Timedora

Environmental Fate Modeling: Mass Balances and Concentration Over Time

A practical introduction to tracking chemical mass and concentration through environmental compartments, from basic balances to time-dependent estimates.

Start with the Balance

Define the system, then balance its

A -balance model follows a chemical across a clearly defined system boundary. For a well-mixed compartment with constant volume VV, the of chemical is M=VCM=VC, where CC is its concentration. The general accounting rule is

dMdt=input rate−output rate+formation rate−loss rate.\frac{dM}{dt}=\text{input rate}-\text{output rate}+\text{formation rate}-\text{loss rate}.

The signs reflect whether a process adds chemical to the compartment or removes it. A transfer across the boundary is an input or output; formation adds chemical within the system, while transformation or other losses remove it.

For a flowing water body with flow rate QQ, inflow concentration CinC_{in}, and first-order reaction loss with rate constant kk, the balance is

VdCdt=QCin−QC−kVC.V\frac{dC}{dt}=Q C_{in}-Q C-kVC.

Here, QCinQ C_{in} is the incoming chemical per time, QCQ C is the outgoing per time, and kVCkVC is the reaction-loss rate. Check units term by term: if QQ is in m3/day\mathrm{m^3/day} and concentration is in mg/m3\mathrm{mg/m^3}, then QCQC is in mg/day\mathrm{mg/day}. The reaction term also has units of per time when kk is in inverse days. Clear boundaries, assumptions, and consistent units help make the model interpretable and reveal setup errors.

Find the

At , dC/dt=0dC/dt=0. Applying that condition to the flowing-water balance gives

0=QCin−QCss−kVCss⟹Css=QCinQ+kV.0=Q C_{in}-Q C_{ss}-kVC_{ss} \quad\Longrightarrow\quad C_{ss}=\frac{Q C_{in}}{Q+kV}.

The denominator includes both flushing, represented by QQ, and reaction loss, represented by kVkV. Dividing by volume shows their concentration-scale rates: flushing is Q/VQ/V, while transformation is kk. Thus both can affect the steady concentration; the result is not determined by inflow concentration alone.

Use this steady-state estimate when inputs and conditions are sufficiently stable and the system has had enough time to approach equilibrium. If conditions are changing or the time scale is too short for that adjustment, a time-varying model is more appropriate.

First-Order Loss and

From proportional loss to exponential decline

In a first-order model, the rate of concentration loss is proportional to the concentration still present:

dCdt=−kC,\frac{dC}{dt}=-kC,

where kk has units of inverse time. With no new chemical entering a well-mixed compartment, solving this equation from initial concentration C0C_0 gives

C(t)=C0e−kt.C(t)=C_0e^{-kt}.

The exponential form means the same fraction is lost over equal time intervals, rather than the same fixed amount. The first-order follows by setting C(t)=C0/2C(t)=C_0/2:

12=e−kt1/2⟹t1/2=ln⁡2k,k=ln⁡2t1/2.\frac{1}{2}=e^{-kt_{1/2}} \quad\Longrightarrow\quad t_{1/2}=\frac{\ln 2}{k}, \qquad k=\frac{\ln 2}{t_{1/2}}.

For the supplied example, the initial concentration is 8 mg/L8\ \mathrm{mg/L} and the is 10 days. After 20 days, two half-lives have passed, so

C(20)=8(12)2=2 mg/L.C(20)=8\left(\frac{1}{2}\right)^2=2\ \mathrm{mg/L}.

Equivalently, k=ln⁡(2)/10≈0.0693 day−1k=\ln(2)/10\approx0.0693\ \mathrm{day^{-1}}, and substituting this into the exponential equation gives the same prediction. The approach is convenient when elapsed time is a whole number of half-lives; the exponential form applies directly to other times.

Interpret the fitted loss rate carefully

A fitted describes the loss processes represented in the observations or model. If concentration declines because of both degradation and physical removal such as flushing, the fitted apparent rate may combine them. It should not be interpreted as degradation alone unless those other losses have been separated.

Also, a constant is a property of first-order decay. Under non-first-order behavior, the time needed for concentration to halve is not necessarily the same at different concentrations.

Dynamic Compartments and Estimation

Track change and connect compartments

A dynamic model applies a balance as concentration changes over time. For the flowing water body, assume constant inflow and outflow, constant volume, and first-order loss. Dividing its balance by VV gives

dCdt=QVCin−(QV+k)C.\frac{dC}{dt}=\frac{Q}{V}C_{in}-\left(\frac{Q}{V}+k\right)C.

The term Q/VQ/V is the flushing rate on a concentration basis. Define the combined removal rate as λ=Q/V+k\lambda=Q/V+k. With constant inflow concentration, the solution is

C(t)=Css+(C0−Css)e−λt,Css=(Q/V)Cinλ.C(t)=C_{ss}+(C_0-C_{ss})e^{-\lambda t}, \qquad C_{ss}=\frac{(Q/V)C_{in}}{\lambda}.

This describes concentration moving from its initial value C0C_0 toward the steady-state value. It does not necessarily decline toward zero: if input continues, the concentration approaches the balance between input and removal. If input stops, set Cin=0C_{in}=0; then flushing and reaction both contribute to the decline. This solution depends on the stated assumptions of constant flow, volume, and inflow, together with first-order loss.

A uses one balance for each well-mixed environmental unit. For example, a model might represent water and sediment, with chemical moving from water to sediment, degrading in either, and potentially returning from sediment to water. A transfer out of one compartment is an input to the other. When balances are summed across the whole system, internal transfers cancel: they move within the boundary rather than adding or removing it from the total system. The coupled equations still need to represent each transfer and loss in the compartment where it occurs.

Estimate concentrations systematically

A practical workflow keeps the equations tied to a well-defined question:

  1. Define the question and boundary. Specify the chemical, location, time period, and compartments included.

  2. List processes. Identify sources, inflows, outflows, transfers, degradation, and any formation.

  3. Choose the model structure. Use a single compartment when it adequately represents a well-mixed system; use multiple compartments when transfer or storage between media matters.

  4. Set parameters and units. Record volume, flow, input concentration or rate, and rate constants. Convert a to a rate constant when using first-order equations.

  5. Specify initial conditions. State the initial concentration or in each compartment.

  6. Solve and check. Use an analytical solution when available, or a numerical time-step method for more complex systems. Check units, nonnegative concentrations, and whole-system accounting.

  7. Test assumptions and uncertainty. Explore how estimates change when uncertain inputs, rates, or compartment assumptions vary.

For the supplied lake example, the initial concentration is 5 μg/L5\ \mathrm{\mu g/L}, the first-order is 30 days, and water replacement is negligible over the period considered. Convert the to a rate constant:

k=ln⁡230≈0.0231 day−1.k=\frac{\ln 2}{30}\approx0.0231\ \mathrm{day^{-1}}.

With no continuing input and negligible flushing, the model is C(t)=5e−0.0231t μg/LC(t)=5e^{-0.0231t}\ \mathrm{\mu g/L}. After one , 30 days, the prediction is 2.5 μg/L2.5\ \mathrm{\mu g/L}; after 60 days, two half-lives, it is 1.25 μg/L1.25\ \mathrm{\mu g/L}. These estimates rely on no important transfer to other compartments and constant first-order behavior. If flushing is significant, include its loss; if sediment exchange matters, a water-only model may be inadequate. More generally, estimates are only as reliable as their assumptions and input data.