Bioaccumulation models are an important and widely-used tool for assessing ecosystem health with regards to heavy metal contamination. However, these models do not usually account for the potentially significant effect of temperature-dependency in metal uptake. In this study, we explored the role of temperature-dependency in heavy metal bioaccumulation by developing and comparing two kinetic-based copper bioaccumulation models for a common estuarine oyster (Saccostrea glomerata): (i) a standard first-order model that ignores temperature effects; and (ii) a modified first-order model that uses a standard temperature function to account for the temperature-dependency of the uptake rate constant. The models were calibrated within a Bayesian framework so that parameters could be treated as random variables and any uncertainty propagated through to the model output. A 12-month biomonitoring study was carried out within Moreton Bay, Queensland, Australia to provide time-series data for the modelling. Results of the modelling showed that the two bioaccumulation models provided comparable fits of the biomonitoring field data. However, dependent on the time of year and monitoring period selected, the copper uptake rate would vary dramatically due to temperature effects, which could result in an overestimation or underestimation of the copper uptake rate. Finally by calibrating the bioaccumulation models within a Bayesian framework, these models were able to utilize prior knowledge of the model parameters as part of the calibration process and also account for the uncertainty and variability in the bioaccumulation predictions. The ability to account for uncertainty and variability is an important consideration when undertaking environmental risk assessments especially in coastal waterways where there are strong seasonal variations.