These Jupyter Notebooks are part of the Hartree National Centre for Digital Innovation (HNCDI) Explain course on Geospatial Data and Modeling.
In this repository, a series of notebooks have been developed to guide users through the GeoDN modeling and data discovery capabilities. A video series accompanies this course material.
This repository is a place to explore the capabilities of GeoDN, to suggest new features and raise bugs or issues. See the contributing section for more details.
To sign up to the Geospatial Data and Modeling Explain course, please visit the Hartree Centre Training Portal.
The course is split in to two sections, A Practical Guide to Geospatial Data and Fundamentals of Geospatial Data and Modeling. Fundamentals of Geospatial Data and Modeling includes two parts, Part 1 - Geospatial Data Discovery for Climate Risk requirements and Part 2 - Geospatial Foundation Models and Workflows.
The course will be launched in January 2024 and will be hosted on STFC's Learning Management service. As part of the course, users will be given access to the GeoDN platform via Open Data Hub where these notebooks will be available to run.
Course content:
Requirements:
- Prior knowledge of python.
- Basic understanding of jupyter notebook, perhaps undertake an introduction to jupyter notebooks course.
- Optional to complete Beginner's Guide to Geospatial Data course.
Target audience:
Data analysts and data scientists, business analysts and risk analysts who want to build models to leverage geospatial data (e.g. weather, climate, satellite data), and potentially combine with their own datasets (for example about location of property or infrastructure). Target domains include supply chain, asset management, insurance, or climate risk.
Course content:
Part 1: Geospatial Data Discovery for Climate Risk
- Query Analyse Rainfall
- Flood model and real flood extent with buildings
- Impact Functions with OpenStreetMaps and Flood Outlines
Part 2: Geospatial Foundation Models and Workflows
- Burn scar fine-tuning
- Workflow development/testing and sharing
- Running a workflow from the catalogue
Requirements:
- Practical Guide to Geospatial Data.
- Experience with Python.
- Previous geospatial data science experience is not necessary.
Target Audience:
Data analysts and data scientists, business analysts and risk analysts who want to build models to leverage geospatial data (e.g. weather, climate, satellite data), and potentially combine with their own datasets (for example about location of property or infrastructure). Target domains include supply chain, asset management, insurance, or climate risk.
Blair Edwards, Anne Jones, Katerina Reusch, Paolo Fraccaro, Rosie Lickorish, Junaid Butt & Geoffrey Dawson.