Modeling the migration of chemicals from food contact materials to food: The MERLIN-expo/VERMEER toolbox.

Evaluating the migration of chemicals from food contact materials (FCM) into food is a key step in the safety assessment of such materials. In this paper, a simple mechanistic model describing the migration of chemicals from FCM to food was combined with quantitative property-property relationships (QPPRs) for the prediction of diffusion coefficients and FCM-Food partition coefficients. The aim of the present study was to evaluate the performance of these operational models in the prediction of a chemical's concentration in food in contact with a plastic monolayer FCM. A comparison to experimental migration values reported in literature was conducted. Deterministic simulations showed a good match between predicted and experimental values. The tested models can be used to provide insights in the amount and the type of toxicological data that are needed for the safety evaluation of the FCM substance. Uncertainty in QPPRs used for describing the processes of both diffusion in FCM and partition at the FCM-Food interface was included in the analysis. Combining uncertainty in QPPR predictions, it was shown that the third quartile (75th percentile) derived from probabilistic calculations can be used as a conservative value in the prediction of chemical concentration in food, with reasonable safety factors.

A probabilistic model for assessing uncertainty and sensitivity in the prediction of monochloramine loss in French river waters.

Monochloramine (NH2Cl) is increasingly used as alternative disinfectant to free chlorine in industrial plants. After use in cooling systems, the waters are released to the environment and residual NH2Cl may be discharged into the receiving waters. As NH2Cl is suspected to exhibit toxicity towards aquatic organisms, a proper risk assessment of its occurrence in environmental waters is needed to prevent adverse effects on wildlife. For this purpose, a comprehensive model simulating monochloramine loss in natural riverine waters was developed. This model incorporates the following processes: (i) autodecomposition; (ii) reaction with nitrite and bromide; (iii) oxidation with Dissolved Organic Carbon (DOC); (iv) oxidation with organic fraction of Suspended Particulate Matter (SPM); (v) reactions in bottom sediments and (vi) volatilization. The model was also designed to conduct uncertainty and sensitivity analysis. It was tested on several French rivers submitted to discharges of monochloraminated effluents and on several seasonal conditions. Uncertainty analysis allowed evaluation of confidence intervals related to NH2Cl half-lives in natural waters. It was shown that simulation intervals are in good agreement with experimental data obtained on the same rivers. Sensitivity analysis using an EFAST variance decomposition approach allowed identification of the most influential parameters on half-life determination. It was shown that the kinetic rate describing rapid reaction of NH2Cl with DOC is by far the most sensitive parameter, demonstrating the predominance of such reactions in the loss process. Variables or parameters involved in temperature dependence (temperature and activation energy) can also significantly influence model results. To a lesser extent, wind velocity is the most sensitive parameter explaining uncertainty in the prediction of volatilization, with a high level of interactions with other parameters, showing that loss through volatilization can be essential in some specific conditions only. This study then identified the most important research priorities for improving the prediction of NH2Cl half-lives in natural rivers.

Evaluation of DGT and DGT-PROFS modeling approach to estimate desorption kinetics of Cs in soils.

The aim of this paper is to assess the suitability of DGT to extract kinetic rates of desorption of cesium (Cs) from soils. For this purpose, laboratory experiments with a natural soil spiked with Cs were carried out under three different contamination conditions, reflecting either an increase in Cs contamination level or an ageing of the contamination within the soil. The experimental results, i.e. the Cs accumulation kinetics onto DGT probes were interpreted by the DGT-PROFS model. The latter calculates the partitioning of Cs between two particulate pools, describing weak and strong interactions respectively, as well as kinetic rates describing exchange reactions. Experimental conditions did not show any major impact on desorption rates, suggesting that desorption kinetics were not significantly affected by contamination level and ageing. Instead, the distribution of Cs among weak and strong sites was shown to be the predominant factor governing the differences observed in the remobilization of Cs to porewater among experimental conditions. The DGT technique combined with the DGT-PROFS modelling approach was proved to be efficient in estimating desorption kinetic rates of Cs in soils.

A comprehensive probabilistic approach for integrating natural variability and parametric uncertainty in the prediction of trace metals speciation in surface waters.

