A Rao-Blackwellized particle filter for joint parameter estimation and biomass tracking in a stochastic predator-prey...

Area of Science:

• Ecology
• Mathematical Biology
• Computational Ecology

Background:

• Predator-prey dynamics are fundamental to ecosystem stability and function.
• Accurate estimation of functional responses and population biomass is crucial for ecological modeling.
• Stochastic models offer a more realistic representation of ecological interactions.

Purpose of the Study:

• To develop and validate a novel particle filtering method for ecological applications.
• To estimate the behavioral parameter (effective search rate) in a Lotka-Volterra functional response.
• To forecast population biomass in a predator-prey system using field data.

Main Methods:

• A stochastic predator-prey model with a Lotka-Volterra functional response was employed.
• A particle filtering approach combining sequential Monte Carlo and analytical integration was developed.
• The method was tested using both synthetic datasets and empirical data from an acarine predator-prey system.

Main Results:

• The particle filtering method successfully estimated the behavioral parameter of the functional response.
• Accurate forecasting of population biomass was achieved using both simulated and real-world ecological data.
• The method demonstrated robust performance in tracking dynamic population changes.

Conclusions:

• Particle filtering provides an effective tool for parameter estimation and population forecasting in ecological studies.
• The proposed method enhances our ability to understand and predict predator-prey dynamics.
• This approach has significant implications for ecological monitoring and management strategies.