How To Set Up a Google Earth Engine Cloud Project

Author(s): Ruiz Rivera

Originally published on Towards AI.

In the realm of environmental monitoring and geospatial analysis, Google Earth Engine (GEE) stands out as a game-changing platform. For researchers, data scientists, and environmental professionals, creating a Google Earth Engine Cloud project account opens up a world of possibilities in handling and analyzing vast amounts of geospatial data. Before the dawn of cloud computing platforms like Microsoft’s Azure and Google Cloud Platform (GCP), research of this magnitude was impossible with just a laptop due to the computational power required to process this data. However, with the GCP Earth Engine, we now have the remote sensing capabilities of accessing a multi-petabyte catalog of satellite imagery and geospatial datasets for the potential to conduct planetary-scale analysis! And if that weren’t enough, users would also have access to decades of historical imagery and scientific datasets with just a few clicks or lines of code, depending on our interface.

If you want to see an example of how you could leverage the Earth Engine platform for spatial analysis, feel free to check out our article on mapping a wildfire perimeter here.

In it, you will be able to imagine how access to timely satellite imagery can have an impact on helping emergency responders fight a fire effectively in a remote area of the Canadian province of British Columbia (BC). For context on why this work was so impactful, BC had experienced an unprecedented surge in wildfire severity and frequency in the past few years, causing significant damage to its social and ecological systems. In 2023, British Columbians saw a record-breaking 2.84 million hectares burned, more than doubling the previous record set in 2018 (BC Public Service, 2023). These fires forced mass evacuations and likely resulted in substantial wildlife losses. Early detection of wildfires, especially in remote areas, is crucial but challenging due to BC’s vast and diverse terrain. However, the advancements in satellite technology and remote sensing that we’re introducing through Earth Engine offer new possibilities for wildfire detection and perimeter estimation. Therefore, the article focuses on using Sentinel-2 satellite data from the Google Cloud Earth Engine to build wildfire perimeters. The process involves extracting satellite data, applying normalized burn ratio transformations, and visualizing the results using the Python-based mapping package folium. This approach aims to enhance wildfire detection capabilities and provide a broader understanding of remote sensing techniques in this critical field.

So if you’re ready to journey down this rabbit hole, here’s how to get started.

1. Sign up for access to Google Earth Engine

a) Visit the Google Earth Engine sign-up page

b) Follow the instructions to request access:

• For this exercise, we recommend registering the project as for noncommercial, unpaid usage;
• Select the project type that most resembles your usage;
• If you don’t already have an existing Google Cloud Project you’d want to register for the Earth Engine API, we recommend the option to “Create a new Google Cloud Project” that you can name accordingly.

2. Create a Google Cloud Project

a) Go to the Google Cloud Console projects page;

b) Click on “Create a project” at the top of the page;

c) Provide a name for your project name in the appropriate field and specify the location or organization if necessary;

d) Click “create” at the bottom of the page.

3. Enable the Earth Engine API

a) In the Google Cloud Console, navigate to “APIs & Services” > “Library”;

b) Search for “Earth Engine”;

c) Click on “Earth Engine API” and then “Enable”;

4. Enter the project id

a) Once you’ve created a Google Cloud Project, return to the projects page, and you should see your project listed under “Resources”;

b) Beside the “name” column is a column named ID. Copy the ID value associated to your project and enter it in the project argument of the ee.Initialize() function below;

# Trigger the authentication flow.
ee.Authenticate()

# Initialize the library.
ee.Initialize(project="ee-hughlynch-demo")

c) A few pop-up links should open up, follow the instructions to authenticate your access to Earth Engine’s cloud compute.

*Note: You may need to wait for approval after signing up for Earth Engine access. The process usually takes a few days to a week.

That’s it, folks. We now have everything we need to start our remote sensing journey using the Earth Engine platform. Amidst the increase of natural disasters we’ve collectively observed in recent years, from record-breaking wildfires, floods, or widespread deforestation, there is an urgent need for advanced monitoring and detection systems. Google Earth Engine and Google Cloud offer powerful solutions to address these challenges, providing researchers, data scientists, and environmental professionals with the tools to make a significant impact.

By leveraging the vast satellite imagery catalog, computational power, and advanced analytics capabilities of these platforms, we can revolutionize our approach to managing wildfires and other such natural disasters. From early detection and accurate perimeter mapping to comprehensive environmental analysis, the possibilities are vast and the integration of these technologies is not only beneficial but essential. In bridging the gap between space-based observation and on-the-ground efforts, we’re not only enhancing our ability to protect communities and ecosystems but also paving the way for more informed, data-driven decision-making in environmental management. The combination of Google Earth Engine’s geospatial capabilities and Google Cloud’s analytics prowess will undoubtedly play a crucial role in our ongoing efforts to understand, predict, and mitigate the impact of natural disasters and environmental crimes globally.

The journey to create more resilient and responsive strategies begins with harnessing these powerful tools, and the time to act is now.

References

• BC Public Service. (2023, December 7). Wildfire Season Summary. https://www2.gov.bc.ca/gov/content/safety/wildfire-status/about-bcws/wildfire-history/wildfire-season-summary

Join thousands of data leaders on the AI newsletter. Join over 80,000 subscribers and keep up to date with the latest developments in AI. From research to projects and ideas. If you are building an AI startup, an AI-related product, or a service, we invite you to consider becoming a sponsor.

Published via Towards AI