The main objectives of this study were to evaluate global uncertainty in the prediction of Distribution coefficients (Kds) for several Trace Metals (TM) (Cd, Cu, Pb, Zn) through the probabilistic use of a geochemical speciation model, and to conduct sensitivity analysis in speciation modeling in order to identify the main sources of uncertainty in Kd prediction. As a case study, data from the Loire river (France) were considered. The geochemical speciation model takes into account complexation of TM with inorganic ligands, sorption of TM with hydrous ferric oxides, complexation of TM with dissolved and particulate organic matter (i.e. dissolved and particulate humic acids and fulvic acids) and sorption and/or co-precipitation of TM to carbonates. Probability Density Functions (PDFs) were derived for physico-chemical conditions of the Loire river from a comprehensive collection of monitoring data. PDFs for model parameters were derived from literature review. Once all the parameters were assigned PDFs that describe natural variability and/or knowledge uncertainty, a stepwise structured sensitivity analysis (SA) was performed, by starting from computationally 'inexpensive' Morris method to most costly variance-based EFAST method. The most sensitive parameters on Kd predictions were thus ranked and their contribution to Kd variance was quantified. Uncertainty analysis was finally performed, allowing quantifying Kd ranges that can be expected when considering all the sensitive parameters together.

A critical review of frameworks used for evaluating reliability and relevance of (eco)toxicity data: Perspectives for an integrated eco-human decision-making framework.

Considerable efforts have been invested so far to evaluate and rank the quality and relevance of (eco)toxicity data for their use in regulatory risk assessment to assess chemical hazards. Many frameworks have been developed to improve robustness and transparency in the evaluation of reliability and relevance of individual tests, but these frameworks typically focus on either environmental risk assessment (ERA) or human health risk assessment (HHRA), and there is little cross talk between them. There is a need to develop a common approach that would support a more consistent, transparent and robust evaluation and weighting of the evidence across ERA and HHRA. This paper explores the applicability of existing Data Quality Assessment (DQA) frameworks for integrating environmental toxicity hazard data into human health assessments and vice versa. We performed a comparative analysis of the strengths and weaknesses of eleven frameworks for evaluating reliability and/or relevance of toxicity and ecotoxicity hazard data. We found that a frequent shortcoming is the lack of a clear separation between reliability and relevance criteria. A further gaps and needs analysis revealed that none of the reviewed frameworks satisfy the needs of a common eco-human DQA system. Based on our analysis, some key characteristics, perspectives and recommendations are identified and discussed for building a common DQA system as part of a future integrated eco-human decision-making framework. This work lays the basis for developing a common DQA system to support the further development and promotion of Integrated Risk Assessment.

Modelling the exposure to chemicals for risk assessment: a comprehensive library of multimedia and PBPK models for integration, prediction, uncertainty and sensitivity analysis - the MERLIN-Expo tool.

MERLIN-Expo is a library of models that was developed in the frame of the FP7 EU project 4FUN in order to provide an integrated assessment tool for state-of-the-art exposure assessment for environment, biota and humans, allowing the detection of scientific uncertainties at each step of the exposure process. This paper describes the main features of the MERLIN-Expo tool. The main challenges in exposure modelling that MERLIN-Expo has tackled are: (i) the integration of multimedia (MM) models simulating the fate of chemicals in environmental media, and of physiologically based pharmacokinetic (PBPK) models simulating the fate of chemicals in human body. MERLIN-Expo thus allows the determination of internal effective chemical concentrations; (ii) the incorporation of a set of functionalities for uncertainty/sensitivity analysis, from screening to variance-based approaches. The availability of such tools for uncertainty and sensitivity analysis aimed to facilitate the incorporation of such issues in future decision making; (iii) the integration of human and wildlife biota targets with common fate modelling in the environment. MERLIN-Expo is composed of a library of fate models dedicated to non biological receptor media (surface waters, soils, outdoor air), biological media of concern for humans (several cultivated crops, mammals, milk, fish), as well as wildlife biota (primary producers in rivers, invertebrates, fish) and humans. These models can be linked together to create flexible scenarios relevant for both human and wildlife biota exposure. Standardized documentation for each model and training material were prepared to support an accurate use of the tool by end-users. One of the objectives of the 4FUN project was also to increase the confidence in the applicability of the MERLIN-Expo tool through targeted realistic case studies. In particular, we aimed at demonstrating the feasibility of building complex realistic exposure scenarios and the accuracy of the modelling predictions through a comparison with actual measurements.

