Astronomical survey data

Astronomical survey data is the observational data produced by a telescope or mission that systematically images or measures large areas of the sky repeatedly, rather than targeting one known object at a time.

Raw images or spectra flow through an automated processing pipeline that calibrates, catalogs, and cross-matches detected objects, producing structured catalogs of positions and brightness, and from repeated visits to the same patch of sky, light curves built from photometry for variable or moving objects. The scale involved, often billions of individual detections, requires high-performance computing and, increasingly, machine learning to classify objects and flag anomalies automatically rather than by manual inspection.

Modern surveys are the primary source of new astronomical discoveries, exoplanets, asteroids, and transient events like supernovae, because they observe so much of the sky so often that rare events are eventually caught somewhere in the data. This differs from a targeted observation of a single known object, which trades that breadth for much greater depth on one target. The central pitfall is selection effects: brighter and closer objects are far easier to detect, and any statistical study built on a survey catalog needs to account for what the survey could and could not see, rather than treating the catalog as a complete census of the sky.