Mass balance modelling of priority toxic chemicals within the great lakes toxic chemical decision support system: RateCon model results for Lake Ontario and Lake Erie

https://doi.org/10.1016/j.envsoft.2004.03.013Get rights and content

Abstract

The great lakes toxic chemical decision support system (GLTCDSS) has been developed in order to integrate all of the available information on Tier I and II chemicals in support of designing, implementing and post-auditing the zero discharge and virtual elimination strategies. A modified version of the rate constant model (RateCon) (J. Great Lakes Res. 20 (1994) 625), which is one of three mass balance models incorporated into the system, has been run for anthracene, benzo(a) pyrene (B(a)P), dieldrin, fluoranthene, lead, DDE, DDT, hexachlorobenzene (HCB), mercury, mirex, polychlorinated biphenyls (PCBs), dioxins (TCDDs), pentachlorophenol, and furans (TCDFs) for Lake Ontario. A subset of these has been run for Lake Erie, due to more limited loadings data being available. The steady-state model results indicate that both lakes are still net sinks for metals such as mercury and lead, and that for concentrations in the water column, the model predicted values fall within the observed ranges plus or minus the confidence intervals for PCBs, HCB, B(a)P, mirex, DDE, lead, and anthracene. Currently there is still insufficient data to allow a simple screening level model such as RateCon to be applied for octachlorostyrene, chlordane, toxaphene, heptachlor, 1,4-dichlorobenzene, 3,3′-dichlorobenzidine, hexachlorocyclohexane, 1,6-dinitropyrene, 1,8-dinitropyrene, and tributyl tin, and many of the PAH's. Concentrations to loadings calculations for Lake Erie indicate that either the lake is very far from steady-state and is responding to historically higher loadings or the estimated loadings are extremely underestimated for PCBs, lead, and fluoranthene, assuming that the measured concentrations at the outlet (Fort Erie) are representative of whole lake values.

Section snippets

Software availability

    Availability:

    The Great lakes Toxic Chemical Decision Support System software is currently only available to research partners of Environment Canada. The RAISON for Windows Decision Support System may be acquired through a license with the National Water Research Institute's NWRI Software.

    Contact details:

    E-mail: [email protected] or Fax: +1-905-336-4430.

Great lakes toxic chemical decision support system

The RAISON for Windows Decision Support System shell that this system is based upon has evolved over the past decade at the National Water Research Institute of Environment Canada (Lam and Swayne, 1993, Lam et al., 1995, Lam et al., 1998, Lam et al., 2004). Its development has been client driven and its design is devoted to providing generic software tools for fast prototyping and practical implementation of environmental decision support systems. The architecture is open in design. All of the

RateCon model

The RateCon model is a modified version of the model developed by Mackay et al. (1994). It consists of a whole lake model and a food chain model. All expressions for rates of chemical transport and transformation are written in ‘rate constant’ form, which allows for easy comparison of magnitudes of diverse processes. The differential mass balance equations that describe the amount of chemical in water (MW) and in the sediment (MS), in units of kg/year are:dMW/dt=EL+EA+MSk5−MW(k1+k2+k3+k4)=MWkIWd

Steady-state model applications

The model may be operated in a number of steady-state modes. There are four options:

  • 1.

    Forwards, loading to concentration: this is the most common mode and is used to calculate concentrations in all the compartments using loadings input by the user.

  • 2.

    Backwards, concentrations in both water and sediments to loading: this mode allows the user to determine what the loadings need to be in order to arrive at the water and sediment concentrations observed in the lake.

  • 3.

    Backwards, concentration in water only

Unsteady-state model applications

Historical loadings as well as projections of loadings (based upon current loadings reductions for each of the load categories) to the year 2020 have been developed for Lake Ontario for PCBs. Historical loadings data have been obtained or calculated from Thomas et al., 1987, Rapaport and Eisenreich, 1988, Halfon and Oliver, 1990. No recent attempts have been found in the literature to determine the historical loads. The projections of loadings to the year 2020 were developed based upon recent

Discussion

It is interesting to compare the 1995 steady-state results for Lake Ontario (Table 4) with those for Lake Erie (Table 13). Total mass of chemical in the system is predicted to be greater in Lake Ontario than in Lake Erie for PCBs, HCB, dieldrin, lead, lead, DDE, fluoranthene, and anthracene The reverse is true for B(a)P, mercury, and DDT, reflecting the different magnitudes and sources of loadings for these chemicals in the two basins. Total mass in the system is predicted to be approximately

Conclusions

Currently there is still insufficient data to allow a simple screening level model such as RateCon to be applied for octachlorostyrene, chlordane, toxaphene, heptachlor, 1,4-dichlorobenzene, 3,3′-dichlorobenzidine, hexachlorocyclohexane, 1,6-dinitropyrene, 1,8-dinitropyrene, and tributyl tin, and many of the PAH's. This includes loadings data, basic physical-chemical properties required as model coefficients, and ambient data required for model calibration and verification. This means that not

Acknowledgements

The authors wish to thank Dr Don Mackay for making available the source code to the rate constant model. We are most grateful to Dr Thomas Tseng (Environmental Protection Branch – Ontario Region, Environment Canada) and Mr Ian Smith (Ontario Ministry of Environment) for their involvement in the creation and development of this study. The Environmental Protection Branch – Ontario Region, Environment Canada, provided partial funding for this study. RAISON team members are thanked for their

References (100)

  • T.J. Kelly et al.

