Mara Hauck

Empirical evaluation of spatial and non-spatial European-scale multimedia fate models: results and implications for chemical risk assessment

Multimedia environmental fate models are commonly-applied tools for assessing the fate and distribution of contaminants in the environment. Owing to the large number of chemicals in use and the paucity of monitoring data, such models are often adopted as part of decision-support systems for chemical risk assessment. The purpose of this study was to evaluate the performance of three multimedia environmental fate models (spatially- and non-spatially-explicit) at a European scale. The assessment was conducted for four polycyclic aromatic hydrocarbons (PAHs) and hexachlorobenzene (HCB) and compared predicted and median observed concentrations using monitoring data collected for air, water, sediments and soils. Model performance in the air compartment was reasonable for all models included in the evaluation exercise as predicted concentrations were typically within a factor of 3 of the median observed concentrations. Furthermore, there was good correspondence between predictions and observations in regions that had elevated median observed concentrations for both spatially-explicit models. On the other hand, all three models consistently underestimated median observed concentrations in sediment and soil by 1-3 orders of magnitude. Although regions with elevated median observed concentrations in these environmental media were broadly identified by the spatially-explicit models, the magnitude of the discrepancy between predicted and median observed concentrations is of concern in the context of chemical risk assessment. These results were discussed in terms of factors influencing model performance such as the steady-state assumption, inaccuracies in emission estimates and the representativeness of monitoring data.

Model and input uncertainty in multi-media fate modeling: Benzo[a]pyrene concentrations in Europe

This paper evaluates the contribution of (i) uncertainty in substance properties, (ii) lack of spatial variability, (iii) intermodel differences and (iv) neglecting sorption to black carbon (BC) to the uncertainty of Benzo[a]pyrene (BaP) concentrations in European air, soil and fresh water predicted by the multi-media fate model Simplebox. Uncertainty in substance properties was quantified using probabilistic modeling. The influence of spatial variability was quantified by estimating variation in predicted concentrations with three spatially explicit fate models (Impact 2002, EVn BETR and BETR Global). Intermodel differences were quantified by comparing concentration estimates of Simplebox, Impact 2002, EVn BETR and the European part of BETR Global. Finally, predictions of a BC-inclusive version of Simplebox were compared with predictions of a BC-exclusive version. For air concentrations of BaP, the lack of spatial variability in emissions was most influential. For freshwater concentrations of BaP, intermodel differences and lack of spatial variability in dimensions of fresh water bodies were the dominant sources of uncertainty. For soil, all sources of uncertainty were of comparable magnitude. Our results indicate that uncertainty in Simplebox can be as large as three orders of magnitude for BaP concentrations in the environment and would be substantially underestimated by focusing on one source of uncertainty only.

How Many Environmental Impact Indicators Are Needed in the Evaluation of Product Life Cycles

Numerous indicators are currently available for environmental impact assessments, especially in the field of Life Cycle Impact Assessment (LCIA). Because decision-making on the basis of hundreds of indicators simultaneously is unfeasible, a nonredundant key set of indicators representative of the overall environmental impact is needed. We aimed to find such a nonredundant set of indicators based on their mutual correlations. We have used Principal Component Analysis (PCA) in combination with an optimization algorithm to find an optimal set of indicators out of 135 impact indicators calculated for 976 products from the ecoinvent database. The first four principal components covered 92% of the variance in product rankings, showing the potential for indicator reduction. The same amount of variance (92%) could be covered by a minimal set of six indicators, related to climate change, ozone depletion, the combined effects of acidification and eutrophication, terrestrial ecotoxicity, marine ecotoxicity, and land use. In comparison, four commonly used resource footprints (energy, water, land, materials) together accounted for 84% of the variance in product rankings. We conclude that the plethora of environmental indicators can be reduced to a small key set, representing the major part of the variation in environmental impacts between product life cycles.

