With Downsizing of WFIRST, CAA Asks if Coronagraph is Really Needed

The National Academies’ Committee on Astronomy and Astrophysics (CAA) was briefed yesterday by the co-chairs of a committee that reviewed the Wide-Field Infrared Survey Telescope (WFIRST) mission.  Their report already prompted NASA to require the WFIRST project team to downsize the mission. Some CAA members questioned whether removing the coronagraph — an instrument added by NASA after WFIRST was chosen as a priority by the scientific community — could be one of the options.

The WFIRST Independent, External Technology-Management-Cost Review (WIETR) committee was tasked by NASA Headquarters to assess the project’s progress. One of the key questions was “are the scope and cost/schedule understood and aligned?”

Their answer is no.

The WIETR committee was co-chaired by Peter Michelson of Stanford and Orlando Figueroa, a retired NASA official who once headed NASA’s robotic Mars program at NASA Headquarters and was deputy director of science and technology at Goddard Space Flight Center.

Goddard manages the WFIRST project.

WFIRST was identified as the top priority for a large space telescope by the most recent astronomy and astrophysics Decadal Survey produced by the National Academies.  It is being designed to study dark energy and dark matter and search for planets around other stars (exoplanets). The Decadal Survey committee estimated its cost at $1.6 billion in FY2010 dollars.

WFIRST will be the follow-on to the James Webb Space Telescope (JWST) that is currently scheduled for launch in 2019.  JWST has suffered many cost overruns and schedule delays, including one announced just last month, slipping the launch from October 2018 to the March-June 2019 time period.  CAA was briefed on that delay earlier this week.

NASA is striving to avoid the same fate for WFIRST and is carefully tracking its progress even though it is still in it earliest stages.  NASA calls it Phase A, formulation.   NASA has made significant changes to WFIRST’s design compared to what the Academies recommended in the Decadal Survey, however, and the pricetag is growing.

Among the changes is the addition of a coronagraph.  Coronagraphs are disks within a telescope that essentially create an eclipse, blocking the light from the center of a star so that the area around it can be studied in more detail to determine if, for example, there are planets in the vicinity.

NASA asked the Academies to review the program a few years ago resulting in a 2014 report chaired by Caltech’s Fiona Harrison.  It was reviewed again by the Academies last year as part of an overall assessment of how NASA is implementing the Decadal Survey. That “mid-term review” was chaired by MIT’s Jackie Hewitt.  While agreeing on the scientific value of a coronagraph, both raised concerns about the additional cost and technical risk that will be incurred by adding one to this mission, especially since the technology is not mature.

The Hewitt report recommended that NASA initiate an independent technical, management and cost assessment of the project, which led to NASA’s decision to establish WIETER.  As with the Harrison and Hewitt reports, WIETER’s review uncovered a number of issues.

The WIETR committee held its first meeting in August and formally submitted its report in October.  NASA Associate Administrator for Science Thomas Zurbuchen accepted it on October 19. In a memo that day, Zurbuchen directed Goddard to downsize or “downscope” the mission to keep the project life cycle cost at $3.2 billion.  That was the cost estimated by project managers when WFIRST reached a milestone NASA calls Key Decision Point-A (KDP-A), which triggers a project’s entry into Phase A (formulation).

NASA flight projects progress through a series of reviews.  A project must pass one before proceeding to the next.  WFIRST is now at the point where a decision must be made whether to proceed from Phase A  into Phase B, preliminary design and technology development.  That requires passing the KDP-B milestone.  If Phase B is successful, then it would proceed to Phase C (KDP-C), implementation.

It is only at KDP-C that NASA makes a commitment to the Administration and Congress about a project’s cost and schedule.  It forms the benchmark against which cost overruns and schedule delays are judged.

WFIRST is still in its early stages, so changing design, requirements, and cost estimates are not especially surprising.  However, it has experienced some significant design changes and resulting cost increases that could forebode even more difficult challenges ahead.

Hence, the decision to initiate the WIETR study.  The Hewitt committee’s recommendation was that it be completed prior to KDP-B.

Zurbuchen has decided not to make the WIETR report public at this time, but the Powerpoint slides used by Michelson and Figueroa to brief CAA are posted on the CAA’s website.

A major question is how the project’s cost is growing and whether it fits within NASA’s expected resources.  A dizzying array of cost numbers arises from those slides and the discussion at CAA, which included not only the CAA members and WIETR’s two co-chairs, but NASA Astrophysics Division Director Paul Hertz, and Zurbuchen (participating remotely).

Some of the estimates are adjusted for inflation, others are not.  Some were calculated using a 50 percent confidence level, while others use 70 percent.  (NASA requires that cost estimates use 70 percent at KDP-C, but not at this phase.)  Another variable is whether WFIRST is managed as a Class A or Class B mission as defined by NASA’s risk classification policies. Currently it is a Class B mission, but WIETR concluded that is “not consistent” with NASA policy and the mission is more akin to a Class A mission, which is more robust and consequently incurs more cost.

In his memo, Zurbuchen directed Goddard to reduce the project’s cost to $3.2 billion (the KDP-A estimate) from $3.6 billion (the current projected cost as calculated by project managers using equivalent parameters).

He also directed that the risk classification “be consistent with the findings in the WIETR report.” Apparently that does not necessarily mean making it a Class A mission.  At the CAA meeting, Hertz explained that the boundaries between the classes are flexible and NASA will look at what is needed and the cost implications.

Hertz pointed out that the downscoping of the mission inevitably means that less science will be achieved.  Zurbuchen’s memo stipulates that reductions are to be taken from the wide-field instrument, science investigations, and the coronagraph.

As the Harrison and Hewitt reports made clear, the technology needed for the coronagraph is not mature.  Indeed, the instrument is being flown on WFIRST as a technology demonstration mission with no science requirements.  Science may be conducted using it, but there are no science requirements driving its design.

Several CAA members asked whether NASA would consider removing the coronagraph to lower costs instead of reducing science capability in order to meet the $3.2 billion target.  Since it is a technology demonstration mission, for example, they asked why it could not be tested on a separate spacecraft.

CAA member Tom Young cautioned that demonstrating a challenging technology on a large, expensive flagship mission like WFIRST “is an oxymoron,” adding that he is not aware that NASA has ever done that before.  Young is a former Director of NASA’s Goddard Space Flight Center.  He was a member of the Decadal Survey study committee, the Harrison study committee, and the Hewitt study committee.

Hertz agreed that to the best of his knowledge NASA has never before conducted a technology demonstration on a flagship mission.

Young added that if NASA is intent on putting it on WFIRST, then it should be used as a scientific instrument, not just as a technology demonstrator: “you’re either all in or not.”

Hertz insisted that coronagraphs are important for future astrophysics missions and testing one on WFIRST is the “least expensive, quickest way to do it.”  He said that NASA had studied a standalone mission, Exo-C, and the estimated cost was $1 billion.

Zurbuchen said that he decided to keep the coronagraph on WFIRST because it is an important technology to demonstrate and because of where “all the stakeholders” are on the issue.

Zurbuchen has given Goddard four months to downscope the project, with a report due to him in February. He noted that there are still Key Decision Points WFIRST must pass through so there will be other opportunities to discuss these issues, but for now he is committed to having “this exciting piece of technology” as part of the mission.

Hertz said the plan is to move WFIRST into Phase B in the late March-early April 2018 time frame. He stressed that during Phase B, a range of costs, from high to low, will be developed; a specific cost is not required until KDP-C.

Note: this article was updated to clarify that NASA’s decision to establish WIETR was based on a recommendation in the Hewitt report.