Wetland Change in Argentina

For an independent study in applications of remote sensing for sustainable urban planning, I used Google Earth Engine and global wetlands datasets to train a Random Forest model on Landsat 7 spectral images and assess wetlands change from 2000 to 2020.

My model achieved 95% accuracy and 65% recall, and suggests that the Iberá wetlands have declined significantly in the past 20 years.

Here, you can see my code in Google Colab:

Additionally, for a wetland ecology class, I put together a presentation on how remote sensing is used for wetland monitoring: