Cloud-Based Remote Sensing with Google Earth Engine: Process and Prospects from a Large Edited Open-Access Book

Google Earth Engine is one of the foremost platforms for processing satellite imagery in the cloud. With many thousands of regular users, Earth Engine fills several roles in the Earth Science Community. These include: as a data storage platform for petabytes of freely available satellite data; as a platform for developing programs in JavaScript and Python for image interpretation; and as an information distribution tool through user-built interfaces. With a generous set of documentation, examples, and tutorials, it provides a good start and reference for learning the platform.

Despite these strengths, we have found that it remains quite challenging for our undergraduate and graduate students to move to higher-level use cases efficiently and effectively. With our contacts in the academic, private, and student communities, we have undertaken a large project to produce a detailed book of Earth Engine tutorials. Contracted by Springer and freely available to everyone, the book Cloud Based Remote Sensing with Google Earth Engine: Fundamentals and Applications is recently released to the public. Working entirely as volunteers, about 100 authors completed the 55-chapter, 1200-page, 250,000 word manuscript in about six months. In this presentation, we review some of the successes and challenges that might be of interest to others considering large volunteer efforts.

In our efforts to ensure the book's quality, we evolved several helpful techniques that could be of interest. We solicited volunteer reviewers; more than 300 detailed reviews were done, with comments both in an online survey and as comments in the live document. To ensure consistency with Google's best practices, code was reviewed at Google; after that, a team of two undergraduates executed every line to remove lingering bugs or missed steps. The form will be made live for future real-world users to comment and find problems. We are undertaking an effort to crowdsource translations of the book with machine learning translation and volunteer improvement. Finally, we are leading users through the book chapter by chapter on Zoom, recording the presentations by the authors themselves for a YouTube channel. Through these coordinated group efforts, we hope that the book will be useful to a wide range of users, and that the lessons learned will be of interest to others.