Development of a standard documentation protocol for communicating exposure models.

An important step in building a computational model is its documentation; a comprehensive and structured documentation can improve the model applicability and transparency in science/research and for regulatory purposes. This is particularly crucial and challenging for environmental and/or human exposure models that aim to establish quantitative relationships between personal exposure levels and their determinants. Exposure models simulate the transport and fate of a contaminant from the source to the receptor and may involve a large set of entities (e.g. all the media the contaminants may pass though). Such complex models are difficult to be described in a comprehensive, unambiguous and accessible way. Bad communication of assumptions, theory, structure and/or parameterization can lead to lack of confidence by the user and it may be source of errors. The goal of this paper is to propose a standard documentation protocol (SDP) for exposure models, i.e. a generic format and a standard structure by which all exposure models could be documented. For this purpose, a CEN (European Committee for Standardisation) workshop was set up with objective to agree on minimum requirements for the amount and type of information to be provided on exposure models documentation along with guidelines for the structure and presentation of the information. The resulting CEN workshop agreement (CWA) was expected to facilitate a more rigorous formulation of exposure models description and the understanding by users. This paper intends to describe the process followed for defining the SDP, the standardisation approach, as well as the main components of the SDP resulting from a wide consultation of interested stakeholders. The main outcome is a CEN CWA which establishes terms and definitions for exposure models and their elements, specifies minimum requirements for the amount and type of information to be documented, and proposes a structure for communicating the documentation to different users.

Perspectives for integrating human and environmental exposure assessments.

Integrated Risk Assessment (IRA) has been defined by the EU FP7 HEROIC Coordination action as "the mutual exploitation of Environmental Risk Assessment for Human Health Risk Assessment and vice versa in order to coherently and more efficiently characterize an overall risk to humans and the environment for better informing the risk analysis process" (Wilks et al., 2015). Since exposure assessment and hazard characterization are the pillars of risk assessment, integrating Environmental Exposure assessment (EEA) and Human Exposure assessment (HEA) is a major component of an IRA framework. EEA and HEA typically pursue different targets, protection goals and timeframe. However, human and wildlife species also share the same environment and they similarly inhale air and ingest water and food through often similar overlapping pathways of exposure. Fate models used in EEA and HEA to predict the chemicals distribution among physical and biological media are essentially based on common properties of chemicals, and internal concentration estimations are largely based on inter-species (i.e. biota-to-human) extrapolations. Also, both EEA and HEA are challenged by increasing scientific complexity and resources constraints. Altogether, these points create the need for a better exploitation of all currently existing data, experimental approaches and modeling tools and it is assumed that a more integrated approach of both EEA and HEA may be part of the solution. Based on the outcome of an Expert Workshop on Extrapolations in Integrated Exposure Assessment organized by the HEROIC project in January 2014, this paper identifies perspectives and recommendations to better harmonize and extrapolate exposure assessment data, models and methods between Human Health and Environmental Risk Assessments to support the further development and promotion of the concept of IRA. Ultimately, these recommendations may feed into guidance showing when and how to apply IRA in the regulatory decision-making process for chemicals.

Perspectives for integrating human and environmental risk assessment and synergies with socio-economic analysis.

For more than a decade, the integration of human and environmental risk assessment (RA) has become an attractive vision. At the same time, existing European regulations of chemical substances such as REACH (EC Regulation No. 1907/2006), the Plant Protection Products Regulation (EC regulation 1107/2009) and Biocide Regulation (EC Regulation 528/2012) continue to ask for sector-specific RAs, each of which have their individual information requirements regarding exposure and hazard data, and also use different methodologies for the ultimate risk quantification. In response to this difference between the vision for integration and the current scientific and regulatory practice, the present paper outlines five medium-term opportunities for integrating human and environmental RA, followed by detailed discussions of the associated major components and their state of the art. Current hazard assessment approaches are analyzed in terms of data availability and quality, and covering non-test tools, the integrated testing strategy (ITS) approach, the adverse outcome pathway (AOP) concept, methods for assessing uncertainty, and the issue of explicitly treating mixture toxicity. With respect to exposure, opportunities for integrating exposure assessment are discussed, taking into account the uncertainty, standardization and validation of exposure modeling as well as the availability of exposure data. A further focus is on ways to complement RA by a socio-economic assessment (SEA) in order to better inform about risk management options. In this way, the present analysis, developed as part of the EU FP7 project HEROIC, may contribute to paving the way for integrating, where useful and possible, human and environmental RA in a manner suitable for its coupling with SEA.

