Theo P. Traas

Determination of organotin compounds in the foodweb of a shallow freshwater lake in The Netherlands

An extensive study on the presence of nine organotin compounds (OTs) in a freshwater foodweb was made, using newly developed analytical procedures in order to obtain insight in accumulation and degradation processes. Tributyltin (TBT), Triphenyltin (TPT) and their degradation products were detected. Zebra mussels, eel, roach, bream, pike, perch, and pike perch and cormorant showed high OT body concentrations.At the lower trophic levels, phenyltin concentrations were higher in benthic species while butyltin concentrations were higher in pelagic species. This indicates that TBT is passed on primarily via the water, while TPT is passed on to a larger extent via the sediment.At the higher trophic levels, net bioaccumulation of TPT was greater than that of TBT, resulting in relatively higher TPT concentrations. High concentrations of biodegradation products of TBT, but not of TPT, were found in the livers of fish and birds, which indicates that TBT is more easily metabolized than TPT.A comparison with literature data of fish lethal body concentrations revealed that fish in the field may be endangered. With birds, the highest concentrations of OTs were present in liver and kidney and not in subcutaneous fat, which confirms that OTs accumulate via different mechanisms than traditional lipophilic compounds. As a whole the OT concentrations found in the foodweb may be considered to be quite alarming.

A freshwater food web model for the combined effects of nutrients and insecticide stress and subsequent recovery

A microcosm experiment that addressed the interaction between eutrophication processes and contaminants was analyzed using a food web model. Both direct and indirect effects of nutrient additions and a single insecticide application (chlorpyrifos) on biomass dynamics and recovery of functional groups were modeled. Direct toxicant effects on sensitive arthropods could be predicted reasonably well using concentration-response relationships from the laboratory with representative species. Model predictions showed that nutrient additions alone caused only small effects on toxicant fate and effects probably due to the relatively high dissipation rate of chlorpyrifos. Enhancement of eutrophication effects by the insecticide was relatively small and seemed to be additive. The recovery of some affected functional groups was hampered in the indoor microcosms due to their isolation from outdoor seed populations. Introducing recolonization scenarios in the model simulated dose-dependent recovery. Recolonization increased the recovering rate after exposure to the pesticide. Modeling can extend the use of microcosms as a link between laboratory and field as this allows the prediction of effects and recovery of ecosystems for concentrations that have not been experimentally tested.

Ecotoxicological environmental risk limits for total petroleum hydrocarbons on the basis of internal lipid concentrations

A method is described for deriving ecotoxicological environmental risk limits (ERLs) for total petroleum hydrocarbons (TPH). Toxicity data for two oil types (light and heavy) to benthic organisms and corresponding estimated internal lipid concentrations, calculated by equilibrium partitioning, are used as a measure of toxicity by narcosis. It is assumed that uptake by organisms takes place from the aqueous phase, and for partitioning, both oil droplets or coating and organic carbon of sediment are taken into account. To distinguish between the different fractions of TPH, the method used is based on a fraction analysis approach in which aliphatic and aromatic compounds are regarded separately and both are further divided into different fractions. A toxic unit approach is applied to these fractions to take additivity into account. Lethality of the lighter oil type (internal concentration 28-204 mmol/Llipid) was in good agreement with data on internal concentrations retrieved from the literature. For the heavier oil type the observed toxicity was slightly higher and can probably be attributed to physical soiling of the organisms by oil or oxygen depletion due to biodegradation of the oil. For deriving ERLs, chronic endpoints are considered. The most sensitive chronic endpoints appear to be similar for both types of oil. The distribution of estimated total internal concentrations for chronic endpoints (1.38-149 mmol/Llipid) is used as a basis for the ERLs. The resulting ERLs for the mixture of TPH are comparable with ERLs for single compounds.

Mapping the potentially affected fraction (PAF) of species as a basis for comparison of ecotoxicological risks between substances and regions

Abstract In order to prioritize environmental policies, it is necessary to compare ecotoxicological risks of different substances, different polluted areas (which may have different mixtures of toxic substances) and to compare between effects of toxic substances and other environmental stressors (excess nutrients, acidification). The concept of ecotoxicological risk forms the basis of setting environmental quality standards for toxic substances in the Netherlands. An environmental concentration is considered to present an acceptable risk if not more than 5% of all species is exposed above their no-observed effect concentration (NOEC). This paper is concerned with the inverse problem, to calculate the potentially affected fraction of species (PAF), defined as the fraction of species exposed to a concentration above their NOEC. It is argued that the PAF is more meaningful as a basis for comparisons of ecotoxicological risks than the commonly used ratio of environmental concentration to no-effect concentration. The latter is of limited value, as the same ratio may have widely different effects for different substances. Cumulative distributions of NOECs were determined from literature data. By combining maps of environmental concentrations (from interpolated measurements or from model results) with estimates of field bioavailability, the PAF was determined for Cd, Cu, Zn and Pb and (in terms of toxic pressure) the most important pesticides. The results show the utility of the approach in ranking substances and areas. They also show the importance of uncertainties: these are mainly caused by uncertainties in toxicological data, in emission data for pesticides and in estimation of bioavailability of the heavy metals.

