📄 evolutionary biology

Eco-evolutionary feedbacks generate bistability in population persistence under gradual environmental change

By integrating a quantitative genetic model with eco-evolutionary feedbacks, this study demonstrates that population persistence under gradual environmental change exhibits bistability, meaning extinction can occur even below the critical rate of change if initial population size and genetic variance are insufficient to overcome transient dynamics.

Original authors: Shen, H., Xu, K.

Published 2026-08-09

📖 7 min read🧠 Deep dive

CC BY 4.0

Original authors: Shen, H., Xu, K.

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ⚕️ This is an AI-generated explanation of a preprint that has not been peer-reviewed. It is not medical advice. Do not make health decisions based on this content. Read full disclaimer

Imagine a world where the climate is slowly turning up the heat, the seasons are shifting, and the perfect conditions for life are constantly on the move. This is the reality many species face today. In the scientific fields of ecology and evolutionary biology, researchers are obsessed with a single, high-stakes question: Can a population evolve fast enough to keep up with these changes, or will it simply run out of time and die out?

To understand the answer, we need to look at three moving parts. First, there's the environment, which is like a moving target that keeps changing its location. Second, there's the population, a group of living things trying to survive. Third, there's evolution, the process where the population changes its traits (like body size or heat tolerance) to match that moving target. If the population can change fast enough, it survives. If it falls too far behind, it dies. For a long time, scientists thought there was a simple "speed limit" for environmental change: as long as the environment didn't change faster than this limit, the population would be safe, provided it had enough time to catch up.

But a new study suggests the story is a bit more complicated and a lot more dramatic. It turns out that even if the environment is changing slowly enough to theoretically allow survival, a population might still crash and burn. Why? Because of a tricky feedback loop between how many individuals are left, how much genetic variety they have, and how far behind they are. The researchers found that a population's fate isn't just about the speed of the change; it's also about where it started. A small group with low genetic diversity might spiral into extinction even when a larger, more diverse group would have survived the exact same conditions. It's a race where the starting line matters just as much as the finish line.

The Race Against a Moving Finish Line

Imagine you are running a race, but the finish line isn't sitting still. It's on a treadmill that's slowly speeding up. To win, you have to run faster and faster to keep up. In this paper, the "runners" are populations of organisms, and the "treadmill" is the changing environment.

For years, scientists believed that if the treadmill didn't speed up too fast, the runners would eventually find a comfortable pace and keep running forever. They thought there was a specific "critical speed" for the treadmill. If the treadmill went slower than that speed, everyone was safe. If it went faster, everyone was doomed.

However, the authors of this study, Kuangyi Xu and Hao Shen, decided to look closer at the runners' legs and their energy levels. They built a mathematical model (a computer simulation) to track three things at once: how big the population is, how much genetic variety they have, and how far behind they are from the finish line.

They discovered something surprising: Just because the treadmill is slow enough to allow a win, doesn't mean the runners will win.

The Trap of the "Extinction Vortex"

The paper reveals a phenomenon called bistability. In plain English, this means there are two possible outcomes for the same race, depending entirely on how the runners start.

Imagine two groups of runners facing the exact same slow-moving treadmill.

Group A starts with a huge crowd and lots of different running styles (high genetic variation). They are close to the finish line. They adjust their pace, find a rhythm, and survive.
Group B starts with a tiny crowd and very few different running styles. They are also behind the finish line.

Here is the twist: Even though the treadmill is moving slowly enough for Group B to theoretically catch up, they might still lose. Why? Because they are small and slow to adapt. As they struggle to keep up, they fall further behind. Falling behind makes them less fit, which causes their numbers to drop even more. A smaller population means they lose even more genetic variety (like losing different running shoes from their inventory). With fewer options, they can't adapt fast enough, they fall further behind, and they shrink until they disappear.

The authors call this an "extinction vortex." It's a spiral where being small and slow makes you smaller and slower, until there's nothing left.

The "Temporary Boost" Illusion

One of the most playful and confusing parts of this discovery is that you can't always trust what you see in the short term. The paper shows that a population might look like it's recovering when it's actually doomed, or look like it's dying when it's actually going to make it.

Think of it like a roller coaster.

The False Hope: A small population might get a temporary boost. Maybe a few lucky individuals survive, and the population size jumps up for a moment. It looks like they are winning! But because they started with too little genetic variety, that boost is just a fling. They eventually run out of steam, the genetic diversity collapses, and they crash.
The False Alarm: Conversely, a population might take a hard hit, dropping in numbers and looking like it's about to die. But if they started with enough genetic variety, they might bounce back, adapt, and eventually find their rhythm on the treadmill.

The authors' simulations show that a temporary increase in population size or genetic variation does not guarantee survival. You have to know the "starting state" of the population to know if that temporary boost is a sign of life or a last gasp.

Why This Matters for Real Life

This isn't just about math on a screen. The authors suggest that when we try to predict if a species will survive climate change, we can't just look at how fast the climate is changing. We have to look at the species' current condition.

If a species is already small, or if it has very little genetic variety (maybe because it's been isolated or hunted), it might be in the "danger zone" even if the environment is changing slowly. It might be sitting on a cliff edge, waiting for a small push to fall off. On the other hand, a large, diverse population might be able to handle a much faster rate of change.

The study also looked at what happens if the environment changes twice. Imagine a population is doing fine, then the environment gets a little worse, and then it gets a sudden, sharp shock. The paper suggests that if the population was already struggling with the slow change, that sudden shock might push it over the edge, even if the shock itself wasn't huge. It's like a rubber band that's already stretched tight; a tiny extra pull snaps it.

The Bottom Line

The main takeaway from Xu and Shen's work is that survival isn't just about the speed of the storm; it's about the strength of the ship.

