According to the Food and Agriculture Organization of the United Nations (FAO, 2020), forests cover more than 31% of the total land area in the Earth. Forest ecosystem provides shelter for animals and maintains biodiversity, but also supplies the oxygen and plays a key role in global carbon cycle while storing circa 861 gigatons of carbon totally (WRI, 2021) from which 42% in live biomass (above and below ground biomass). More particularly, tropical forests are known as the greatest carbon stock containers, they represent only 30% of total global forest but contain about 50% of the world’s forest carbon stock.
During this webinar, we will use Sentinel-2 data to estimate the value of above-ground biomass in a forest area in Ethiopia. We will calculate different Sentinel-2 derived vegetation indices which will be used as predictors to linear and non-linear machine learning regression models. At the end of the exercise, we will evaluate the results and produce biomass maps for the area of interest, which you can compare later with available online global biomass maps. As reference values of above-ground biomass, we will use the publicly available database developed for the years 2010, 2017 and 2018 by the ESA Biomass Climate Change Initiative (BCCI), whereas processing of the image of Sentinel-2 data will be done with the snappy module in Python applied in the Jupyter Notebook environment.
This webinar will also show you how to benefit from the RUS Service to download, process, analyse and visualize the free data acquired by the Copernicus satellites.
You can replay this webinar through our RUS Copernicus Training channel available on Youtube.
You can also retrieve the corresponding training support in the Train with RUS section of the RUS portal.