Mobility of Cd and Cu in formulated sediments coated with iron hydroxides and/or humic acids: a DGT and DGT-PROFS modeling approach.

The diffusive gradients technique in thin films (DGT) was used to investigate the kinetic resupply of Cd and Cu to pore water from the solid phase. For the sake of simplification, experiments were performed using formulated sediments that differed in the presence or absence of humic acids (HA) and/or of iron hydroxides (i.e., goethite and ferrihydrite). The effects of the time after the contamination of the solid phase (aging effect) on formulated sediments that were coated with goethite and HA and spiked with Cd were also evaluated. Kinetic DGT results were interpreted using the newly developed, multi-compartmental model DGT-PROFS. Due to Cu humate formation, the addition of HA slightly increased the Cu concentration in the pore water independent of the effect of the iron hydroxide coating on the formulated sediments and slightly decreased that of Cd. The impact of 8-190d of aging resulted in a significant decrease in the Cd concentration of the pore water over an increasing incubation time. Modeling our results with DGT-PROFS led to the following conclusions concerning the impact of HA and iron hydroxides on Cd and Cu availability. First, in the presence of HA and absence of iron hydroxides, Cd is associated mainly with weak sites, while Cu is bound to strong sites. Similarly, in the presence of both iron hydroxides and HA, Cu appeared to be more heavily associated with the strong sites than did Cd. When the incubation time increased from 8 to 190d, a proportion of Cd initially adsorbed onto weak sites transferred to the strong sites, suggesting that the adsorption of Cd on sediments is controlled partially by slow kinetic processes.

Probabilistic multicompartmental model for interpreting DGT kinetics in sediments.

Extensive research has been performed on the use of the DIFS (DGT-Induced Fluxes in Soils and Sediments) model to interpret diffusive gradients in thin-film, or DGT, measurements in soils and sediments. The current report identifies some areas where the DIFS model has been shown to yield poor results and proposes a model to address weaknesses. In particular, two major flaws in the current approaches are considered: (i) many studies of accumulation kinetics in DGT exhibit multiple kinetic stages and (ii) several combinations of the two fitted DIFS parameters can yield identical results, leaving the question of how to select the 'best' combination. Previously, problem (i) has been addressed by separating the experimental data sets into distinct time segments. To overcome these problems, a model considering two types of particulate binding sites is proposed, instead of the DIFS model which assumed one single particulate pool. A probabilistic approach is proposed to fit experimental data and to determine the range of possible physical parameters using Probability Distribution Functions (PDFs), as opposed to single values without any indication of their uncertainty. The new probabilistic model, called DGT-PROFS, was tested on three different formulated sediments which mainly differ in the presence or absence of iron oxides. It was shown that a good fit can be obtained for the complete set of data (instead of DIFS-2D) and that a range of uncertainty values for each modeling parameter can be obtained. The interpretation of parameter PDFs allows one to distinguish between a variety of geochemical behaviors, providing useful information on metal dynamics in sediments.

Probabilistic distribution coefficients (K(d)s) in freshwater for radioisotopes of Ag, Am, Ba, Be, Ce, Co, Cs, I, Mn, Pu, Ra, Ru, Sb, Sr and Th: implications for uncertainty analysis of models simulating the transport of radionuclides in rivers.

The objective of this study was to provide operational probability density functions (PDFs) for distribution coefficients (K(d)s) in freshwater, representing the partition of radionuclides between the particulate and the dissolved phases respectively. Accordingly, the K(d) variability should be considered in uncertainty analysis of transport and risk assessment models. The construction of PDFs for 8 elements (Ag, Am, Co, Cs, I, Mn, Pu and Sr) was established according to the procedure already tested in Durrieu et al. [2006. A weighted bootstrap method for the determination of probability density functions of freshwater distribution coefficients (K(d)s) of Co, Cs, Sr and I radioisotopes. Chemosphere 65 (8), 1308-1320]: (i) construction of a comprehensive database where K(d)s values obtained under various environments and parametric conditions were collected; (ii) scoring procedure to account for the 'quality' of each datapoint (according to several criteria such as the presentation of data (e.g. raw data vs mean with or without replicates), contact time, pH, solid-to-liquid ratio, expert judgement) in the construction of the PDF; (iii) weighted bootstrapping procedure to build the PDFs, in order to give more importance to the most relevant datapoints. Two types of PDFs were constructed: (i) non-conditional, usable when no knowledge about the site of concern is available; (ii) conditional PDFs corresponding to a limited range of parameters such as pH or contact time; conditional PDFs can thus be used when some parametric information is known on the site under study. For 7 other radionuclides (Ba, Be, Ce, Ra, Ru, Sb and Th), a simplified procedure was adopted because of the scarcity of data: only non-conditional PDFs were built, without incorporating a scoring procedure.

