How to pay for the OIR study recommendations?

Maria Womack (@StarzanPlanets) is a physics professor at the University of South Florida. From 2011-2015 she worked as a ‘rotating’ astronomy program director to the National Science Foundation and her research includes multi-wavelength spectroscopy of comets and exoplanets.

This is the third in a series of three guest posts on the recently released National Research Council (NRC) report often referred to as the “Optical and infrared (OIR) optimization study in the LSST era.” The first post introduces two NRC decadal reports referenced by the OIR Study; the second summarizes the committee’s motivation and top-level recommendations; the third addresses the estimated yearly cost for the recommended actions and how NSF might pay for them.

The OIR Optimization Study report only rarely breaks out individual costs, but estimates that in all, these recommendations can be met with an NSF investment of \$10-20 million per year.  NSF’s Division of Astronomical Sciences (AST) first responses to the study are generally positive, although cautious, due to depending on future budgets, as you can read in the Jun 2015 AAAC presentation slides and Sep 1, 2015 Dear Colleague Letter addressing the report.

It is important to realize that NSF’s response to most of these recommendations depends on whether the AST budget increases, possibly significantly, since there is no “extra money” in the current budget, which has been relatively flat at ~\$240 million since 2010. The NSF Town Hall presentation at the Jan 2015 AAS gives AST budget information, including approximate breakdowns into categories for recent years, along with predictions. In the Sep 1, 2015 DCL response at the link above, NSF states that “…the activities recommended for NOAO are beyond the planned scope and budget that is to be funded by NSF starting in FY 2016. Thus, augmentations to NOAO’s budget and scope would be required, after consideration of tradeoffs with other funding needs.” Here, we look at how NSF might fund the recommendations and some possible “tradeoffs” on the table.

Happily, some of the recommendations in #3, 4 and 6 can be met without an NSF AST funding increase, nor without impinging on the Astronomy and Astrophysics Research Grants (AAG) Program.  For example, some of these proposed instrumentation improvements are already allowed to compete for ATI and Major Research Instrumentation (MRI) funds.  As such, they do not require any new initiatives or moneys from NSF, but they do depend a large part on having a successful merit review (and thus, may displace other highly-ranked proposals).  Note that an extreme precision RV spectrometer (EPDS), which the OIR report recommended, is already incorporated into the NASA-NSF Exoplanet Observational Research partnership (NN-EXPLORE). This partnership started in early 2015, and hence, is already in the works – but the spectrometer will be paid for by NASA and not NSF. NN-EXPLORE is something I worked hard on during my last year at NSF as a ‘rotating’ program director.

Let’s consider a scenario in which the NSF annual budget stays at around \$240 – \$245 million for the near future.  It is very difficult, if not impossible, for NSF to substantially reduce or divest facility budgets in just a few years, although the agency is pursuing this. Furthermore, the facilities have already endured budget reductions and are fiscally lean. With a flat budget, the money to do all of the recommendations would probably have to come from the individual investigator and mid-scale grant programs. Since the mid-scale innovations program (MSIP) is specifically called out in this report for a significant increase in funding, this leaves the individual investigator programs, of which the AAG is the largest, and thus, has the greatest ability to act as “capacitor” for NSF AST’s budgetary needs.  

Some recommendations, such as the telescope and data exchange program, expanded MSIP, and Giant Segmented Mirror Telescopes (GSMTs) investments may very well exert pressure on the individual investigator grants programs. To understand this better, consider that the AAG program’s budget has been essentially flat at ~\$45M since 2010 and it makes ~120 new awards per year across all four areas: Planetary, Stellar, Galactic, and Extragalactic & Cosmology (this is also shown in the 2015 AAS Town Hall presentation).  Each new \$1M of a new initiatives will likely correspond to 2-3 new AAG awards that won’t be awarded that year (assuming an average 3 year total award amount \$380,000). Therefore, \$20M/year in new initiatives, which is what the committee estimates it would cost to do all the recommendations, could come at the expense of ~50 AAG awards from a year when only 120 are typically awarded.  Thus, funding even half of the recommendations without an NSF AST budget increase would likely put significant pressure on the AAG grants program and may push success rates into single digits.

If NSF AST maintains a flat budget for the next several years, would you support NSF paying for the OIR Study recommendations, even if it meant dramatic reduction in the amount of grants funded (possibly to you) each year? The OIR Optimization Study committee was specifically instructed not to discuss the AAG, and focus on top-level optimization of ground-based OIR operation, but how do you think AAG awards contribute to the needs of the OIR community in the LSST era?