Per-Capita Rates and Net Population Change

From Cell-Level Rates to Population Fluxes

A per-capita rate describes how frequently an event occurs for an individual cell, on average, per unit time. A population-level flux describes how many such events occur across the population per unit time. Converting between them requires both the per-capita rate and the number of cells exposed to that rate. The resulting division and death fluxes can be large even when their difference, the instantaneous net change in cell count, is small.

Building on Population States, Flows, and Cell-Count Accounting, consider one well-mixed cell population with size \(N(t)\). The focus here is the immediate accounting of division and death events: how their per-cell rates become population fluxes, and what their difference says about the direction and rate of change at a particular instant. It does not, by itself, specify a population’s full time course.

The rates \(k_{\mathrm{div}}\) and \(k_{\mathrm{death}}\) are not population counts. Their units are, for example, \(\mathrm{day}^{-1}\). Multiplying one of them by \(N\), measured in cells, gives a population flux with units of cells per day (or division events per day for division). The net rate \(\dot N\) has units of cells per day. When \(N(t)>0\), dividing the net rate by the current population gives the instantaneous per-capita net growth rate:

This is an instantaneous accounting identity under the stated event convention. A positive difference means the population is increasing at that instant; a negative difference means it is decreasing; a zero difference means the two fluxes balance at that instant. None of these conclusions alone says what the population will do at later times, because the rates can change and the population’s state can affect them.

Why the Difference Gives Net Change

The key is to count the change in the population caused by each event, not simply to count the cells involved. A death removes one cell, a change of \(-1\). A division event replaces one cell with two, a net change of \(+1\), not \(+2\). Thus the division-event flux contributes \(+J_{\mathrm{div}}\) cells per unit time, while the death flux contributes \(-J_{\mathrm{death}}\) cells per unit time.

A short-interval argument makes this precise. Suppose that, over a sufficiently small interval \(\Delta t\), each cell has probability \(k_{\mathrm{div}}\Delta t+o(\Delta t)\) of dividing and probability \(k_{\mathrm{death}}\Delta t+o(\Delta t)\) of dying. Here \(o(\Delta t)\) denotes terms that become negligible relative to \(\Delta t\) as \(\Delta t\) approaches zero. Assume rates are approximately constant over this interval, the per-cell rates apply to the cells present, and simultaneous events on the same cell have negligible probability to first order. Each of the \(N(t)\) cells can contribute an expected \(+1\) from division or \(-1\) from death. Therefore:

Divide by \(\Delta t\) and take the limit as \(\Delta t\) approaches zero. This gives the instantaneous expected net rate, or the corresponding deterministic population-balance equation when the model represents the population by a continuous state:

This derivation identifies the assumptions behind the equation: the two rates are nonnegative, apply per cell, and describe the relevant division and death events over the interval. The equation also assumes no other process changes this population count during that interval. If cells enter, leave, differentiate into another counted state, or are removed by an additional process, that process needs its own flux term.

Net Growth Does Not Reveal Gross Turnover

The difference \(k_{\mathrm{div}}-k_{\mathrm{death}}\) determines the per-capita net rate at the instant, but it does not determine either flux separately. Populations with the same size and the same net rate can have very different rates of division and death. That distinction matters when the question concerns how much turnover is occurring, not only whether the population is momentarily increasing or decreasing.

There is a convention worth making explicit in any model or dataset. In this tutorial, \(k_{\mathrm{div}}\) counts division events per cell per unit time. One event transforms one cell into two and therefore increases the total count by one. If a rate is instead defined as the per-cell rate of daughter-cell production, it counts two produced daughters per division event under this idealized event description. That production rate must not be substituted into the equation as though it were a division-event rate without checking the definition. Clear event accounting prevents a factor-of-two error.

Instantaneous Interpretation and Checks

The sign of \(k_{\mathrm{div}}-k_{\mathrm{death}}\) classifies the direction of net change for a positive population. When it is positive, the division flux exceeds the death flux; when negative, death exceeds division; when zero, the two fluxes are equal. Zero net change does not mean that nothing is happening: both fluxes can be large and exactly balanced. Nor does a zero difference prove a stable long-term state. It is a statement about the rates at the specified instant and state.

A practical check is to calculate the net rate in two ways: subtract the per-capita rates and multiply by \(N\), then separately calculate the two fluxes and subtract them. The results should agree:

If they do not, check the arithmetic, the units, and whether one input is a division-event rate or a daughter-cell production rate. Also verify that the rates refer to the same time point, population, and time unit. For example, a rate per hour cannot be combined directly with one per day without conversion.

Alternatives, Diagnostics, and Limits

When only the net change is needed, the net per-capita rate \(k_{\mathrm{div}}-k_{\mathrm{death}}\) is a compact summary. When turnover or mechanism matters, retain the separate division and death rates and their fluxes. Reporting only the net rate loses information about gross activity, as the comparison of populations A and B showed. Reporting only one flux is also insufficient to determine the net change unless the other contributing fluxes are known.

Before using the calculation, check that \(N\) is the number of cells in the population to which both rates apply, that both rates use the same time unit, and that their definitions specify the counted event. The basic balance is incomplete if the chosen population also gains or loses cells through processes other than division and death. A mismatch between observed changes and the calculated balance can indicate a missing flux, inconsistent rate definitions, or failure of the assumptions that the rates represent the population at that instant.

The balance describes a population-level rate, not the exact next event in a small population. In a small or highly variable population, the observed count over a short interval can differ from its expected change because events occur discretely and randomly. Treating \(N\) as continuous is then an approximation to the population’s average behavior, not a claim that a fraction of a cell divides or dies. Finally, rates may differ among cells or change with time or population state; a single pair of rates should not be treated as universal unless the model or measurements justify that simplification.

• Common mistake: counting two new cells at division. A division event replaces one parent with two daughters, so the total count increases by one. With a division-event rate, its contribution is \(+J_{\mathrm{div}}\), not \(+2J_{\mathrm{div}}\).
• Common mistake: treating zero net change as zero turnover. Zero means the division and death fluxes balance. Both can be nonzero.
• Common mistake: interpreting a per-capita rate as a population flux. A rate in inverse time must be multiplied by the relevant population count to obtain a population-level flux.
• Common mistake: projecting an instantaneous rate as a long-term trend. The calculated sign applies at the specified instant. Longer-term behavior requires information about how rates and population size evolve.