sitsbook

This book provides a comprehensive appraisal of satellite image time series analysis on Earth observation data cubes, an emerging discipline for measuring change on the planet. It introduces sits, an open-source R package designed for big Earth observation data analysis using satellite image time series. The resource highlights the power of combining Earth observation data cubes with time series for monitoring changes and gaining insights that single snapshots cannot provide.

The sits API enables users to build regular data cubes from major cloud services, including Amazon Web Services, Microsoft Planetary Computer, and the Copernicus Data Space Ecosystem. It incorporates advanced functionalities such as training sample quality measures, various machine learning and deep learning classification algorithms, and Bayesian post-processing methods for smoothing and uncertainty assessment. The package also supports best-practice accuracy assessments to evaluate results effectively.

Designed for remote sensing experts in Earth Sciences, the sits package allows users to apply advanced data analysis techniques with only basic programming knowledge. A Python API is also available, interfacing with the R API, which allows Python users to directly run sits and convert its data structures for use with Python data.frames and xarrays.