While long-term monitoring is essential to improve our knowledge of marine ecosystems health, it remains challenging to summarise complex ecological data in order to characterise and understand biodiversity trends. To compile monitoring data across large numbers of species, scientists and policymakers mainly rely on diversity and species richness indices. This task may prove complicated however, as many indices exist and no individual metric undoubtedly emerges as the best overall. Here, using data from zooplankton surveys from 1998 to 2014, we examined year-to-year changes in copepod communities in two littoral ecosystems of Western Europe - the Arcachon Bay and the Gironde estuary - that share similar climate, but with different local ecological processes, especially hydrological conditions. We tested the ability of commonly used α and β-diversity metrics, such as species richness, Pielou's evenness or Jaccard's index, to mirror year-to-year changes in species abundances and we detected a synchronous change in both copepod abundances and α-diversity that took place circa 2005 in the two sites. In response to changes in environmental conditions such as nutrients, salinity, river discharge or particulate matter, two opposite biodiversity trends were observed, with a decrease in copepod diversity in the Arcachon Bay but an increase in the downstream part of the Gironde estuary. Although diversity metrics allowed us to well detect trends, the use of multivariate approaches such as principal component analysis provided important information on how and why diversity fluctuates. Our study provides evidence that long-term monitoring programmes must be encouraged for optimising management and conservation actions such as the Marine Strategy Framework Directive and that more local comparative studies need to be initiated for better characterising diversity trajectories at very fine scales at which ecologists often work.< Réduire