PARAMETER UNCERTAINTY IN MODELING BIOACCUMULATION FACTORS OF FISH

We quantified the uncertainty due to biota-related parameters in estimated bioaccumulation factors (BAFs) of persistent organic pollutants for fish through Monte Carlo simulations. For this purpose, the bioaccumulation model OMEGA (Optimal Modeling for EcotoxicoloGical Applications) was parameterized based on data from the existing literature, analysis of allometric data, and maximum likelihood estimation. Lipid contents, fractions of food assimilated, the allometric rate exponent, normalized food intakes, respiration and growth dilution rates, and partial mass transfer resistances in water and lipid layers were included as uncertain parameters. The uncertainty in partial resistances was particularly important in the estimation of the rate constants for chemical intake from water by fish. Uncertainties in the fractions of food assimilated and partial water layer resistances from and to food were particularly important in the estimation of the rate constants of chemical intake from food. The uncertainty in the model outcomes for the bioaccumulation factors for fish was a factor of 10 (ratio of 95th and fifth percentile estimates), which was mainly caused by the uncertainty in the lipid fraction. For chemicals with a K(OW) of 10(3) to 10(6), the uncertainty in the lipid contents of fish accounted for more than 50% of the uncertainty in the estimated bioaccumulation factor. For chemicals with a high K(OW) (10(7) and higher), the fractions of food assimilated and partial resistances also contributed to uncertainty in the estimated bioaccumulation factor (up to 60%). A case study showed that uncertainty in estimated BAF for nonpersistent substances can be dominated by uncertainty in the rate constants for metabolic transformation.

How to quantify uncertainty and variability in life cycle assessment: the case of greenhouse gas emissions of gas power generation in the US

This study quantified the contributions of uncertainty and variability to the range of life-cycle greenhouse gas (LCGHG) emissions associated with conventional gas-fired electricity generation in the US. Whereas uncertainty is defined as lack of knowledge and can potentially be reduced by additional research, variability is an inherent characteristic of supply chains and cannot be reduced without physically modifying the system. The life-cycle included four stages: production, processing, transmission and power generation, and utilized a functional unit of 1 kWh of electricity generated at plant. Technological variability requires analyses of life cycles of individual power plants, e.g. combined cycle plants or boilers. Parameter uncertainty was modeled via Monte Carlo simulation. Our approach reveals that technological differences are the predominant cause for the range of LCGHG emissions associated with gas power, primarily due to variability in plant efficiencies. Uncertainties in model parameters played a minor role for 100 year time horizon. Variability in LCGHG emissions was a factor of 1.4 for combined cycle plants, and a factor of 1.3 for simple cycle plants (95% CI, 100 year horizon). The results can be used to assist decision-makers in assessing factors that contribute to LCGHG emissions despite uncertainties in parameters employed to estimate those emissions.

How to address data gaps in life cycle inventories: A case study on estimating co2 emissions from coal-fired electricity plants on a global scale

One of the major challenges in life cycle assessment (LCA) is the availability and quality of data used to develop models and to make appropriate recommendations. Approximations and assumptions are often made if appropriate data are not readily available. However, these proxies may introduce uncertainty into the results. A regression model framework may be employed to assess missing data in LCAs of products and processes. In this study, we develop such a regression-based framework to estimate CO2 emission factors associated with coal power plants in the absence of reported data. Our framework hypothesizes that emissions from coal power plants can be explained by plant-specific factors (predictors) that include steam pressure, total capacity, plant age, fuel type, and gross domestic product (GDP) per capita of the resident nations of those plants. Using reported emission data for 444 plants worldwide, plant level CO2 emission factors were fitted to the selected predictors by a multiple linear regression model and a local linear regression model. The validated models were then applied to 764 coal power plants worldwide, for which no reported data were available. Cumulatively, available reported data and our predictions together account for 74% of the total worlds coal-fired power generation capacity.

