Environmental fate modeling for water managers

The chemical behaviour of priority substances in the environment is complex. After release the chemicals can be transported through evaporation, diffusion, sorption to solids and advection with water and air currents. The bad news is that the transport processes are different for each of the 33 priority substances, but the good news are that you can use

environmental fate models

to cope with most of this complexity without really understanding it.

1. Introduction
The models have been developed since the 1960s and currently they are in wide use in remediation planning, licensing of new chemicals and identification of emission sources. They are a good starting point for getting a grip of the processes that affect the chemical concentrations in your water body.

A word of warning before we start: there are a few pitfalls in using these models. If the main problem is metals (cadmium, nickel or mercury) or PAH substances bound to soot (black carbon particles) you are going to need some technical assistance. The transport processes of these substances can be very local (depending on the water quality) so we recommend that you consult the local university or a consultant specialized in these substances. However for the remaining 29 priority substances, environmental fate modeling is deceptively simple.

2. What is an environmental fate model?
From an environmental chemists' point of view, environmental fate model is a scientific instrument for studying the relative importance of environmental processes. But from your point of view as a water manager, environmental fate models are best thought of as black boxes. You put information in and get an estimate (or a guess) about the PS concentrations in your water-body and about the timescales of change (Figure 1). For many cases this is enough, but if you want to get the most out of modeling find an environmental chemist specialized in modeling. If you want to understand how an environmental fate model works, consult the DSS handbook and look at the references. This simple manual will skip all the science and will just show how to use the models and what to look for in the results.

Figure 1: Environmental fate model as a black box. If you want details about the insides of that box, look at the DSS. Otherwise, read on.

model from Trent

University website

www.socopse.eu

Now you are ready to start modeling. Start the QWASI model and let's look at an example.

3.2 Modelling PS transportation with QWASI
Figure 2 shows a screenshot of the QWASI model. Model usage is straightforward: you go through the boxes in the left by clicking at them one at a time, fill in the data, click “compute" and look at the results. That's it.

Figure 2: The main view of the QWASI model software. You get the items of interest (presented in Figure 1) from the places marked with the little arrows.

In this example we will look at nonylphenol. The next stage is to define your water body to the model. Click “Environmental properties". What you should see is a window (Figure 3) with three tabs: “LakeProperties", “Flows" and “Transfer coefficients".

There are quite many parameters that you need to define, but don't panic. Some of these parameters are only important for some chemicals. Consult the sensitivity segment diagram in the DSS if you are uncertain about where it is okay to take some shortcuts in parameterization. You will find a form for gathering all the necessary parameters at the end of this paragraph. Either print it, fill it and get back to modeling afterward or select one of the default environments.

Figure 3: The input-windows for defining the water body. In the “transfer coefficients” tab it is only necessary to change one value (Rain rate), the default values are usually good for the other parameters.

Figure 4: Menu for defining the emissions to the water body. The emissions are divided into three categories: direct discharges, inflow and deposition from air.

3.3 Interpreting results
The most convenient way of looking at the results is through a mass flow diagram (click the “Diagram" button). A result for nonylphenol is shown for a bay area in the case study of Vantaa, Finland (Figure 5). There is a lot of information in the mass flow diagram, but you only need to focus on a few things:

Q: What is the concentration/emission relationship?
A: 138 ng/l/100 kg/year = 1.38 (ng/l)/(kg/year)

Q: What kind of emissions result in exceeding the EQS?
A: For nonylphenol the EQS is 300 ng/l, so based on the concentration/emission relationship, you would need 218 kg/year of emissions.

Q: What is the overall residence time?
A: 404 hours. Not that much.

Q: What are the main transport processes?
A: Here outflow 74 kg and evaporation 21 kg play the main role, so the problem is transported downstream.

Q: Where is the most of the chemical found?
A: Sediments store 75% of total mass.

Figure 5: Mass flow diagram of nonylphenol transport in the bay of Vanhankaupunginlahti outside Helsinki. (Note that the emission amount is illustrative.)

Figure 6: In the mass balance diagram one of the most revealing figures is the residence time of the chemical in the compartment where most of the substance is found. In this case it is the sediments, where the residence time is 0.775 years.

In our example in order to get the amount to drop to 10 % in sediments:
M/M0 = 10% = exp(-t/0.775y)
-->  -t/0.775 y = ln(0.1) = -2.3
-->  t = 2.3 * 0.775 y = 1.78 year

One would have to wait for 1.78 years for the concentrations to drop to 10% after all emissions have ceased.

Based on this simple example we know that it is unlikely that nonylphenol would cause problems in the bay area. First of all, the emissions would have to be quite large (200 kg/year to a 6 km2 bay). Secondly the outflow is so strong that most of the chemical is rapidly transported to the sea. Finally the system will probably respond very quickly to reductions in the emissions (concentration reduction to 10% in less than 2 years).

4. What about time series and scenarios and stuff?
If by now, you have the feeling that this is all too simple and not really realistic, please continue reading. Otherwise, put the model results to use. The time dependant models are useful in situations where you have an estimate about the emission trends in the near future (20-30 years) and the emissions are changing very rapidly (e.g. after the banning of TBT).

In the previous paragraph we applied the QWASI model to water management. The model is easy to use, but it has two limitations: (a) it does not consider what happens after the chemical has been emitted and (b) it cannot be used for simulating responses to changing emission scenarios. In this paragraph we will look at a slightly more complex model, the CozmoPOP 2.0 model (downloadable at www.scar.utoronto.ca/~wania/downloads2.html), which does not have these limitations.

The only change in chemical input data is the need for a degradation rate in soil and temperature dependencies of the physical chemical properties. The first one is simple, but the second is not. For most cases you can estimate that the physico-chemical properties remain constant, in others you might want to consult a chemist.

The environmental properties required by the CozmoPOP model are lot more detailed than in the QWASI model (water balance, organic carbon concentrations in soil, forest growth). You should allocate 1-3 weeks for gathering this data or try to check if the local meteorological/hydrological institute or local university has gathered this data already.

Below (Figure 7 and Figure 8) are shown the relation between emission scenarios and concentrations for alpha-HCH and the instantaneous mass balance in the Baltic catchment area. These time series are a good way of understanding the time scales. It is also worthwhile to note that the chemical concentrations vary within the year as much as 30-50% because of meteorological conditions. This stresses the importance of monitoring the water phase more often than once a year (in sediments the fluctuation is much smaller).

As you can see from the figures, for alpha-HCH the time required for the concentrations to reach a level of 10% of the peak is 15-20 years. And the main processes in governing the environmental fate are emissions, evaporation from forests, deposition to fresh water, evaporation from water surface and advection to the sea. Sediment burial is meaningless for alpha-HCH (but important for many of the more hydrophobic substances).

Figure 7: An example of CozmoPOP output for alpha-HCH in the Baltic proper catchment area. As you can see, the concentrations fluctuate within years because of weather conditions.

Figure 8: An instantaneous mass balance of alpha-HCH in the Baltic catchment area. The instantaneous mass balances are useful when the sediments and soils have stored substantial amounts of pollutants and the emissions have ended. Then the system is not in equilibrium.