Hack the land out of them: Obtaining land cover classifications from GLOBE Observer photographs (for Appendix F).

A picture can be worth a thousand words and this describes the data necessary to identify and track global land cover change. Citizen science apps and social media are generating enormous amounts of data in the form of photographs uploaded to the cloud. Photographs submitted via the Globe Observer app provide the link between space and Earth, fostering an understanding of Earth’s physical processes and natural phenomenon. The volume of photographs is outpacing human ability to organize, characterize, evaluate and classify data for search, comparison, change detection, and prediction. This is where the hack comes in. We propose a year of activities to explore and create machine learning (ML) tools to segment, characterize, classify, and label components of land cover photographs from the GLOBE Observer (GO) land cover app. We propose to recruit a broad diversity of participants of New Mexican, American and international students, satellite images are already capable of being processed to classification standards by ML tools, providing the opportunity to prove the concept that similar approaches can be developed for photographs. In contrast to imagery analysts processing data on a computer, our approach includes outdoor observation and sampling of landscapes in situ, making the connection between the land, the human observers and machine learning palpable.

Designing a series of simultaneous virtual learning workshops will collate participants into teams, and facilitate engagement by introducing the shared languages of photos, land interpretation and code (-), to outline approaches and targets of the hackathon (1). Recruitment will focus on participants from across the Southwest. Targeted recruiting of female and Hispanic participants from New Mexico State University and the tribal colleges in New Mexico, in addition to programs associated with AmericaView will result in teams from locations across America. The online platform (Website) will allow participation from any location without time restraints. Packaging the invitation to those who embrace a sense of Earth stewardship will foster the creation and establishment of local service projects to perpetuate the networking and collaboration established in the hack environment. Data preparation chooses image replicates from the Globe Observer data set (e.g., download images, identify locations in close proximity over a range of ecosystems). Preparation of workshops includes designing outreach and application materials, and completing the IRB training. Designing workshops starts with preparing data and other resources for collaboration, practicing GO activities, recording video and creating homework assignments. Organizing data and sample locations will involve collaboration through Jupyter notebooks. Design of social components for daily schedules includes meet-and-greet, physical exercise, breaks, games, and meals.

This is a mature technology (TRL 7+) — the realistic path in is usually NASA's Technology Transfer Program: licensing an existing NASA patent, or a Space Act Agreement to use NASA facilities/expertise directly. NASA also runs a startup licensing program with no upfront fee for companies formed to commercialize a specific NASA technology.

None of these are guaranteed paths for this specific project — TechPort itself doesn't have an "apply" button. Reaching out to the contact(s) above with a specific question is usually the fastest way to find out what's actually open.