Modeled and monitored variation in space and time of pcb-153 concentrations in air, sediment, soil and aquatic biota on a european scale

We evaluated various modeling options for estimating concentrations of PCB-153 in the environment and in biota across Europe, using a nested multimedia fate model coupled with a bioaccumulation model. The most detailed model set up estimates concentrations in air, soil, fresh water sediment and fresh water biota with spatially explicit environmental characteristics and spatially explicit emissions to air and water in the period 1930-2005. Model performance was evaluated with the root mean square error (RMSE(log)), based on the difference between estimated and measured concentrations. The RMSE(log) was 5.4 for air, 5.6-6.3 for sediment and biota, and 5.5 for soil in the most detailed model scenario. Generally, model estimations tended to underestimate observed values for all compartments, except air. The decline in observed concentrations was also slightly underestimated by the model for the period where measurements were available (1989-2002). Applying a generic model setup with averaged emissions and averaged environmental characteristics, the RMSE(log) increased to 21 for air and 49 for sediment. For soil the RMSE(log) decreased to 3.5. We found that including spatial variation in emissions was most relevant for all compartments, except soil, while including spatial variation in environmental characteristics was less influential. For improving predictions of concentrations in sediment and aquatic biota, including emissions to water was found to be relevant as well.

Resource Footprints are Good Proxies of Environmental Damage

Environmental footprints are increasingly used to quantify and compare environmental impacts of for example products, technologies, households, or nations. This has resulted in a multitude of footprint indicators, ranging from relatively simple measures of resource use (water, energy, materials) to integrated measures of eventual damage (for example, extinction of species). Yet, the possible redundancies among these different footprints have not yet been quantified. This paper analyzes the relationships between two comprehensive damage footprints and four resource footprints associated with 976 products. The resource footprints accounted for >90% of the variation in the damage footprints. Human health damage was primarily associated with the energy footprint, via emissions resulting from fossil fuel combustion. Biodiversity damage was mainly related to the energy and land footprints, the latter being mainly determined by agriculture and forestry. Our results indicate that relatively simple resource footprints are highly representative of damage to human health and biodiversity.

Modelling bioaccumulation of semi-volatile organic compounds (SOCs) from air in plants based on allometric principles.

A model was developed for gaseous plant-air exchange of semi-volatile organic compounds. Based on previous soil-plant modelling, uptake and elimination kinetics were scaled as a function of plant mass and octanol-air partition ratios. Exchange of chemicals was assumed to be limited by resistances encountered during diffusion through a laminar boundary layer of air and permeation through the cuticle of the leaf. The uptake rate constant increased and the elimination rate constant decreased with the octanol-air partition ratio both apparently levelling off at high values. Differences in kinetics between species could be explained by their masses. Validation on independent data showed that bio-concentration factors of PCBs, chlorobenzenes and other chemicals were predicted well by the model. For pesticides, polycyclic aromatic hydrocarbons and dioxins deviations occurred.

Including exposure variability in the life cycle impact assessment of indoor chemical emissions: the case of metal degreasing.

The present paper describes a method that accounts for variation in indoor chemical exposure settings and accompanying human toxicity in life cycle assessment (LCA). Metal degreasing with dichloromethane was used as a case study to show method in practice. We compared the human toxicity related to the degreasing of 1m(2) of metal surface in different exposure scenarios for industrial workers, professional users outside industrial settings, and home consumers. The fraction of the chemical emission that is taken in by exposed individuals (i.e. the intake fraction) was estimated on the basis of operational conditions (e.g. exposure duration), and protective measures (e.g. local exhaust ventilation). The introduction of a time-dependency and a correction for protective measures resulted in reductions in the intake fraction of up to 1.5 orders of magnitude, compared to application of existing, less advanced models. In every exposure scenario, the life cycle impacts for human toxicity were mainly caused by indoor exposure to metal degreaser (>60%). Emissions released outdoors contributed up to 22% of the life cycle impacts for human toxicity, and the production of metal degreaser contributed up to 19%. These findings illustrate that human toxicity from indoor chemical exposure should not be disregarded in LCA case studies. Particularly when protective measures are taken or in the case of a short duration (1h or less), we recommend the use of our exposure scenario-specific approach.