About Geo-harmonizer

Implementation plan 2020–2022.

Geo-harmonizer: EU-wide automated mapping system for harmonization of Open Data based on FOSS4G and Machine Learning
Duration: September 2019 – July 2022
Total budget: 1.8M EUR
Funding source: 2018 CEF Telecom Call – Public Open Data (CEF-TC-2018-5)
General programme: Connecting Europe Facility 2014–2020

The GeoHarmonizer project aimed at reducing problems of national data with using seamless complex (geographical) data over the entire extent of the EU, and “opening data” through:

• Using Open Data licenses, 
• Enabling wider public access to the data by not only scientists and specialists, but also non-professionals, 
• Facilitating increased use of EC-funded data without imposing any expectations on users to possess specialised or costly infrastructure, 
• Working closely with the national authorities, organizations and NGOs including the existing EC-funded systems such as DIAS, Copernicus programme and similar. 

Geo-harmonizer tools and data are meant to generate decision-ready layers such air quality and pollution, potential natural vegetation, potential for producing energy from solar insolation, wind energy and similar. To calculate these value-added decision-ready maps, we use topographic data (DEMs), Earth observation (EO) data, hydrological and meteorological data that we map using automated mapping systems largely based on Machine Learning algorithms available in the Open Source software.

Our Mission

The Copernicus land data products and similar data products funded by European Commission play a central role in the reference data framework that determines the quality and traceability of all Copernicus products and services. But many of these layers suffer from limited reproducibility and completeness. By building and implementing Geo-harmonizer we contributed to the Copernicus product in producing new added-value products. For example, we tested automated mapping methods for producing annual land cover maps at 30 m resolution for the complete period 2000–2020, the “OpenStreetMap+” (improved EU version of the OpenStreetMap) and a number of other value-added layers. Such products can then be directly used both for national and regional projects across Europe. 

The Digital Service Infrastructure we developed and released to the public connects state-of-the-art remote sensing data sources (Sentinel, Landsat, TanDEM-X and similar), machine learning frameworks, cloud and High Performance Computing. We expect that it also results in significant reduction of processing and delivery time. We further believe that automation of the harmonization of layers will also decrease the costs of generation of value added data important for energy, climate, protection of the environment and sustainable use of natural resources. By implementing GeoHarmonizer, we demonstrated the power of the Copernicus programme in combination with Open data projects such as OpenStreetMap and Open Source software projects.

Our mission consisted of automating the production and harmonization of patchy data products to make them more available / accessible through web-GIS and similar technology.

Geo-Harmonizer aimed at partnering with multiple similar international and national projects:

• OpenEO: We leveraged knowledge gained in the EU-H2020 openEO project (https://openeo.org, “A common open source interface between Earth Observation data infrastructures and front-end applications; H2020 H2020 grant 776242, Call EO-2-2017: EO Big Data Shift) in which mundialis is a partner with focus on backend development for massive data processing and analysis.
• Sentinel-2 for Agriculture: (http://www.esa-sen2agri.org/). The Sen2-Agri project is designed to develop, demonstrate and facilitate the Sentinel-2 time series contribution to the satellite EO component of agriculture monitoring. The project will demonstrate the benefit of the Sentinel-2 mission for the agriculture domain across a range of crops and agricultural practices.
• Climate – ADAPT: (https://climate-adapt.eea.europa.eu/) is a partnership between the European Commission and the European Environment Agency that supports Europe in adapting to climate change giving access to and sharing data and information on: expected climate change in Europe, current and future vulnerability of regions and sectors, etc.
• Candela: (http://candela-h2020.eu/) is a cross-disciplinary project combining the AI researchers with the EO technology sector. It has similar objectives as the GeoHarmonizer project but is much broader.
• OpenML.org: The Open Machine Learning project aims at building an open, organized, online ecosystem for machine learning. It creates tools and software to run ML in open source programming environments such as R.