The echoes of ecological collapse: how noise pollution disturbs trophic dynamics and food web stability

Anthropogenic noise is increasingly recognized as a pervasive environmental stressor, yet predicting its ecological impacts remains challenging due to the complex interplay between its physiological influencing pathways and the life-stage-dependent responses of marine organisms. Here, we identify three critical ecological asymmetries — in predator feeding preference, trophic-level sensitivity, and ontogenetic vulnerability — that fundamentally govern how noise disturbance propagates through structured communities. This study employs a stage-structured biomass model comprising a resource, a stage-structured consumer (divided into juvenile and adult), and a top predator to investigate how noise pollution affects community structure and its stability through four key pathways: reduced food intake, increased energy expenditure, elevated mortality, and reduced reproductive output. Our results show that the energy-related pathways, including the food intake reduction and metabolism increase, consistently exert stronger effects in driving bistability than other pathways. However, the system stability under noise disturbance is shaped not only by the disturbance pathways, but also by predator feeding preferences and the asymmetries in noise sensitivity across both trophic levels and life stages. Specifically, the three ecological asymmetries are: (1) predator feeding preference for juvenile consumers, (2) stronger destabilization when noise affects top predators rather than intermediate consumers, and (3) greater sensitivity of adult consumers compared to juveniles. The noise-induced biomass overcompensation (i.e., an increase in population biomass beyond pre-disturbance levels) and subsequent regime shifts are driven by stage-mediated feedbacks and trophic interactions. Model comparisons further highlight that these impacts are not absolute but dependent on model structure. Stage-structuring can unveil emergent bistability absent in non-structured frameworks. These findings provide a mechanistic framework for predicting the ecological consequences of noise pollution, emphasizing the necessity of incorporating ontogenetic structure and species interactions into ecological risk assessments.