New Climate Change Initiative: High Resolution Land Cover Map from the European Space Agency
* Register (or log in) to the AI4G Neural Network to add this session to your agenda or watch the replay
Land cover is one of the main environmental climate variables that is highly correlated with climate change. The role of land for storing carbon and its future potential for offsetting carbon emissions is an area of active research. In the framework of the European Space Agency (ESA) Climate Program, the Climate Change Initiative (CCI) High Resolution Land Cover project is aimed to generate high resolution (10/30m) land-cover and change detection maps at subcontinental/regional level to study the role of the spatial resolution in climate modelling research.
This webinar presents the architectures and the methodologies developed for implementing the full processing chains that have been developed to process Earth Observation data and generate the products. Special focus will be given to i) the data processing and machine learning approaches used, ii) the adopted cascade paradigm to multitemporal data classification and iii) to the procedures used for model training and map validation. These steps are very complex considering the requirements to work at large scale with multisensor (multispectral and SAR) images and heterogeneous data quality moving since 1990 to nowadays. The primary products of the project consist of: (i) land-cover maps at subcontinental scale at 10m (generated for 2019 only) as reference static input to the climate models, (ii) a long-term record of regional land cover maps at 30m for the historical analysis every 5 years (generated in the period 1990-2019), and (iii) change information at 30m at yearly scale consistent with historical land-cover maps.
The products have been generated in three crucial regions for climate change studies (the Amazon basin, the Sahel band in Africa and the northern high latitudes od Siberia). In the webinar they will be presented together with a demonstration of the procedure for their download from the ESA portal and a description of their format and properties.
This live event includes a 30-minute networking event hosted on the AI for Good Neural Network. This is your opportunity to ask questions, interact with the panelists and participants and build connections with the AI for Good community.