While extreme climatic events (ECEs) are predicted to become more frequent, reliably predicting their impacts on consumers remains challenging, particularly for large consumers in marine environments. Many studies that do evaluate ECE effects focus primarily on direct effects, though indirect effects can be equally or more important. Here, we investigate the indirect impacts of the 2011 “Ningaloo Niño” marine heatwave ECE on a diverse megafaunal community in Shark Bay, Western Australia. We use an 18‐year community‐level data set before (1998–2010) and after (2012–2015) the heatwave to assess the effects of seagrass loss on the abundance of seven consumer groups: sharks, sea snakes (multiple species), Indo‐pacific bottlenose dolphins (Tursiops aduncus), dugongs (Dugong dugon), green turtles (Chelonia mydas), loggerhead turtles (Caretta caretta), and Pied Cormorants (Phalacrocorax spp.). We then assess whether seagrass loss influences patterns of habitat use by the latter five groups, which are under risk of shark predation. Sharks catch rates were dominated by the generalist tiger shark (Galeocerdo cuvier) and changed little, resulting in constant apex predator density despite heavy seagrass degradation. Abundances of most other consumers declined markedly as food and refuge resources vanished, with the exception of generalist loggerhead turtles. Several consumer groups significantly modified their habitat use patterns in response to the die‐off, but only bottlenose dolphins did so in a manner suggestive of a change in risk‐taking behavior. We show that ECEs can have strong indirect effects on megafauna populations and habitat use patterns in the marine environment, even when direct effects are minimal. Our results also show that indirect impacts are not uniform across taxa or trophic levels and suggest that generalist marine consumers are less susceptible to indirect effects of ECEs than specialists. Such non‐uniform changes in populations and habitat use patterns have implications for community dynamics, such as the relative strength of direct predation and predation risk. Attempts to predict ecological impacts of ECEs should recognize that direct and indirect effects often operate through different pathways and that taxa can be strongly impacted by one even if resilient to the other.