Skip to content
MC

LabInteractive demonstrators

Explore the models. Understand the mechanisms.

Each module states what it is: a published model transcribed from the public code, a simplified demonstrator or a teaching diagram. None replaces the published analyses.

Compare trajectories under different assumptions.

Three models, three scales of reading. The first two are the published systems, transcribed from the notebooks; the third is the discrete teaching model of chapter 2.

Observethe mechanism

Sterilisation is not always perfect. This model follows three populations (sterile males S, wild females F and wild males M) and distinguishes two forms of residual fertility.

δ: share of released males that remain fully fertile. ε: share of matings with a sterile male that remain fertile.

S˙=(1−δ) σ−μSS\dot S = (1-\delta)\,\sigma - \mu_S SF˙=r(1−ρ) M+εηSk+M+ηS F1+βF−μFF\dot F = r(1-\rho)\,\frac{M + \varepsilon\eta S}{k + M + \eta S}\,\frac{F}{1+\beta F} - \mu_F FM˙=rρ M+εηSk+M+ηS F1+βF−μMM+δσ\dot M = r\rho\,\frac{M + \varepsilon\eta S}{k + M + \eta S}\,\frac{F}{1+\beta F} - \mu_M M + \delta\sigma

Explorethe parameters

500 /day
1 %
0 %
  • No releases
  • Releases, perfect sterility
  • Releases with residual fertility
Wild female density over 300 days. Without releases it stays at 983. With σ = 500 sterile males per day and perfect sterility it reaches 18.3; with δ = 1 % and ε = 0 % it reaches 175.2.02004006008001000050100150200250300time (days)wild females F

Understandthe results

At t = 300 days, the model gives 18.3 females with perfect sterility and 175.2 with residual fertility, against 983 without releases.

Raise δ: even with intense releases, the solid blue curve can settle above zero, a reduction rather than a suppression.

Courtois, Mailleret, Touzeau, van Oudenhove & Grognard (2025). Bulletin of Mathematical Biology 87, 25.

TypeScript port (RK4, step 0.1 day) of the notebook's modele_SIT function, with its fixed parameters and the initial condition F₀ = M₀ = 983. Checked: reproduces the notebook's “time dynamics” figure (σ = 500 and 5,000). Teaching exploration: it does not replace the published analyses.

From principle to trajectory, in five steps.

Releasing sterile males does not automatically suppress a population: competition, residual fertility and remating can change the outcome.

Step 1/5 · Wild males and females breed: each mating can produce offspring.

  • wild male
  • sterile male
  • female
  • mated with a sterile male

Steps 1–4: illustrative diagram. Step 5: residual-fertility model (Courtois et al., 2025), parameters from the public notebook.

Release often, or release a lot? Impulsive releases.

With a constant weekly budget (200 sterile males per week), is it better to release often and few, or rarely and many?

7 d
Release
200 ind.
Oscillation
435 – 635

The longer the period, the more the sterile male population oscillates: it spends part of the time at a low level. This is one side of the frequency–amplitude trade-off studied in chapter 5 of the thesis.

  • Simulation (pulses)
  • Analytical periodic solution
Sterile male population with releases of 200 individuals every 7 days. It converges to a periodic solution oscillating between 435 and 635.0100200300400500600700020406080100time (days)sterile males

Formula transcribed from Notebook_part_impulsive_releases.ipynb (μS = 0.054 day⁻¹). Teaching demonstrator.

From individuals to the population.

A population model describes aggregated numbers through equations. An individual-based model simulates each insect (its lifespan, its matings, its egg laying) and the population dynamics emerge from these rules.

Sperm-use rule

200

Equivalent to 10,000 individuals per day at the scale of the original model.

Individual scale — females and the content of their spermatheca

Day 0: .

  • fertile
  • fertile + sterile
  • sterile
  • unmated

Population scale

day 0 / 100 · replicate no. 1

  • Larvae
  • Females
  • Wild males
Individual-based simulation, last rule, 200 sterile males per day. Day 0: 0 larvae, 0 females, 0 wild males, 0 sterile males.0200400600800020406080100time (days)individuals (1/50 scale)

Methodological note

Simplified JavaScript adaptation of the Mesa model in script_ABM_batchrun.py(opens in a new tab). Kept: lifespans (sterile males 1–36 days, wild males 31–129 days, females 21–144 days), larval mortality by stage, emergence after 6–11 days, sex ratio 0.5, probability of mating with a wild male W/(W + ηS) with η = 0.59, refractory periods, Poisson egg laying (6 eggs/day), larval carrying capacity and the “first”, “last” and “mixed” rules.

Simplifications

Numbers scaled down by a factor of 50, daily releases, no harvest, “choice” and “preference” rules not included, a single replicate shown. This is a demonstrator: its outputs are not comparable with the article's results, obtained with 100 repetitions per scenario.

Code and reproducibility

Research does not stop at the figures.

The scripts and notebooks of my thesis make it possible to retrace the numerical analyses and visualisations used in the manuscript. They are an essential part of the documentation and reproducibility of the work.
Code explorerthesisData / Chapter 3 - Residual Fertility/Script_Manuscrit_SIT_RF.ipynbOpen on GitLab (new tab)

System of differential equations

verbatim excerpt

def modele_SIT(etat,t,params):    x = etat    μF,μM,μS,ρ,β,r,k,σ,η,𝜖,𝛿 = params    # association of each equation to a vector element    S = x[0] # sterilized males    F = x[1] # wild females    M = x[2] # wild males    # definition of each differential equation    dS = - μS*S + (1-𝛿  )*σ    dF = - μF*F + r*(1-ρ)*((M+𝜖*η*S)/(k+M+η*S))*(1/(1+β*F))*F    dM = - μM*M + r*ρ*((M+𝜖*η*S)/(k+M+η*S))*(1/(1+β*F))*F + 𝛿*σ    return [dS, dF, dM]

Cell 28. This system is ported to TypeScript in the model explorer.