5  Model description

You are reading the work-in-progress of this report. This chapter is currently a dumping ground for ideas, and I don’t recommend reading it.

This description follows the ODD protocol for Agent-Based Models (ABMs) (Overview, Design concepts, and Details). See Grimm et al. (2010), Müller et al. (2013), and Grimm (2020) to learn more.

5.0.1 Overview (O)

5.0.1.1 Purpose and patterns

The purpose of this Agent-Based Model (ABM) is to investigate how anthropogenic climate change impacts the health and nutrition of Brazilian children under five years old who are monitored in the Primary Health Care of Brazil’s public health system (SUS). The model seeks to elucidate the dynamic interactions between food systems, health outcomes, and environmental changes, focusing on the vulnerability of young children to both malnutrition and obesity.

5.0.1.2 Entities, state variables, and scales

5.0.1.2.1 Entities
  • Children
    • Age (continuous)
    • Weight-for-height (categorical: underweight, normal, overweight)
    • Health status (categorical: healthy, stunted, wasted, obese, sick)
  • Families
    • Income level (categorical: low, middle, high)
    • Education level (categorical: low, medium, high)
    • Dietary practices (categorical: healthy, unhealthy)
    • Food security status (categorical: secure, insecure)
  • Schools
    • Availability of healthy foods
    • Participation in school feeding programs
    • Educational programs on nutrition
  • Environmental factors
    • Frequency and intensity of extreme climate events (numerical)
    • Deforestation rate (numerical)
    • Levels of greenhouse gas emissions (numerical)
5.0.1.2.2 State variables
  • Children’s nutritional status (e.g., stunting, wasting, obesity rates)
  • Health outcomes (e.g., prevalence of morbidity and mortality)
  • Food security (levels of food availability and affordability in families)
  • Climate indicators (frequency of extreme events, deforestation rates)
5.0.1.2.3 Scales
  • Time step: Monthly, spanning over 10 years
  • Geographical scale: Regions within Brazil monitored by SUS

5.0.1.3 Process overview and scheduling

  • Monthly update
    • Calculate climate impacts.
    • Adjust food system variables based on climate inputs.
    • Evaluate food availability and prices.
    • Update household food security status.
    • Update children’s health and nutritional status.
    • Capture feedback from school nutrition programs.
    • Aggregate health outcomes for the region.

5.0.2 Design concepts (D)

5.0.2.1 Theoretical and empirical background

The model is grounded in the global syndemic, incorporating synergistic interactions between obesity, undernutrition, and climate change. Empirical data is derived from literature on food systems, nutritional status in children, and climate impacts on food production.

Add assumptions, hypotheses, and sources here.

5.0.2.2 Basic principles

The model integrates the socio-ecological model, recognizing the multi-level influences on children’s health outcomes, from individual dietary practices to broader environmental impacts.

5.0.2.3 Individual decision-making

Children and their families make decisions about food consumption based on availability, affordability, and cultural practices. Schools make decisions on food program implementations.

5.0.2.4 Emergence

Emergent phenomena include patterns in children’s health outcomes, regional disparities in malnutrition and obesity rates, and the cumulative impact of climate events on food security.

5.0.2.5 Adaptation

Agents adapt by modifying dietary practices, ?seeking alternative food sources?, and responding to policy interventions (e.g., subsidies for healthy foods).

5.0.2.6 Objectives

The primary objective for families and schools is to ensure the nutritional well-being of children under five, despite changing environmental conditions.

5.0.2.7 Learning

Adaptive behaviors emerge as agents learn from past climate events and adjust their strategies to improve resilience (e.g., diversifying crops, enhancing food storage).

5.0.2.8 Prediction

Families predict food availability based on current climate conditions and adjust consumption and purchases accordingly.

5.0.2.9 Sensing

Agents sense local environmental conditions (e.g., market prices, weather forecasts) and community health trends (e.g., outbreaks of illnesses).

5.0.2.10 Interaction

Interactions occur between families, schools, and environmental factors, such as access to school feeding programs and exposure to extreme climate events.

5.0.2.11 Heterogeneity

Agents are heterogeneous in terms of socio-economic status, educational background, and geographical location, leading to diverse impacts and adaptive capacities.

5.0.2.12 Stochasticity

Stochastic elements include the occurrence and magnitude of extreme climate events and random variations in food prices.

5.0.2.13 Collectives

Collective entities include family groups, school communities, local markets, and the broader environmental system.

5.0.2.14 Observation

Model output includes health and nutritional indicators at individual, family, and regional levels, food security status, and environmental metrics.

5.0.3 Details (D)

5.0.3.1 Implementation details

The model is implemented using the NetLogo framework and data integration from national health and demographic databases and climate models.

5.0.3.2 Initialization

The model initializes with real-world data on children’s health and nutritional status, family demographics, school feeding programs, and baseline climate conditions.

5.0.3.3 Input data

Input data includes climate forecasts, food price indices, nutritional surveys, and demographic statistics.

5.0.3.4 Submodels

  • Climate impact submodel: Simulates the effects of climate change on food production and availability.
  • Nutrition submodel: Tracks children’s health outcomes based on dietary intake and access to food.
  • Food security submodel: Assesses household food availability and affordability.
  • School feeding submodel: Captures the effects of school meal programs on children’s nutrition.