A dynamic model of the fate of organic chemicals in a multilayered air/soil system: development and illustrative application

Use of machine learning to identify key factors regulating volatilization of semi-volatile organic chemicals from soil to air

Volatilization from soil to air is a key process driving the distribution and fate of semi-volatile organic contaminants. However, quantifying this process and the key environmental governing factors remains difficult. To address this issue, the volatilization fluxes of polybrominated diphenyl ethers (PBDEs) and organophosphate esters (OPEs) from soil were determined in 16 batch experiments orthogonally with six variables (chemical property, soil concentration, air velocity, ambient temperature, soil porosity, and soil moisture) and analyzed with machine learning methods. The results showed that gradient-boosting regression tree models satisfactorily predicted the volatilization fluxes of PBDEs (r2 = 0.82 ± 0.07) and OPEs (r2 = 0.62 ± 0.13). Permutation importance analysis showed that partitioning potential of chemicals between soil and air was the most important factor regulating the volatilization of the target compounds from soil. Temperature and soil porosity played a secondary role in controlling the migration of PBDEs and OPEs, respectively, due to higher volatilization enthalpies of PBDEs than those of OPEs and dominant adsorption of OPEs on mineral surface. The effect of soil moisture was negative and positive for the volatilization fluxes of PBDEs and OPEs, respectively. These results suggested different responses in the soil-air diffusive transport of PBDEs and OPEs to high temperature and rainstorm induced by climate change.

Environmental fate of sulfonated-PCBs: Soil partitioning properties, bioaccumulation, persistence, and mobility

Two new classes of PCB metabolites were recently discovered: sulfonated-polychlorinated biphenyls (sulfonated-PCBs) and hydroxy-sulfonated-polychlorinated biphenyls (OH-sulfonated-PCBs). These metabolites, originating from PCB degradation, seem to possess more polar characteristics than their parent compounds. However, no other information, such as their chemical identity (CAS number) or their ecotoxicity or toxicity, is available so far, although more than about one hundred different chemicals were observed in soil samples. In addition, their physico-chemical properties are still uncertain since only estimations are available. Here we show the first evidence on the fate of these new classes of contaminants in the environment, producing results from several experiments, to evaluate sulfonated-PCBs and OH-sulfonated-PCBs soil partition coefficients, degradation in soil after 18 months of rhizoremediation, uptake into plant roots and earthworms, as well as a preliminary analytical method to extract and concentrate these chemicals from water. The results give an overview of the expected environmental fate of these chemicals and open questions for further studies.

Synthesis of a new sulfonated-hexachlorobiphenyl standard for environmental analysis, ecotoxicological, and toxicological studies

Sulfonated-polychlorinated biphenyls (sulfonated-PCBs) are a newly discovered class of PCB metabolites. They were observed for the first time in polar bear serum and lately, in soil, together with hydroxy-sulfonated-PCBs. Their presence is ubiquitous in soils, and their estimated physical chemical properties show high mobility in water, compared to the parent compounds. However, no single pure standards exist so far and therefore their quantification in the environmental matrices is not accurate. Additionally, pure standards are needed to experimentally determine their physical chemical properties, as well as the ecotoxicological and toxicological characteristics. In the present work, the challenging goal of preparing a polychlorinated biphenyl monosulfonic acid was achieved exploring different synthetic approaches, along which the selection of the starting material resulted in a crucial point. Using PCB-153 (2,2'-4,4'-5,5'-hexachloro-1,1'-biphenyl) the synthesis afforded, as the major species, a side compound. On the contrary, the use of PCB-155 (2,2'-4,4'-6,6'-hexachloro-1,1'-biphenyl), a symmetric hexachlorobiphenyl derivative showing chlorine atoms at all the ortho positions, gave the target sulfonated-PCB compound. In this case, sulfonation was successfully carried out through a two-step procedure, involving chlorosulfonylation and the subsequent hydrolysis of the chlorosulfonyl intermediate.

A review of modeling pesticides in freshwaters: Current status, progress achieved and desirable improvements

This study comprises a critical review of modeling of pesticides in surface waters. The aim was to update the status of the use of models to simulate the fate of pesticides from diffuse sources. ISI papers were selected on Scopus and the information concerning the study areas, type of pesticides (herbicides, fungicides and insecticides), the model, and the methodology adopted (i.e., calibration and/or validation, spatial and temporal scales) were analyzed. The studies were carried out in Europe (55.5%), North America (22.3%), Asia (13.9%) and South America (8.3%). The Soil and Water Assessment Tool proved to be the most used model (45.95%). Herbicides were the most modeled pesticides (71.4%), followed by insecticides (18.2%) and fungicides (10.4%). The main herbicides modeled were atrazine, metolachlor, isoproturon, glyphosate, and acetochlor. Insecticides such as chlorpyrifos and metaldehyde. Chlorothalonil, and fungicides (i.e., tebuconazole) were the most widely investigated. Based on published studies, it was found that modeling approaches for assessing the fate of pesticides are constantly evolving and the model algorithms work well with diverse watershed conditions, management strategies, and pesticide properties. Several papers reported concentrations of pesticides exceeding ecotoxicological thresholds revealing that water contamination with pesticides used in agriculture and urban areas is a priority issue of current global concern.

Predicting the regional contamination evolution of DDT for 100-years with a new gridded spatial and dynamic multimedia fate model

In 1996 high dichlorodiphenyltrichloroethane (DDT) concentrations were found in Lake Maggiore (Italy) fish and sediments. DDT was produced by a chemical company located in a subalpine valley (Ossola valley, Piedmont Region, Italy), and ended up in the Toce River, a tributary of Lake Maggiore. In the area surrounding the chemical plant, high DDT concentrations in soil and vegetation were found after subsequent investigations. The quantification of the release from contaminated soil and the following migration toward downwind areas, deposition to the soil, and further evaporation plays an important role in understanding future DDT trends in soil and the atmosphere. To study this phenomenon, soil, and vegetation from Ossola Valley were monitored in 2001 and 2011. The concentration values obtained (soils: 0.05 to 1 μg/g; vegetation 2-100 ng/g), allowed to reconstruct the contamination gradient in the valley and were used to develop and calibrate a spatially resolved multimedia fugacity model. The model accounts for spatial and temporal dynamicity of environmental characteristics such as wind speed and direction, variable air compartment height etc., and simulates the fate and transport of chemicals on a local scale. The dynamic emission of DDT (about 13,000 kg for the 50 y production time) to the air was estimated and utilized for a 100-year simulation (from 1948 to 2048). The results obtained allowed to understand the temporal and spatial pattern of DDT contamination for a long period at a local scale as well as the potential contribution as a source potentially affecting sites at larger distances.

