Director’s Discretionary Proposals with Hubble: An Overview of the Program

A. Fox (afox[at]


Astronomical events can happen unexpectedly. Supernovas explode. Comets fragment. Variable stars flare. Planetary atmospheres evolve. Gravitational wave sources emit electromagnetic counterparts. When time-variable phenomena such as these occur, Hubble can be used for rapid follow-up observations via the Director’s Discretionary (DD) program. Hubble’s great heritage of pushing the frontier of astronomy is well served by a strong and productive DD program. In this Newsletter article, we review this program and review process. We also present statistics of the submitted and accepted DD proposals over the last five Hubble Cycles, to identify the characteristics of successful proposals and the usage of the DD program by the community.

The DD Program

Around 10% of the available observing time on Hubble is reserved for allocation at the discretion of the Director. Historically, a significant part of this percentage has been allocated to very large, community-driven programs including the Hubble Deep Field, Hubble Ultra-Deep Field, the Multi-Cycle Treasury Programs, and the Frontier Fields. Other DD orbits may be used to provide Mission Support observations for other NASA observatories. The remaining part is available for general DD proposals that can be submitted throughout the year to follow-up on new discoveries and time-variable phenomena. Around 100 orbits of Hubble time are allocated per year to these general DD proposals, which are the subject of this article.

The Review Process 

DD proposals can be submitted at any time. They are prepared according to formatting guidelines given online, which describe the page limits and section requirements. Once a DD proposal is submitted, members of the Science Policies Group (a team of several astronomers within the Science Mission Office) perform an initial internal review of the proposal. This involves determining whether the proposal could wait for an upcoming mid-cycle proposal deadline or annual General Observer (GO) proposal deadline with no loss of scientific return; whether the proposal is a resubmission of a recent unsuccessful GO program; and whether the proposal duplicates any other active proposal, such as a target-of-opportunity (ToO) program. If a proposal passes this initial review, it is then sent out for external review, usually to three or four referees, selected to be experts in the field of study. Over the last five cycles, 57% of DD proposals were sent out for external review.

The external referees are asked to comment on the scientific merits of the proposal, specifically on whether the proposal meets the high standard expected for all Hubble observations, whether the observations need to be taken from space as opposed to using a ground-based telescope, and whether the orbit request is justified. Based on the reviews received from the external referees, the Science Policies Group forms a recommendation for the Director, and meets to discuss the merits of the proposal. The Director then makes the final decision.

In some cases, the external referees offer unanimous recommendations, and the final recommendation made to the Director is straightforward. In other cases, the referees may have differing opinions, and the various considerations have to be weighed together to form a final recommendation. Over the last five cycles, the overall DD proposal success rate has been 43%.

DD Observations

Hubble’s scheduling system, in which observations are scheduled in one-week blocks known as calendars, makes rapid-response observations more challenging than with ground-based observatories. Nonetheless, in the most compelling cases, DD observations can be scheduled within a few days of a proposal being submitted (this involves interception and redesign of the current week’s observing calendar). More commonly, they are scheduled within a timescale of weeks to months, depending on the properties of the source and the demands of the science case. DD observations typically have no proprietary period, and the data are thus made publicly available in the archive as soon as they are downlinked from the telescope. DD programs are typically less than five orbits in size, though larger requests can be considered if sufficiently justified.

Statistics of Approved DD Proposals

Using a database of all 171 DD proposals submitted in the period from July 2013 to February 2018, we analyzed the success rates of DD proposals based on a number of different factors. The results are shown in Table 1 and Figure 1. In particular, we looked at the science category, location of Principle Investigator (PI), gender of PI, and Hubble instrument. This allows us to build a global picture of the DD program—who is using it and what they are using it for. Hubble Heritage proposals (designed to take high-quality astronomical images for education and public outreach purposes) are not included in this analysis. The 171 proposals analyzed contain a total orbit request of 1247 orbits.

Table 1: Success Rates of DD Proposals by Category

Proposal Category

Number Submitted

Number Accepted

Success Rate (%)





By science category:




Solar System
























By location of PI:




















By gender of PI:












By instrument*:

























* Note: for multi-instrument proposals, only prime instrument is counted in the statistics.

Table 1 reveals some interesting trends. In terms of science area, the DD program is heavily used by the solar system community, particularly for observations of comets, asteroids, gas-giant planets, trans-Neptunian objects, and planetary moons. Various types of variable stars (such as novas, transients, binaries, X-ray sources, and magnetars), and supernovas (which we treat as a separate category) also account for a significant fraction of DD usage. DD proposals on supernovas have a notably high success rate, at 60%. In contrast, galaxies do not represent a large sector for DD usage, despite their significant fraction of all Hubble GO observations—only one DD proposal on galaxy evolution has been approved in the last five years. The reason for this is that galactic phenomena do not change on short timescales, and so observations can almost invariably wait for the next annual proposal cycle (or mid-cycle deadline). The other scientific categories of DD proposal are active galactic nuclei and intergalactic medium (AGN/IGM), and exoplanets, particularly those that are variable in time.

The gender statistics of DD proposals are also noteworthy. Only 15.8% of DD proposals (27 out of 171) submitted during the last five years were led by women (as opposed to 25.5% of GO proposals led by women through the annual TAC process over the last five cycles). However, the success rate for women-led proposals was slightly higher than for men, at 48.1% (13 out of 27) vs. 41.7% (60 out of 144). The small sample size and consequent large statistical uncertainties should be kept in mind, as should the fact that these successful women-led proposals include several by the same PI. Nonetheless, it is noteworthy that DD proposals do not appear to show the same difference between success rates for men and women PIs that has been identified among Hubble GO proposals over many consecutive cycles (Reid 2016, PASP, 126, 923).

US-based PIs account for the majority of all Hubble DD proposals, submitting 66.1% of the total, whereas PIs from ESA countries account for a further 28.7%, with comparable success rates in both cases. By instrument, WFC3 accounts for the largest fraction of DD proposals, both submitted and accepted, as is the case in the annual Hubble TAC. That being said, all four of Hubble’s principal active instruments (ACS, COS, STIS, WFC3) are regularly used as part of the DD program, and one proposal with the Fine Guidance Sensor (FGS) was approved and successfully executed.

Maximizing Your Chances of Being Awarded DD Time on Hubble 

If you are considering submitting a DD proposal for Hubble observations, ask yourself if your science case can wait until the next mid-cycle or annual proposal deadline, and if it can be done from the ground. If the answer to both these questions is no, you may have a good case for a DD proposal. Stick to the page limits, include a clear abstract, ensure the main scientific goals and the time-criticality of the observations are well articulated, ensure the orbit request is well justified, and, if the proposal is for follow-up of a recent discovery, describe the discovery observations. Finally, check the real-time list of approved DD programs, to make sure nobody has already secured Hubble observations of the same target. 

The full policy regarding DD proposals is given in the Hubble Call for Proposals.

Fox Fig 1
Figure 1:   Distributions of DD Proposal by Category, for submitted proposals.
Fox Fig 2
Figure 2:   Distributions of DD Proposal by Category, for accepted proposals.