3

Progress, Opportunities, and Challenges for Decadal Survey Research Goals and Recommendations

3.1 AN OVERVIEW OF PROGRESS TOWARD THE DECADAL RESEARCH RECOMMENDATIONS

The 2013 heliophysics decadal survey (NRC, 2013) included five top-level research recommendations and two top-level space weather application recommendations, some of which were divided into sub-parts (shown in Table 1.4 in Chapter 1). The present chapter reviews the research recommendations; Chapter 4 reviews the application recommendations.

Figure 3.1 summarizes progress made toward the research recommendations. A detailed description of progress, along with programmatic and other changes that have occurred since publication of the decadal survey (items 2 and 3 in the committee’s statement of task), is provided in the sections below. The committee’s research-related recommendations for the remainder of the survey decadal interval are also included in this chapter (committee tasks 4 and 5).

A key challenge affecting NASA’s implementation of the decadal survey recommendations is that the Heliophysics Division (HPD) budget has not increased to the level expected by the 2013 decadal survey (Figure 3.2). In 2014, NASA published its plan to implement the decadal survey, Our Dynamic Space Environment: Heliophysics Science and Technology Roadmap for 2014-2033 (NASA, 2014). The roadmap’s implementation plan accounted for budget expectations that were significantly lower than that assumed by the survey; the difference in projections amounted to an unplanned deficit of $100 million per year by 2024. Further, the decadal survey’s even higher “enabling budget,” which approached $750 million by the end of 2019, has not been realized. Instead there has been a modest increase to a little below $700 million in government fiscal year (GFY) 2019 (see also OIG 2019 report¹).

Over the last 5 years, the NASA HPD budget rose 14 percent, which is slightly less than the inflation rate. In contrast, the NASA overall budget rose by 23 percent over this time period, and the NASA Science Mission Directorate (SMD) budget rose by 30 percent. Additionally, according to the Office of the Inspector

___________________

¹ See the Office of the Inspector General (OIG) report on NASA’s Heliophysics Portfolio (NASA OIG, 2019). This audit assessed to what extent NASA (1) had an effective strategy for maintaining its heliophysics science capabilities, (2) was controlling costs for its current and planned missions, (3) had implemented appropriate recommendations and action plans, and (4) was effectively coordinating heliophysics activities across federal agencies and the private sector.

General (OIG) report, although Parker Solar Probe and the Global-scale Observations of the Limb and Disk (GOLD) missions were launched on schedule and within cost, ICON (Ionospheric Connection Explorer), Solar Orbiter, and SET (Space Environment Testbeds) have incurred a collective $41 million growth in cost as the consequence of launch-related delays.²

These budgetary factors have contributed to a delay in implementation of the next STP (Solar-Terrestrial Probes) mission recommended in the decadal survey (STP-5/IMAP, Interstellar Mapping and Acceleration Probe) and an inability to start the other recommended STP missions (DYNAMIC, Dynamical Neutral Atmosphere-Ionosphere Coupling; MEDICI, Magnetosphere Energetics, Dynamics and Ionospheric Coupling Investigation) and the recommended Living With a Star (LWS) mission (GDC, Geospace Dynamics Constellation). It should be noted, however, that IMAP (STP-5) and GDC are substantially delayed compared with the Heliophysics roadmap, even though the actual NASA HPD budget exceeds the roadmap forecast.

As of October 2019, the IMAP mission is in its formulation phase; the GDC mission has had only a community-supported concept study; and there are no current plans to start the DYNAMIC or MEDICI missions. Given that neither the 2013 decadal survey nor the roadmap had activity planned for the MEDICI mission in the current decadal interval (Figure 3.3), with respect to future strategic missions, DYNAMIC and GDC missions are the primary focus of study for this midterm assessment (see Sections 3.5 and 3.6, respectively). In addition to mission postponements, implementation of some DRIVE elements (e.g., Heliophysics Science Centers [HSCs]) has also been delayed, which Section 3.3 of this midterm assessment describes in detail.

The comparison of the roadmap notional budget and currently expected budget provided by NASA HQ requires some additional discussion. The “Research” branch in the NASA budget is separated into its compo-

___________________

² ICON launched on October 10, 2019. Its delays were due to a problem with the Pegasus launch vehicle. The OIG report notes that launch vehicle risks were not included in the cost analysis, thus leading to unexpected additional costs. However, not including launch vehicle risks in the cost analysis is consistent with agency practice.

nent parts in the roadmap. This allows us to appreciate the impact of the DRIVE program on the competed (grants) research program; DRIVE represents about a 50 percent increase to competed research when fully funded. The amount of research support available to the community is only about 30 percent of the total research funds indicated in the NASA HQ plot, with the remainder of research funds taken up by infrastructure and management support. The budget presented to this committee by NASA (Figure 3.2, bottom) subsumes the DRIVE elements into this same research line after fiscal year (FY) 2019. (Note that the NASA DRIVE elements are organized differently in Table 3.1, which shows an increase in the budget for the DRIVE elements from $65 million in FY 2015 to $155 million in FY 2020—an increase of 146 percent.)

In addition to budgetary challenges, execution of the survey’s recommended activities has occurred against a backdrop of frequent changes in HPD leadership. The current HPD Director, Nicola Fox, assumed her position in September 2018. Prior to her arrival, there had been six different directors or acting directors since 2011.

Despite management and budgetary challenges, NASA successfully launched three large-class missions—Van Allen Probes, Magnetic Multiscale Mission (MMS), and Parker Solar Probe—(Figure 3.4). Only two Explorer missions (Interface Region Imaging Spectrograph [IRIS], ICON) have been launched in the current decade, a notable difference compared to activity in previous decades, illustrating the motivation for the decadal survey recommendation related to Explorers (see Section 3.4). No medium-class missions have been launched in this decade; however, Solar Orbiter is currently on track to launch in early 2020, and Interstellar Mapping and Acceleration Probe (IMAP) was recently selected as a principal investigator

TABLE 3.1 Budget Actuals for NASA Heliophysics Research Programs (in millions of dollars)

NOTE: Provided by NASA HPD and the summed expected research budget from the NASA HPD roadmap. Note that the roadmap didn’t include any Space Weather Research Funding after FY 2014.

(PI)-led STP mission. Two new Explorer missions and one new Mission of Opportunity (MoO) have been selected in 2019. However, Explorers have not seen a reduction in development cost or time despite recent trends in small satellite manufacturing in the commercial sector.

