Answers are grounded in this paper and cite their sources.
Numerous tools/software exist to gap-fill missing eddy covariance (EC) data, with varying performance depending on study-site dynamics. Disturbed ecosystems like former cutaway-peatlands may be challenging for gap-filling. Researchers using gap-filling spreadsheets may benefit from transitioning to R, but may face challenges if they lack programming skills. To address these, we introduce ‘miniRECgap’, a user-friendly tool in R for effortless gap-filling of EC carbon dioxide flux data using well-known temperature- and light-response functions. ‘miniRECgap’ can model net ecosystem exchange (NEE) via GUI-supported scripts with only five code-lines and minimal inputs. A case-study on one ‘classic’ (forest) and one ‘challenging’ (rehabilitating cutaway-peatland) ecosystem indicated that standard gap-filling (MDS) performed better for the ‘classic’, but not for the ‘challenging’ ecosystem (MDS R 2 = 0.24; ‘miniRECgap’ R 2 = 0.57). For the rehabilitating-peatland, an optimised shallow Artificial Neural Network outperformed other two approaches (R 2 = 0.68). These findings demonstrate the importance of NEE gap-filling for assessing ecosystem-level carbon-dynamics, important for rehabilitating-peatlands.
No takes yet. Share an insight, caveat, or question.
Premrov et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: