Maximising time-series inclusion reduces geographic and taxonomic biases in the Living Planet Index

Measuring how and why biodiversity is changing is critical to protecting it. Among the tools developed to measure biodiversity, one indicator has come under recent scrutiny. The Living Planet Index (LPI) is an indicator based on vertebrate population trends used as evidence for policy and a resource for scientific research; it has a high profile and global reach in the media by conveying a simple message about biodiversity loss1. Toszogyova et al recently published a critique on the indicator’s approach to data inclusion and weighting, asserting that the LPI provides a biased estimate of global vertebrate abundance change. As the scientists behind the ongoing development of the LPI, we appreciate efforts to improve the index. Whilst the authors rightly highlight the sensitivity of the LPI to characteristics of time-series data, their conclusions are not substantiated. Here, we identify aspects of their method which preclude a direct comparison to the published LPI, contest the conclusions drawn and provide a rationale advising against adopting their approach.