    Atmospheric and tributary inputs of toxic substances to Lake Erie

    J. Great Lakes Res

    (1991)
  • D. Mackay

    Modeling the long-term behavior of an organic contaminant in a large lake: application to PCBs in Lake Ontario

    J. Great Lakes Res

    (1989)
  • D. Mackay et al.

    A rate constant model of chemical dynamics in a Lake Ecosystem: PCB's in Lake Ontario

    J. Great Lakes Res

    (1994)
  • H.A. Morrison et al.

    The relative importance of species invasions and sediment disturbance in regulating chemical dynamics in western Lake Erie

    Ecol. Model

    (2000)
  • S. Painter et al.

    Sediment contamination in Lake Erie: a 25-year restrospective analysis

    J. Great Lakes Res

    (2001)
  • R.F. Pearson et al.

    Atmospheric inputs of polychlorinated dibenzo-p-dioxins and dibenzofurans to the great lakes: compositional comparison of PCDD and PCDF in sediments

    J. Great Lakes Res

    (1998)
  • M.R. Richard et al.

    PCB concentrations and congener composition in macrophytes and sediments in the St

    Lawrence River near Cornwall, Ontario. J. Great Lakes Res

    (1997)
  • S. Thompson et al.

    A modelling strategy for planning the virtual elimination of persistent toxic chemicals from the great lakes: an illustration of four contaminants in Lake Ontario

    J. Great Lakes Res

    (1999)
  • M. Velleux et al.

    A screening-level mass balance analysis of mirex transport and fate in the Oswego River

    J. Great Lakes Res

    (1995)
  • ARET (Accelerated Reduction/Elimination of Toxics) Voluntary Commitments to Action on Toxics through ARET, March 1995,...
  • Biberhofer, J.A., 1994. Concentrations and loadings of trace organic compounds measured in the St. Lawrence River...
  • Booty, W.G., 1999. GLTCDSS Technical User's Manual, NWRI Contribution No....
  • W.G. Booty et al.

    Great lakes toxic chemical decision support system

  • Boyd, D., Biberhofer, 1999. Large Volume Sampling at Six Lake Ontario Tributaries During 1997 and 1998. Project...
  • B.J. Brownawell et al.

    The role of resuspension in contaminant cycling in Lake Ontario

    Great Lakes Res. Rev

    (1996)
  • Canviro Consultants, 1988. Thirty Seven Municipal Water Pollution Control Plants: Pilot Monitoring Study. Report...
  • CCPA Reducing Emissions 3, 1994. Emissions Inventory and 5 Year...
  • Cohen, M., et al. 1995. Quantitative Estimation of the Entry of Dioxins, Furans and Hexachlorobenzene into the Great...
  • Data Interpretation Group, River Monitoring Committee. January 2000. Joint Evaluation of the Niagara River...
  • J.V. DePinto et al.

    What other ecosystem changes have zebra mussels caused in Lake Erie: potential bioavailability of PCBs

    Great Lakes Res. Rev

    (1997)
  • J.V. DePinto et al.

    Development of LOTOX2: Solids and PCB calibration and management applications

    Great Lakes Res. Rev

    (1998)
  • DePinto, J.V., Liu, S., Young, T.C., Booty, W.G., 1999. An Update of the Lake Ontario Contaminant Mass Balance Modeling...
  • D. DeVault et al.

    Polychlorinated dibenzofurans and polychlorinated dibenzo-p-dioxins in Great Lakes fish: a baseline and interlake comparison

    Environ. Toxicol. Chem

    (1989)
  • E.P. Dobson et al.

    Increased deposition of organic matter, polychlorinated biphenyls, and cadmium by zebra mussels (Dreissena polymorpha) in western Lake Erie

    Can. J. Fish. Aquat. Sci

    (1998)
  • Eisenreich, S.J., Strachan, W.M.J., 1992. Estimating Atmospheric Deposition of Toxic Substances to the Great Lakes; An...
  • ENVIRODAT, Canada Centre for Inland Waters, Environment...
  • M.E. Fox et al.

    Loadings of PCBs and PAHs from Hamilton Harbour to Lake Ontario

    Wat. Qual. Res. J. Canada

    (1996)
  • K.A. Golden et al.

    Accumulation and preliminary inventory of organochlorines in Great Lakes sediment

    Wat. Sci. Technol

    (1994)
  • Halfon, E, Schito, N., 1993. Organic Contaminants in Lake Ontario, 1968–1991: A review and a Data Base. In: Reducing...
  • Harris, R.C., Gherini, S.A., Hudson, R., 1996, Draft User's Guide and Technical Regional Mercury Cycling Model (R-MCM):...
  • Hornbuckle, K.C., Bogdan, J., Miller, S., 1998. Predicting and Measuring Atmospheric Deposition of Persistent Organic...
  • P.H. Howard et al.

