Invasion lag before explosion
A Socratic walk-through of Invasion lag before explosion — reasoned out one step at a time, not lectured.
The question we started with
THE QUESTION #Why can an introduced species sit harmless for decades and then overrun a continent?
A plant arrives in a new country. For forty years it sits in a corner of one estuary, or one roadside, and everyone who notices it concludes it is harmless. Then, over a single decade, it is everywhere, and the emergency money appears.
The obvious story is that something changed — the climate warmed, a barrier fell, the species evolved. Sometimes that is true. But I want to ask a more uncomfortable question first: if nothing at all had changed, if the population had been growing at exactly the same steady rate the whole time, what would we have seen? Would it have looked any different?
Reasoning it through
REASONING #Let us do the arithmetic rather than argue about it. Suppose a founding population of ten individuals grows at thirty per cent per year — unremarkable for a weed or a beetle — and suppose that rate never changes for fifty years.
After ten years there are about 140. After twenty, roughly 1,900. After thirty, about 26,000. After forty, around 360,000. After fifty, nearly five million.
Now ask yourself: at what point in that history did the species "become invasive"? The growth rate answered that question the same way every year. But the increments did not. In year ten the population added about thirty individuals. In year fifty it added over a million. To a person walking the ground, the first four decades are indistinguishable from a curiosity and the fifth is an invasion — and yet the underlying process is one unbroken exponential.
There is a second half to this, which is the observer. A survey does not measure population; it measures whether the species crosses the threshold of being noticed. Detection is roughly proportional to abundance and to area occupied, so the whole invisible portion of the curve sits below that threshold. The lag is not a phase of the population's life. It is the part of the curve that lies underneath our ability to see it.
Does that not reframe the question? We ask "why the sudden explosion", but the explosion is what constant growth always looks like when plotted against a fixed detection line. The honest surprise would be a population that did announce itself early.
Now, having established that no extra explanation is needed, we can ask which extra explanations are nevertheless real — because ecologists have found several, and they lengthen the lag beyond what pure arithmetic gives.
A small population can grow more slowly than its own potential, or not at all, because of Allee effects: too few mates, too little pollen, too small a group to swamp local predators. Growth is not simply exponential from individual one; the rate itself can be depressed at low density and then rise as density builds, which bends the early curve flatter still.
Sometimes a required partner is missing. Several Ficus species stood in Florida for decades as ornamental trees producing no viable seed, because their species-specific pollinating fig wasps were absent; when the wasps arrived in the 1970s the trees began setting seed and spreading. That is a lag with a mechanical cause, and it ends abruptly.
Sometimes the environment moves the goalposts — a warming winter, a new road network, a disturbance that opens ground. And sometimes the population adapts, since a founding population is a genetic bottleneck that may take many generations to assemble a genotype suited to the new range. Spartina cordgrass in Willapa Bay, Washington is the case usually cited for a long quiet period followed by rapid estuary-wide spread, though disentangling arithmetic from adaptation there is genuinely contested.
So the lag has one guaranteed component and several optional ones. Which do you think matters more for policy?
The analogy
THE ANALOGY #Think of a pond covered by a lily that doubles its area every day, and which will cover the whole pond on day thirty. On which day is the pond half covered? Day twenty-nine. Ask a gardener on day twenty-five, when a mere three per cent is green, and they will tell you truthfully that the lily is nowhere near a problem. They are not being careless. They are reading the increment, and the increment is small right up until it is the whole pond.
a real invasion is not a tidy doubling on a fixed board — the growth rate itself changes with density, weather and the arrival of partners, and the "pond" has irregular edges, refuges and unsuitable ground that make the final stretch messier and often slower than the clean geometry suggests.
Clarifying the model
THE MODEL #Two corrections are worth making explicitly.
First, a lag is not evidence of harmlessness, and its absence is not evidence of safety. Because the informative part of the curve is the rate, not the count, the right question to ask of a small new population is "how fast is it multiplying and spreading", not "how much of it is there". A species doubling annually from a hundred plants is a far worse prospect than a static population of ten thousand.
Second, the exponential does not run forever. Every invasion eventually saturates as suitable habitat fills, which is why occupancy curves usually turn out to be S-shaped rather than endlessly rising. The "explosion" is the steep middle of that S. Recognising that also explains why control is so much cheaper early: effort scales with the population you must remove, and that is precisely the quantity growing exponentially while you deliberate.
A picture of it
THE PICTURE #How to readThe bars are one population growing at a constant thirty per cent per year; the flat line is a rough threshold of being noticed. Read left to right and notice that the first three bars are visually indistinguishable from zero even though each is seven times the one before — the "lag" is simply everything below the line, and the "explosion" is the first bar that rises above it.
What became clearer
WHAT CLEARED #The lag phase is mostly an illusion of perception, not a property of the organism. Constant exponential growth beneath a fixed detection threshold produces exactly the pattern of decades of quiet followed by sudden catastrophe, and biological causes such as Allee effects, missing mutualists and adaptation lengthen that quiet period rather than creating it. The practical consequence is uncomfortable: the moment an invasion becomes obvious is roughly the last moment it is cheap to stop.
Where to go next
ONWARD #- How ecologists actually estimate spread rates from sparse early records, and why the estimates are so uncertain.
- Whether "lag" is better modelled as a delay in the growth rate itself or purely as a detection artefact.
- Why some invasions never explode at all, and what distinguishes them from those that do.
Key terms
TERMS #| Term | What it means |
|---|---|
| Lag phase | the interval between an introduction and the point at which spread becomes conspicuous. |
| Exponential growth | growth by a constant proportion per unit time, producing absolute increments that themselves grow. |
| Allee effect | reduced per-capita growth at low population density, from difficulty finding mates or other cooperative shortfalls. |
| Propagule pressure | the number and frequency of individuals introduced, a strong predictor of establishment success. |
| Mutualist bottleneck | an invasion held back until a required partner, such as a specific pollinator, is also present. |
Every term the collection defines is gathered in the glossary.