[Paper Review] The Hobby-Eberly Telescope Dark Energy Experiment (HETDEX): Description and Early Pilot Survey Results

Gary J. Hill, Karl Gebhardt+12 more|ArXiv.org|Jun 1, 2008

Astronomy and Astrophysical ResearchPhysics and Astronomy97 citations

TL;DR

HETDEX proposes a wide-area survey using the Hobby-Eberly Telescope equipped with the VIRUS instrument to detect 0.8 million Lyman-alpha emitting galaxies at z = 1.9–3.5, enabling a 3-σ detection of dark energy and 0.1%-level curvature measurement via baryonic acoustic oscillations and power spectrum analysis. The pilot survey with VIRUS-P confirms the feasibility of integral field spectroscopy for identifying LAEs and measures key properties of the emission-line population.

ABSTRACT

The Hobby-Eberly Telescope Dark Energy Experiment (HETDEX) will outfit the 10 m HET with a new wide field and an array of 150 integral-field spectrographs to survey a 420 sq. deg. area in the north Galactic cap. Each fiber-coupled unit spectrograph will cover 350-550 nm, simultaneously. This instrument, called VIRUS, will produce ~34,000 spectra per exposure, and will open up the emission-line universe to large surveys for the first time. The survey will detect 0.8 million Lyman-alpha emitting (LAE) galaxies with 1.9

Motivation & Objective

• To directly detect dark energy at high redshift (z ≈ 3) using baryonic acoustic oscillations and power spectrum shape in a 3D galaxy map.
• To achieve 0.9% accuracy on the Hubble parameter H(z) and 0.1% accuracy on angular diameter distance D_A(z) at z ≈ 2.8.
• To measure the curvature of the Universe to 0.1–0.2%, improving current constraints by a factor of ten.
• To validate the use of integral field spectroscopy with VIRUS-P for identifying Lyman-alpha emitting galaxies in a pilot survey.
• To characterize the luminosity function, bias, and contamination of LAEs for the full HETDEX survey.

Proposed method

• Deploying the VIRUS-P prototype spectrograph on the McDonald 2.7 m telescope to conduct a pilot survey of deep fields (COSMOS, GOODS-N, MUNICS-deep).
• Using a fiber-coupled integral field spectrograph with 224 fibers per unit, covering 350–550 nm at 5.7 Å resolution and 1.8 arcsec² per fiber.
• Applying dithering patterns of three exposures per field to achieve a 1/3 fill factor and fill in gaps in the IFU coverage.
• Combining data from multiple dithered exposures to produce high-sensitivity spectra and detect emission-line sources.
• Using deep imaging (AB ~25) to separate Lyman-alpha emitters from [OII] emitters based on equivalent width.
• Applying a non-evolving luminosity function model to compare observed source counts with expected redshift distributions.

Experimental results

Research questions

• RQ1 Can wide-area integral field spectroscopy with VIRUS-P successfully detect and characterize Lyman-alpha emitting galaxies at z ~ 2–3.5?
• RQ2 What is the contamination fraction of low-redshift galaxies in emission-line surveys, and how can it be minimized?
• RQ3 How do the observed number counts of LAEs compare with predictions from a non-evolving luminosity function at z ≈ 3?
• RQ4 To what extent does the pilot survey data support the evolution of the LAE luminosity function from z ≈ 4 to z ≈ 2?
• RQ5 Can the VIRUS-P instrument achieve sufficient sensitivity and throughput to detect LAEs in 2-hour exposures on a 2.7 m telescope?

Key findings

• The pilot survey detected 99 emission-line objects in 40 arcmin² of COSMOS field, including 45 secure Lyman-alpha emitters at z ≈ 3.4.
• The survey achieved a 5-σ line flux sensitivity of 5–6 × 10⁻¹⁷ erg/cm²/s in 2 hours of integration time.
• Analysis shows a hint of evolution in the LAE luminosity function, consistent with a factor-of-two increase from z ≈ 4 to z ≈ 2, though full dataset analysis is needed.
• The pilot survey confirmed the viability of wide-area IFU spectroscopy for identifying LAEs and validated the data reduction pipeline for HETDEX.
• The VIRUS-P instrument successfully detected a z = 3.415 LAE in four dithered fiber positions, demonstrating robust source detection across the IFU.
• The survey indicates that moderate-depth imaging (AB ~25) is effective in separating LAEs from [OII] emitters using equivalent width criteria.