Spatially resolved environmental fate models: A review

Spatially resolved environmental models are important tools to introduce and highlight the spatial variability of the real world into modeling. Although various spatial models have been developed so far, yet the development and evaluation of these models remain a challenging task due to several difficulties related to model setup, computational cost, and obtaining high-resolution input data (e.g., monitoring and emission data). For example, atmospheric transport models can be used when high resolution predicted concentrations in atmospheric compartments are required, while spatial multimedia fate models may be preferred for regulatory risk assessment, life cycle impact assessment of chemicals, or when the partitioning of chemical substances in a multimedia environment is considered. The goal of this paper is to review and compare different spatially resolved environmental models, according to their spatial, temporal and chemical domains, with a closer insight into spatial multimedia fate models, to achieve a better understanding of their strengths and limitations. This review also points out several requirements for further improvement of existing models as well as for their integration.

Estimating temporal and spatial levels of PAHs in air using rain samples and SPME analysis: Feasibility evaluation in an urban scenario

There is a growing interest in evaluating the role of concentration changes of contaminants in temporal and spatial gradients. This is often relevant for fast moving environmental phases such as air and water. In this paper, small volumes of rainwater were sampled as proxy for air concentrations of Polycyclic Aromatic Hydrocarbons (PAHs): rain was collected in three sampling sites (high traffic, restricted traffic and a low traffic zone) in Como. Solid phase micro extraction (SPME) was used for the extraction to reduce required sample volumes, allowing the acquisition of more samples in time. Rain samples highlighted a spatial and temporal variability along a traffic gradient in the Como city, especially for the most abundant PAH, e.g. phenanthrene. Air concentrations were then estimated from rain concentrations. The results show that this is a cheap and promising method, although requiring rainfall/snowfall conditions, that can be used to perform monitoring campaign of air concentrations at a higher temporal and spatial resolution than the adopted standard methods (e.g. high-volume air samplers). The results could be employed for evaluation of the exposure, emission profiles and calibration of fate models.

A new dataset of PCB half-lives in soil: Effect of plant species and organic carbon addition on biodegradation rates in a weathered contaminated soil

This paper presents a new dataset of Polychlorinated Biphenyls (PCBs) half-lives in soil. Data were obtained from a greenhouse experiment performed with an aged contaminated soil under semi-field conditions, collected from a National Relevance Site (SIN) located in Northern Italy (SIN Brescia-Caffaro). Ten different treatments (combination of seven plant species and different soil conditions) were considered together with the respective controls (soil without plants). PCB concentration reduction in soil was measured over a period of 18 months to evaluate the ability of plants to stimulate the biodegradation of these compounds. Tall fescue, tall fescue cultivated together with pumpkin and tall fescue amended with compost reduced more than the 50% of the 79 measured PCB congeners, including the most chlorinated ones (octa to deca-PCBs). However, the data obtained showed that no plant species was uniquely responsible for the effective degradation of all isomeric classes and congeners. The obtained half-lives ranged from 1.3 to 5.6 years and were up to a factor of 8 lower compared to generic HL values reported in literature. This highlighted the importance of cultivation and plant-microbe interactions in speeding up the PCB biodegradation. This new dataset could contribute to substantially improve the predictions of soil remediation time, multimedia fate and the long-range transport of PCBs. Additionally, the half-lives obtained here can also be used in the evaluation of the food chain transfer of these chemicals, and finally the exposure and potential for effects on ecosystems.

ChemFate: A fate and transport modeling framework for evaluating radically different chemicals under comparable conditions

With the ever-increasing development on novel chemicals and materials, with radically different properties and environmental behavior, it is challenging to compare their environmental behavior under similar conditions. For example, pesticides may be non-ionizable organics, ionizable organics, metal ion-based, or nanomaterials. These very different chemicals behave very differently. To date, no single modeling platform can handle all adequately, in a dynamic framework that accounts for actual variability in meteorology, rates of release to the environment, and the actual characteristics of the region of interest. Here we present ChemFate, a framework to address this challenge, by incorporating four different fate and transport models, each to address these four classes of chemicals (non-ionizable organics, ionizable organics, metal ion-based, or nanomaterials). We build upon established models, but have incorporated a number of additional processes. After demonstrating that the individual models comparable favorably with observed data and the previous models, under similar conditions, we conducted a case study with four radically different fungicides, used in the Central Valley, California. We found that although the concentrations of the non-ionizable and ionizable organic fungicide spike after application and temporarily may exceed toxicity thresholds for Daphnia Magna, they do not accumulate over time, while the metal ion and metallic nanoparticle result in increasing accumulation of Cu²⁺, eventually exceeding the toxicity threshold during runoff events. This case study demonstrates the value of a framework that allows the comparison of different classes of chemicals under the same conditions.

New Data Set of Polychlorinated Dibenzo-p-dioxin and Dibenzofuran Half-Lives: Natural Attenuation and Rhizoremediation Using Several Common Plant Species in a Weathered Contaminated Soil

In this paper, a new data set of polychlorinated dibenzo-p-dioxin and dibenzofuran (PCDD/Fs) half-lives (HLs) in soil is presented. Data are derived from a greenhouse experiment performed with an aged contaminated soil under semi-field conditions, obtained from a National Relevance Site (SIN) located in Northern Italy (SIN Brescia-Caffaro). Ten different treatments (combination of seven plant species with different soil conditions) were considered together with the respective controls (soil without plants). The ability of the plants to stimulate the biodegradation of these compounds was evaluated by measuring the PCDD/F concentration reduction in soil over a period of 18 months. The formation of new bound residues was excluded by using roots as a passive sampler of bioaccessible concentrations. The best treatment which significantly reduced PCDD/F concentrations in soil was the one with Festuca arundinacea (about 11-24% reduction, depending on the congener). These decreases reflected in HLs ranging from 2.5 to 5.8 years. Simulations performed with a dynamic air-vegetation-soil model (SoilPlusVeg) confirmed that these HLs were substantially due to biodegradation rather than other loss processes. Because no coherent PCDD/F degradation HL data sets are currently available for soil, they could substantially improve the predictions of soil remediation time, long-range transport, and food chain transfer of these chemicals using multimedia fate models.

