changed 2 years ago
Published Linked with GitHub

Development and Sharing of Open Geodata

   


18 January 2023
ITC Auditorium, Enschede and Online

Event Page
https://itc.nl/big-geodata/meetings/event/development-and-sharing-of-open-geodata/


Image Not Showing Possible Reasons
  • The image file may be corrupted
  • The server hosting the image is unavailable
  • The image path is incorrect
  • The image format is not supported
Learn More →
READ ME FIRST!

This is the official collaborative discussion and feedback document of the meeting.

You can use the document to ask questions or provide comments on specific presentations. It will be also used during the plenary discussion session of the event.

  • Please add your personal information to the Participants section below, with your institution, job title, e-mail address, and a short nickname.
  • Please indicate your nickname in parentheses after your question or comment, so that it doesn't stay anonymous.
  • Please provide your feedback about the meeting under the Feedback section.

Image Not Showing Possible Reasons
  • The image file may be corrupted
  • The server hosting the image is unavailable
  • The image path is incorrect
  • The image format is not supported
Learn More →
Live broadcast is available at https://itc.nl/live on 18 January, 9:45-16:00.


Quick Links:


Introduction

Please do not edit this section.

Open geodata is the idea that certain geospatial and Earth Observation (EO) data should be freely available to everyone to use, re-use and redistribute without restrictions like copyright or patents, but with requirements such as attribution and share- alike. It is considered as a best practice for modern geo-information and EO research in line with Open Science and FAIR principles. However, developing easily accessible (e.g. cloud-optimised) and re-usable (big) geospatial datasets, and sharing them "effectively" with the research community and other stakeholders is challenging and usually requires a team effort including researchers, research software engineers, technology advisors, and community managers.

Collaboration between research institutions to share technical and practical know-how on best practices, as well as lessons learned while developing and sharing open geodata, supports such efforts and helps in developing an open data culture for open research and development communities. The meeting on "Development and Sharing of Open Geodata" aims to initiate such a collaboration between ITC and OpenGeoHub, by bringing both institutions together to present state-of-the-art research activities with a focus on open geodata, discuss the needs and challenges, and pursue collaboration opportunities especially to produce joint datasets and software.

OpenGeoHub is an independent research foundation (a DeepTech organisation) with 20 staff, primarily data scientists and technical experts, located at the Agro Business Park in Wageningen. It focuses on promoting open data and open source software solutions to support regional and global projects, and currently leads the Open Earth Monitor and Eco Data Cube, and is also involved in the Land Carbon Lab project led by WRI.

Programme

Please do not edit this section.

9:00 – 9:10

9:55 – 10:20

Presentation on ITC and its 4 profiling themes (Prof. Dr. Karin Pfeffer, Portfolio Holder Research ITC)

10:20 – 10:45

Presentation on OpenGeoHub Foundation and its open geodata activities (Dr. Tomislav Hengl, Director OpenGeoHub Foundation)

10:45 – 11:00

Coffee break

11:00 – 12:30

Presentations of ITC and OpenGeoHub projects:

12:30 – 13:30

Lunch

13:30 – 14:15

Presentations of ITC and OpenGeoHub projects:

  • AI4SoilHealth: accelerating collection and use of soil health information using AI technology to support the Soil Deal for Europe and EU Soil Observatory (Dr. Ichsani Wheeler, OpenGeoHub)
  • Open geospatial data publication at ITC (Masoome Shariat, ITC-FB)

14:15 – 14:45

Plenary Discussion:
Needs and Challenges of Developing and Sharing Open Geospatial Datasets

14:45 – 15:00

Coffee break

15:00 - 16:00

PhD's in the Spotlight:
Lightning presentations of PhD candidates with a focus on datasets that will be developed during their studies.

