Using the covariance between populations to get a more precise Living Planet Index

Biodiversité et politiques
jeudi 19 déc. 10:10 AM (15 minutes)
Salle EFG
Though a plethora of metrics of biodiversity change exist, almost none explicitly consider relationships between taxa and their responses to global environmental change. The Living Planet Index (LPI) is one of the most prevalent of these metrics, measuring the overall global trend in vertebrate abundances since 1970, based on population time series collected from researchers across the globe. However, the LPI’s uncertainty accumulates over time from the baseline in 1970. This growing imprecision could impede our ability to confidently interpret both the direction and magnitude of biodiversity change, which has serious implications for monitoring progress towards global biodiversity targets. Here, we suggest a methodological approach to address this growing uncertainty via the joint variability between co-occurring populations, or covariance, which provides an opportunity to bolster confidence in measured trends. Specifically, we propose a joint modelling approach to integrate the covariance between co-occurring populations’ growth rates into the computation of the LPI, in order to decrease imprecision. Rather than assuming all population trends vary independently, this approach allows us to explicitly evaluate the covariance between co-occurring populations that would otherwise remain unexplained. Through a comparison of precision between the reported LPI and the LPI obtained using our proposed joint model methodology, we demonstrate that additional information is rendered accessible if we conceptually and methodologically consider the covariation between taxa in our measurements of biodiversity change. Ultimately, this consideration of covariance has potential political, ecological, and communicational implications to improve the overall reliability of our measurement and communication of biodiversity change.

Mon horaire

Ajouter à Mon Horaire comme Favoris