Quantifying uncertainty in the trophic magnification factor related to spatial movements of organisms in a food web

Abstract

Trophic magnification factors (TMFs) provide a method of assessing chemical biomagnification in food webs and are increasingly being used by policy makers to screen emerging chemicals. Recent reviews have encouraged the use of bioaccumulation models as screening tools for assessing TMFs for emerging chemicals of concern. The present study used a food web bioaccumulation model to estimate TMFs for polychlorinated biphenyls (PCBs) in a riverine system. The uncertainty associated with model predicted TMFs was evaluated against realistic ranges for model inputs (water and sediment PCB contamination) and variation in environmental, physiological, and ecological parameters included within the model. Finally, the model was used to explore interactions between spatial heterogeneity in water and sediment contaminant concentrations and theoretical movement profiles of different fish species included in the model. The model predictions of magnitude of TMFs conformed to empirical studies. There were differences in the relationship between the TMF and the octanol-water partitioning coefficient (KOW ) depending on the modeling approach used; a parabolic relationship was predicted under deterministic scenarios, whereas a linear TMF-KOW relationship was predicted when the model was run stochastically. Incorporating spatial movements by fish had a major influence on the magnitude and variation of TMFs. Under conditions where organisms are collected exclusively from clean locations in highly heterogeneous systems, the results showed bias toward higher TMF estimates, for example the TMF for PCB 153 increased from 2.7 to 5.6 when fish movement was included. Small underestimations of TMFs were found where organisms were exclusively sampled in contaminated regions, although the model was found to be more robust to this sampling condition than the former for this system.

Contrasting PCB bioaccumulation patterns among Lake Huron lake trout reflect basin‐specific ecology

“…Recent studies have concluded that a broad understanding of chemical, physical and biological characteristics is required to understand how hazard assessment metrics such as the trophic magnification factor are affected by such properties [41,42]. The results of the present study support this conclusion and emphasize the significance of species growth when undertaking such hazard assessment exercises.…”

Bioaccumulation of decamethylpentacyclosiloxane (D5): A review

“…Warner et al observed that concentrations of D5 in sediment decreased with increasing distance from a wastewater outlet in Adventfjorden in the Svalbard archipelago. McLeod et al used the AquaWeb model to illustrate that large variations in the TMF of PCBs can occur because of spatial gradients in concentration. Spatial gradients in concentration may produce mixed signals regarding the bioaccumulation behavior of chemicals that are not strong biomagnifiers.…”

“…McLeod et al [52] used the AquaWeb model to illustrate that large variations in the TMF of PCBs can occur because of spatial gradients in concentration. Spatial gradients in concentration may produce mixed signals regarding the bioaccumulation behavior of chemicals that are not strong biomagnifiers.…”

Characterizing polychlorinated biphenyl exposure pathways from sediment and water in aquatic life using a food web bioaccumulation model

“…Equation provides the capability to treat ingested sediments differently from ingested biological items that may be present in an animal's food, as specified by the feeding matrix used in the model, and to consider different fractions of respired water, consisting of either porewater or overlying waters. The algorithms describing the model parameter estimates for ingested biological dietary items ( E diet , G f(bio) i , E f(bio) i , and K BF(bio) i ) are described in detail by Arnot and Gobas () and McLeod et al (). However, these previous studies were less explicit in describing how estimates of these parameters are obtained for ingested sediments and are therefore outlined in detail below.…”

“…Where C sed(dry) is the concentration in sediment (ng · g –1 dry weight), and ρ sed is the bulk density of sediments, estimated to be 1.2 g · mL –1 (McLeod et al ). Porewater PCB concentrations (ng · mL –1 ) were calculated by the model according to the equilibrium partitioning algorithm recommended by Arnot and Gobas () and modified to include the IM content as follows: …”