NASA’s CubeSat program has shown impressive growth with 18 missions currently in development or recently launched. In addition to missions, NASA has embraced the DRIVE initiative, creating new programs; for example, H-TIDeS ITD (Heliophysics Technology and Instrument Development for Science: Instrument and Technology Development), LNAPP (Laboratory Nuclear, Atomic, and Plasma Physics), and the much anticipated HSCs, as well as expanding existing programs; for example, guest investigator (GI) programs. NASA also continues to support the Heliophysics Summer School, which was established in 2006 to help train young scientists and has recently created a new program for early-career investigators (ECIP).

The National Science Foundation (NSF) funding profile from 2012-2018 is shown in Figure 3.5. The AGS (Division of Atmospheric and Geospace Sciences) section has seen a 14 percent increase since 2012, with the majority of that increase since 2015. The 2016 NSF Geospace Portfolio review pointed out that 38 percent of the budget goes into operations and maintenance of facilities. The review made recommended closures or reduction in support for several Geospace Science (GS) facilities in order to enable new programs and facilities. In response, the Sondrestrom Incoherent Scatter Radar ceased operations in March 2018.

In 2016, NSF unveiled a set of 10 “Big Ideas,” which are described as “bold, long-term research and process ideas that identify areas for future investment at the frontiers of science and engineering.”³ For 2019, NSF plans to invest $30 million for each Big Idea; some of which—for example, “Mid-scale Research Infrastructure”—provide new opportunities for GS awards, albeit only through success in a highly competitive program.

___________________

³ National Science Foundation, “NSF’s 10 Big Ideas,” https://www.nsf.gov/news/special_reports/big_ideas/index.jsp.

At NSF, construction of the Daniel K. Inouye Solar Telescope (DKIST) is nearing completion and is expected to see first light in late 2019 (see 3.3.2). A Memorandum of Understanding was recently signed between NSF and the National Oceanic and Atmospheric Administration (NOAA) to support continued operations of Global Oscillation Network Group (GONG) for synoptic observations, at least in the short term. Although the NSF CubeSat program has been highly successful and is largely responsible for the current success and growth of CubeSats more generally, the CubeSat solicitation was not offered in 2016 and 2017 as the program was reinvented as the foundation-wide CubeSat Ideas Lab program. Two Missions (VISORS, SWARM-EX) were selected for development in 2019 as a result of the first Ideas Lab workshop. NSF recently revived its Faculty Development in Space Sciences (FDSS) program, selecting six universities to hire new faculty in 2019-2020. A NSF midscale facilities program was recently created, competed across all NSF divisions. NSF has also continued support of the Center for Integrated Space Weather Monitoring (CISM) summer school, now renamed the Boulder Space Weather Summer School.

A detailed assessment of progress towards all of the decadal survey recommendations is provided in the sections below. In Section 3.2, the baseline priority decadal survey recommendation is discussed. Progress towards each element DRIVE is discussed in Section 3.3, and NASA Explorers, STP, and LWS missions are discussed respectively in Sections 3.4-3.6.

3.2 BASELINE PRIORITY FOR NASA AND NSF: COMPLETE THE CURRENT PROGRAM

The baseline recommendation made in the decadal survey was the following:

This recommendation recognizes the importance of studying the coupled Sun-Earth system as a whole, and thus the necessity of a coordinated Heliophysics System Observatory (HSO) which includes NASA’s existing flight missions, missions under development, and NSF’s ground-based facilities. To ensure a robust HSO, the baseline and highest-priority recommendation made to NASA by the survey committee was to complete the missions that were then under development including the Radiation Belt Storm Probes (RBSP) and related Balloon Array for Radiation belt Relativistic Electron Losses (BARREL), MMS, and IRIS Explorer. For NSF, the decadal survey recommended continued development of the Advanced Technology Solar Telescope (ATST), which has since been renamed the Daniel K. Inouye Solar Telescope.

Radiation Belt Storm Probes was launched in August 2012 shortly before completion of the decadal survey, and was subsequently renamed the Van Allen Probes, in honor of James Van Allen who is credited with discovering Earth’s radiation belts. Van Allen Probes is part of the LWS mission line. It has been highly successful: with over 600 related publications since launch and a mission H-index4 of 48, the mission has changed our view of the structure of Earth’s radiation belts and led to unexpected discoveries. Some of these science highlights are described in Chapter 2 and Appendix E. The BARREL MoO carried out six balloon campaigns in support of the Van Allen Probes, with a total of 57 balloons launched from Antarctica and Sweden, revealing new information about electron loss to Earth’s atmosphere. After nearly 7 years of

___________________

⁴ For more on the H-index, see N. Oswald, “Does Your h-index Measure Up?,” BiteSizeBio.com, https://bitesizebio.com/13614/does-your-h-index-measure-up, and references therein.

operation, including 5 years of extended mission operations, the twin Van Allen Probes recently completed their end-of-mission deorbit maneuvers in late 2019.

MMS was launched during the current decadal survey period in March 2015. The 4-spacecraft mission is the latest in the STP line. The prime mission duration was 2 years after the start of science operations in September 2015. The prime mission was highly successful, meeting all of its Level 1 requirements and, to date, producing over 470 publications and achieving a mission H-index of 45. A science highlight from the mission is described in Chapter 2. With the highest time resolution electron measurements and most accurate electric and magnetic field measurements in the reconnection electron diffusion regions at the magnetopause and in the magnetotail, MMS has revolutionized our understanding of reconnection physics.

IRIS is a NASA Small Explorer (SMEX) mission launched in June 2013 and designed to investigate the physics of the Sun’s chromosphere, transition region, and corona. IRIS is the highest resolution observatory to provide spectra and images with seamless coverage from the photosphere into the corona. The unique combination of spectra and images at 0.33 arcsec resolution in the far ultraviolet (including C II and Si IV lines) and 0.4 arcsec resolution in the near ultraviolet (including the Mg II h and k lines), at a cadence as high as 2s, allows the tracing of mass and energy through the critical interface between the solar surface and the corona. An integral part of the IRIS science investigation is the development and public release of advanced numerical models to allow detailed statistical comparisons between IRIS observations and synthetic variables from the simulations. During more than 6 years of operation, IRIS has enabled crucial research on each of the four key science goals of the solar and space physics decadal survey, as well as many of the research focus areas of the Heliophysics roadmap. The application of machine learning techniques, combined with the extensive database of IRIS observations, has also revolutionized our diagnostic capabilities of the solar chromosphere, a key region in the solar atmosphere that will be the focus of NSF’s new 4m DKIST telescope. IRIS has produced over 315 publications and achieved a mission H-index of 35.

The missions above were in advanced stages of development at the time of the decadal survey. Thus, per the survey’s task statement, these missions were not included in prioritization exercises. However, the decadal survey committee did review two missions that were in earlier planning stages.⁵ Solar Orbiter is a European Space Agency (ESA)-NASA partnership that was targeted for 2017 launch with the objective of investigating connections between the solar surface, corona, and inner heliosphere from a distance of 62 solar radii. Solar Probe Plus was a mission under development that would fly closer to the Sun than ever before, discovering how the corona is heated and how the solar wind is accelerated. Both of these missions are part of the LWS program. Solar Probe Plus was renamed Parker Solar Probe (PSP) after Eugene Parker in May 2017, just over a year before its successful launch on August 12, 2018. The spacecraft will make 24 orbits around the Sun during its prime mission (2018-2025), diving closer and closer to the Sun. PSP has so far completed three perihelion passes (November 5, 2018, April 4, 2019, September 1 2019) and has set the record for closest approach to the Sun and fastest human-made object. An early science result is highlighted in Appendix E. A special issue of The Astrophysical Journal on PSP early results has over 50 submitted papers and is expected out in 2020.

NASA’s primary contribution to the ESA-led Solar Orbiter mission is the launch services aboard an Atlas V, with the current plan to launch Solar Orbiter in February 2020. NASA also supports the mission science and instrument hardware for the Heliospheric Imager (SoloHI), Heavy Ion Sensor, Energetic Particle Detector, Solar Wind Plasma Analyser, and Spectral Imaging of the Coronal Environment (SPICE). The Solar

___________________

⁵ Originally, the survey’s task statement excluded from consideration Solar Probe Plus. Midway through the survey, NASA requested that the survey committee comment on the scientific rationale for the mission in the context of scientific developments since the publication of the 2003 decadal survey. In addition, the survey committee was asked to provide appropriate programmatic or cost triggers as part of the anticipated decision rules to guide NASA in the event of major technical, cost, or programmatic changes during the development of Solar Probe Plus.

Orbiter will be in an elliptical orbit about the Sun with closest approach near the orbit of Mercury (0.3 AU). The Solar Orbiter will study the dynamics and energetics in the inner heliosphere that complement solar observations by NASA’s PSP, Solar Dynamics Observatory (SDO), IRIS, and Solar Terrestrial Relations Observatory (STEREO) satellites and DKIST ground-based observatory. The Solar Orbiter’s orbital inclination will be raised to 25° over its 7-year mission (and up to 34° for an extended mission), thus providing new glimpses of the solar pole’s magnetic fields that are crucial drivers for the solar dynamo 22-year cycle.

Construction of the ground-based solar telescope DKIST, formerly called ATST, was just underway when the decadal survey was published. With a 4-m aperture, DKIST is by far the largest optical solar telescope in the world and will provide extremely high-resolution measurements of the Sun. Construction is nearing completion, with first light expected Fall 2019. DKIST status and operations are discussed further in Section 3.3.2 below.

Finding 3.1 Completion of the program of record as recommended in the decadal survey, combined with new tools and data analysis approaches, has resulted in significant scientific advances (see Chapter 2) and has added important elements to the HSO.

3.3 IMPLEMENT THE DRIVE INITIATIVE

The initiation of DRIVE was recommended to maximize the science return from NASA heliophysics missions and NSF large solar and space physics ground-based facilities by coordinating existing research programs and making specific, cost-effective augmentations. Specifically, DRIVE aims to “diversify” observing platforms, “realize” the scientific potential of existing assets, “integrate” observing platforms into successful investigations, “venture” forward with new technologies, and “educate” the future heliophysics workforce. Among the specific items highlighted in the decadal survey are increased opportunities for small satellite projects, implementation of an NSF mid-scale facilities line, creation of heliophysics science centers, and increased investment in instrument development.

At NASA, the Research & Analysis (R&A) programs continue to be a major source of support for science research and have taken over much of the research formerly under Mission Operations and Data Analysis (MO&DA) within the mission lines. This is especially true for the large number of heliophysics missions in their extended mission phase. Proposal success rates for non-technological R&A programs at NASA hover around 20-25 percent as of this writing, which is an approximately 5-10 percent increase over rates several years ago. At NSF, operations of solar, space physics, and geospace facilities and data analysis are funded out of the AGS and Division of Astronomical Sciences (AST) under the Directorate for Geosciences and Directorate for Mathematics and Physics Sciences, respectively.

The decadal survey made 16 specific sub-recommendations under the top-level DRIVE recommendation. Progress for DRIVE is summarized in Figure 3.1. NASA research spending exceeded the 2014 Heliophysics roadmap expected spending during the first part of the decade (Table 3.1). While HPD lags behind other NASA SMD divisions in growth, the research grants program is healthy. This is consistent with the decadal survey rules of the road for spending priorities.

TABLE 3.2 National Science Foundation (NSF) DRIVE Funding (in millions of dollars)

SOURCE: Presented to the committee by the NSF Geospace Section Head.

Table 3.2 shows the breakdown of DRIVE funding by the NSF Geospace program for 2012-2018. The modest increase in funding for GS is spread relatively evenly across the programs, although the Space Weather program has seen a reduction.

A detailed discussion of progress on each DRIVE sub-element is provided in Sections 3.3.1 to 3.3.5, along with a discussion of the relevant programmatic and other changes that have occurred since the decadal survey was published. In Section 3.3.6, recommendations are made to the agencies for implementing DRIVE through the remainder of the decade.

3.3.1 DRIVE Diversify

Diversify observing platforms with microsatellites and midscale ground-based assets.

