The Hubble Metrics and Bibliographic Database

Jill Lagerstrom (library@stsci.edu), Daniel Apai, Karen Levay, and Elizabeth Fraser

For nearly two decades, the Hubble Space Telescope has provided astronomers with scientific data from new observations. New knowledge based on the analysis of these observations is communicated in many ways among astronomers and to the public. The ultimate, official record of research and discovery is the publication of results in peer-reviewed journals.

To assess and help maximize the scientific impact of Hubble, a new metrics team at the Institute—Daniel Apai, Jill Lagerstrom, Karen Levay, and Elizabeth Fraser—identifies and tracks Hubble-based papers in peer-reviewed journals. To qualify, a paper must use Hubble data as part of its scientific analysis. The team is assembling a database comprising approved proposals, refereed publications based on the data, and all papers that cite them. We can use this database to trace and quantify the evolution of astronomical projects from the proposal phase through execution and publication, to their impact on the astronomical literature. This knowledge, in turn, provides useful information for refining science policies, such as strategies for time allocation.

Figure 1.  The number of refereed articles per year. “Not archival” denotes guest-observer papers. “AR” denotes papers based on archival data; “Partly Archival” are papers that combined GO and archival data.

Figure 1. The number of refereed articles per year. “Not archival” denotes guest-observer papers. “AR” denotes papers based on archival data; “Partly Archival” are papers that combined GO and archival data.

Figure 2.  Number of refereed publications per Hubble instrument, with important events highlighted.

Figure 2. Number of refereed publications per Hubble instrument, with important events highlighted.

Observatories have traditionally characterized their contributions by counting not only the number of papers they have published, but also the number of papers that have cited them.  We refer to the former as “productivity” and to the latter as “impact.” As of September 2009 we have identified more than 8,300 papers, and more than 287,000 citations.

We completed the database for 1998–2008 in August 2009, and our study of it reveals several interesting points. One is the strong demonstration of the increasing role of the Hubble archive (see White, R. L., et al. 2009, in Astro2010: The Astronomy and Astrophysics Decadal Survey, Position Papers, no. 64), served through the Multi-mission Archive at STScI (MAST). Figure 1 shows that approximately half of recent papers that analyze Hubble data use archival data, rather than only observations made by the authors of the paper.

Throughout its history, Hubble’s instruments have been upgraded, replaced, and repaired by five servicing missions. Figure 2 shows the historical productivity of Hubble’s various instruments, including the Fine Guidance Sensor, which has produced exciting results despite not being an official scientific instrument. For almost ten years, Wide-Field Planetary Camera 2 was Hubble’s most productive instrument, up to the installation of the Advanced Camera for Surveys, which provided superior imaging for many projects.

The bibliographic database offers an opportunity to evaluate the impact of proposals across different topics and as a function of program size. The time-allocation committee (TAC) has historically selected a wide range of programs of various sizes, ranging from those that require just a few Hubble orbits to others that require hundreds. Smaller programs generally observe a small number of specific targets, while larger programs are generally designed to create surveys. The analysis of our database shows that the smallest programs receive the highest number of citations per orbit (8 cits/orbit). However, the largest programs, while less productive per unit of telescope time (2 cits/orbit), are the most efficient in producing highly cited data sets.