Project By: Avani Adhikari
This project was developed as a final assignment for the "AI for Urban Sustainability" course. The primary objective is to assess and visualize the environmental and social vulnerabilities of Kathmandu, Nepal, by leveraging open-source datasets and geospatial AI techniques. The project was done in 4 parts:
-
Using building footprint data from OpenStreetMap, a binary Random Forest Model was used to create the building footprint density of the city.
-
Using the Digital Elevation Model from NASA, flooding was modeled to create an inundation dataset.
-
Using Landsat 8 data from Google Earth Engine, NDVI values for the entire city were calculated.
-
Finally, all of these datasets were normalized and combined with wealth distribution data from the 2021 Nepal census to create a vulnerability map. The final product of this can be seen below.