Progress Toward Decadal Survey Recommendations

The decadal survey made several recommendations to develop an increasing diversity of observing platforms, both on the ground and in space. Heliophysics has a long history of successful suborbital experiments that make use of sounding rockets and balloons, and their importance was reaffirmed in the decadal survey.⁶ Sounding rockets are the only platform that can make in situ measurements in the mesosphere and lower ionosphere (40-150 km), between the altitudes accessible by balloons and low Earth orbit (LEO) satellites. Balloons can carry heavy payloads high in the atmosphere (typically up to approximately 50 km) and have been instrumental in solar physics and particle precipitation studies. Rockets and balloons have also been used to augment larger NASA missions.⁷

The technology and launch opportunities for CubeSats as science missions have grown at a rapid pace. The NSF Directorate for Geospace Science was first to implement a modest CubeSat research-education program in 2008. For example, one of the highly successful NSF Geospace CubeSats is the Colorado Student Space Weather Experiment with more than 20 science papers, including one in Nature (Li et al., 2017). The decadal survey recognized the potential for space science at low cost using CubeSats.

___________________

⁶ See Appendix C of NRC (2013).

⁷ For example, BARREL recently made supporting measurements of precipitation in support of the Van Allen Probes mission (e.g., Woodger et al., 2015) and the solar EUV underflight calibration rocket flights supported SDO.

NASA SMD released the first grant solicitation for scientific CubeSats as part of ROSES 2013.⁸ The first NASA Heliophysics CubeSat science mission, the Miniature X-ray Solar Spectrometer (MinXSS), was launched in December 2015 to the International Space Station (ISS) and had a highly successful mission until May 2017 when it re-entered Earth’s atmosphere. MinXSS was the first CubeSat mission to demonstrate precision 3-axis pointing control of better than 10 arcsec, which enables new scientific observations that require fine pointing control. MinXSS also made new observations of the solar X-ray spectrum to study flare energetics and coronal heating (Woods et al., 2017).

NASA support for scientific CubeSats has steadily grown, and there are now 18 CubeSat science missions funded in the NASA HPD. As of August 2019, six NASA-funded CubeSats have been launched and another 12 are in development. Of the six NASA CubeSats launched, three have achieved full success, two have achieved partial success, and one is still in commissioning. The small augmentation to Research Opportunities in Space and Earth Science (ROSES) of $10 million for all SMD CubeSats in 2014 has now expanded to include a dedicated SmallSats and Rideshare Opportunities line with a $9 million per year budget in the Heliophysics Flight Opportunities for Research and Technology (H-FORT) program, initiated in ROSES 2019.⁹ The era of science exploration with CubeSats has just begun, and the community is already looking ahead toward constellations of small satellites to provide global coverage with much higher time cadence than what single, larger satellites have traditionally accomplished. NASA Headquarters (through the H-TIDES program) originally managed NASA Heliophysics CubeSat missions but oversight support has since expanded in 2019 to the NASA Goddard Space Flight Center Wallops Small Satellite Project Office.

Finding 3.2 CubeSat missions are intended to be low-cost, higher-risk exploratory missions. The number of CubeSat science missions has increased significantly in this decade. While recognizing the challenge of managing a rapidly increasing number of CubeSat projects, NASA will need to ensure that managerial oversight does not translate into the imposition of additional reviews and reporting requirements to the level of larger missions.

Suborbital projects are also supported by the NASA ROSES omnibus solicitation. Figure 3.6 shows the number of suborbital and CubeSat H-TIDeS selections from 2013-2018. The overall selection rate of both suborbital and CubeSat projects has increased, and the inclusion of CubeSats in the solicitation has not had a significant impact on the number suborbital projects selected.

___________________

⁸ ROSES-2013 separated the “Geospace Science” program, which formerly included Low Cost Access to Space (LCAS), into a number of distinct programs, including Heliophysics Technology and Instrument Development for Science (H-TIDeS). Appendix B.3 for this element states, “This program has three main research thrusts, (1) payloads on balloons, sounding rockets, or as secondary, rocket-class payloads, including CubeSats and International Space Station payloads, on flights of opportunity collectively referred to as Low-Cost Access to Space (LCAS), (2) Instrument and Technology Development (ITD) that may be carried out in the laboratory and/or observatory, and (3) enabling Laboratory Nuclear, Atomic, and Plasma Physics (LNAPP).”

⁹ In 2019 NASA created a separate ROSES element, H-FORT, for “Flight Opportunities for Research and Technology” that covered the LCAS (sub-orbital rockets and balloons) and SmallSat and Rideshare Opportunities. H-TIDeS continues as the program for Laboratory Nuclear, Atomic, and Plasma Physics (LNAPP) and Instrument Technology Development (ITD).

In summary, NASA Heliophysics has implemented a robust and growing CubeSat program. As development takes 3-4 years per mission, the realization of science results and benefits from these new CubeSat science missions is expected during the later half of the heliophysics decade (2019-2023). Additionally, the total number of rocket, balloon, and CubeSat selections has met or exceeded the new start rate of six per year that was recommended in the decadal survey.

For NSF, no CubeSats were selected for funding in 2013 and 2014, while three were selected in 2015. There is also a notable gap in 2016 and 2017, during which the CubeSat solicitation was not released. Two new CubeSat missions were selected for funding at the end of 2018. Following a recommendation that was made in the NSF portfolio review, a cross-division¹⁰ initiative to spur CubeSat innovation—CubeSat Ideas Lab—was created in 2019. The vision of the Ideas Lab is to “support research and engineering technology development efforts that will lead to new science missions in geospace and atmospheric sciences using self-organizing CubeSat constellations/swarms” (NSF, 2019a). Through this program, NSF initiated a study of constellation concepts and recently selected two 3-satellite constellation projects.

Finding 3.3 NSF’s CubeSat Program had no new solicitations in 2016 and 2017 and has not received a significant augmentation. However, the new CubeSat Ideas Lab initiative, if continued, will reinstate the program to the level that was recommended in the decadal survey.

___________________

¹⁰ The Ideas Lab is organized by the Division of Atmospheric and Geospace Sciences (AGS) in the Directorate for Geosciences (GEO), the Division of Computer and Network Systems (CNS) in the Directorate for Computer and Information Science and Engineering (CISE), and the Division of Electrical, Communications and Cyber Systems (ECCS) and the Division of Engineering Education and Centers (EEC) in the Directorate for Engineering (ENG).

