Follow forest and mangrove evolution

With climate change, there are more and more forest fire. In addition with human deforestation… the planet is in danger. The issue to tackle is tremendous. This specific tool allows you to assess the state of mangroves and forests trough time using machine learning methods and tree canopy coverage filter.

Objectives

Follow the evolution of forest and mangrove and identify areas where the vegetation is wiping out. Control the tree canopy coverage percentage and the degradation stratification in order to raise awareness in people mind and lunch different project. The purpose is to monitor those areas for people or companies who would like to implement conservation and reforestation project. A biomass map is in development. .

Solution

Interactive mapping highlighting vegetation disaster due to human or natural disaster.

This graphical interface allows the users to monitor mangrove evolution through time and estimate the degradation rate through a 17-degradation level. Every result is based on Mundi DIAS images that we collect from Sentinel-1 and Sentinel-2 data.

Using Machine learning Random Forest on a model, the backend algorithm can predict the degradation level in other areas and different times. The users can add a learning model with some requirement. According to the model you choose, the output can be different, the best choice is to use a model where its zone is near to the zone you want a result. (Ground similarity). Beside, biomass mapping is in development to estimate the biomass rate (Mg/ha) of an area, also base on Machine Learning.

Using FCOVER (tree canopy coverage), the backend algorithm is able to give you on a 3-level degradation the deforested ratio 0-10%, 10-50%, 50-100% tree canopy

The solution is therefore a good way to have an estimation of how degraded the area is. Once it has been quantified, the project could start on a solid footing.