📄 evolutionary biology

Seasonal fluctuations in fitness result in severe reductions in effective population size

By simulating seasonally fluctuating selection in *Drosophila* populations using realistic genomic and demographic parameters, this study demonstrates that such temporal selection can reduce the genome-wide effective population size by approximately 50%, with the magnitude of reduction primarily driven by the allele frequency amplitude of the most strongly fluctuating loci.

Original authors: Johnson, O. L., Tobler, R., Schmidt, J. M., Huber, C. D.

Published 2026-04-01

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Johnson, O. L., Tobler, R., Schmidt, J. M., Huber, C. D.

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 ⚕️ This is an AI-generated explanation of a preprint that has not been peer-reviewed.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

The Big Idea: Why "Seasonal Swings" Shrink the Genetic Pool

Imagine a population of fruit flies (specifically Drosophila melanogaster) living in a world where the seasons change dramatically. In the summer, being "tan" might help you survive the heat. In the winter, being "pale" might help you hide from the cold.

This paper asks a simple but profound question: What happens to the genetic diversity of a population when the rules of survival keep flipping back and forth every few months?

The authors found that these constant changes act like a giant, invisible vacuum cleaner, sucking out about 50% of the population's genetic "effective size." In plain English, even if there are a million flies, the genetic variety behaves as if there are only 500,000.

The Analogy: The "Musical Chairs" of Survival

To understand why this happens, let's use the game of Musical Chairs.

The Normal Game (Stable Environment): Usually, in evolution, if you have a good trait, you stay in the game. The music stops, you find a chair, and you pass your genes to the next round.
The Fluctuating Game (This Study): Now, imagine the music stops, and suddenly the rules change. The chairs you were sitting on are removed, and new chairs appear in a different spot.
- In Summer, the "Tan" flies are the winners. They get to sit in the chairs and reproduce. The "Pale" flies are standing up, getting tired, and having fewer babies.
- Then, Winter hits. Suddenly, the "Tan" chairs disappear, and "Pale" chairs appear. The "Tan" flies are now the ones standing up, struggling, and having fewer babies. The "Pale" flies get to sit and reproduce.

The Problem: Every time the season changes, the "winners" of the previous season become the "losers" of the current one. They are forced to start from scratch. This creates a massive bottleneck. The population isn't just shrinking because of the cold; it's shrinking because the reproductive success of individuals is swinging wildly from "super-parent" to "almost no offspring" depending on the weather.

The Key Findings

1. The "Worst Offender" Rule

The researchers discovered that the overall health of the population's genetic diversity isn't determined by the average change in the flies. Instead, it's determined by the single trait that changes the most.

Analogy: Imagine a classroom where most students' grades go up and down by a few points. But one student, "Super-Student," has grades that swing from 100% to 0% every week. That one wild swing causes the whole class's average stability to crash.
The Science: The paper shows that the gene (locus) with the biggest seasonal swing in frequency is the one that drags down the entire population's effective size.

2. It's Not About How Many Flies There Are

You might think, "If we have a huge population of 10 million flies, surely the seasons won't matter as much."

The Surprise: The study found that it doesn't matter if you have 20,000 flies or 1,000,000. The "seasonal squeeze" reduces the genetic diversity by roughly the same percentage (about 50%) regardless of the total crowd size.
Analogy: Whether you have a small crowd or a stadium full of people, if the DJ keeps changing the music so fast that half the crowd is always dancing the wrong way, the "vibe" (genetic diversity) of the whole room suffers equally.

3. The "Boom and Bust" Cycle

In nature, fruit fly populations often explode in summer and crash in winter. The researchers simulated this "boom and bust."

The Result: Even when the population size changes naturally with the seasons, the genetic damage caused by the changing selection pressures is still massive. The population is constantly being "reset," preventing new genetic combinations from sticking around long enough to build up.

Why Does This Matter?

For decades, scientists have tried to solve a mystery called Lewontin's Paradox. This is the observation that species with huge populations (like bacteria or flies) don't have much more genetic diversity than species with tiny populations (like humans or cheetahs). You'd expect the big populations to be genetic goldmines, but they aren't.

This paper suggests a new culprit: Seasonal Fluctuations.

The Takeaway: Nature isn't just a steady stream; it's a rollercoaster. The constant up-and-down of "who is the fittest right now" creates a bottleneck that prevents genetic diversity from building up, even in massive populations.

Summary in One Sentence

Just as a constantly changing rulebook in a game prevents any team from building a consistent winning strategy, constantly changing seasons prevent fruit fly populations from maintaining their full genetic potential, effectively cutting their genetic "strength" in half.

1. Problem Statement

Effective population size ( $N_e$ ) is a fundamental parameter in population genetics that dictates the strength of genetic drift and the maintenance of genetic diversity. While $N_e$ is often viewed through a demographic lens, it is heavily influenced by selection. Previous research has established that positive selection (selective sweeps) and purifying selection (background selection) reduce $N_e$ , while balancing selection can increase it.

However, the impact of fluctuating selection—where selection pressures change direction or intensity over time (e.g., seasonally)—on genome-wide $N_e$ remains poorly understood. Empirical studies in Drosophila melanogaster have identified hundreds of loci undergoing adaptive oscillations across seasons, yet theoretical models have largely relied on abstract or weakly parameterized assumptions. It is unknown how the cumulative effect of hundreds of seasonally fluctuating loci, their epistatic interactions, and demographic factors combine to alter $N_e$ in natural populations.