In addition to the space-based suborbital and CubeSat platforms discussed above, the decadal survey recognized the importance of ground-based facilities. Facilities such as solar radio arrays, radars, riometers, and magnetometer arrays provide a more global view of the geospace system, and allow for long-term monitoring. All of these instruments also provide important context for single-point spacecraft measurements. While recognizing the importance of these facilities, the decadal survey also pointed out that there is a critical funding gap between relatively small to moderate ground-based projects and the very large facilities funded by the agency (such as DKIST). Two key midscale initiatives—the Frequency Agile Solar Radiotelescope and the Coronal Solar Magnetism Observatory (COSMO)—were identified as priority midscale ground-based infrastructure. At the time, no opportunities existed for funding projects of this size.¹¹

In response to broad community interest, the NSF has created a budget line for mid-scale research infrastructure¹² as one of its 10 “Big Ideas.” Two new solicitations were announced in 2018: the Midscale Infrastructure 1 (Mid-scale RI-1) program solicited projects in the $6 million to $20 million range (proposal deadline May 20, 2019), and the Mid-scale RI-2 program targeted projects in the $20 million to $70 million range (proposal deadline August 2, 2019). The Mid-scale RI-1 program also supports design and development programs in amounts down to $600,000. In addition, the threshold for eligibility for the Major Research Equipment and Facilities Construction line was reduced from 10 to 5 percent of an annual directorate budget, or roughly $70 million. These programmatic initiatives by NSF are necessary steps to address the long-standing need of the research community for a more balanced portfolio of research infrastructure in solar and space physics. However, since these opportunities are competed across multiple NSF divisions, the likelihood of more than one proposal in solar and space physics being selected is expected to be low. The AST and AGS divisions need to make the necessary investments to position priority initiatives to compete for midscale project funds successfully. Chapter 6 discusses how the solar and space physics community could improve their chances of being awarded a mid-scale facility.

Finding 3.4 NASA and NSF have provided a number of opportunities for the science community to add to the array of diverse observing platforms that enable heliophysics science, including a robust and growing NASA CubeSat program, continuation of a strong suborbital program, and creation of a NSF midscale facilities program.

The Changing Landscape Related to Diverse Observing Platforms

Since the decadal survey was published, there have been significant developments related to small satellites, particularly in the commercial sector. New additions to LEO activity are the planned large satellite constellations, or mega-constellations, promising continuous, global communication and Internet services. Such proposed constellations include SpaceX with 4,000 satellites, Samsung with 4,200 satellites, and OneWeb with 720 satellites (Radtke et al., 2017). This presents potential new opportunities for science,

___________________

¹¹ At the time of survey publication, the NSF equipment and facilities program supported investments in both small and very large facilities. NSF maintained a major research instrumentation program for instrument development projects (less than $4 million per year) and the Major Research Equipment and Facilities Construction program for large infrastructure projects (greater than 10 percent of an annual directorate budget, of order $140 million).

¹² Creation of the mid-scale line followed the recommendations of the National Science Board (NSB, 2018a).

including leveraging the technology development, increased rideshare opportunities, and commercial data buy opportunities.

A number of studies have been carried out to assess the scientific potential of small satellites. The 2015 National Academies of Scienes, Engineering, and Medicine report Achieving Science with CubeSats (NASEM, 2016), recognized that CubeSats have already had a scientific impact, and they have significant potential in specific areas of heliophysics research.¹³ An Explorer MoO, SunRISE, was recently selected for an extended Phase A concept study utilizing a small constellation of CubeSats to study solar radio bursts. Thus, CubeSats are already being proposed for larger missions. The recent Committee on Space Research (COSPAR) roadmap study for small satellites in space sciences outlined the science potential and opportunities for leveraging developments in the commercial sector (Millan et al., 2019).

On the issue of CubeSats, the 2016 NSF Geospace portfolio review (Lotko et al., 2016) recommended an increased emphasis on scientific mission concepts and instrument development, and less emphasis on engineering of CubeSat buses and communication systems. This recommendation was intended to encourage NSF as a whole to develop a proper home for the CubeSat program that embodies both science and technology, with potential applications beyond the Geospace Section. In response, Geospace leadership at NSF has taken an innovative approach to advancing the CubeSat program by teaming with the Engineering Directorate (ENG) and the Directorate for Computer and Information Science and Engineering (CISE) to create the CubeSat Ideas Lab (described above).

3.3.2 DRIVE Realize

Realize scientific potential by sufficiently funding operations and data analysis.

Progress Toward Decadal Survey Recommendations

The 2013 decadal survey DRIVE/Realize recommendations had elements directed both to NSF and to NASA. Recommendations addressed to each agency are discussed separately below, followed by consideration of the role of data science in meeting DRIVE/Realize recommendations.

DRIVE/Realize for NSF

NSF’s role in DRIVE includes support of essential ground-based facilities for obtaining synoptic data sets, such as the GONG solar magnetic maps regularly used for global coronal and solar wind analyses, modeling, and forecasting. These facilities continue to struggle to survive in spite of their widespread use. In addition, with the construction of DKIST, a major, new need for infrastructure support to maintain and utilize this state-of-the-art research tool must also be managed.

With respect to synoptic observations, FY 2016 support for the National Solar Observatory (NSO) included a one-time $2.50 million investment in GONG to increase its robustness for future space weather

___________________

¹³ For Heliophysics, the report noted that CubeSats can provide measurements from high risk orbits, augment large facilities with targeted supporting measurements, and have the potential to enable constellation missions.

predictions. NSO is in the process of upgrading the GONG facility with this funding, with completion expected in FY 2020. As part of the NSF plan to ramp up DKIST operations support, NSO’s synoptic program was cut from about $4 million per year to $2 million per year. This is partially mitigated by an NSF and NOAA interagency agreement in 2016 whereby NOAA is providing approximately $800,000 per year in funding support for GONG operations.¹⁴ However, concerns remain that the GONG and other synoptic observations by NSO are at high risk of ceasing operations in 2021 when the NOAA-NSO agreement ends. NSO proposals to the Air Force and NSF to enhance the synoptic instrumentation for research and space weather operations were declined in 2019. To maintain and grow the synoptic program beyond the NOAA-NSO agreement, NSO would need additional funding sources prior to 2021. As one example, such support could be acquired through the NSF Mid-Scale RI program.

Finding 3.5 A plan exists to support NSO’s synoptic observations in the short term. The long-term plan past 2021 for supporting these synoptic observations is unclear. To address this would require immediate attention.

DKIST is a ground-based, advanced-technology solar telescope operating at optical and infrared wavelengths. As an NSF AST facility operated by NSO, DKIST will be the flagship ground-based solar facility for the foreseeable future. Located on the summit of Haleakala on Maui, Hawaii, this $344 million construction project is nearing completion,¹⁵ and operations of DKIST should commence in summer of 2020.

