AIMES Analysis

This repository contains code for analyzing the relationship between climate variability modes and the basic reproduction number ($R_0$) of Aedes-borne diseases across different regions and the whole globe. The analysis utilizes the climate component of disease transmission in AIMES (Aedes-borne dIsease Monitoring of Environmental Suitability), described in Corvillo et al., 2025, and focuses on correlation and causality between climate indices and its $R_0$ values, with time series decomposition to separate different components (trend, seasonal, decadal, and remainder).

Setup and Requirements

1. Clone this repository
2. Install required R packages if needed:

install.packages(c("s2dv", "ncdf4", "viridis", "dplyr", "zoo", "signal", "ggplot2", "cowplot", "tidyr", "purrr", "reshape2"))

3. Install required Python packages if needed:

pip install numpy scipy matplotlib xarray cartopy

4. Download the data sources from this link, and extract the zip file in 0_data_and_functions
5. The complete analysis can be run using the provided shell script:

./execute_analysis.sh

Analysis Workflow

The analysis is based on three main data sources:

1. $R_0$ data: The basic reproduction number for vector-borne diseases, obtained from the AeDES2 Monitoring System.
2. Temperature data: Surface temperature from AIMES's observational datasets (Era5 + Era5Land + CPC Unified Global + GHCN-CAMS).
3. Climate variability indices: Various climate variability indices generated with the temperature data, including:

• Atlantic 3 (ATL3)
• Indian Ocean Basin Mode (IOB)
• Indian Ocean Dipole (IOD)
• El Niño 3.4
• North Pacific Meridional Mode (NPMM)
• South Atlantic Subtropical Dipole (SASD)
• South Indian Ocean Dipole (SIOD)
• South Pacific Meridional Mode (SPMM)
• Tropical North Atlantic (TNA)

And the analysis consists of three main steps:

1. Timescale Decomposition:

• Detrend temperature data
• Decompose $R_0$ time series into trend, seasonal, decadal, and remainder components
• Generate climate indices for correlation and causality analyses

2. Correlation and Causality Analyses:

• Analyze relationships between climate variability modes and $R_0$
• Determine causal relationships between climate indices and $R_0$
• Identify the dominant climate modes affecting each region

3. Output Processing and Visualization:

• Merge correlation and causality outputs
• Generate visualizations for different regions and seasons
• Create summary maps and plots of relationships between climate indices and disease dynamics

The analysis generates various outputs in the 4_outputs/ directory:

1. Data files:

• Detrended temperature and $R_0$ data (.nc)
• Climate indices data (.dat)
• Correlation and causality maps (.nc)

2. Figures (.png, .eps):

• Temperature trend plots
• Time series decomposition plots
• Correlation maps for different regions and seasons
• Causality maps for different regions and seasons
• Summary visualizations

Repository Structure

aimes_analysis/
├── 0_data_and_functions/                # Core data and functions used across analyses
│   ├── data/                               # Input data for analysis
│   │   ├── r_nought/                          # $R_0$ data files from AeDES2
│   │   └── tas/                               # Temperature data files from AIMES's climate datasets
│   └── functions/                          # Utility functions
│       ├── r_functions.R                      # Utility functions for R scripts
│       └── python_functions.py                # Utility functions for Python scripts
├── 1_timescale_decomposition/           # Analysis of time components
│   ├── temp_detrend.R                      # For detrending of temperature values
│   ├── timescale_decomposition.R           # $R_0$ time series decomposition
│   └── write_indices.R                     # Generation of climate indices
├── 2_correlation_and_causality/         # Correlation and causality analyses
│   ├── correlation_4_r0.py                 # Correlation analysis script
│   └── causality_4_r0.py                   # Causality analysis script
├── 3_merge_and_plot/                    # Output processing and visualization
│   ├── output_merge.R                      # Merge analysis outputs
│   └── output_plotting.py                  # Generate visualizations
├── 4_outputs/                           # Analysis outputs
│   ├── data/                               # Generated data files
│   │   ├── climate_indices/                # Climate indices data
│   │   ├── correlation_and_causality/      # NetCDFs for analysis outputs
│   │   ├── detrended_vars/                 # Detrended variables
│   └── figures/                            # Generated plots and visualizations
└── execute_analysis.sh                  # Shell script to run the full analysis