Hierarchical Progressive Survey (HiPS) format: moving from visualisation to scientific analysis

TL;DR

This paper assesses the scientific utility of Hierarchical Progressive Survey (HiPS) skymaps for broadband photometry. It introduces AstroBrowser, a modular toolkit with a web interface, backend API, and libraries to retrieve HiPS data and perform automated aperture photometry, validated against the Herschel Reference Sample across UV to FIR. The study finds that nine of ten HiPS maps reproduce the original data quality and that the automated photometry matches public catalog fluxes within a few percent, while a PACS 100 μm map exhibits a systematic offset due to mixed-resolution data products. It concludes that HiPS enables robust, science-grade analyses with some metadata improvements and calls for reference datasets and formal validation to maximize data reuse.

Abstract

Context. In the current era of multi-wavelength and multi-messenger astronomy, international organisations are actively working on the definition of new standards for the publication of astronomical data, and substantial effort is devoted to make them available through public archives. Aims. We present a set of tools that allow user-friendly access and basic scientific analysis of observations in Hierarchical Progressive Survey (HiPS) format, and we use them to gauge the quality of representative skymaps at ultraviolet, optical, and infrared wavelengths. Methods. We apply a fully-automatic procedure to derive aperture photometry in 10 different bands for the 323 nearby galaxies in the Herschel Reference Sample (HRS), and compare its results with the rigorous analyses involving specialised knowledge and human intervention carried out by the HRS team. Results. Our experiment shows that 9 of the 10 skymaps considered preserve the original quality of the data, and the photometric fluxes returned by our pipeline are consistent with the HRS measurements within a few per cent. In the case of Herschel PACS maps at 100 μm, we uncovered a systematic error that we ascribe to an inconsistent combination of data products with different spatial resolution. For the remaining skymaps, the estimated statistical uncertainties provide a realistic indication of the differences with respect to the HRS catalogue. Conclusions. In principle, the currently available HiPS skymaps in Flexible Image Transport System (FITS) format allow to carry out broadband photometric analyses with an accuracy of the order of a few percent, but some level human intervention is still required. In addition to assessing data quality, we also propose a series of recommendations to realise the full potential of the HiPS format for the scientific analysis of large astronomical data sets.

Figures (6)

• Figure 1: AstroBrowser components diagram.
• Figure 2: Herschel SPCS 500 catalogue overlay on M31 superimposed on SPIRE 500 HiPS. The shape size is mapped to the flux metadata column, and the shape hue is mapped to the snrnoise metadata column.
• Figure 3: Interactive DataExplorer panel view on GALEX NUV HiPS targeting M31.
• Figure 4: Comparison between the flux measured from the HiPS skymaps and the values reported in the HRS catalogues for GALEX Cortese+12, SDSS Cortese+12, and Herschel Ciesla+12Cortese+14. Error bars in the vertical axis represent the uncertainties returned by our pipeline (they do not include HRS errors); measurements above the median $S/N$ are highlighted in black. Horizontal dotted lines and grey bands denote the median and $16-84$ percentiles of the flux ratio (see Figure \ref{['fig:flux_comparison']} for the full probability distribution).
• Figure 5: Probability distribution of the ratio between the flux measured from the HiPS skymaps and the values reported in the HRS catalogues. Cumulative fractions and differential probability densities are shown in the top and bottom panels, respectively.