Planning for operations of DKIST continues in parallel to the construction effort. Operations staffing is ramping up, and the DKIST Science Support Center on Maui has been completed. The DKIST Data Center, located in Boulder, Colorado, will process, store and distribute approximately 3 PB/year of fully calibrated data to the user community (i.e., Level 1 data). The data center completed its design phase in 2019 and is now entering the implementation phase. The community, led by the DKIST Science Working Group, is preparing the critical science plan, which captures high-priority observations to be conducted with DKIST during the initial operations phase.

The DKIST steady state operating cost is estimated to be $21.6 million per year, 13 percent higher than that estimated for the current AURA-NSO Cooperative Agreement (CA) for FY 2015-2024. However, by re-profiling the budget, NSO expects to remain within the 10-year CA budget. Under this plan, DKIST Data Center will provide Level 1 data, but the production of derived data products (Level 2 data) for detailed scientific research—e.g., magnetic field, temperature, and velocity in the solar atmosphere—are not included in current science operations planning.

NSF has recently provided supplemental funds to NSO for 2 years ($7 million total¹⁶) to define Level 2 data products and to develop processing algorithms under a plan that involves NSO scientists, postdocs, and graduate students from the community. The prospects for providing Level 2 data products to the community as part of steady state operations are not clear. The committee has concerns that there is no funding identified to routinely process or to improve Level 2 products past 2020. Continuation of the DKIST

___________________

¹⁴ Information from the FY 2019 NSF Budget Request to Congress. See https://www.nsf.gov/about/budget/fy2019/pdf/40t_fy2019.pdf.

¹⁵ All major site construction is complete. All large mechanical structures, including the Telescope Mount and the Coudé rota-tor, have been integrated and tested. The DKIST 4-meter primary mirror and the secondary mirror have been installed and aligned to specifications. The telescope has achieved first light pointing at stars and planets. Integration of instrument systems, including the polarization calibration unit, is now progressing (Rimmele, 2019). The integration, test, and commissioning phase of the project will continue into 2020 with the installation of optics to complete the optical path to the Coudé instrument laboratory and implementation of adaptive optics and four first-light instruments. The first high-resolution solar images from DKIST’s Visible Broadband Imager are scheduled to be obtained in fall of 2019 (V.M. Pillet, personal communication, 2019).

¹⁶ See NSF, “Directorate for Mathematical and Physical Sciences,” pp. MPS-1-MPS-20 in FY 2020 NSF Budget Request to Congress, March 18, 2019, https://www.nsf.gov/about/budget/fy2020.

Level 2 development is important and is motivated by the 2013 decadal survey statement: “Realizing the full scientific potential of solar and space physics assets … requires investment in their continuing operation and in effective exploitation of data.”

Finding 3.6 The scientific success of DKIST will depend on Level 2 and higher data processing. The committee is concerned that provision of robust Level 2 data products to the user community is not part of steady-state operations planning and no resources have been allocated by NSF for Level 2 data products and their development past 2020.

DKIST, like other NSF facilities in the AST division, will include a proprietary period during which data will not be publicly available. Additionally, there does not appear to be a plan for development of analysis tools to facilitate broad use of the data. In contrast, NASA’s SDO mission, which was recommended in the 2001 decadal survey of astronomy and astrophysics (NRC, 2001) at the same time that DKIST was recommended, made data available to the international community in near-real time. From the first, there was a library of SDO software tools that could be applied to aid in the production of scientific results. Further, NASA funds were available to carry out the SDO data analysis. DKIST will be a giant step forward in understanding how magnetic fields are generated and dispersed over the solar surface, how flares occur, how prominences are formed, and how coronal mass ejections are driven. In order to realize this scientific potential, DKIST data will be combined with data from NASA, ESA, and JAXA (Japan Aerospace Exploration Agency) missions. The mission science teams will expend a great deal of effort to supply collaborative observations that are freely available, while the DKIST observations are proprietary.

Finding 3.7 DKIST is the flagship observatory of NSF solar astronomy. DKIST funding past 2020 supports primarily DKIST operations and its data center, but with limited support for research. Substantial research funding, of more than $5 million per year, from NSF needs to be available in anticipation of the number of science proposals that will be submitted. Coordinated efforts that use DKIST along with NASA, ESA, and JAXA mission data will lead to scientific breakthroughs, requiring adequate support.

At NSF, the construction of large facilities is typically funded by programs outside of the divisions (e.g., AGS or AST); however, maintenance and operating costs must be covered by the division budget. Every new facility comes with a large operations and maintenance cost within a fixed divisional budget. This has led to closure of the Sondrestrom facility and funding challenges for the Arecibo Observatory. As another example, DKIST construction funds came from the NSF facilities budget, while its operations budget and grants program will be in the AST budget—a budget that will not be incremented because of the new major facility. This results in the paradox that the world’s most scientifically powerful ground-based solar telescope will reduce the funding available to support that very telescope’s scientific potential.

A recent National Science Board (NSB) study¹⁷ was conducted on operations and maintenance costs for NSF facilities. This study found that, because operations and maintenance costs have not been a major budgetary problem for NSF as a whole, impacts at the divisional level might not be apparent. Moreover, choices made at the divisional level out of budgetary necessity, such as maintenance deferral, descoping of science, and underutilization, may not be in alignment with NSF’s strategic priorities. The report recommended that (1) the NSB and the NSF director should continue to enhance agency-level ownership of the facility portfolio through processes that elevate strategic and budgetary decision-making, (2) NSF and NSB should reexamine what share of NSF’s budget should be devoted to research infrastructure, and (3) NSB

___________________

¹⁷ NSB, Study of Operations and Maintenance Costs for NSF Facilities, NSB-2018-17, https://www.nsf.gov/nsb/publications/2018/NSB-2018-17-Operations-and-Maintenance-Report-to-Congress.pdf.

and NSF should develop model funding and governance schemes for the next generation of partnerships at the agency, interagency, and international levels. Such recommendations could be achieved, for example, by having separate maintenance and operational budgets for any facilities developed with the support of NSF funding at the agency level rather than at the division level.

Finding 3.8 The operations and maintenance model for NSF’s large facilities has had significant impacts on the AGS and AST budgets.

DRIVE/Realize for NASA

In order to realize the scientific potential of the HSO, the decadal survey recommendations to NASA included increased MO&DA funding and institution of a mission-specific GI program.

