Hubble Legacy Fields (HLF) Data Release 2.0 for the GOODS-S region (HLF-GOODS-S)

• Overview
• Information on Observations
• Data Products, Zero Points, and Download Options

Highlights of the Hubble Legacy Fields (HLF) V2.0 Release

• Includes essentially all ultraviolet, optical and infrared data taken by Hubble over 14 years across the Extended CDF-South region including the GOODS-South, HUDF, CANDELS, ERS, HUDF09/12, UVUDF, HDUV and many other programs.
• The largest High Level Science Product (HLSP) data release to date, incorporating 7491 individual HST exposures, totaling 6.4 Msec.
• All images use a coordinate grid tied to the Gaia DR2 reference frame; we also provide results on the original GOODS-S coordinate system.
• The WFC3/UV dataset includes data from the 2009 ERS program which has never been released as a HLSP.
• The ACS/WFC and WFC3/IR data have also been organized and made available as separate data products for twelve one year epochs (from 2002-2007 and from 2009- 2016) as separate data products.
• A photometric catalog is provided for eachof the 13 HST bandpasses used in the HLF dataset.
• The HLF dataset includes 5 deep regions, the original HUDF (now as the XDF dataset) and four newer regions (HLF-HUDF parallels) that are strikingly deep in subsets of the 13 filters.

Overview

This describes the V2.0 version data release of the Hubble Legacy Fields (HLF) project for the GOODS-South (ECDF-S) region from the HST archival program AR-13252. The V2.0 version of the HLF dataset provides WFC3/UV, ACS/WFC and WFC3/IR source catalogs and images covering a 25 x 25 arcminute area plus (smaller) cutouts of four deep areas in the GOODS-South field. All the data covering the original HUDF field within the GOODS-South area were previously released as a high level science product as the eXtreme Deep Field (XDF — sometimes referred to as the XDF/HUDF field).

HLF Release: The release combines exposures from Hubble's two main workhorse cameras, the Advanced Camera for Surveys Wide Field Channel (ACS/WFC) and the Wide Field Camera 3 (WFC3), including data from both the infraRed channel (WFC3/IR) and the UV/visible channel (WFC3/UVIS). These data were taken over more than a decade between mid-2002 to the end of 2016. The HLF includes essentially all the near-uv (WFC3/UV F225W, F275W, F336W), optical (ACS/WFC F435W, F606W, F775W, F814W and F850LP filters) and infrared (WFC3/IR F098M, F105W, F125W, F140W and F160W filters) filters. The HLF includes data taken by Hubble over the original CDF-South region including the GOODS-South, CANDELS, ERS, the Extended CDF-S (ECDF-S), numerous SNe followup programs, and many other programs. Given that this dataset combines all images in the archive on the ECDF-S to date from numerous different programs, the HST AR-13252 proposal identified the product under a single global name "Hubble Legacy Field".

Astrometric Frame: A particularly challenging aspect, given the large area, was to establish a robust astrometric framework. A global astrometric solution was bootstrapped from the smaller datasets. All the HLF image mosaics have been produced using a coordinate grid tied to an absolute Gaia DR2 reference frame. We also provide an alternate coordinate system using the same tangent point as the original GOODS-S dataset.

Exposure Time: The total HST on-target exposure time of all the data incorporated in the HLF-GOODS-S is 6.4 MSec. The final HLF GOODS-S dataset derives from 7491 exposures taken by 20 different HST programs. This on-target exposure time is equivalent to that of about 2665 orbits (assuming that the exposure in a typical orbit is 2400 s). This total exposure corresponds to nearly ~75% of a full HSTresource cycle! The HLF dataset also includes four very deep pointings from observations made in parallel when HUDF was primary (the HLF-HUDFP1-4 parallel fields). Unlike the XDF/HUDF, these parallel fields have a smaller set of filters and have varying depths, but still constitute a resource of unusually deep fields.

Catalog: A catalog of all the sources in the dataset, with photometry for sources in each of the 13 filters, is also made available with this release. The photometric catalog includes 103,098 objects, selected in either of two near-IR filters (F850LP and F160W). The photometric results are measured in 0.7′′ diameter apertures and corrected to total fluxes based on the F160W curve of growth.

HLF Team: The HLF Team is The HLF Team is Garth Illingworth, Daniel Magee, Rychard Bouwens, Pascal Oesch, Kate Whitaker, Ivo Labbe, Pieter van Dokkum, Mauro Stefanon, Bradford Holden, Marijn Franx, and Valentino Gonzalez. The catalog was led by Katherine Whitaker. Please credit Illingworth et al (2019) when using the HLF HSLP data and products, and and Whitaker et al (2019) for the catalog. The UVUDF dataset was graciously provided by Harry Teplitz, Anton Koekemoer, Marc Rafelski, Norman Grogin and the UVUDF Team. If you use any of the UV data, please cite Teplitz et al. (2013) and Rafelski et al. (2015).