Issues and practices in the use of effects data from FREDERICA in the ERICA Integrated Approach.

The ERICA Integrated Approach requires that a risk assessment screening dose rate is defined for the risk characterisation within Tiers 1 and 2. At Tier 3, no numerical screening dose rate is used, and the risk characterisation is driven by methods that can evaluate the possible effects of ionising radiation on reproduction, mortality and morbidity. Species sensitivity distribution has been used to derive the ERICA risk assessment predicted no-effect dose rate (PNEDR). The method used was based on the mathematical processing of data from FRED (FASSET radiation effects database merged with the EPIC database to form FREDERICA) and resulted in a PNEDR of 10 microGy/h. This rate was assumed to ascribe sufficient protection of all ecosystems from detrimental effects on structure and function under chronic exposure. The value was weighed against a number of points of comparison: (i) PNEDR values obtained by application of the safety factor method, (ii) background levels, (iii) dose rates triggering effects on radioactively contaminated sites and (iv) former guidelines from literature reviews. In Tier 3, the effects analysis must be driven by the problem formulation and is thus highly case specific. Instead of specific recommendations on numeric values, guidance on the sorts of methods that may be applied for refined effect analysis is provided and illustrated.

A dynamic model for assessing radiological consequences of tritium routinely released in rivers. Application to the Loire River.

A dynamic model for assessing the transfer of tritium in a food chain was applied to the Loire River, where 14 nuclear power plants situated on five different sites operate. The model considers several potential exposure pathways in the aquatic and terrestrial ecosystems: transfer of tritium through the aquatic food chain (especially fish); use of river water for agricultural purposes (irrigation) and transfer of radionuclides through the terrestrial food chain (vegetables, meat, milk); subsequent internal exposure of humans due to ingestion of contaminated foodstuffs. For biological environmental compartments, the transfer of tritium to organic matter (i.e. OBT) was simulated. For each of the parameters introduced in this model, a probability density function, allowing further uncertainty and sensitivity analyses, was proposed. Uncertainty/sensitivity analyses were performed to determine a confidence interval for the mean annual dose to critical groups and to identify the parameters responsible for the uncertainty and subsequent research priorities.

A weighted bootstrap method for the determination of probability density functions of freshwater distribution coefficients (Kds) of Co, Cs, Sr and I radioisotopes.

The objective of the study was to provide global probability density functions (PDFs) representing the uncertainty of distribution coefficients (Kds) in freshwater for radioisotopes of Co, Cs, Sr and I. A comprehensive database containing Kd values referenced in 61 articles was first built and quality scores were affected to each data point according to various criteria (e.g. presentation of data, contact times, pH, solid-to-liquid ratio, expert judgement). A weighted bootstrapping procedure was then set up in order to build PDFs, in such a way that more importance is given to the most relevant data points (i.e. those corresponding to typical natural environments). However, it was also assessed that the relevance and the robustness of the PDFs determined by our procedure depended on the number of Kd values in the database. Owing to the large database, conditional PDFs were also proposed, for site studies where some parametric information is known (e.g. pH, contact time between radionuclides and particles, solid-to-liquid ratio). Such conditional PDFs reduce the uncertainty on the Kd values. These global and conditional PDFs are useful for end-users of dose models because the uncertainty and sensitivity of Kd values are taking into account.

Conceptual approaches for the development of dynamic specific activity models of 14C transfer from surface water to humans.

