Concerns about reported harvests in European forests

Ceccherini et al.1 quantify change using map pixel counts, rather than using a statistically rigorous sampling approach that is more appropriate for the estimation of area change7. [...]although Ceccherini et al.1 considered false positives (incorrect detection of forest loss) in their sample analyses, they did not consider false negatives (undetected forest loss). [...]analyses, which address both omission and commission errors, offer accurate and unbiased results of forest change. [...]sample reference data tailored to the specific purpose of a given study can be used to discriminate proportions of loss due to natural disturbances within the overall forest loss rates12. [...]we are confident that natural disturbances were not correctly excluded. [...]information and knowledge are crucial to develop science-based, climate-smart forestry strategies18 to ensure that European forests continue to be an important carbon sink and a key ecosystem service provider in relation to the protection of biodiversity and the development of the bioeconomy. https://doi.org/10.1038/s41586-021-03292-x Received: 3 July 2020 Accepted: 26 January 2021 Published online: 28 April 2021 Check for updates Acknowledgements We thank G. Ceccherini and co-authors for immediately making available all original material, processing codes and results of their study upon request.