by Tim Van Achte
Together with Yves D’Eer (Campus Vesta team member within E2mC), I attended a satellite data workshop in Ghent in March 2018. Organised by the ever inspiring city platform #AppsForGhent, this activity formed part of a bigger initiative #ghentinspace. We looked at how these practices can be valuable in emergency management at the city level, in function of the research and development on sensors, social media and crowdsourcing we are doing as part of the E2mC project.
Call for Experts • We are looking for Emergency Planning / Response / Management Experts to provide us with your valuable insights within an innovative European project • learn more
AppsForGhent is a “city of people” initiative, offering a range of activities built around a conference and one large hackaton event. Every year, the annual hackaton results in prototypes of smart and innovative applications, made by and for citizens, local initiatives and local organisations. Completely based on “real world” use cases, I was not at all surprised to discover that some of these Apps for Ghent have already managed to scale up to actual city wide applications gaining in popularity. Developed and shared in an open source philosophy, I noticed that these apps, web services, modules and other technological components result in a direct positive impact to the daily life and work. They benefit the many communities forming the Belgian city I live and work in: Ghent.
Thanks to speakers of the European Environment Agency, Terrascope and “Informatie Vlaanderen” (former AGIV) we witnessed one staggering development after the other, all around the idea of smartly deploying various satellite based technologies. The bottom line of the workshop for me was that we are not always aware of the powerful technologies and data sources we have at our disposal. For free, and in an open data, open source philosophy. For instance, during the hands-on part of this workshop, we experienced the power of:
- Combining satellite based open data into GIS layers (Note: GIS stands for Geographic Information System, it resembles a range of applications, systems, services and data formats that enables anyone to work with – and exchange – so called “geospatial” data, which simply means: data on maps).
- Performing our own analysis on these map layers.
- Importing pre-analysed satellite based data.
- Performing our own calculations on this data and produce this personalised output on new maps.
(Translated: Ludvig Forslund teaches us more about @CopernicusEU data!)
Various use cases – inspired by real city challenges – were covered, such as:
- Calculate the anticipated yield and optimal position of solar panels on rooftops, based on collected height data, slope level of the roof surface, etc. and providing this information as a service to citizens.
- Analysing changes in the population density fabric between 2006, 2012 and 2018, to visualise an enriched view on local community evolutions in the city.
- Locate where agricultural areas are turning into urban areas.
- Predict food shortages based on the evolution of vegetation.
- Long term damage assessment of bridges and other structures.
We’re sharing some of our takeaways on this fascinating subject:
- Initiatives like Terrascope are making great advancements and setting the example in providing more user friendly ways to get working with satellite data. Typically, satellite data can take terrabytes, is greedy for processing power, and requires very specific technical knowhow and tech savvy experts. Terrascope is successfully countering these issues with low level web based tools, virtual machines, code libraries, etc. Open access to raw data should always be accompanied by resources and services like these. Lowering the many technological thresholds is crucial when organisations want citizens and organisations to embrace open data into their life.
- Open data providers such as the European Environment Agency are doing a great job on providing more than the actual satellite acquisitions to the public. We learned that (pre-)analysed data is shared as well. It was great to discover the true potential of geospatial data formats, so called “shapefiles”. There is much more to these files than a collection of georeferenced ‘polygons’ or ‘point’ on a map. Each polygon/point is also a full record, and comes with attributes of raw or analysed data. This results in a “map layer as a database”. Filtering and querying this database enables endless possibilities. We’re curious to see some of this potential come to life during the #ghentinspace hackaton (also organised by AppsForGhent).
- In all three open data portals from the different governmental levels present in this workshop (Europe, Belgium, Flanders) there is much more to discover than visual satellite imagery. Radar and even laser technology based datasets should be able to kickstart equally interesting innovative services. The ambitious project of “Informatie Vlaanderen” to provide a computer model of Flanders is an example that sparked my curiosity. (Their step by step approach: first provide a 2D model, then a 3D one in the sense that cubes represent buildings, then a next level 3D model in the sense that for each building the roof is modelled on top, etc.)
- We learned that – in average – only 25% of photographic satellite based images actually show the ground level, due to the presence of clouds. You’ll realize how this statistic is evident the next time you step into an airplane 😉 For emergency management services that rely heavily on satellite data (such as Copernicus EMS) this realization demonstrates more than ever the need to investigate other additional sources of imagery. Literally looking at the area impacted by the disaster from other perspectives where possible. Eye witness reports shared on social media, drones, UAV’s (unmanned aerial vehicles), public webcams, press content, relevant online content submitted by the crowd, people as sensors and many other crowdsourced information sources are the extra eyes on the ground you will need when clouds spoil the view from above. All of these can contribute to improve decision making in emergency management. Two of these, social media and crowdsourcing components is exactly what the E2mC project is focussing on. A second consequence of this realization is that I now have even more respect for the work of those people that provide “cloud free” mosaic pictures from many separate pieces of satellite imagery 😉
(A tweet containing a good example of a cloud free “mosaic” composed of many different high resolution satellite pictures. More examples below. Truly beautiful views from above!)
The workshop opened our eyes to a lot of possibilities concerning the smart usage of Copernicus satellite data in different domains. Where the discussed examples mainly focused on the general topic of improving life in cities, open data can definitely initiate improvements in the crisis management domain. It’s without a doubt a large piece of the puzzle to make cities more resilient and crisis ready.
Don’t hesisate to contact me if you want me to share more information or links on any of the topics discussed in this article!
Thank you to the City of Ghent and Apps For Ghent to invite us for this workshop. Learn more about their 2018 hackaton on space data, you can participate as visitor or as hacker.
It’s always a good idea to follow these accounts on Twitter 😉
Tim Van Achte
for Campus Vesta in the E2mC project
More useful resources and tweets:
The Campus Vesta team in this project is composed of: Bert Brugghemans, Tim Van Achte, Kenny Meesters, Yves D’Eer and Kathleen Van Heuverswyn.
The E2mC Project is co-funded by the European commission / H2020 Programme / Grant Agreement No. 730082