Modelling herbicides mobility in amended soils: Calibration and test of PRZM and MACRO

Addition of organic residues to soil is a current farming practice but it is not considered in the modelling studies for pesticide risk assessment at regulatory level despite its potential impact on the pesticide dynamics in soil. Thus, the objective of this work was to examine and to compare the ability of PRZM and MACRO pesticide fate models to simulate soil water content, and bromide (Br^-, tracer), chlorotoluron and flufenacet concentrations in the soil profiles (0-100 cm) of one agricultural soil, unamended (control soil, S), amended with spent mushroom substrate (S + SMS) or amended with green compost (S + GC). Based on a two-year field-scale dataset, the models were first calibrated against measurements of water and solutes contents in the soil profiles (first year) and then tested without any further model calibration by comparison with the field observations of the second year. In general, the performance of MACRO to simulate the whole dataset in the three soil treatments was higher than that of PRZM. MACRO simulated satisfactorily the water dynamics along the soil profiles whereas it was poorly described by the capacity model PRZM. Both models predicted very well the Br^- mobility in control and amended soils after dispersion parameters were fitted to observations. No calibration was necessary to reproduce correctly herbicides vertical distribution in the control soil profile. In the amended soils, MACRO simulations were highly correlated to the observed vertical distribution of flufenacet and chlorotoluron, but calibration of the K_d of chlorotoluron was needed. On the contrary, modelling with PRZM required calibration of K_d and DT₅₀ of both herbicides to obtain an acceptable agreement between observations and predictions in the amended soils. K_d and DT₅₀ calibration was based on the initial dissolved organic carbon contents (DOC) of amended soils. It allowed to take into account the processes that decrease the herbicides sorption on the soil and enhance their bioavailability, but that are not described in PRZM and MACRO (such as the formation of herbicide-DOC mobile complexes). This work showed that models such as PRZM and MACRO are able to simulate the fate of pesticides in amended soils. However, before using these models as predictive tools in large amended soil conditions, and especially in the regulatory context, further modelling studies should focus on other pedoclimatic-pesticides-organic residues combinations, and on longer periods.

Plants radically change the mobility of PCBs in soil: Role of different species and soil conditions

The mobility of Polychlorinated Biphenyls (PCBs) in soil cultivated with different plant species was evaluated by means of a column experiment to investigate the specific plant influence on PCB environmental fate and the potential for leaching. The soil was collected at a National Relevance Site for remediation located in Northern Italy (SIN Brescia-Caffaro) and underwent a rhizoremediation treatment for 18 months with different plant species (Festuca arundinacea, Cucurbita pepo ssp pepo and Medicago sativa). The same but unplanted soil was also considered as control for comparison. The columns were leached with tap water and PCB concentrations were measured in the leachate after 7 days of soil/water contact. Soil previously cultivated with different plant species exhibited statistically different behavior in terms of chemical leaching among the different fractions. Total PCB bulk concentrations ranged from 24 to 219 ng/L. Leachate samples were enriched in tetra- to hepta-PCBs. While PCB concentrations in the dissolved phases varied within a factor of 2 between controls and treatments, PCB associated to particulate organic carbon (POC) differed by more than one order of magnitude. More specifically, Medicago sativa enriched the soil with POC doubling PCB leaching with respect to the other plant species and the unplanted controls.

Modelling peak exposure of pesticides in terrestrial and aquatic ecosystems: importance of dissolved organic carbon and vertical particle movement in soil

In the present work, an existing vegetation/air/litter/soil model (SoilPlusVeg) was modified to improve organic chemical fate description in terrestrial/aquatic ecosystems accounting for horizontal and vertical particulate organic carbon (POC) transport in soil. The model was applied to simulate the fate of three pesticides (terbuthylazine, chlorpyrifos and etofenprox), characterized by increasing hydrophobicity (log K_OW from about 3 to 7), in the soil compartment and more specifically, their movement towards surface and groundwater through infiltration and runoff processes. The aim was to evaluate the role of dissolved organic carbon (DOC) and POC in the soil in influencing the peak exposure of pesticides in terrestrial/aquatic ecosystems. Simulation results showed that while terbuthylazine and chlorpyrifos dominated the free water phase (C_W-FREE) of soil, etofenprox was mainly present in soil porewater as POC associated chemical. This resulted in an increase of this highly hydrophobic chemical movement towards groundwater and surface water, up to a factor of 40. The present work highlighted the importance of DOC and POC as an enhancer of mobility in the water of poor or very little mobile chemicals. Further studies are necessary to evaluate the bioavailability change with time and parameterize this process in multimedia fate models.

Rhizoremediation of weathered PCBs in a heavily contaminated agricultural soil: Results of a biostimulation trial in semi field conditions

This paper describes the results of a rhizoremediation greenhouse experiment planned to select the best plant species and soil management for the bioremediation of weathered polychlorinated biphenyls (PCBs). We evaluated the ability of different plant species to stimulate activity and diversity of the soil microbial community leading to the reduction of PCB concentrations in a heavily contaminated soil (at mg kg-1 dw level), of the national priority site for remediation (SIN) "Brescia-Caffaro" in Italy. Biostimulation was determined in large size (6kg) pots, to reflect semi-field conditions with a soil/root volume ratio larger than in most rhizoremediation experiments present in the literature. In total, 10 treatments were tested in triplicates comparing 7 plant species (grass and trees) and 5 soil/cultivation conditions (i.e., only one plant species, plant consociation, redox cycle, compost or ammonium thiosulfate addition) with the appropriate unplanted controls. After 18months of biostimulation the overall reduction of total PCBs varied between 14 and 20%. Microbial analysis revealed a shift in the microbial community structure over time and showed that all the planted treatments significantly enhanced microbial hydrolytic activity and the abundance of bacterial populations, including potential PCB degraders, in the soil surrounding plant roots. The plant species most effective in reducing the contaminant concentrations were Festuca arundinacea cultivated adding compost or in consociation with Cucurbita pepo ssp. pepo and Medicago sativa cultivated with Rhizobium spp. and mycorrhizal fungi; they reduced total PCB concentrations of about 20% and showed the significant depletion of a high number of PCB congeners (29, 37 and 23, respectively, out of the 79 measured). Our results suggest that these plant species are particularly efficient in increasing soil PCB bioavailability and in stimulating microbial degradation. They could be used in field rhizoremediation strategies to enhance the natural attenuation process and reduce PCB levels in historically contaminated sites.

