Agro-informatics, a powerful approach integrating information technology with agricultural data management, analysis, and application, offers a revolutionary solution to tackle critical global challenges like improving nutrition, agricultural productivity and sustainability, rural livelihoods, and global food security.
Agro-informatics depends on reliable, timely, high-resolution, integrated location-based information across diverse sectors, including Food Security, Crops, Vegetation, Livestock, Agricultural Trade and Production, Land Cover/Use, Soil, Water, Fisheries, Forestry, Socioeconomic, Climate, Topography, Environment, Emergencies and Infrastructure. However, the complexity and fragmentation of heterogeneous data sources present a significant challenge.
Empowered by big data and the cloud, agro-informatics at scale is revolutionising food and agriculture through data-driven insight. Near real-time data analysis can identify potential threats such as droughts, pests, and diseases, enabling timely interventions and mitigating their impact.
This presentation explores how the Food and Agriculture Organization (FAO) is tackling these challenges, building a unified environment for accessing, managing, visualising and analysing diverse datasets. We delve into the approach and architecture, detailing the tools, infrastructure and techniques used to collect, store, process, integrate, federate, and analyse diverse agricultural data sets. We also acknowledge the challenges associated with developing and operating such a platform at scale. We showcase the transformative power of agro-informatics at scale by exploring use cases in early warning, decision support, intervention design, monitoring, and evaluation. Finally, we outline the data governance and operation of the platform.
Date
27 November 2024, 13:00-14:00 CET
Speaker
Karl Robert MORTEO
Lead Developer, Digital FAO and Agro-informatics Division
Karl has over 30 years of experience developing and supplying IT solutions & services. He is currently working at the Food and Agriculture Organization, where he is currently leading the development of the flagship agro-informatics platform. This platform brings together technical units from across the Organization and integrates data from the United Nations system, international organisations, NGOs, space agencies, public data providers, and the private sector. Previously, Karl has delivered IT solutions at the global, regional, and national scale in animal health, livestock, fisheries, crops, plant genetics and pests, food security, climate, trade and markets, soil, land, and water. He has also worked in telecoms, merchant banking and the space industry.
Karl is a Cloud Architect, Scrum Master, PRINCE2 project manager and an ITIL practitioner. He graduated from the University of Leeds with a bachelor's degree in Computational Science and completed a master’s degree in Intelligent Knowledge Based Systems at Cranfield Institute of Technology.
Video
This video player uses cookies
Questions and Answers
In the beginning you mentioned hexagon grids. Are they supported? Currently we do not support hexagon or more specifically H3. We have been treating data in such format as individual shape files and processing them as vector plus tabular data. What we're actively working on is to make them well known areas in the platform so that they they have a clear separation between the data warehouse. So if you're using the H3 code, you would be able to then access data. And that means that you will also be able to get the automatically generated statistics for those hexagons. We're increasingly seeing data being published in that format. We we hope to support better next year.
This answers my question.
Thank you for your feedback
Who is authorized to use the data available on the platform? There's no simple answer to that. The FAO data in general is licensed under Creative Commons and it is the default license for data generated by ourselves. Unfortunately, not all data can be made available in that format, so an example is some of the sensitive animal health data or data that identifies specific location of a teak tree or a pesticide store.
The way that the sharing mechanism works is that there area organizations on the platform, and they decide who is in that group and they decide which data that's owned by that group is public and only visible to members of the group. Let's say somebody in the University of Twente has an organization and they published some data. They would have then the owner of the data, and they can decide which data will be public and which data will be kept private.
This answers my question.
Thank you for your feedback
Are there certain criteria that needs to be met to publish data on the platform? There are two concepts regarding data publishing. One of them is that it appears in the catalogue so that people can go and find it, and the other one that it appears in an application, and so they're slightly different. You might have a data set that's in a catalogue, but it's not in an application. Secondly, the data on the platform requires an FAO sponsor, so we don't typically charge people to upload data to the platform unless we need to do data engineering work. The purpose of the sponsors is in fact to provide the trust and the accountability for the data that's eventually published.
This answers my question.
Thank you for your feedback
