Résumé: Biodiversity in estuarine ecosystems suffers from the impact of environmental changes and human activities. This mainly involves changes in temperature, salinity, pollution, habitat degradation or loss and fishing activities. The diversity of species communities is traditionally assessed on the basis of their species richness and composition. However, there is growing interest in taking into account complementary components dealing with species differences (e.g. taxonomic relatedness). In spite of their social, ecological and economic importance, the diversity of tropical estuarine fish assemblages has rarely been monitored by means of a multi-component approach under different human pressure and environmental conditions. We analysed the diversity of exploited fish communities (both target and non-target species) sampled during scientific surveys within four estuarine complexes in the state of Pernambuco, Brazil: Itapissuma, Suape, Sirinhaém, and Rio Formoso. A total of 122 species were collected within 34 samples. Overall, diversity indices and species models fitting dominance-evenness profiles mainly revealed differences between assemblages from Itapissuma, being the largest estuary with wide areas of mangrove, and the other estuaries. While assemblages from Itapissuma generally encompassed more species and individuals than the other estuaries, species were more closely related from a taxonomic point of view. In addition, a Double Principal Coordinate Analysis (DPCoA) established a typology of assemblages, useful for management purposes, and linked to particular fish families: it highlighted differences between Itapissuma, Suape, Sirinhaém and Rio Formoso. This method combines matrices of species abundances and differences (here taxonomic distances according to the Linnean classification). It was particularly accurate with a first factorial plane explaining 73% of the total inertia, while only 17% was achieved by a traditional Principal Component Analysis (PCA). Overall, this study provides an assessment of the state of fish assemblage diversity in Pernambuco estuaries where contrasted human and environmental conditions occur. It underscores the accuracy of using a multi-component diversity approach, with a multivariate analysis that is not yet widely used, for monitoring the diversity of estuaries for ecosystem-based fisheries management purposes.