2. Methodology

The authors employed forward-time population genetic simulations using the software SLiM (v4.0.1) to model multilocus adaptation under seasonal environmental shifts.

The Segregation Lift (SL) Model: The study utilized a recently developed model (Wittmann et al., 2017) where individual fitness is determined by a seasonal score ( $z$ ). This score is the sum of contributions from multiple adaptive loci. Fitness ( $w$ ) is calculated via a power transformation ( $w = (1+z)^y$ ), where the exponent $y$ controls the intensity of epistasis (specifically, diminishing-returns epistasis).
Simulation Parameters:
- Demography: Simulations were run with a constant census population size ( $N_c$ ) of 1 million diploid individuals (downscaled to 500k for computational efficiency in some runs) over 36,000 generations (equivalent to 18,000 generations in the downscaled model).
- Seasonality: The environment alternated between Summer and Winter, with 10 generations per season.
- Loci: Initial simulations tested 100 to 500 seasonal loci with varying epistasis intensities ( $y$ ranging from 0.5 to 20). Allele frequencies started at 50%.
- Empirical Calibration: To mimic natural D. melanogaster, the authors derived specific parameters (number of loci and epistasis intensity) using multiple linear regression models fitted to data from recent empirical studies (e.g., Bergland et al. 2014; Machado et al. 2021; Rudman et al. 2022).
$N_e$ Estimation: The study calculated the variance effective population size ( $N_e$ ) by tracking the temporal variance in allele frequencies of neutral markers (inserted at 100kb intervals on unlinked chromosomal arms) over a full seasonal cycle. This allowed the authors to measure the genome-wide impact of selection on linked neutral diversity.
Robustness Checks: The authors tested various scenarios, including:
- Varying census population sizes ( $N_c$ from 20k to 1M).
- Seasonal "boom-bust" demography (population size fluctuating between 1M in summer and 100k in winter).
- Capping offspring numbers (limiting reproductive variance to 20 offspring per parent) to address potential artifacts of the SL model.

3. Key Contributions

Quantification of Fluctuating Selection Impact: This is the first study to provide direct, empirically informed estimates of how polygenic seasonal selection reduces $N_e$ in a realistic genomic context.
Predictive Modeling: The authors developed regression models that accurately predict allele frequency amplitudes based on the number of segregating loci and epistasis intensity ( $R^2 > 0.96$ ).
Identification of the Primary Driver: They demonstrated that the reduction in $N_e$ is not driven by the mean fluctuation of all loci, but is strongly correlated with the maximum allele frequency amplitude of the single most fluctuating locus.
Decoupling from Census Size: The study shows that the proportional reduction in $N_e$ is largely independent of the census population size ( $N_c$ ), challenging the idea that fluctuating selection alone can explain "Lewontin's Paradox" (the weak correlation between $N_e$ and $N_c$ across species).

4. Key Results

Severe Reduction in $N_e$ : Under empirically realistic scenarios for D. melanogaster, seasonal fluctuating selection reduces the effective population size by approximately 50% relative to the census size ( $N_e/N_c \approx 0.5$ ). Reductions ranged from 7% to 76% depending on the specific parameter set.
Role of Maximum Amplitude: A strong linear correlation ( $R^2 = 0.702$ ) was found between the maximum seasonal allele frequency amplitude and the magnitude of $N_e$ reduction. Loci with the most extreme oscillations (e.g., ~35–45% amplitude observed in nature) drive the bottleneck effect.
Mechanism: The reduction is caused by serial reproductive bottlenecks. As seasons change, individuals with genotypes favored in the previous season become maladapted, while previously maladapted individuals gain a reproductive advantage. This creates high variance in reproductive success (offspring number) at the transition points between seasons, effectively shrinking $N_e$ .
Stabilization: Allele frequency oscillations stabilize rapidly (within ~1,200–2,000 generations), and the resulting $N_e$ reduction stabilizes quickly thereafter.
Demographic Independence: The proportional reduction in $N_e$ remained consistent across different constant population sizes. Even in "boom-bust" scenarios, the reduction scaled with the harmonic mean population size, maintaining the ~50% reduction relative to the effective size.
Offspring Capping: Limiting the maximum number of offspring per parent (to prevent unrealistic variance) slightly reduced the magnitude of $N_e$ loss in extreme scenarios but did not negate the ~50% reduction observed in moderate scenarios.

5. Significance

Re-evaluating Genetic Diversity: The findings suggest that standard models of genetic diversity may significantly overestimate $N_e$ in species subject to strong seasonal selection (like insects and other ectotherms). The "hidden" variance in reproductive success caused by seasonal adaptation is a major, previously under-appreciated constraint on genetic variation.
Lewontin's Paradox: While fluctuating selection causes significant $N_e$ reductions, the study concludes it is insufficient to fully resolve Lewontin's Paradox on its own because the reduction is largely independent of $N_c$ . Instead, it likely acts in concert with linked selection and demographic history.
Evolutionary Dynamics: The results highlight that fluctuating selection creates a persistent "drag" on genetic diversity, distinct from the transient effects of selective sweeps. This has implications for conservation genetics and predicting adaptive potential in rapidly changing climates.
Methodological Advance: By successfully recapitulating empirical allele frequency oscillations and linking them to $N_e$ via forward simulation, the study provides a robust framework for future investigations into temporal selection in diverse taxa.