Flows of dioxins and furans in coastal food webs

Inverse modeling, sensitivity analysis, and applications of linear system theory

Tuomo M. Saloranta*, Tom Andersen, Kristoffer Næs

*Corresponding author for this work

    Research output: Contribution to journalArticle

    12 Citations (Scopus)


    Rate constant bioaccumulation models are applied to simulate the flow of polychlorinated dibenzo-p-dioxins and dibenzofurans (PCDD/Fs) in the coastal marine food web of Frierfjorden, a contaminated fjord in southern Norway. We apply two different ways to parameterize the rate constants in the model, global sensitivity analysis of the models using Extended Fourier Amplitude Sensitivity Test (Extended FAST) method, as well as results from general linear system theory, in order to obtain a more thorough insight to the system's behavior and to the flow pathways of the PCDD/Fs. We calibrate our models against observed body concentrations of PCDD/Fs in the food web of Frierfjorden. Differences between the predictions from the two models (using the same forcing and parameter values) are of the same magnitude as their individual deviations from observations, and the models can be said to perform about equally well in our case. Sensitivity analysis indicates that the success or failure of the models in predicting the PCDD/F concentrations in the food web organisms highly depends on the adequate estimation of the truly dissolved concentrations in water and sediment pore water. We discuss the pros and cons of such models in understanding and estimating the present and future concentrations and bioaccumulation of persistent organic pollutants in aquatic food webs.

    Original languageEnglish
    Pages (from-to)253-264
    Number of pages12
    JournalEnvironmental Toxicology and Chemistry
    Issue number1
    Publication statusPublished - Jan 2006


    • Bioaccumulation
    • Dioxins
    • Modeling
    • Sensitivity analysis

    Fingerprint Dive into the research topics of 'Flows of dioxins and furans in coastal food webs: Inverse modeling, sensitivity analysis, and applications of linear system theory'. Together they form a unique fingerprint.

  • Cite this