Google Earth Engine
Workshops sessions are recorded and posted to the UA Datalab Youtube Channel about 1 week after the live session.
Google Earth Engine is a cloud-based platform that enables users to process and analyze geospatial data. It provides access to a massive collection of satellite imagery, weather data, and other geospatial datasets. Users can use Earth Engine to visualize data, create maps, and run geospatial analysis. Google handles all of the cyberinfrastructure allowing researchers to focus on their geospatial projects. It is a truly ground-breaking platform that allows planetary scale data analysis and facilitates collaboration.
Earth Engine is free to use for anyone with a Google account. It is important to note that Google accounts issued through universities are not likely to work. This is because usage of Earth Engine is tied to setting up a Google Cloud project and many university ITs departments have disabled the ability to create Google Cloud projects from a university account. So to get access to Earth Engine, you need to use a personal Google account. They are of course free to obtain. With a valid personal Google account, you can get started with Earth Engine by going to this address https://code.earthengine.google.com/register and setting up a free Google Cloud project.
For more detailed information on setting up an account, please check out this blog.
The main interface for working in Earth Engine is the Code Editor. It is a web-based integrated development environment (IDE) that executes commands using JavaScript. Details on the Code Editor can be found here in the official Earth Engine documentation. A tutorial on basic usage of the Code Editor can be found here
Earth Engine has curated many community developed tutorials for using Code Editor to do many things. They are a great way to get started!
Users get an allocation of 250 GB of free storage to host your own data within Earth Engine.
Code and data assets can be worked on collaboratively in Code Editor. It just takes a few clicks to share with colleagues.
The official Earth Engine data catalog has over 70 petabytes and 800+ curated geospatial datasets, including 40 years of satellite imagery. All the datasets are public to use and can be brought into your analysis with ease.
Another data catalog of note is the Awesome GEE Community Data Catalog. It is a large collection all contributed by Earth Engine users and curated by Samapriya Roy.
If you prefer coding with Python, but you can use the GEE Python library to interact with Earth Engine from your local machine. Through this modality, you are still using Google Cloud to access the datasets and perform heavy computational tasks such as image processing.
Alternatively, you can execute Python commands using Google Colab. Go here for an example of using Colab including how to authenticate your account and use the Python API.
A vast collection of Jupyter Notebooks using Earth Engine can be found in this Github Repo
Within Earth Engine, supervised classification methods such as CART, RandomForest, NaiveBayes and Support Vector Machine are supported in the Code Editor and Python API.
Deep learning and neural network machine learning are generally not supported within Earth Engine. There are, however, a few narrow use cases that are possible. For example, it is possible to use data within Earth Engine to train neural network models within the framework TensorFlow. It is also possible to do prediction with GEE from an outside model. Please see the documentation on ML for more information.
An amazing feature of Earth Engine is the ability to create interactive web apps based on your geospatial analysis. This is a great way to show colleagues or clients a slick presentation of data, charts, and maps. You can host the website on a public url that anyone in the world can access without an Earth Engine account. Check out this example app showcasing drone imagery.
Official Earth Engine Documentation
___
