Creating optical-based Earth Observation Applications
kindly offered byAgricultural University of Athens
Date / Time: Saturday 101 November / 14:20-16:20 EET @ Workshop room (1st floor Right - Room 105)
The purpose of the workshop is to help attendees understand how optical satellite images work, how they are processed to extract results from wildfire events and predict the severity of a potential wildfire event.
During the workshop, attendees will learn about the data sources that are needed to predict the severity of wildfire events, like EFFIS, NASA Power, Sentinel-2, etc.
They will also learn how to process ESA’s COPERNICUS Sentinel-2 data using ESA’s SNAP software. SNAP is an open source common architecture for ESA Toolboxes ideal for the exploitation of Earth Observation data.
Background info and Definitions
Forest wildfires constitute a pressing global concern, causing widespread environmental, economic, and social damage. The increasing frequency and intensity of these events have been attributed to factors such as climate change, deforestation, and human activities. Accurate and timely prediction of wildfire severity, particularly the size of the area that a fire might cover, is crucial for effective disaster management, resource allocation, and mitigation efforts.
Remote sensing data, especially from satellite platforms such as Sentinel-1 and Sentinel-2, as well as meteorological data from sources like NASA Power or European Centre for Medium-Range Weather Forecasts (ECMWF), provide a wealth of information that can be harnessed to develop machine learning and deep learning models for wildfire severity estimation.
EFFIS – European Forest Fire Information System – supports the services in charge of the protection of forests against fires in the EU and neighbor countries and provides the European Commission services and the European Parliament with updated and reliable information on wildland fires in Europe.
NASA Power – Provides solar and meteorological data sets from NASA research for support of renewable energy, building energy efficiency and agricultural needs.
Sentinel-2 – is an Earth observation mission from the Copernicus Programme that systematically acquires optical imagery at high spatial resolution over land and coastal waters.
Sentinel-Hub (optional) – is an API that allows users to make WMS and WCS web requests to download and process satellite images from various data sources.
The objective of the workshop is to make the atendee capable of understanding and creating optical-based Earth Observation Applications.
- Access to Copernicus Sentinel-2 satellite images.
- Correct satellite images for the effects caused by the atmosphere.
- Vegetation Mapping.
- Burn Area Mapping.
- Burn severity forecasting with Machine Learning (ML).
The Graph Builder module is used extensively during the course in order to create workflows. Example XML workflow files are provided within the workshop to better help students practice at their own computers! Example Sentinel-2 images are also included in the workshop, so the students can test the contents of the course presented directly at their computers!
Regarding the burn severity forecasting with machine learning, the EO4Wildfires dataset will be provided, along with sample code to run ML experiments during the workshop.
- Students / Engineers of computer science
- Students / Engineers interested in Earth Observation technologies and applications
- Students / Engineers in the forestry domain
Prerequisites on Audience:
- Basic knowledge of earth observation
- Laptop with installed ESA’s SNAP software, and Python with pytorch
- Presentation in pdf format
- Burn area mapping workflow in XML format
- Example code for EO4Wildfires
- Access Multispectral Sentinel-2 Images
- Getting the Sentinel-2 images to learn multispectral Earth Observation Applications
- Dataset preparation
- Atmospheric Corrections
- Atmospherically correct Sentinel-2 images
- Comparison between corrected (L2A) and uncorrected (L1C) images
- Resampling L2A images
- Vegetation and Burn Area Mapping
- Sampling spectral signatures
- Leaf Area Index
- Improved masking before mapping
- Normalized Burn Ratio Index
- Burn area mapping workflow
- Wildfire Severity Prediction
- Introduction to the EO4Wildfires dataset
- Dataset exploration
- Example code ML experiments
The work presented in this course has been conducted in the framework of SILVANUS project. This project has received funding from the European Union’s Horizon 2020 research and innovation program under grant agreement No 101037247.
The contents of this course are the sole responsibility of the authors and can in no way be taken to reflect the views of the European Commission.