Speaker
Description
The ATLAS telescopes have largely operated using a traditional discovery mode in which data from each telescope are processed separately and NEOS are discovered from a series of four "quad" observations from one telescope. Since they operate independently, sky coverage (and therefore NEO discovery) are subject to losses from poor weather; for example, a quad of observations can be spoiled by a single bad observation with clouds, or an entire night of ineffective observations can be executed because of thin clouds over an entire part of the sky for the entire night. The data processing for the ATLAS telescopes has also operated independently, with separate processing streams for each telescope -- the system cannot easily combine the data from multiple telescopes to discover objects. For example, a faint object might appear only three times in a quad, not enough to confirm as a detected object, but another telescope might also have three faint observations, so in theory the combined detections can be confirmed by testing the large number of potential linkages.
With the expansion of ATLAS to a five-observatory system with geographic and weather diversity, we have embarked on a two-pronged strategy to enhance the discovery capability of ATLAS network of telescopes. In this talk, we present ADIOS (the ATLAS Dynamic Inter-Observatory Scheduler), a program that can simultaneously schedule all operating ATLAS telescopes for maximum NEO discovery in the face of weather and other observational losses. ADIOS employs an NEO population model and awareness of local sky conditions to decide where to observe at any given instant to maximize the NEO discovery rate. ADIOS can avoid patches of sky where discovery is unlikely, and ADIOS can divert another ATLAS observatory to complete a quad spoiled by weather or twilight.
ADIOS is complemented by another software component called PUMA (Position Using Motion and Acceleration), a performant orbit fitter that can evaluate thousands of short-arc (~1-2 day) linkages per second on a single CPU. We can use PUMA to test linkages of faint three-detection tracklets across all ATLAS observatories, effectively improving the sensitivity of the ATLAS survey. We present results of PUMA cross-observatory linkages with the ATLAS telescopes and a pilot effort to cross-link ATLAS with the Catalina Sky Survey 703 telescope. This effort dovetails with similar projects in the field such as the Juric et al. 2024 HelioLinC linking service B612's ADAM precovery service that incorporate data across observatories.