Carbon-14 is a particularly interesting radionuclide from the perspective of dose estimation. Many nuclear facilities, including power reactors, release 14C into the environment, and much of this is as 14CO2. This mixes readily with stable CO2, and hence enters the food chain as fundamental biomolecules. This isotopic mixing is often used as the basis for dose assessment models. The present model was developed for the situation of 14C releases to surface waters, where there are distinct changes in the water 14C activity concentrations throughout the year. Complete isotopic mixing (equilibrium) cannot be assumed. The model computes the specific activity (activity of 14C per mass of total C) in water, phytoplankton, fish, crops, meat, milk and air, following a typical irrigation-based food-chain scenario. For most of the biotic compartments, the specific activity is a function of the specific activity in the previous time step, the specific activity of the substrate media, and the C turnover rate in the tissue. The turnover rate is taken to include biochemical turnover, growth dilution and mortality, recognizing that it is turnover of C in the population, not a tissue or an individual, that is relevant. Attention is paid to the incorporation of 14C into the surface water biota and the loss of any remaining 14CO2 from the surface water-air interface under its own activity concentration gradient. For certain pathways, variants in the conceptual model are presented, in order to fully discuss the possibilities. As an example, a new model of the soil-to-plant specific activity relationship is proposed, where the degassing of both 14C and stable C from the soil is considered. Selection of parameter values to represent the turnover rates as modeled is important, and is dealt with in a companion paper.

A dynamic model for assessing radiological consequences of routine releases in the Loire river: parameterisation and uncertainty/sensitivity analysis.

A dynamic model for assessing the transfer of several radionuclides ((58)Co, (60)Co, (110 m)Ag, (134)Cs, (137)Cs, (54)Mn and (131)I) in a food-chain was applied on the Loire river, where 14 nuclear power plants situated on five different sites operate. The model considers the following potential exposure pathways: (i) transfer of radionuclides through the aquatic food chain and the subsequent internal exposure of humans due to ingestion of contaminated water and/or fish; (ii) use of river water for agricultural purposes (irrigation), transfer of radionuclides through the terrestrial food chain and the subsequent internal exposure of humans due to ingestion of contaminated foodstuffs; (iii) internal exposure due to inhalation of dust originating from resuspension of contaminated soil particles; (iv) external exposure from radionuclides present in the river or deposited on the river sediments or the soil. For each of the parameters introduced in this model, a probability density function, allowing further uncertainty and sensitivity analysis, was proposed. Uncertainty/sensitivity analysis were performed to: (i) compare calculations to empirical data; (ii) determine a confidence interval for the mean annual dose to critical groups; and (iii) identify the parameters responsible for the uncertainty and subsequent research priorities.

Kinetics of the adsorption and desorption of radionuclides of Co, Mn, Cs, Fe, Ag and Cd in freshwater systems: experimental and modelling approaches.

The sorption and release kinetics of 54Mn, 58Co, 59Fe, 109Cd, 110mAg and 134Cs by freshwater suspended particles were investigated to better identify the biogeochemical processes involved and to obtain suitable data for improving models describing radionuclide migration in freshwater streams. In order to observe any seasonal variability in the interaction of radionuclides with natural particles, experiments were performed both in winter and in summer during a phytoplanktonic bloom. Two kinetic models are compared in this paper: the "one-step reversible" model, based on the hypothesis that the transfer of radionuclides between water and suspended solids is governed by a reversible reaction, and the "two-successive-step reversible" model, which assumes two distinct types of sites or reactions on the solid phase. The "one-step reversible" model is generally unable to describe properly the exchange kinetics; this result shows that at least two processes are generally involved in radionuclide exchange between water and suspended particles. On the contrary, a model involving the existence of two successive reversible reactions properly simultes both the sorption and release kinetics. The determination of the kinetic coefficients allows quantitative assessment of the relative importance and kinetics of the processes. In particular, it has been shown that, except for Cs and Cd, major fractions of the radionuclides are associated at equilibrium with particulate sites involving strong interactions. The kinetics to reach this equilibrium depend on seasonal conditions, especially for Co and Mn: the transfer of Co and Mn to particulate sites involving strong interactions is much slower in winter. The distribution of the radionuclides between water and particulate sites involving weak interactions also shows seasonal variations for Co, Mn, Fe and Ag: the capacity to associate radionuclides is much higher in summer for Co and Mn, while the inverse tendency is observed for Fe and Ag. For Cs and Cd, no significant seasonal differences were observed.