Monitoring Victorian Forestry using Free and Open Earth Observation Data - A Tutorial for the Digital Earth Australia Sandbox

Summary

This repository contains learning materials for a tutorial that teaches participants how to use free and open satellite data to monitor forestry activities in the state of Victoria, Australia. The materials make use of a free and open computing platform, the Digital Earth Australia Sandbox, along with analysis-ready Sentinel-2 data for Australia. The materials have been designed to be delivered by an instructor over three one-hour sessions, either in-person, online, or both (hybrid mode).

Context

Earth observation data from satellites are an important source of information for a wide variety of environmental issues. As both the quantity and quality of Earth observation data increases – due to the rapid growth in satellites equipped with high-resolution sensors – it becomes possible to obtain detailed information about surface features: vegetation, soils and minerals, and water bodies. Further classification can occur such as determining the type and quantity of vegetation, water content, or plant health. Through regular re-imaging of regions of the Earth, it becomes possible to measure environmental changes over short- (days to weeks) and longer-term (months to years) timescales.

The tutorial poses that the participant has started working as a graduate data analyst at the Victorian Forest Monitoring Program, and must report back to their team about how different logging events appear in satellite imagery.

Learning objectives

For people getting their first exposure to the field of Earth observation, there can be many challenges: understanding how to work with large quantities of data; learning how to tie satellite observations to real-world environmental changes; mastering computer programming skills to automate data processing; and communicating their findings to inform those who make critical environmental decisions. This tutorial has been designed to guide participants through each of these challenges, providing a real-world scenario where they must analyse satellite imagery and report on their findings.

By completing the tutorial, participants will gain:

• Experience in applying best-practice methods for environmental monitoring with satellite imagery.
• Practice in conducting a real-world project, where data is collected, analysed, and interpreted.
• Exposure to Python, one of the leading programming languages for data analysis.
• Familiarity with the Digital Earth Australia platform – Australia's free and open analysis platform for satellite data, provided by Geoscience Australia.

Audience

This material was designed for university students with a general science background.

Depnding on participants' familiarity with remote sensing, environmental monitoring, and programming, instructors may need to provide participants with additional information before beginning the practical. To support this, we provide some additional materials on

• Environmental monitoring (short video)
• The Digital Earth Australia Sandbox (short video)
• Programming with Python (Jupyter notebook)

Running this tutorial

To run the tutorial, we recommend that instructors have knowledge in:

• Remote sensing, including familiarity with multispectral data and band indices.
• Python, including geospatial libraries such as geopandas and xarray.
• Digital Earth Australia, including the Digital Earth Australia Sandbox.

This material can be run in-person, online, or in a hybrid environment. We recommend a class size of up to 20 people with two instructors. In a hybrid environment, we recommend that one instructor attend in person, and one attend online.

For more details, see the instructions README.

Contributing

We welcome contributions back to the repository. To contribute, please make a fork of the repository, and then submit a pull request with your changes.

Acknowledgements

This resouce was developed by Dr. Caitlin Adams and Professor Chris Fluke (Swinburne University of Technology). The resource was desined for the Space Environment, Data, Visualisation and Applications course course at Swinburne University of Technology. The HyFlex implementation was supported by a Teaching and Learning Innovation Grant from the Australian Council of Environmental Deans and Directors.

We are grateful to Geoscience Australia for providing the Digital Earth Australia Sandbox and associated Python libraries, which are the foundation for this resource.

Citations

Krause, C., Dunn, B., Bishop-Taylor, R., Adams, C., Burton, C., Alger, M., Chua, S., Phillips, C., Newey, V., Kouzoubov, K., Leith, A., Ayers, D., Hicks, A., DEA Notebooks contributors 2021. Digital Earth Australia notebooks and tools repository. Geoscience Australia, Canberra. https://doi.org/10.26186/145234