Modelling the Dynamics of an Aedes albopictus Population

We present a methodology for modelling population dynamics with formal means of computer science. This allows unambiguous description of systems and application of analysis tools such as simulators and model checkers. In particular, the dynamics of a population of Aedes albopictus (a species of mosquito) and its modelling with the Stochastic Calculus of Looping Sequences (Stochastic CLS) are considered. The use of Stochastic CLS to model population dynamics requires an extension which allows environmental events (such as changes in the temperature and rainfalls) to be taken into account. A simulator for the constructed model is developed via translation into the specification language Maude, and used to compare the dynamics obtained from the model with real data.

💡 Research Summary

The paper introduces a formal, computer‑science‑based methodology for modelling the dynamics of an Aedes albopictus (Asian tiger mosquito) population. Traditional ecological models rely on differential or difference equations that treat population change as a smooth, deterministic process, often ignoring the stochastic influence of environmental factors such as temperature and rainfall. To overcome these limitations, the authors adopt the Stochastic Calculus of Looping Sequences (Stochastic CLS), a process‑algebraic formalism originally designed for describing biochemical systems.

A key contribution is the extension of Stochastic CLS with “environmental events.” In the extended framework, the rates of the stochastic rewrite rules are not fixed constants but functions of external variables (e.g., temperature, precipitation). For instance, the rule governing the transition from egg to larva carries a rate that increases when temperature lies within a favorable range, while a rule for larval mortality incorporates rainfall intensity. By embedding these functions directly into the rewrite semantics, the model can react in real time to any time‑varying climate data set.

The authors translate the extended Stochastic CLS specification into Maude, a high‑performance rewriting engine. The Maude implementation receives as input an initial multiset representing the numbers of eggs, larvae, pupae, and adults, together with a time‑series of temperature and rainfall measurements. At each simulation step Maude evaluates the current environmental event, updates the corresponding rates, and applies the stochastic rewrite rules to generate a new population state. This approach yields a discrete‑time stochastic simulation that faithfully reproduces the life‑cycle stages of A. albopictus while preserving the probabilistic nature of each transition.

To validate the model, the authors compare simulation outputs with field data collected over several seasons in a temperate region. The simulated trajectories capture the characteristic summer peak and winter trough of mosquito abundance. Moreover, the model accurately reproduces abrupt population surges following heat waves and sharp declines during prolonged dry periods, demonstrating that the explicit treatment of environmental events substantially improves predictive performance.

Beyond simulation, the formal nature of the model enables automated verification. By coupling the Maude specification with a model‑checking tool, the authors verify safety properties such as “the adult population never falls below a critical threshold within a 30‑day horizon under any admissible climate scenario.” This capability provides public‑health officials with rigorous, scenario‑based risk assessments that are difficult to obtain from purely statistical models.

The discussion highlights the broader applicability of the framework. Because the life‑cycle stages are encoded as rewrite rules and environmental dependencies are expressed as parameterised rate functions, the same methodology can be adapted to other vector species, to incorporate additional climatic variables (humidity, wind), or to evaluate control strategies (insecticide spraying, release of sterile males). The authors propose future work on integrating finer‑grained climate models, accounting for genetic variation among mosquito populations, and coupling the formal model with geographic information systems for spatially explicit predictions.

In summary, the paper demonstrates that extending Stochastic CLS with environmental events and implementing the resulting model in Maude yields a transparent, reproducible, and analytically tractable tool for mosquito population dynamics. The approach bridges the gap between formal computer‑science techniques and ecological modelling, offering a powerful platform for both scientific investigation and evidence‑based public‑health decision making.