Environmental quality criteria for organic chemicals predicted from internal effect concentrations and a food web model.

Environmental quality criteria (EQC) for hydrophobic organic chemicals were calculated with a model for bioaccumulation in food webs. The model was calibrated and verified using polychlorinated biphenyl concentrations in food webs of shallow lakes. The EQCs in water and sediment were derived based on internal effect concentrations (IECs) for several modes of toxic action. By reverse calculation with the food web model for each organism in the web, a different water or sediment concentration is calculated corresponding to the IEC in each organism. A statistical procedure with an acute-to-chronic value is used to derive chronic EQCs based on bioaccumulation. The model-based chronic EQCs were compared with previously established EQCs. The EQCs calculated with the food web model generally are within an order of magnitude of the previously derived EQCs based on toxicity data on individual chemicals. Some previously derived EQCs are much lower than model predictions and usually based on small samples of toxicity data such as no-observed-effect concentrations (NOECs) with large assessment factors. When faced with data gaps, it is proposed to use model-based chronic EQCs for (polar) narcotic chemicals. Other modes of action require a different model concept to account for receptor-based toxicity.

Estimating the effect on soil organisms of exceeding no-observed effect concentrations (NOECs) of persistent toxicants

In estimating the effects of toxic substances on ecosystems we generally lack information on the sensitivity (expressed as a no-observed effect concentration, NOEC) of individual species in the field, and have to rely on information from laboratory test species, expressed as a frequency distribution of NOECs. In this case we can express toxic stress as the fraction of organisms that is exposed above its NOEC: the potentially affected fraction (PAF). This paper describes a model of the soil food web and the effect of toxic stress by persistent pollutants. The model predicts that in the absence of competition, individual species disappear from the foodweb at toxic concentrations 3–5 times their NOEC. With competition present, species affected by toxic stress are replaced by less sensitive ones. This has a twofold effect: species disappear from the foodweb at a lower concentration because loss of competitiveness occurs well before absolute extinction, but the replacement of disappearing species implies that the effect on total biomass and diversity becomes only noticeable at a PAF level near 100%. Model predictions are in good agreement with observations on nematode communities in experimental fields contaminated with copper and zinc. The model serves to illustrate why overall measures of ecosystem functioning (total biomass, production, diversity) are affected by toxic stress only at high levels of pollution, which is particularly true for systems with a high diversity. This apparent robustness masks a considerable genetic ‘erosion’, i.e. the disappearance of sensitive species or genotypes. The PAF is a good indicator of the latter effect.

Uncertainty Analysis and Risk Assessment Combined: Application to a Bioaccumulation Model

A bioaccumulation model has been formulated for contaminant accumulation in meadow-ecosystems. This type of model generally is parameter-rich, which poses problems for risk prediction due to parameter uncertainty. A procedure is needed to deal with model uncertainty and risk-assessment simultaneously. The model was subjected to uncertainty analysis, leading to probability distributions of all model output variables. Uncertainty measures were calculated using a linear regression model. The probability that environmental standards or No Observed Effect Concentrations are exceeded, was derived from the same distributions, used for the uncertainty analysis. Effects of different toxicant loading scenarios on these probabilities were calculated. The procedure discussed here facilitates the use of complex ecosystem models for risk-assessment.

Verification of a REACH Environmental Prioritization System Against Regulatory Risk Indices

ABSTRACT A prioritization method that was developed to rank chemical substances on the basis of their environmental impact was applied to 230 new substance notifications from earlier European chemical legislation (67/548/EEC). The method encompasses three steps: (1) assigning an environmental hazard score, (2) assigning an environmental exposure score, and (3) combining these two scores into one priority score. In this study, the resulting scores ranged between 4 (highest priority) and 15 (lowest priority). The scores were compared with results from risk assessments available for 138 of the 230 substances. For most substances in these assessments, a priority score of 12 or higher was associated with no or limited risk, while a score of 11 or lower represented high risk or led to the conclusion that risk reduction measures were required. This categorization applied to all but 15 of the 138 substances. The method was also used for ranking the first 15 Substances of Very High Concern (SVHC) under REACH regulation, to compare the priority scores of new substances to those of the SVHC. In sum, the prioritization method seems to be valuable to identify substances of concern with respect to the environment.