Friedman, S., Averbuch, N., Nevo, T. and Moshelion, M. (2026), Integrating Load‐Cell Lysimetry and Machine Learning for Prediction of Daily Plant Transpiration. Plant, Cell & Environment, 49: 410‐429. https://doi.org/10.1111/pce.70222.
In Table 3 of the above article, the Tukey HSD significance letters were inadvertently transferred incorrectly during manuscript preparation, resulting in all models being labelled with the same letter (‘a’), although the statistical analysis indicates that some models should be assigned different letters.
Table 3.
Comparison of evaluation scores among the four methods model.
| Model | RMSE | R2 | Run time (sec) |
|---|---|---|---|
| Decision tree | 104.51a | 0.83a | 0.01a |
| Random forest | 86.93b | 0.88b | 1.65a |
| XGBoost | 85.19b | 0.89b | 0.24a |
| Neural Network | 84.63b | 0.89b | 57.99b |
Note: The average of 10‐fold cross‐validations are presented. An ANOVA test wasconducted to statistically evaluate performance differences among the 10‐foldmodels scores, followed by Tukey's Honest Significant Difference (HSD), differences are indicated by letters.a p < 0.05.421.
Importantly, all numerical values reported in Table 3 (RMSE, R², and runtime) are correct, and the Results text accurately describes the correct statistical conclusions. The error is therefore limited solely to the letter annotation used to indicate significant differences among models and does not affect the interpretation or conclusions of the study.
The correct version of Table 3, in which the Tukey HSD lettering properly reflects the statistical analysis, is presented below:
We apologize for this error.