2025-07-24 –, Main Room 4
As we move deeper into the “big data astronomy” era, the need for fast, stable, homogenous data reduction pipelines is more pressing. I will present the recent development of a pure Julia pipeline for the APOGEE instrument that takes 3D non-destructive readout images to 1D wavelength calibrated stellar spectra components. I will emphasize implementations of new/old methods of general interest to the JuliaAstro community and the desiderata to facilitate both daily and large HPC reductions.
As we move deeper into the “big data astronomy” era, the need for fast, stable, homogenous data reduction pipelines is more pressing and the performance, extensibility, and readability of Julia makes it a good candidate for astronomical pipelines. This talk will focus on recent efforts within the Sloan Digital Sky Survey that have led to a new pure Julia pipeline. First, I will outline implementations of both old and new methods for conventional analyses of near-infrared detectors: including non-destructive readout (SUTR), 1/f noise correction, pixel response calibrations (flats/darks), trace extraction, and wavelength calibration. Then, I will emphasize how the speed and simplicity of LinearAlgebra in Julia facilitates Bayesian component separation on large data volumes of spectra, which allows us to avoid conventional sky “subtraction,” instead jointly modeling all of the components contributing to the spectra.
Andrew Saydjari is a NASA Hubble Fellow in the Department of Astrophysical Sciences at Princeton. Saydjari’s research focuses on combining astrophysics, statistics, and high-performance coding to study the chemical, spatial, and kinematic variations in the dust that permeates the Milky Way. This involves developing Bayesian methods and data reduction pipelines for spectroscopic and imaging surveys containing millions and billions of stars, respectively, usually implemented in Julia.