The HARMONIZE Toolkit is an open-source, reproducible suite of digital tools designed to collect, harmonize, and integrate diverse data sources across varying spatial and temporal scales. Its primary goal is to support public health decision-making in climate change hotspots across Latin America and the Caribbean.

The toolkit enables users to integrate data from a wide range of sources including ground-based sensors, drone imagery, satellite observations, climate reanalysis and forecasts, socioeconomic surveys, and health surveillance systems. Each tool consists of (semi-)automated workflows in R or Python tailored to specific data types, allowing users to process data to desired spatial and temporal resolutions (e.g., generating monthly dengue case counts by province in Brazil).

By harmonizing multiple spatio-temporally aligned data streams into unified, multi-source datasets (e.g. monthly dengue incidence, extreme precipitation events, and poverty rates by province) the toolkit facilitates the exploration of links between health outcomes and environmental or socioeconomic drivers. Tools are designed to accommodate both experienced programmers and users with limited coding experience, with comprehensive documentation and user guides to support broad adoption.

By creating multi-source, temporally and spatially harmonized datasets, the HARMONIZE Toolkit empowers researchers, public health professionals, and policy-makers to uncover actionable insights, strengthen health systems and build climate resilience.

Tools

• data4health
  Processes  health data (e.g., numbers of symptomatic individuals or confirmed disease cases), usually sourced from surveillance systems.
  
  
• clim4health
  Processes  climate data (e.g., precipitation and temperature) collected via satellite remote sensing, weather stations, or reanalysis and forecast datasets.
  
  
•  socio4health
  Processes  socioeconomic data (e.g., population distribution by income, education, gender, ethnicity, or age), typically sourced from surveys or censuses.
  
  
•  land4health
  Processes land use and land cover data (e.g., rural vs. urban classification, vegetation cover) primarily using satellite imagery.
  
  
•  drone4health
  Processes drone imagery (e.g., to detect environmental features like standing water or waste sites that contribute to disease vector proliferation).
  
  
• cube4health
  Facilitates the integration, processing, and publication of drone, climate, and health data into accessible data cubes for analysis and visualization.