[Paper Review] Mathematical Modeling on Obligate Mutualism: Interactions between leaf-cutter ants and their fungus garden

Yun Kang, Michael Makiyama+2 more|arXiv (Cornell University)|Feb 11, 2011

Insect and Arachnid Ecology and Behavior4 citations

TL;DR

This study develops a novel mathematical model based on Michaelis-Menton kinetics to analyze the obligate mutualism between leaf-cutter ants and their fungus garden during early colony development. It reveals that division of labor and initial population sizes are critical for coexistence, with sensitivity analysis showing that fungal growth rate and ant population parameters are more influential than conversion or death rates.

ABSTRACT

We propose a simple mathematical model by applying Michaelis-Menton equations of enzyme kinetics to study the mutualistic interaction between the leaf cutter ant and its fungus garden at the early stage of colony expansion. We derive the sufficient conditions on the extinction and coexistence of these two species. In addition, we give a region of initial condition that leads to the extinction of two species when the model has an interior attractor. Our global analysis indicates that the division of labor by workers ants and initial conditions are two important {factors} that determine whether leaf cutter ants colonies and their fungus garden survive and grow can exist or not. We validate the model by doing the comparing between model simulations and data on fungal and ant colony growth rates under laboratory conditions. We perform sensitive analysis and parameter estimation of the model based on the experimental data to gain more biological insights on the ecological interactions between leaf cutter ants and their fungus garden. Finally, we give conclusions and {discuss} potential future {work}.

Motivation & Objective

• To understand the population dynamics of leaf-cutter ants and their fungus garden during the early ergonomic growth stage.
• To address the lack of theoretical models for obligate mutualisms, especially those involving complex behavioral traits like division of labor.
• To validate the model against empirical data from laboratory colonies and estimate hard-to-measure parameters.
• To investigate the stability and coexistence conditions of the mutualistic system using global dynamical analysis.
• To explore the impact of life-stage-specific behaviors and initial conditions on colony survival and growth.

Proposed method

• Adapts Michaelis-Menten kinetics to model the functional and numerical responses in the ant-fungus interaction.
• Constructs a two-species system (ants and fungus) with density-dependent growth and consumption terms based on Holland and DeAngelis' consumer-resource framework.
• Incorporates division of labor by modeling worker ants' roles in fungus cultivation and leaf harvesting as distinct functional responses.
• Performs global stability analysis to derive sufficient conditions for extinction and coexistence of both species.
• Uses parameter estimation and sensitivity analysis to calibrate the model with experimental data on colony growth rates.
• Validates model predictions against laboratory data, particularly focusing on discrepancies in weeks 6–9 to suggest model limitations.

Experimental results

Research questions

• RQ1 How do division of labor and initial population sizes influence the coexistence or extinction of leaf-cutter ants and their fungus garden?
• RQ2 What are the sufficient conditions for stable coexistence versus mutual extinction in this obligate mutualism?
• RQ3 How do key parameters such as fungal growth rate, ant population growth, and death rates affect colony development outcomes?
• RQ4 Why does the model show a poor fit with data during weeks 6–9, and what does this imply about the biological processes during this phase?
• RQ5 Which parameters have the greatest influence on model output, and how can they be estimated from empirical data?

Key findings

• The model identifies a threshold initial population size for both ants and fungus required for coexistence, with failure to meet this threshold leading to mutual extinction.
• Sensitivity analysis shows that the half-saturation constant $ b $ has the largest influence on model output, followed by fungal growth rate $ r_f $ and ant growth rate $ r_a $.
• Parameter estimation reveals that conversion rate $ r_c $ and ant death rate $ d_a $ are extremely small and highly sensitive to initial guesses, suggesting minimal impact on early-stage dynamics.
• The model fits well with experimental data during early colony growth, but shows inconsistency between weeks 6 and 9, indicating potential limitations in capturing later developmental stages.
• The presence of multiple attractors in the model suggests that population dynamics may be highly unstable during early colony development, depending on initial conditions.
• The study concludes that a more detailed model incorporating life stages (eggs, larvae, pupae) or stochasticity is needed to better capture the complexity of early colony dynamics.