Identification of Sulfonated and Hydroxy-Sulfonated Polychlorinated Biphenyl (PCB) Metabolites in Soil: New Classes of Intermediate Products of PCB Degradation?

In this paper we describe the identification of two classes of contaminants: sulfonated-PCBs and hydroxy-sulfonated-PCBs. This is the first published report of the detection of these chemicals in soil. They were found, along with hydroxy-PCBs, in soil samples coming from a site historically contaminated by the industrial production of PCBs and in background soils. Sulfonated-PCB levels were approximately 0.4-0.8% of the native PCB levels in soils and about twice the levels of hydroxy-sulfonated-PCBs and hydroxy-PCBs. The identification of sulfonated-PCBs was confirmed by the chemical synthesis of reference standards, obtained through the sulfonation of an industrial mixture of PCBs. We then reviewed the literature to investigate for the potential agents responsible for the sulfonation. Furthermore, we predicted their physicochemical properties and indicate that, given the low pK_a of sulfonated- and hydroxy-sulfonated-PCBs, they possess negligible volatility, supporting the case for in situ formation from PCBs. This study shows the need of understanding their origin, their role in the degradation path of PCBs, and their fate, as well as their (still unknown) toxicological and ecotoxicological properties.

Accumulation and fate processes of organochlorine pesticides (OCPs) in soil profiles in Mt. Shergyla, Tibetan Plateau: A comparison on different forest types

Previous work documented that forest plays an important role in the deposition of persistent organic pollutants (POPs) in the southeast Tibetan Plateau (TP) due to the "forest filter effect". However, forest types in the southeast TP are entirely different and the influence on POPs fate and forest filter effect by different forests remains unclear. This study focused on the distribution and transfer of organochlorine pesticides (OCPs) in soil of different forest types (quercus, birch, fir, and spruce dominated forests) in Mt. Shergyla, southeast TP under similar environmental and meteorological conditions. Total levels of ∑HCHs, ∑DDTs and HCB in soils ranged from < LOD to 2.25 ng/g dry weight (dw), < LOD-10.2 ng/g dw, and < LOD-0.95 ng/g dw, respectively. Concentrations of OCPs in humus layers were significantly higher than those in mineral layers in the four forest types. Relatively higher ∑DDTs concentrations were found in soil profile of broadleaved birch forest, while higher concentrations of ∑HCHs and HCB were found in soil profile of coniferous fir forest, and the same trend was observed in fresh leaf samples. Air-to-ground fluxes and mobility of OCPs in the four forest types were also evaluated. Relatively higher fluxes were found in fir forests than in other forest types, suggesting that fir forest could be more effective to transfer OCPs from the air into soil in the southeast TP. The findings in this study would be helpful for improving model simulation of POPs fate in different forest ecosystem.

A review of multimedia transport and fate models for chemicals: Principles, features and applicability

The frequent use of chemicals has caused ecosystems and humans to be threatened due to their discharge into the environment. Multimedia environmental fate models could provide a comprehensive picture of transport behaviour and fate for organic chemicals in multiple environmental media. They have been designed and widely used for chemical risk assessment, chemical ranking and management support, and determination of chemical bioaccumulation. This study reviewed the principles, features and applicability of recent commonly used multimedia fate models from peer-reviewed literature. Fugacity-based and concentration-based models are now widely adopted for use in chemical fate evaluation, while they are more appropriate for volatile and semi-volatile chemicals. Or the fugacity-based models can use aquivalence equilibrium criterion to cations, anions and involatile chemicals. The MAMI and SESAMe models based on activity approach are applicable to neutral and ionizable molecules. However, interactions of ionic species with other water solutes are not taken into account in these models. Additionally, they could not directionally simulate how chemicals transported form one grid to another. Future attention should be focused on the reliability of transfer behaviour and fate of ionizable chemicals, as integrating the advantages of these two kinds of models into a reconstructed one may be a better choice. In a word, environmental multimedia models have been beneficial tools for chemical control and management, risk and effect estimation, and decision supporting.

A review of the predictive models estimating association of neutral and ionizable organic chemicals with dissolved organic carbon

Dissolved organic carbon (DOC) plays a key role in environmental transport, fate and bioavailability of organic chemicals in terrestrial and aquatic ecosystems. Predicting the association of contaminants to DOC is therefore crucial in modelling chemical exposure and risk assessment. The models proposed so far to describe interaction mechanisms between chemicals and DOC and the most influential variables have been reviewed. The single-parameter linear free energy relationships (sp-LFERs) and the poly-parameter linear free energy relationships (pp-LFERs) in the form of linear solvation energy relationships (LSERs) currently available in literature for estimating dissolved organic carbon/water partition (K_DOC) and distribution (D_DOC) coefficients for organic chemicals were discussed, and limits of the existing approaches explored. For neutral chemicals many predictive equations are currently available in literature, but the quality of the input data on which they are based is often questionable, due to the lack of an unequivocal definition of DOC among different references and to the different and often unreliable K_DOC measurement method. For ionizable chemicals instead there is a substantial lack of predictive approaches that need to be fulfilled since just few models are nowadays available to predict D_DOC of ionized species. This paper reviews the current approaches for neutral and ionizable chemicals proposing guidelines to select conditions for obtaining reliable data and predictive equations for an improved estimation of K_DOC and D_DOC.

