This course is a skills-based follow-on to GEO391-Innovation in Earth Observation, a seminar that reviewed key limitations facing Earth Observation (EO), and the recent developments that are challenging these limitations. In this course, students will work, within the broader context of several active research projects, on developing and applying several specific EO methods that were reviewed in GEOG391. These methods are:

• Scaling-up crop growth and productivity estimates derived from automated in situ sensors and UAS imagery up to smallsats;
• Processing imagery using cloud-based computational platforms, such as Google Earth Engine and Radiant Earth.
• An active learning approach (combining crowdsourcing and machine learning) to mapping agricultural land cover;

We will learn a range of new skills, including:

• Programmatic access to sensor and image-serving APIs, as well as cloud-based earth observation platforms, using Javascript, python, and R;
• Use of AWS computing instances;
• Postgres/PostGIS databases;
• UAS flight planning and image processing with PIX4D software.

After an initial introduction to the various toolsets we will be using, students will form project teams (~3-5 people each) to tackle further development and application of one of the three project areas. These projects (described here in more detail) will be assessed by means of a formal in-class presentation and team-authored final project.

Class is held in JC105 on Mondays and Wednesdays from 12-13.15.

Office hours: Tuesdays 1-3 pm in Jefferson 105.

• General Information
• Syllabus
• Overview of Projects
• Content of Introductory Modules