Chemicals must be bioavailable for there to be a potential for exposure and consequent risk to human or environmental health. Passive sampling devices (PSDs) are used to quantify the time-integrated concentration of bioavailable contaminants. We demonstrate that PSDs can be paired with the zebrafish developmental toxicity bioassay to produce site-specific, temporally resolved information about the toxicity of environmental samples. Furthermore, modeling associations between the chemical components of environmental mixtures and the toxic outcomes they elicit can link bioactive compounds to biological effects. This research also shows that PSDs can be used as direct biological surrogates in a risk assessment model. We were able to determine spatial and seasonal variations in exposure and risk from the consumption of polycyclic aromatic hydrocarbons (PAH) in organisms from the Portland Harbor Superfund that were not detected in the Public Health Assessment for the area. Additionally, PSDs are a tool that we were able to rapidly deploy after the Deepwater Horizon oil spill. We quantified biologically relevant PAH contamination on a large spatial scale, over a long period of time when the chemicals of concern were present at relatively low dissolved concentrations, their impact on certain areas was sporadic and their presence and toxicological significance were not easily visualized. The research presented here can be applied to improve environmental monitoring, mixture toxicity assessment and risk assessment.