Predicting dissolved organic carbon partition and distribution coefficients of neutral and ionizable organic chemicals

Estimating K_DOC (dissolved organic carbon/water partition coefficient) and D_DOC (dissolved organic carbon/water distribution coefficient) of neutral and ionizable organic chemicals is a crucial task for assessing mobility, modelling transport, environmental fate of a variety of chemicals and for evaluating their bioavailability in terrestrial and aquatic environments. A critical literature search of reliability-selected K_DOC and D_DOC values was performed to setup novel predictive relationships for K_DOC and D_DOC of neutral and ionizable organic chemicals. This goal was pursued by using: 1) LSER (linear solvation energy relationship) models to predict K_DOC for neutral chemicals using Abraham solute parameters calculated for different DOC sources (all DOC sources together, soil porewater, surface water, wastewater and Aldrich humic acid (HA)); 2) linear regressions for predicting D_DOC of organic acids from the octanol/water partition coefficient (Log K_OW or Log P) and the dissociation constant (pKa), accounting separately for the contribution of the neutral and ionic fraction. The proposed models predicted Log K_DOC and D_DOC values within a root mean square deviation (RMSD) generally smaller than 0.3 log units.

How good are the predictions of mobility of aged polychlorinated biphenyls (PCBs) in soil? Insights from a soil column experiment

A column leaching experiment was performed to evaluate the influence of some relevant environmental factors (soil/water contact time, temperature, saturation) on mobility of aged polychlorinated biphenyls (PCBs) in soil together with transport mediated by dissolved organic carbon (DOC) and mobile organic carbon (OC) coated fine particles/colloids. Consecutive fractions of leachates were collected after a variable pre-equilibration time (2, 5, 7, 48 days), using leaching solutions with different DOC content (tap water vs. Aldrich humic acid), in saturated vs. field capacity conditions and at different temperatures (25 °C vs. 15 °C). The data obtained were compared to the predicted values using a multimedia model (SoilPlusVeg) to evaluate model behaviour. Contact time and temperature determined a relevant effect on DOC and particle/colloid availability, with significant variations in leachate concentrations (up to 1 order of magnitude), typically overlooked by most environmental fate models. Results obtained at different temperatures show a modulation of the DOC/particles production with temperature and therefore the role of temperature changes in the environmental scenarios (e.g. seasonal variations). Transport of PCBs enhanced by Aldrich DOC was not linearly correlated to chemical hydrophobicity but revealed a threshold to ~Log K_OW 6.5, likely because of the slow sorption kinetics of more hydrophobic chemicals. Additionally, variation of the saturation conditions (e.g. drying-wetting cycles) can determine contamination peaks at the beginning of an irrigation/rainfall event because of the soil/water equilibration. Model simulations, even when including DOC in the water phase, but not accounting for the particle/colloidal transport and sorption/desorption kinetics, mismatched the ratio of dissolved vs. DOC-associated and particle-associated PCBs and substantially underpredicted concentrations, especially for the high chlorinated congeners. The results indicated that some of the common assumptions and paradigms in fate modelling of such hydrophobic compounds should be revisited and models updated.

Pesticide fate in cultivated mountain basins: The improved DynAPlus model for predicting peak exposure and directing sustainable monitoring campaigns to protect aquatic ecosystems

Agricultural activities can involve the use of plant protection products (PPPs) and the use of such chemicals can occur near surface waters bodies, thus creating a potential for adverse effects on aquatic ecosystems. In mountain watersheds, where runoff fluxes are particularly rapid due to side slopes, exposure is generally characterized by short but intense concentration peaks. Monitoring campaigns are often inadequate or too expensive to be carried out and modelling tools are therefore vital for exposure assessment and their use is encouraged by current legislation. However, currently adopted models and scenarios (e.g., FOCUS for PPPs) are often too conservative and/or "static" to accurately capture exposure variability, and the need for more realistic and dynamic tools is now one of the major challenges for risk assessment. In a previous work, the new fate model DynAPlus was developed to improve pesticide fate predictions in cultivated mountain basins and was successfully evaluated against chlorpyrifos water concentrations measured in a mountain stream in Northern Italy. However, the need for some model improvements (e.g., the inclusion of dissolved organic matter and macrophytes in water) was highlighted. In this work, DynAPlus was improved by replacing the water-sediment unit with ChimERA fate, a recently-published model capable of predicting bioavailable chemical concentrations in shallow water environments accounting for the presence and temporal variations of particulate/dissolved organic carbon and primary producers. The model was applied to preliminarily characterize the risk associated to the use of four PPPs (two insecticides and two fungicides) in a sub-basin of the Adda River (Valtellina Valley, Northern Italy), surrounded by apple orchards. Results revealed the potential magnitude of exposure peaks for the four PPPs and suggested that monitoring campaigns should prioritize, in the selected case study, chlorpyrifos, etofenprox and fluazinam. The potential role of DynAPlus in providing more realistic exposure predictions for ecological risk assessment, as well as for planning efficient monitoring campaigns and help pesticide management practices, was also stressed.

Improving the SoilPlusVeg model to evaluate rhizoremediation and PCB fate in contaminated soils

Tools to predict environmental fate processes during remediation of persistent organic pollutants (POPs) in soil are desperately needed since they can elucidate the overall behavior of the chemical and help to improve the remediation process. A dynamic multimedia fate model (SoilPlusVeg) was further developed and improved to account for rhizoremediation processes. The resulting model was used to predict Polychlorinated Biphenyl (PCB) fate in a highly contaminated agricultural field (1089 ng/g d.w.) treated with tall fescue (Festuca arundinacea), a promising plant species for the remediation of contaminated soils. The model simulations allowed to calculate the rhizoremediation time (about 90 years), given the available rhizoremediation half-lives and the levels and fingerprints of the PCB congeners, to reach the legal threshold, to show the relevance of the loss processes from soil (in order of importance: degradation, infiltration, volatilization, etc.) and their dependence on meteorological and environmental dynamics (temperature, rainfall, DOC concentrations). The simulations showed that the effective persistence of PCBs in soil is deeply influenced by the seasonal variability. The model also allowed to evaluate the role of DOC as a possible enhancer of PCB degradation as a microorganism "spoon feeder" of PCBs in the soil solution. Additionally, we preliminary predicted how the contribution of PCB metabolites could modify the PCB fingerprint and their final total concentrations. This shows that the SoilPlusVeg model could be used in selecting the best choices for a sustainable rhizoremediation of a POP contaminated site.