In making this recommendation, the decadal survey refers both to MO&DA funding for mission extensions and the importance of a stable general GI program. However, the suggested funding increase appears to refer only to MO&DA for extended missions. Table 3.3 shows the 2014 Heliophysics roadmap extended mission operations budgets for 2013 and 2017 versus the FY 2017 actuals taken from the 2018 NASA Office of the Inspector General report (NASA OIG, 2018). The GI program budget is not included. Note that NASA HPD had no missions in prime operations phase in 2017. The comparison between 2013 and 2017 may in part reflect how mission costs are bookkept. Nevertheless, the MO&DA funding for some missions was higher than projected in the roadmap budget for FY 2017.

In the years leading up to the last decadal survey, support for the GI program was sporadic. The decadal survey pointed out the importance of a stable GI program and also recommended creation of a mission-specific directed GI element in order to address cuts in both Phase E mission funding and cuts in the general GI program that occurred in the prior decade (e.g., see Box 4.2 in the decadal survey).

Funding for the GI program has increased from $9 million in 2015 to an expected $21 million in 2020 (Table 3.1), compared with a notional approximately $8 million in the 2014 roadmap. In 2013, the ROSES GI element solicited both general proposals and proposals that focused on the Van Allen Probes mission. ROSES-2014 GI was also open to general and mission-specific (Van Allen Probes/BARREL and IRIS) proposals. In ROSES-2016 and 2017, however, the GI program was separated into two different elements, an open GI element and a mission-specific (MMS) element. ROSES-2018 included only an open GI program. The intended ICON/GOLD GI element was delayed; the solicitation states, “This Program element has been delayed to ROSES-2019, at which point the data streams will be stable for both missions.” However, ROSES-2019 also did not include this program element, presumably due to the ICON launch delay, though

GOLD has been operating for well over a year. The ROSES solicitation in 2019 did include both the open GI program and an outer heliosphere element that supports analysis of the Interstellar Boundary Explorer (IBEX), Voyager, and other relevant heliospheric data, such as from New Horizons and Cassini. However, while a healthy GI program is critical in enhancing the scientific potential of missions, particularly in their extended phase, it does not support the mission team and primary science objectives, and thus should not be viewed as a replacement for adequate Phase E funding.

The Changing Landscape Related to Realizing Scientific Potential

Broad community involvement in NASA Heliophysics missions is critical for realizing their maximum scientific potential. The GI program, while extremely valuable, has traditionally enabled such participation primarily after launch. The recent implementation of the decadal survey recommendation to make STP missions PI-led (Section 3.5) could have the unintended side effect of reducing this kind of community involvement in strategic missions. It is critical to maintain and enhance community involvement during the earlier phases of mission development. This will (1) expand the diversity of perspectives and ideas for accomplishing the mission science goals, (2) engage the community earlier so they are familiar with the mission and can be more productive immediately after launch, and (3) enhance the diversity of mission teams.

The GI program as it is traditionally implemented is not the best way to address this issue because GIs are not viewed as part of the mission team. Better mechanisms may exist that provide an opportunity for

scientists to be engaged as members of the mission team earlier in the process without having to compete against proposals that use data from already-operating missions. A recent IMAP mission paper (McComas et al., 2018) outlines a plan for community engagement, welcoming participation from outside scientists. However, this participation is unfunded, potentially excluding members of the community. Funding will be provided during Phase-E (after launch) through a mission GI program, and the mission plans both a student collaboration and future leaders component to involve early-career scientists. Nevertheless, HPD can learn from past experiences and other divisions to insure broad and diverse participation.

The model used by NASA’s Planetary Science Division, their Participating Scientist (PS) Program,¹⁸ provides a useful example that HPD could learn from and consider for future missions. The PS Program provides a mechanism by which scientists can participate in team meetings and contribute ideas early in the mission. A similar model has been used successfully for previous Heliophysics missions. For example, the MMS and TIMED missions had interdisciplinary scientists (IDS). However, the IDS model has not been routinely implemented for all HPD strategic missions. It should be emphasized that, to be successful, such a program must be implemented with care; in the new PI-led model for STP missions, the PI is responsible for meeting Level 1 requirements. Thus, such a program must be implemented in a way that is value-added and does not impose additional requirements on the PI and team.

Finding 3.9 A model similar to the PS Program used in the Planetary Science Division would contribute to realizing the scientific potential of Heliophysics missions by ensuring broad and diverse community participation.

DRIVE/Realize and the Role of Data Science in Solar and Space Physics

Another development mentioned in the decadal survey, but which has become even more pressing in recent years, is the size of data sets and our ability to efficiently store, retrieve, and analyze the data in a reproducible way. DKIST is expected to produce 25 terabytes of data a day for some 40 years, amounting to hundreds of petabytes throughout its lifetime (Berukoff et al., 2015). Existing NASA satellites already produce large amounts of data. For example, the SDO produces 1.5 terabytes of data a day (Pesnell et al., 2012) and has accumulated a few petabytes of data to date. Moreover, simulations also have higher spatial and temporal evolution than ever before. The advent of these incredibly large and complex data sets, along with sophisticated data-processing techniques and relatively inexpensive computing power, created what NSF calls “the data revolution” (NSF, 2019c).

Science Platforms: Developing a Modern Data Infrastructure and Workflow Using Common Standards

In order to efficiently explore and analyze large and complex data sets, the solar and space physics community will need to develop a modern data infrastructure and workflow to store, retrieve, and process large data sets (e.g., Bauer et al., 2019). This will require a change in scientific workflow; instead of moving data to a local machine to analyze, users move their software to an external computing environment and perform their analysis there, minimizing data transfer. Several institutes have developed science platforms to analyze large data sets in astronomy, such as the National Optical Astronomy Observatory Data Lab (Fitzpatrick et al., 2014) and the Large Synoptic Survey Science Platform (Dubois-Felsmann et al., 2019).

___________________

¹⁸ The benefits of the Participating Scientist program are described in Grebowsky et al. (2015).

Scientific Software: Incentivizing and Supporting the Development and Adoption of General Purpose Open-Source Software Tools

Open-source software packages such as scikit-learn (Pedregosa et al., 2011), which include machine learning and data mining algorithms, have enabled nearly all of the machine learning and data mining studies within solar and space physics over the last 5 years (Burrell et al., 2018). Many of these studies also used open-source libraries for efficient data analysis, such as cloud computing and parallel processing. In addition to these general computing applications, the number of open-source software packages specific to the solar and space physics community has grown considerably over the last 5 years,¹⁹ such as space weather open-source applications in the Community Coordinated Modeling Center (CCMC).