What's new in V2.0

A major change from the HLF-GOODS-S V1.5 data release to the V2.0 release is the inclusion of three UV datasets. These sets constitute a substantial body of UV data, totaling 280 orbits of HST WFC3/UVIS imaging, or about 0.5 Msec of data. The three UV datasets are the Early Release Science (ERS) observations (Windhorst et al 2011), the UltraViolet Ultra-Deep Field (UVUDF) dataset (Teplitz et al. (2013), Rafelski et al. (2015)) and the Hubble Deep UltraViolet (HDUV) legacy dataset Oesch et al 2018, see also Vanzella et al 2016). These datasets were updated and astrometrically-matched to the V1.5/V2.0 release of the HLF-GOODS-S. The ERS dataset required a full processing since that has not be available in processed form on MAST. Thus the HLF provides for the first time the release of the WFC3/UV 2009 ERS dataset as a HLSP.

The steps that were taken to assemble the HST UV data along with the HST optical and near-IR data are described in the paper that discusses the assembly of the HLF- GOODS-S dataset (Illingworth et al 2019).

As noted above the V2.0 release now includes a catalog of over 100,000 sources with photometry in all 13 filters (or those sources for which the S/N was adequate for inclusion). See Whitaker et al 2019.

Additionally, we have organized the ACS/WFC and WFC3/IR data into twelve one year epochs from the fourteen years of observations on the ECDF-S, starting with the first epoch from May 2002 to April 2003 (there were two years where no significant additional data was obtained due to ACS/WFC failure in January 2007). See HLF_v2.0_goodss_epoch_by_filter.pdf.

The HLF-GOODS-S Field

The HLF-HUDF Parallel Deep Fields

back to top

Observations

Description

Observations included in this version of the HLF-GOODS-S Field were taken from July 2002 to October 2016 from 20 different HST programs.

Programs Used for HLF-GOODS-S Field

Data Processing

The process begins with individual flt files requested from the MAST HST archive. These images have already been subject to bias correction, dark subtraction, flat-field correction, and cosmic ray rejection. The flt files obtained from the MAST are reasonably well calibrated, given that the on-the-fly pipeline processing by STScI already takes advantage of darks and flat fields constructed from latest on-orbit calibration data. But further effort was needed to ensure that the images used in the construction of the HLF were of a uniform high quality.

All images used to produce the HLF were first visually inspected to identify any data quality issues (loss of guiding, excessive background, pointing accuracy) and any image which could not be corrected was rejected for processing. During this visual inspection we also identify images affected by satellite trails and optical ghosts from filter reflections in the ACS/WFC generated by bright stars and updated the data quality array to ensure these artifacts were masked during final processing.

Due to the slow but clear degradation in the performance of the ACS wide-field camera CCDs (ACS/WFC), we needed to correct later data for charge transfer efficiency (CTE) degradation. In particular, for ACS/WFC data taken after HST Servicing Mission 4 (SM4) we used images that have been corrected for CTE losses and bias striping

ACS/WFC Data

The HLF ACS/WFC dataset were reduced by the HLF data processing pipeline HLFRED (built on DrizzlePac) which can be found on Github. The steps undertaken by HLFRED are described in Illingworth, Magee et al 2019.

WFC3/UV and WFC3/IR Data

The UVUDF Data

We would like to thank Harry Teplitz, Anton Koekemoer, Marc Rafelski, Norman Grogin and the UVUDF team for graciously providing the flt images that were used in generating the UV dataset on the HUDF. These data required additional processing that is unavailable in the standard WFC3 calibration pipeline provided by STScI (See the README file for the V2.0 UVUDF data). If you use these data, please cite Teplitz et al. (2013) and Rafelski et al. (2015).

The HLF WFC3/UV and WFC3/IR dataset were also reduced by the HLF data processing pipeline HLFRED The HLFRED processing of the WFC3/IR observations in the HLF follows the procedure as described in (Magee, Bouwens & Illingworth 2011) but with a few additional corrections in order to improve the reductions.

Additional Data Processing Steps

While most WFC3/IR flt files requested from MAST can be used readily, in some cases CALWF3 can falsely flag nearly all pixels as cosmic rays, due to a varying background during the exposure. For WFC3/IR MultiAccum mode observations, CALWF3 assumes that accumulating background counts over the entire observation is a linear function. This assumption may not be the true for all observations. Depending on circumstances of the observation the background count rate may vary over the duration of the observation. In order to determine if the background count rate is sufficiently non-linear, we compared the average exposure time in a flt file's TIME array extension with the exposure time listed in the header EXPTIME keyword. If the we find the exposure time varies by more than 2% we correct the background signal in the raw image using code developed by the Frontier Fields team. This additional step assures the background count rate is linear before reprocessing the image with CALWF3. For more info on time-varying background see Instrument Science Report WFC3 2014-03.

We also mask out pixels on WFC3/IR images that were affected by source persistence. This masking was performed by utilizing persistence models generated by the STScI WFC3 Persistence Project. A persistence model is created for each WFC3/IR exposure which incorporates an internal persistence model (persistence within a visit) and an external persistence model (persistence from earlier visits). Using this model of the total persistence we flagged all pixels above a threshold of 0.2 electrons/s thus ensuring that our final reductions are not significantly affected by source persistence.

back to top