This is my final project for my University diploma in GIS and Spatial Analysis that I have obtained from the University of Grenoble in 2025. The aim of the project was to visualise the geographical areas (covering 106 countries where malaria is present) that are likely to become more or less conducive to optimal conditions for mosquito populations and the malaria transmission cycle in 2050, compared with 2020.

I did a statistical analysis and geospatial analysis at the regional level (Adm1) for the 106 countries. I implemented a scoring system based on three variables (temperatures, precipitations, and humidity) as these three variables have a large impact on the reproduction cycles of mosquitoes. Each area got a score assigned for 2020 and 2050, the final analysis was to identify the areas where the score increase (greater risk) and the ones where the score decreased (lower risk).

Malaria incidence rate (2020)
Score evolution (2020-2050)

Data sources:

  • Malaria Atlas Project
  • Geoboundaries
  • Copernicus Climate Data Store
  • Global Data Lab
  • Groupe intergouvernemental d’experts sur l’évolution du climat
  • Humanitarian Data Exchange
  • ESRI

Final report (in French) here. 1-pager infographic (in French) here.

This is a student project realised under significant time constraint, it is therefore a start and is far from perfect. I had already identified possible improvments to be made at the time of the delivery:

  • Go into greater depth, but focus on a specific country or region rather than the global scale
  • Add data and statistics on population density
  • Use more criteria (e.g., elevation, season)
  • Use available time-series data on incidence rates and meteorological data to better visualize trends over time
  • Take surface water into account (coastal ponds, floodplains, irrigation systems, ponds, and pools), as it is an equally important factor because it provides a habitat for mosquitoes to lay their eggs. Estimating future surface water levels, however, is very challenging.
  • Altitude: It is difficult to obtain a Digital Elevation Model for the entire globe. Although this data exists, the files are very large, and downloads often must be done by region, which makes it less practical. This study therefore does not take this factor into account.
  • Wider use of PostGIS and an extensible, modular data model to allow for the addition of new annual data (incidence rates, meteorological data)