Pesticide fate modelling in conservation tillage: Simulating the effect of mulch and cover crop on S-metolachlor leaching

Conservation tillage practices mainly based on cover crops and no-tillage with accumulation of crop residues at the soil surface (mulch) modify the environmental fate of pesticides. However, only few pesticide fate models are able to consider mulch of crop residues as well as the effect of intermediate cover crops. Thus, the objective was to develop an approach to model the effects of crop residues left at the soil surface and cover crops on the fate of pesticides. This approach consisted in (1) considering the crop residues as a soil layer with specific physical, hydrodynamic and pesticide-reactivity properties close to that of a high organic content soil layer, and (2) introducing a correction factor of the potential evapotranspiration, estimated through a calibration step, to take into account the reduction of soil evaporation by the presence of a mulch. This approach was developed using MACRO as support pesticide model. To assess the model performances, we used the data from a field experiment designed in an irrigated maize monoculture under conservation tillage. Soil water content, water percolates, soil temperature and S-metolachlor herbicide concentrations in the leachate at 1m depth were measured during two years. The approach chosen to simulate the mulch effects allowed MACRO to make acceptable predictions of the observed water percolation, soil temperature and to a less extent herbicide leaching. However, it showed a poor performance to simulate the soil water content. Results are discussed in terms of further modelling options to better assess the environmental risks of pesticides under conservation tillage. This approach remains to be tested against various soils, crops, pesticides and types of mulch.

Predicting pesticide fate in small cultivated mountain watersheds using the DynAPlus model: Toward improved assessment of peak exposure

The use of plant protection products (PPPs) in agricultural areas implies potential chemical loadings to surface waters, which can pose a risk to aquatic ecosystems and human health. Due to the spatio-temporal variability of PPP applications and of the processes regulating their transport to surface waters, aquatic organisms are typically exposed to pulses of contaminants. In small mountain watersheds, where runoff fluxes are more rapid due to the steep slopes, such exposure peaks are particularly likely to occur. In this work, a spatially explicit, dynamic model for predicting pesticide exposure in surface waters of cultivated mountain basins (DynAPlus) has been developed. The model has been applied to a small mountain watershed (133km²) located in the Italian Eastern Alps and characterized by intensive agriculture (apple orchards) around the main river and its tributaries. DynAPlus performance was evaluated for chlorpyrifos through experimental monitoring, using samples collected during the 2011 and 2012 productive seasons. The comparison between predictions and measurements resulted in a good agreement (R²=0.49, efficiency factor 0.60), although a more accurate spatial information in the input scenario (e.g., field-specific applications, rainfall amount, soil properties) would dramatically improve model performance. A set of illustrative simulations performed for three PPPs highlighted the potential role of DynAPlus in improving exposure predictions for ecological risk assessment and pesticide management practices (e.g., for active ingredient and application rate selection), as well as for planning efficient monitoring campaigns and/or interpreting monitoring data. However, some model improvements (e.g., solid erosion and transport) and a more thorough model validation are desirable to enlarge the applicability domain.

Environmental fate and exposure models: advances and challenges in 21st century chemical risk assessment

Environmental fate and exposure models are a powerful means to integrate information on chemicals, their partitioning and degradation behaviour, the environmental scenario and the emissions in order to compile a picture of chemical distribution and fluxes in the multimedia environment. A 1995 pioneering book, resulting from a series of workshops among model developers and users, reported the main advantages and identified needs for research in the field of multimedia fate models. Considerable efforts were devoted to their improvement in the past 25 years and many aspects were refined; notably the inclusion of nanomaterials among the modelled substances, the development of models at different spatial and temporal scales, the estimation of chemical properties and emission data, the incorporation of additional environmental media and processes, the integration of sensitivity and uncertainty analysis in the simulations. However, some challenging issues remain and require research efforts and attention: the need of methods to estimate partition coefficients for polar and ionizable chemical in the environment, a better description of bioavailability in different environments as well as the requirement of injecting more ecological realism in exposure predictions to account for the diversity of ecosystem structures and functions in risk assessment. Finally, to transfer new scientific developments into the realm of regulatory risk assessment, we propose the formation of expert groups that compare, discuss and recommend model modifications and updates and help develop practical tools for risk assessment.

Do environmental dynamics matter in fate models? Exploring scenario dynamics for a terrestrial and an aquatic system

Nowadays, there is growing interest in inserting more ecological realism into risk assessment of chemicals. On the exposure evaluation side, this can be done by studying the complexity of exposure in the ecosystem, niche partitioning, e.g. variation of the exposure scenario. Current regulatory predictive approaches, to ensure simplicity and predictive ability, generally keep the scenario as static as possible. This could lead to under or overprediction of chemical exposure depending on the chemical and scenario simulated. To account for more realistic exposure conditions, varying temporally and spatially, additional scenario complexity should be included in currently used models to improve their predictive ability. This study presents two case studies (a terrestrial and an aquatic one) in which some polychlorinated biphenyls (PCBs) were simulated with the SoilPlusVeg and ChimERA models to show the importance of scenario variation in time (biotic and abiotic compartments). The results outlined the importance of accounting for planetary boundary layer variation and vegetation dynamics to accurately predict air concentration changes and the timing of chemical dispersion from the source in terrestrial systems. For the aquatic exercise, the results indicated the need to account for organic carbon forms (particulate and dissolved organic carbon) and vegetation biomass dynamics. In both cases the range of variation was up to two orders of magnitude depending on the congener and scenario, reinforcing the need for incorporating such knowledge into exposure assessment.

Rhizoremediation half-lives of PCBs: Role of congener composition, organic carbon forms, bioavailability, microbial activity, plant species and soil conditions, on the prediction of fate and persistence in soil

Polychlorinated biphenyls (PCBs) are persistent organic pollutants widely produced and used in many countries until the increasing concern about their environmental risk lead to their ban in the 1980s. Although their emissions decreased, PCBs are nowadays still present in the environment and can be reemitted from reservoir compartments such as contaminated soils. In the last two decades, there has been a growing interest in bioremediation technologies that use plants and microorganisms (i.e. rhizoremediation) to degrade organic chemicals in contaminated sites. Different studies have been conducted to investigate the potential of plant-microbe interactions in the remediation of organic chemical contaminated soils. They range from short-term and laboratory/greenhouse experiments to long-term and field trials and, when correctly set up, they could provide useful data such as PCB rhizoremediation half-lives in soil. Such type of data are important input parameters for multimedia fate models that aim to estimate the time requested to achieve regulatory thresholds in a PCB contaminated site, allowing to draw up its remediation plan. This review focuses on the main factors influencing PCB fate, persistence and bioavailability in soil including PCB mixture congener composition, soil organic carbon forms, microorganism activity, plant species and soil conditions. Furthermore, it provides an estimate of rhizoremediation half-lives of the ten PCB families starting from the results of literature rhizoremediation experiments. Finally, guidance to perform appropriate experiments to obtain comparable, accurate and useful data for fate estimation is proposed.

