Cloud-native geospatial datacube workflows with open-source tools

Welcome to Cloud-native geospatial datacube workflows with open-source tools! This book demonstrates steps of typical scientific workflows involving earth observation data with a focus on cloud-optimized data formats, larger-than memory data, and manipulating multi-dimensional datasets to prepare them for analysis.

We focus on data derived from different types of satellite imagery that are publicly available in cloud-hosted repositories such as AWS S3. The tutorials are composed of Jupyter Notebooks that demonstrate how to work with these datasets in Python using software packages from the popular Pangeo ecosystem that are built on and around the Xarray data model.

The goal of this book is to reduce barriers to entry to working with earth observation data for scientific analysis. It features two stand-alone tutorials, each detailing steps involved in a typical workflow, from data access and organization to exploratory data analysis and visualization.

Underpinning these examples is a focus on understanding the different components of n-dimensional, gridded datasets, how they relate to the tools we use to work with them (in this case, the Python package Xarray), and how a strong understanding of a scientific dataset within the context of your chosen data model can enable more efficient and intuitive analysis.