrew Richardson: providing critical suggestions for the analysis
• Thomas Wutzler: sharing knowledge about REddyProc
• Darby Bergl: sharing experience in building Binder
• Brian Wang and Weijie Zhang: reviewing the code
• Oscar Zimmerman: serving as the first audience for the tutorials