The CO2PEAT Project Introduces the 'miniRECgap' R-Package – Research on Simple Gap-Filling of Missing Eddy Covariance CO₂ Flux Measurements from a Peatland Undergoing Rehabilitation.

The Eddy covariance (EC) is a well-known technique to investigate the ecosystem exchange of greenhouse gases (GHGs) between the biosphere and the atmosphere. It often experiences data gaps due to various reasons and gap-filling flux data can be challenging in some ecosystems. The CO2PEAT project is excited to announce research on the ‘miniRECgap’ R-package, which is designed for simple gap-filling of missing EC CO₂ flux measurements, with the manuscript currently under review in a scientific journal.

The ‘miniRECgap’ package enables robust gap-filling via so-called ‘classic’, traditional robust, and validated modelling approaches for EC CO₂ flux data (selected popular temperature- and light-response functions). The submitted manuscript under review focuses on introducing the ‘miniRECgap’ to the research community and includes a crude evaluation of its performance compared to two other selected approaches: the Marginal Distribution Sampling (MDS) approach and the optimized shallow Artificial Neural Network (ANN) approach.

The ‘miniRECgap’ offers graphical user interface (GUI) supported scripts for modelling and gap-filling CO₂ flux data. The package is meant to be accessible to users with varying levels of R programming skills, including those who are new to the R environment.

This research highlights the application of the ‘miniRECgap’ on an Irish cutaway peatland undergoing rehabilitation, and importance of assessing the ecosystem CO₂ flux data to gain further insights into ecosystem carbon dynamics. ‘miniRECgap’ hopes to be an additional valuable resource for researchers focused on EC flux data modelling.

Parts of this research have already gained recognition at several conferences, including:

• EGU24 (Austria, Vienna, April 2024): Premrov, A., Yeluripati, j., Saunders, M. 2024. Introducing the ’miniRECgap’ package with GUI-supported R-scripts for simple gap-filling of Eddy Covariance CO₂ flux data. EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-6475 DOI: https://doi.org/10.5194/egusphere-egu24-6475.
• IGRM24 (Galway, Ireland, March 2024): Premrov, A., Yeluripati, J., Barry, S., McCorry, M., Slevin, R., Harcourt Bates, A. & Saunders, M. 2024. Insights into the application of an artificial neural network approach for gap-filling the CO₂ Eddy Covariance flux data from Cavemount Bog. IGRM24 - Irish Geoscience Research Meeting 2024, Galway, Ireland, 1-3 Mar 2024. Book of abstracts (p. 67): IGRM24 Book of Abstracts.pdf.
• ICOS24 Science Conference (Versailles, France, September 2024): Premrov, A, Yeluripati, J., Saunders, M. 2024. Insights into hyperparameter-optimisation for shallow artificial neural network used in Eddy Covariance CO₂ flux data gap-filling, ICOS Science Conference 2024, 10th - 12th September 2024, Versailles Palais des Congrès, France, and online https://www.icos-cp.eu/news-and-events/science-conference/icos2024sc/all-abstracts.

The ‘miniRECgap’ R-package version v01.0 is already freely available to the public:

• Via GitHub: https://github.com/APremrov/miniRECgap
• Published on Zenodo: Premrov, Alina, 2024. 'miniRECgap': R-package for gap-filling of the missing eddy covariance CO₂ flux measurements using selected classic nonlinear environmental response functions via simple user-friendly GUI supported R scripts. (v0.1.0). Copyright © Trinity College Dublin 2024. https://doi.org/10.5281/zenodo.13228228; https://doi.org/10.5281/zenodo.13228227.

Find out more about CO2PEAT Project Team at Project Team & Contributors | Co2peat.