Modelling Population Dynamics in Realistic Landscapes with Linear Elements: A Mechanistic-Statistical...

Area of Science:

• Mathematical Biology
• Ecology
• Spatial Dynamics

Background:

• Species distribution and spread are influenced by landscape features.
• Linear corridors like roads can significantly impact species movement and population dynamics.
• Existing models often simplify landscape complexity, potentially limiting predictive accuracy.

Purpose of the Study:

• To propose and develop a general reaction-diffusion modeling approach for species dynamics in 2D landscapes with 1D corridors.
• To create a hybrid 2D/1D model integrating population dynamics in both landscape areas and corridors.
• To validate the model using the invasive spread of Aedes albopictus in France, utilizing highway networks as corridors.

Main Methods:

• Developed a hybrid 2D/1D reaction-diffusion model with homogeneous coefficients.
• Employed a mechanistic-statistical approach coupling population dynamics with a probabilistic observation model.
• Utilized maximum-likelihood estimation to fit model parameters to real-world infestation data at the departmental level.

Main Results:

• The 2D/1D model demonstrated a better fit and predictive power compared to standard homogeneous reaction-diffusion models.
• The model successfully captured the range expansion dynamics of Aedes albopictus.
• Highways were identified as significant corridors facilitating the spread of Aedes albopictus in mainland France.

Conclusions:

• The hybrid 2D/1D modeling approach is effective for studying species dynamics in complex landscapes with corridors.
• Corridor effects, such as highways, play a crucial role in species range expansion and require explicit consideration in ecological models.
• The findings support the importance of landscape connectivity in invasive species management and ecological studies.