Interdecadal changes in the community, population and individual levels of the fish fauna of an extensively modified estuary

This study examined inter‐period changes over two to three decades in the fish fauna of an urbanized estuary experiencing rapid population growth and a drying climate (Swan–Canning Estuary, Western Australia). Responses were compared at the fish community level (species composition; 1978–2009 in the shallows and 1993–2009 in deeper waters) and at the population and individual levels of an estuarine indicator species, black bream Acanthopagrus butcheri (biomass–abundance and per capita mass at age, respectively; 1993–2009). All three levels showed distinct shifts from earlier to later periods, but their patterns, sensitivity and breadth differed. Community composition changed markedly in the shallows of the lower‐middle estuary between the late 1970s and all later periods and moderately between more disparate periods from 1995 to 2009. Several species trends could be linked to the increasing salinity of the estuary or declining dissolved oxygen levels in its middle–upper reaches. Community changes were, however, small or insignificant in the shallow and deeper waters of the upper estuary and deeper waters of the middle estuary, where environmental perturbations are often most pronounced. This may reflect the resilience of the limited suite of species that typify those reaches and thus their lack of sensitivity in reflecting longer‐term change at the coarser level of mean abundance. One such species, the selected indicator, A. butcheri, did, however, show marked temporal changes at both the population and individual levels. Biomass decreased markedly in deeper waters while increasing in the shallows from earlier to later periods, presumably reflecting an onshore movement of fish, and per capita body mass in the 2+, 3+ and 4+ year classes fell steadily over time. Such changes probably indicate deteriorating habitat quality in the deeper waters. The study outcomes provide support for a multifaceted approach to the biomonitoring of estuaries using fishes and highlight the need for complementary monitoring of relevant stressors to better disentangle cause–effect pathways.