    Handbook of Environmental Degradation Rates

    (1991)
  • M.J. Howdashell et al.

    Historical input and degradation of toxaphene in Lake Ontario sediment

    Environ. Sci. Technol

    (1996)
  • Lake Ontario LaMP, Stage 1: Problem Definition. Draft, April...
  • D.C.L. Lam et al.

    An expert system approach of integrating hydrological datbase. Models and GIS: application of the RAISON system

  • Lam, D.C.L., Booty, W.G., Wong, I.W.S., Kay, D., Kerby, J.P., 1995. RAISON Decision Support System: Examples of...
  • D.C.L. Lam et al.

    An integrated acid rain model for Canada using the RAISON for Windows system

    Wat. Qual. Res. J. Can

    (1998)
  • Lam, D.C.L., Leon, L.F., Hamilton, S., Crookshank, N., Bonin, D., Swayne, D.A. Multi-model integration in a decision...
  • Leonardelli, S., 1996. Industrial releases within the Great Lakes Basin: An Evaluation of NPRI and TRI Data,...
  • L'Italien, S., 1993. Organic Contaminants in the Great Lakes 1986–1990. Environmental Quality Branch, Inland Waters...
  • Cited by (15)

    • Multi-box mass balance model of PFOA and PFOS in different regions of San Francisco Bay

      2020, Chemosphere
      Citation Excerpt :

      These variation rates are calculated by multiplying the mass of pollutant in a given moment (kg) by a series of rate constants (i.e. not time depending) depicting the main processes of gains and losses of a pollutant in a given compartment for a designated time period (d−1). Due to their simplicity and good performance, rate constant models have been applied to predict the long term concentrations and compartment distribution of different pollutants for entire aquatic ecosystems (Booty et al., 2005; Gobas et al., 2018; Xiao et al., 2008). However, this approach fails to provide spatial distribution of pollutants within the ecosystem, and more complex calculation efforts must be performed to obtain this information (Kaur et al., 2012).

    • Development of a system dynamics model for polycyclic aromatic hydrocarbons and its application to assess the benefits of pollution reduction

      2018, Ecotoxicology and Environmental Safety
      Citation Excerpt :

      Such models provide a powerful framework for understanding the behavior of chemicals in the environment (MacLeod et al., 2011) and have been adopted by industry and government agencies to support decision making (MacLeod et al., 2010). For example, the California Environmental Protection Agency in the United States sponsored the development of CalTOX, a multimedia contaminant fate and exposure model for assessing hazardous waste site; the System for the Evaluation of Substances of the European Union was built upon a multimedia chemical fate model; the US Environmental Protection Agency (EPA) developed a number of models to aid in decision making (CEMC, 2010); multimedia models were incorporated into the EPI Suite collection of exposure assessment tools (MacLeod et al., 2010); and multimedia models are used to develop strategies for reducing the levels of high-priority contaminants in the Great Lakes region of North America (Booty et al., 2004). Current multimedia mass-balance models provide a “tool box” that can be applied to different environmental problems and form an integrated framework for introducing environmental chemistry concepts to researchers.

    • Mass balance modeling to elucidate historical and continuing sources of dioxin into an urban estuary

      2013, Chemosphere
      Citation Excerpt :

      Almost none of the aforementioned types of models focus on elucidating historical and modern day sources, and very few studies have gathered the data necessary to develop such models due to the significant expense and time involved. As of this writing, few published studies have discussed methods and approaches that could be used for associating loads with specific sources and even fewer studies specifically targeting dioxins (e.g., Gobas et al., 1998; Booty et al., 2005; Chen and Chang, 2006; Warren et al., 2007). In this paper, development of a mass balance spreadsheet model for dioxin that is linked to a hydrodynamic and a water quality model of the Houston Ship Channel (HSC) and Upper Galveston Bay (UGB) in Texas is presented.

    • Development and test application of a screening-level mercury fate model and tool for evaluating wildlife exposure risk for surface waters with mercury-contaminated sediments (SERAFM)

      2008, Environmental Modelling and Software
      Citation Excerpt :

      Models are widely recognized as powerful tools for addressing environmental contamination problems with limited data. The field of mercury modelling has been presented using mass balances (Booty et al., 2005; Kotnik et al., 2002; Henry et al., 1995; Vette et al, 2002; Macleod et al., 2005) and mixed-batch reactors (Širca et al., 1999; Hudson et al., 1994; Driscoll et al., 1995; Knightes and Ambrose, 2004; Leonard et al., 1995) and a series of mixed-batch reactors for rivers (Ambrose et al., 2005). The Mercury Cycling Model (MCM) family of process-based models describe mercury fate and transport in lakes (Hudson et al., 1994), but are currently proprietary and not available for general public use.

    View all citing articles on Scopus
    View full text