  • Space-time modeling of vegetation in a changing environment (Carmelo Bonannella, OpenGeoHub)
  • Global long-term daily 1 km surface soil moisture dataset with physics-informed machine learning (GSSM 1km) (Qianqian Han, ITC)
  • Big and Useful Maps: Land use / land cover classification with high thematic depth (Martijn Witjes, OpenGeoHub)
  • Microwave Remote Sensing for soil moisture estimation and vegetation characterization with Physics-Informed Machine Learning (Ting Duan, ITC)
  • Mapping land potential / tracking land degradation using EO data (Julia Hackländer, OpenGeoHub)
  • HRVQA: A Visual Question Answering Dataset for High-Resolution Aerial Images (Kun Li, ITC)
  • Mapping soil health indicators with EU soil database (Xuemeng Tian, OpenGeoHub)
  • An overview of the process for publishing data: Spatiotemporal dynamics of domestic wastewater (Nestor De la Paz Ruiz, ITC)

16:00 - 17:00

Drinks

Speakers

Please add your name, institution, job title, e-mail address, and nickname to the list.

Presentations

Please ask your questions or provide your comments under each presentation by indicating your nickname.

ITC and Profiling Themes

OpenGeoHub Foundation and Open Geodata Activities

  • rolf: Clearly, open source software brings you in contact with intelligent people because its use requires intelligence. Or reversely, some of the commercial software has a dummified interface. ("Every function hides behind a button.")

  • rolf: Change "cool hobbies" to "true passions."

  • RZM: Thank you for the presentation. Academia is more than research so I am wondering about your views on Open Educational resources (also because of your experience with summer schools)

  • Yijian: yes, i agree with Raúl, NWO is currently taking it up as well to acknowledge more than research. A question to Tom is that: How do you view the 'technical threshold' to enable the open science? (there are many knowledge accumulated using for example matlab etc., it would be great the Open Science is inclusive for those software not initially desgined open)

Open Earth Monitor

Towards a Soil-Plant Digital Twin

  • Tom: "Digital Twin" cannot predict everything that will happen in physical world. Perhaps, we should say 'simulate' instead of 'predict'.
  • Yijian: Fully agree with this observation. This is coming from the general definition of 'digital twin' in industries. For Digital Twin Earth, there are different sectoral sectors (biodiversity, hydrology and Water, Food and Agriculture, etc.https://digital-strategy.ec.europa.eu/en/library/destination-earth), and ECWMF is trying to build two digital twins to address the impact of extreme climates on these sectors (https://stories.ecmwf.int/destination-earth/index.html).

EcoDataCube

  • rolf: very interesting work. I am keen to understand what drove the decision to aggregate to (only) 4 dates per year. Some applications in agriculture would like to see more, and perhaps not at constant frequency, but more takes in some part of the year.
  • Yijian: The optimization of all elements in your processing chain sounds really cool! I am wondering if there is a paper indicating these perspectives?
  • Marcin: in top of what WebGIS soft ecodatacube.eu is build on?

Glacier Outlines using Deep Learning

IDEAMAPS

AI4SoilHealth

Open Geodata at ITC

PhD's in the Spotlight

Space-time modeling of vegetation in a changing environment

  • Yijian: what is the key feature determining the vegetation distribution? And is the human activity factored into your modeling?

Global long-term daily 1 km surface soil moisture dataset with physics-informed machine learning

Big and Useful Maps: Land use / land cover classification with high thematic depth

  • Yijian: Similar question to the first talk. What is the key features controlling your classification? And what are physical mechanisms behind such feature importance? If the 'distance to coast' is important in your algorithm, what is the physical explanation of that importance?

Microwave Remote Sensing for soil moisture estimation and vegetation characterization with Physics-Informed Machine Learning

  • Martijn: what do you mean by 'physics-informed'?

Mapping land potential / tracking land degradation using EO data

HRVQA: A Visual Question Answering Dataset for High-Resolution Aerial Images

Mapping soil health indicators with EU soil database

An overview of the process for publishing data: Spatiotemporal dynamics of domestic wastewater

Discussion

Please ask your questions or provide your comments on the needs and challanges of developing and sharing open geospatial datasets by indicating your nickname.

Feedback

Please provide your feedback about the meeting under this section. Anonymous feedback is also very welcome.

  • A good overall workshop today, intensive, but worthwhile with interesting speakers and applications. We need to give good thought on what is possible as follow-up.
Select a repo