The Old Idea: If the environment changes slower than a certain speed, the population is safe.
The New Finding: Even if the environment changes slowly, a population can still go extinct if it starts too small or with too little genetic variety.
The Catch: You can't tell if a population is doomed just by watching it for a little while. A temporary rise in numbers might be a trap, and a temporary drop might be a recovery in disguise.

The authors emphasize that to truly understand extinction risk, we need to know the "eco-evolutionary state" of a population—its size, its genetic health, and how far behind it is—before we can predict its future. It's a reminder that in the race against a changing world, where you start matters just as much as how fast the finish line is moving.

Technical Summary: Eco-evolutionary feedbacks generate bistability in population persistence under gradual environmental change

Problem Statement
The persistence of natural populations in gradually deteriorating environments is a central concern in ecology and evolutionary biology, particularly in the context of climate change. Existing theoretical frameworks have primarily focused on identifying a "critical rate" of environmental change ( $v_c$ ) beyond which extinction is inevitable. These models generally assume that if the rate of environmental change is below this threshold, a viable equilibrium exists where populations can track a moving phenotypic optimum, thereby guaranteeing persistence. However, this study addresses a critical gap: the existence of a viable equilibrium does not guarantee that a population can reach it from its initial state. Previous qualitative observations suggested that large phenotypic lags could trigger a feedback loop of demographic decline and loss of genetic variance (an "extinction vortex"), yet quantitative models of gradual change have rarely explored whether this leads to bistability—where identical environmental conditions yield divergent outcomes (persistence vs. extinction) depending on initial eco-evolutionary states.

Methodology
The authors develop a deterministic quantitative genetic model for a haploid population with overlapping generations, subject to density-dependent logistic growth. The model explicitly incorporates feedbacks among three dynamic variables:

Population size ( $N$ ): Governed by logistic growth where the intrinsic growth rate is reduced by "lag load" (mismatch between mean trait and optimum) and "variance load" (phenotypic variation).
Mean trait evolution ( $\mu$ ): The population tracks a phenotypic optimum $\theta(t)$ that moves linearly over time ( $\theta(t) = \theta_0 + vt$ ), potentially with an initial abrupt shift.
Genetic variance ( $V$ ): Modeled via an equation accounting for mutation input ( $V_M$ ), reduction by stabilizing selection, and loss due to genetic drift (inversely proportional to $N$ ).

The system tracks the dynamics of the phenotypic lag ( $\Delta = \theta - \mu$ ), genetic variance ( $V$ ), and population size ( $N$ ). The authors analyze the existence of critical rates, equilibrium states, and the basins of attraction for these equilibria using mathematical analysis (Appendices S1 and S2) and numerical simulations.

Key Contributions and Results

Bistability Below the Critical Rate: The study demonstrates that even when the rate of environmental change ( $v$ ) is below the critical threshold ( $v < v_c$ ) where extinction is certain, the system exhibits bistability. There are two locally stable equilibria: extinction ( $N=0$ ) and a viable state where the population persists with a constant lag.
Dependence on Initial Conditions: Whether a population reaches the viable equilibrium depends critically on its initial eco-evolutionary state. Specifically, if the initial population size ( $N_0$ ) and genetic variance ( $V_0$ ) fall below a critical threshold (defined by a basin boundary), the population will go extinct despite the environmental change rate being theoretically survivable.
Non-Linear Transient Dynamics: The initial state dictates the trajectory toward equilibrium. Populations may exhibit temporary increases in size or variance before declining to extinction, or conversely, show temporary declines before rebounding to persistence. Crucially, a transient increase in population size or genetic variance does not guarantee ultimate survival; populations starting with insufficient initial resources may appear to recover temporarily before collapsing.
Scenario Analysis:
- Mutation-Selection-Drift Balance: When initial variance is determined by population size, stronger stabilizing selection increases the minimum population size required for persistence, potentially rendering persistence impossible if selection is too strong.
- Sudden Shifts in Rate: A population at equilibrium under a slow rate may fail to persist if the rate suddenly increases, even if the new rate is below $v_c$ , particularly if accompanied by an abrupt shift in the phenotypic optimum.
- Temporary Environmental Spikes: Populations can survive a temporary environmental rate exceeding $v_c$ if the duration is short. However, if the duration is too long, the population may cross the basin boundary into extinction, leading to "extinction debt" even after conditions return to normal.

Significance and Claims
The paper claims that characterizing the current demographic and evolutionary state of a population is as critical as estimating the rate of environmental change for predicting extinction risk. The authors argue that the traditional focus on critical rates is insufficient because it ignores the accessibility of viable equilibria.

The study suggests that:

Predictive Power: Reliable predictions of population persistence require data on initial population size, standing genetic variation, and the degree of maladaptation (lag), not just the rate of environmental change.
Interpretation of Dynamics: Observed demographic trajectories (e.g., temporary recoveries) cannot be taken as definitive indicators of evolutionary rescue; the underlying initial conditions determine the ultimate fate.
Theoretical Advancement: The work establishes that bistability arises naturally from the canonical eco-evolutionary feedback loop among demography, genetic drift, and adaptation, without requiring non-quadratic fitness functions or density-dependent selection strength (mechanisms proposed in previous studies).

The authors maintain a modest stance, noting that their deterministic model serves as a proof-of-concept. They acknowledge that incorporating stochasticity would likely lower the critical rate and expand the basin of extinction, but expect the qualitative conclusion of bistability to remain robust. They also note that their assumption of a Gaussian fitness function and polygenic adaptation may simplify complex genetic architectures, but the core mechanism of feedback-driven bistability is expected to hold.