SoilPlusVeg: An integrated air-plant-litter-soil model to predict organic chemical fate and recycling in forests

Current modelling approaches often ignore the dynamics of organic chemicals uptake/release in forest compartments under changing environmental conditions and may fail in accurately predict exposure to chemicals for humans and ecosystems. In order to investigate the influence of such dynamics on predicted concentrations in forest compartments, as well as, on air-leaf-litter fluxes, the SoilPlusVeg model was developed including a forest compartment (root, stem, leaves) in an existing air-litter-soil model. The accuracy of the model was tested simulating leaf concentrations in broadleaf woods located in Northern Italy and resulted in satisfying model performance. Illustrative simulations highlighted the "dual behaviour" of both leaf and litter compartments. Leaves appeared to behave as "filters" of air contaminants but also as "dispensers", being deposition flux exceeded by volatilization flux in some periods of the day. Similarly, litter seemed to behave as a dynamic compartment which could accumulate and then release contaminants recharging air and vegetation. In just 85days, litter could lose due to volatilization, diffusion to depth and infiltration processes, from 6% to 90% of chemical amount accumulated over 1year of exposure, depending on compound physical and chemical properties. SoilPlusVeg thus revealed to be a powerful tool to understand and estimate chemical fate and recycling in forested systems.

Differentiating current and past PCB and PCDD/F sources: The role of a large contaminated soil site in an industrialized city area

Cities and contaminated areas can be primary or secondary sources of polychlorinated biphenyls (PCBs), dibenzo-p-dioxins (PCDDs), dibenzofurans (PCDFs), and other chemicals, into air and soil and can influence the regional level of some of these pollutants. In a contaminated site, the evaluation of such emissions can be crucial in the choice of the remediation technology to be adopted. In the city of Brescia (Northern Italy), more than 100 ha of agricultural areas were contaminated with PCBs, PCDD/Fs and heavy metals, originating from the activities of a former PCB factory. In order to evaluate the current emissions of PCBs and PCDD/Fs from the contaminated site, in a location where other current sources are present, we compared measured and predicted air concentrations, resulting from chemical volatilization from soils as well as fingerprints of Brescia soils and of soils contaminated by specific sources. The results confirm that the contaminated area is still a current and important secondary source of PCBs to the air, and to a lesser extent of PCDFs (especially the more volatile), but not for PCDDs. PCBs in soils have fingerprints similar to highly chlorinated mixtures, indicating contamination by these mixtures and/or a long weathering process. PCB 209 is also present at important levels. PCDD fingerprints in soil cannot be related to current emission sources, while PCDFs are compatible to industrial and municipal waste incineration, although weathering and/or natural attenuation may have played a role in modifying such soil fingerprints. Finally, we combined chemical and microbiological analyses to provide an integrated approach to evaluate soil fingerprints and their variation in a wider perspective, which accounts for the mutual effects between contamination and soil microbiota, a pivotal hint for addressing in situ bioremediation activities.

Modeling long-term uptake and re-volatilization of semi-volatile organic compounds (SVOCs) across the soil-atmosphere interface

Soil-atmosphere exchange is important for the environmental fate and atmospheric transport of many semi-volatile organic compounds (SVOCs). This study focuses on modeling the vapor phase exchange of semi-volatile hydrophobic organic pollutants between soil and the atmosphere using the multicomponent reactive transport code MIN3P. MIN3P is typically applied to simulate aqueous and vapor phase transport and reaction processes in the subsurface. We extended the code to also include an atmospheric boundary layer where eddy diffusion takes place. The relevant processes and parameters affecting soil-atmosphere exchange were investigated in several 1-D model scenarios and at various time scales (from years to centuries). Phenanthrene was chosen as a model compound, but results apply for other hydrophobic organic compounds as well. Gaseous phenanthrene was assumed to be constantly supplied to the system during a pollution period and a subsequent regulation period (with a 50% decline in the emission rate). Our results indicate that long-term soil-atmosphere exchange of phenanthrene is controlled by the soil compartment - re-volatilization thus depends on soil properties. A sensitivity analysis showed that accumulation and transport in soils in the short term is dominated by diffusion, whereas in the long term groundwater recharge and biodegradation become relevant. As expected, sorption causes retardation and slows down transport and biodegradation. If atmospheric concentration is reduced (e.g. after environmental regulations), re-volatilization from soil to the atmosphere occurs only for a relatively short time period. Therefore, the model results demonstrate that soils generally are sinks for atmospheric pollutants. The atmospheric boundary layer is only relevant for time scales of less than one month. The extended MIN3P code can also be applied to simulate fluctuating concentrations in the atmosphere, for instance due to temperature changes in the topsoil.

Investigating the need for complex vs. simple scenarios to improve predictions of aquatic ecosystem exposure with the SoilPlus model

A spatially-explicit version of the recent multimedia fate model SoilPlus was developed and applied to predict the runoff of three pesticides in a small agricultural watershed in north-eastern Italy. In order to evaluate model response to increasing spatial resolution, a tiered simulation approach was adopted, also using a dynamic model for surface water (DynA model), to predict the fate of pesticides in runoff water and sediment, and concentrations in river water. Simulation outputs were compared to water concentrations measured in the basin. Results showed that a high spatial resolution and scenario complexity improved model predictions of metolachlor and terbuthylazine in runoff to an acceptable performance (R(2) = 0.64-0.70). The importance was also shown of a field-based database of properties (i.e. soil texture and organic carbon, rainfall and water flow, pesticides half-life in soil) in reducing the distance between predicted and measured surface water concentrations and its relevance for risk assessment.

Accumulation kinetics and equilibrium partitioning coefficients for semivolatile organic pollutants in forest litter

Soils are important stores of environmentally cycling semivolatile organic contaminants (SVOCs) and represent relevant atmospheric secondary sources whenever environmental conditions favor re-emission. The exchange between air and soil is controlled by resistances posed by interfacial matrices such as the ubiquitously distributed vegetation litter. For the first time, this study focused on the experimental characterization of accumulation parameters for SVOCs in litter under real field conditions. The logarithm of the litter-air equilibrium partitioning coefficient ranged 6.8-8.9 and had a similar dependence on logKOA as that of plant foliage and soil data. Uptake and release rates were also KOA dependent with values (relevant for real environmental conditions) ranging 30,000-150,000 d(-1) and 0.0004-0.0134 d(-1), respectively. The overall mass transfer coefficient v controlling litter-air exchange (0.03-1.4 cm s(-1)) was consistent with previously reported data of v for foliage in forest canopies after normalization on leaf area index. Obtained data suggest that litter holds the potential for influencing atmospheric fugacity in proximity to soil, likely affecting overall exchange of SVOCs between the soil reservoir and the atmosphere.

Challenges for exposure prediction in ecological risk assessment

Evaluating organism exposure in the ecosystems is a difficult task and can be carried out measuring or predicting concentrations in the environment. Although current regulatory approaches favor a modeling approach, they either use a static representation of the environment and of the chemical discharge or a simplified dynamic approach (e.g., dealing with pesticides). Improving the ecological realism of exposure prediction offers a number of challenges. Some are related to the understanding of basic mechanisms such as bioavailability and the determination of internal exposure or the need to develop new paradigms for polar and ionized chemicals. Other issues are the need to provide monitoring data to understand the environmental fate of chemical mixtures, polar and ionized chemicals and metabolites, to understand the complexity of exposure in spatially and temporally variable environments. Exposure models require the development of suitable approaches to simulate the complexity of exposure in the ecosystems including the development of a variety of temporal and spatial scenarios and the integration of submodels (such as aquatic and terrestrial food webs). Finally, the integration of dynamic exposure and effect models is envisaged to fully carry out a more realistic ecological risk assessment.

Polycyclic aromatic hydrocarbons in soils from the Tibetan Plateau, China: distribution and influence of environmental factors

Thirty four sampling sites along an elevation transect in the Tibetan Plateau region were chosen. Soil cores were divided into several layers and a total of 175 horizon soil samples were collected from July to September 2011, for determination of polycyclic aromatic hydrocarbons (PAHs). The measured PAHs concentration in surface soils was 56.26 ± 45.84 ng g(-1), and the low molecular weight PAHs (2-3 rings) predominated, accounting for 48% and 35%. We analyzed the spatial (altitudinal and vertical) distribution of PAHs in soil, and explored the influence of related environmental factors. Total organic carbon (TOC) showed a controlling influence on the distribution of PAHs. PAH concentrations declined with soil depth, and the composition patterns of PAHs along soil depth indicated that the heavy PAHs tended to remain in the upper layers (0-10 cm), while the light fractions were transported downward more easily. PAHs inventories (8.77-57.92 mg m(-2)) for soil cores increased with mean annual precipitation, while the topsoil concentrations decreased with it. This implies that an increase in precipitation could transfer more PAHs from the atmosphere to the soil and further transport PAHs from the topsoil to deeper layers.

Integration of an atmospheric dispersion model with a dynamic multimedia fate model: development and illustration

Growing attention is devoted to understand the influence of the short-term variations in air concentrations on the environmental fate of semivolatile organic compounds (SVOCs) such as polycyclic aromatic hydrocarbons (PAHs). These variations are ascribable to factors such as temperature-mediated air-surface exchange and variability of planetary boundary layer (PBL) height and structure. But when investigating the fate of SVOCs at a local scale, further variability can derive from specific point source contributions. In this context, a new modeling approach (AirPlus) which integrates a previously developed model (AirFug) with an air dispersion model (AERMOD) is presented. The integrated model is illustrated for two PAHs in a Northern Italy scenario. Results show how chemical contributions deriving from background advective inflows, local emissions and a point source interact in an hourly-varying meteorological scenario to determine air concentration rapid changes and the consequent response of the soil compartment.

Climatic, biological, and land cover controls on the exchange of gas-phase semivolatile chemical pollutants between forest canopies and the atmosphere

An ecophysiological model of a structured broadleaved forest canopy was coupled to a chemical fate model of the air-canopy exchange of gaseous semivolatile chemicals to dynamically assess the short-term (hours) and medium term (days to season) air-canopy exchange and the influence of biological, climatic, and land cover drivers on the dynamics of the air-canopy exchange and on the canopy storage for airborne semivolatile pollutants. The chemical fate model accounts for effects of short-term variations in air temperature, wind speed, stomatal opening, and leaf energy balance, all as a function of layer in the canopy. Simulations showed the potential occurrence of intense short/medium term re-emission of pollutants having log K(OA) up to 10.7 from the canopy as a result of environmental forcing. In addition, relatively small interannual variations in seasonally averaged air temperature, canopy biomass, and precipitation can produce relevant changes in the canopy storage capacity for the chemicals. It was estimated that possible climate change related variability in environmental parameters (e.g., an increase of 2 °C in seasonally averaged air temperature in combination with a 10% reduction in canopy biomass due to, e.g., disturbance or acclimatization) may cause a reduction in canopy storage capacity of up to 15-25%, favoring re-emission and potential for long-range atmospheric transport. On the other hand, an increase of 300% in yearly precipitation can increase canopy sequestration by 2-7% for the less hydrophobic compounds.

Guidance on the selection of efficient computational methods for multimedia fate models

Dynamic multimedia fate models (MFMs) have to deal with the temporal and spatial variation of physical-chemical properties, environmental scenarios, and chemical emissions. In such complex simulation tools, an analytical solution is not practically feasible, and even a numerical approach requires a suitable choice of the method in order to obtain satisfying speed and reliability, particularly when certain combinations of modeling scenarios and chemical properties occur. In this paper, considering some examples of a wide range of realistic chemical and scenario properties, some sources of stiffness in MFM equations are pinpointed. Next, a comparison of the performances of several numerical schemes (chosen as representatives of three wide classes) is performed. The accuracy and the computational effort required by each method is evaluated, illustrating the general effectiveness of automatically adapted timesteps in numerical algorithms and the pros and cons of implicit timestepping. The results show that automatic error control methods can significantly improve the quality of the computed solutions and most often lead to relevant savings in computing time. Additionally, explicit and implicit methods are compared, indicating that an implicit method of medium order (around 5) is the best choice as a general purpose MFM computing engine.