However, these packages developed by members of the community remain largely unfunded. At present, funding to support digital infrastructure is often donation based—for example, the Linux Foundation’s Core Infrastructure Initiative, NumFOCUS, Mozilla’s Open Source Support program, the Free Software Foundation, the Sloan Foundation, the Ford Foundation, and the Moore Foundation. For example, SunPy received $265.00 from NumFOCUS in 2017.²⁰ This increased to $3,120 in 2018 (NumFOCUS, 2018). According to statistics from OpenHub, a service that tracks open-source software, the cost of producing the SunPy code base using paid software developers (with an annual salary of $75,000) would take about 7 years and cost approximately $500,000.²¹

In 2018, the National Academies published the report Open Source Software Policy Options for NASA Earth and Space Sciences (NASEM, 2018) and recommended increased support from NASA SMD for open source software development.²² Some funding opportunities are beginning to appear. The NASA ROSES-2019 Heliophysics Data Environment Enhancements program solicits proposals entirely for the development of open-source software and encourages the community to adhere to a set of standards and workflows to maximize interoperability and reduce duplicate efforts. NSF is also responding with its Cyberinfrastructure for the Geosciences program. Continued support is essential, as is recognition within the reviewing community that software and associated computing hardware cost are significant and critical

___________________

¹⁹ About 50 such open source software packages (e.g., Annex et al., 2018) exist today—such as SunPy (SunPy Community et al., 2015), SpacePy (Morley et al, 2014), and PlasmaPy (Plasma Py Community et al., 2018). The foundation of the open source scientific programming stack—a collection of five packages for array manipulation (NumPy; Van Der Walt et al., 2011), time series analysis (Pandas; McKinney et al., 2010), plotting (matplotlib; Hunter et al., 2007), numerical methods (SciPy; Jones et al., 2001) and development environments (Pérez et al., 2007)—contributed significantly to the rapid development of general-purpose tools for the solar and space physics community.

²⁰ The 2017 annual report from NumFOCUS, a 501(c)(3) public charity which serves as a fiscal sponsor for many open-source scientific software packages.

²¹ Figures cited on “The Black Duck Open Hub” at: https://www.openhub.net/p/sunpy/estimated_cost/.

²² Among the key recommendations of Open Source Software Policy Options for NASA Earth and Space Sciences (NASEM, 2018) were the following (p. 4):

for much of solar and space physics research. Accepting these costs in proposals, and questioning proposals that claim to be able to carry out research without these resources, can quickly change the working environment in the heliophysics community.

Education and Training: Participating in Workshops, Conferences, and Courses to Learn Modern Statistical and Computational Techniques

Gleaning meaningful scientific results from large and complex data sets requires a new kind of scientist—a data scientist—well-versed in both their physical domain and also in modern statistical and computational techniques (VanderPlas, 2014; Faris et al., 2011). Yet, much of the solar and space physics community is still unfamiliar with data science. Additional opportunities to educate and train the community with such modern data science techniques are needed. Agencies can help by sponsoring workshops and university training programs. An example of an existing program is the Large Synoptic Survey Telescope Data Science Fellowship Program, which teaches data skills not easily addressed by current astrophysics programs.

Interdisciplinary Collaboration: Establishing Opportunities for Solar and Space Physicists to Collaborate with Data Scientists, Statisticians, and Computer Scientists

To effectively implement modern statistical and computational techniques, the solar and space physics community will need support for engaging in interdisciplinary collaboration with data scientists, statisticians, and computer scientists specializing in machine learning and data mining. For example, funding agencies could encourage and sponsor the development of interdisciplinary data science centers—such as the three Moore-Sloan Data Science Environments (the Berkeley Institute of Data Science, the University of Washington eScience Institute, and the New York University Center for Data Science)—and interdisciplinary grant programs, such as those that compete under the NSF Harnessing the Data Revolution Big Idea.

Finding 3.10 A modern data infrastructure, support for the development of software tools, education about data science methods, and interdisciplinary collaboration are needed to realize the scientific potential of the large and complex data sets being produced today.

3.3.3 DRIVE Integrate

Integrate observing platforms and strengthen ties between agency disciplines.

Progress Toward Decadal Survey Recommendations

In ROSES-2013, NASA created the LNAPP program within H-TIDeS. This program is currently funded at $0.6 million per year (Table 3.1) with an average of two new selections per year (Figure 3.7 below). The LNAPP program is separate from the existing joint NSF-Department of Energy (DOE) program, so this does not completely address the decadal survey recommendation, and the level of investment does not meet the decadal survey target. In particular, LNAPP supports laboratory experiments, but there is currently no program at NASA supporting development of computer codes or tools that support laboratory plasma science.

Connections between the heliophysics community and the plasma physics community are growing. There are several plasma laboratories that focus on experiments motivated by questions that have arisen in space physics. DOE is supporting laboratories and personnel that work with the outside community to develop new experimental investigations. These facilities are eager to support new users, thus there is a real opportunity for the heliophysics community. NASA can facilitate progress by making efforts to better coordinate with DOE and by enabling the community to take advantage of these opportunities. Currently, Plasma 2020,²³ a decadal assessment of plasma science, is being conducted by the National Academies; NASA may find new opportunities arising from this assessment.

Finding 3.11 Laboratory research, from plasma physics to spectroscopy, is a critical, foundational component for heliophysics research. The NASA LNAPP program is a positive step toward increasing opportunities for laboratory experiments, but it does not fully address the decadal survey recommendation, specifically the need for increased NASA-DOE collaboration.

Significant progress has not been made towards this recommendation. For example, Sun-as-a-star and planetary magnetospheric research falls between the AGS and AST divisions. Another example is the science of the outer heliosphere. Recent observations of the outer heliosphere by NASA satellites raise fundamental science questions pertaining to the structure of shocks, where and how magnetic reconnection takes place, and how particles are accelerated, all of which are subjects integral to the Sun-Earth-heliosphere system science program. However, there is still no clear home for outer heliosphere research at NSF.

Finding 3.12 The placement of solar and space physics in multiple divisions and directorates arises from the cross-cutting relevance of the science. However, there are very few cross-divisional funding opportunities at the agencies. This makes it difficult for proposers to obtain funding for basic research on subjects that are not clearly aligned with one division. Proposals that cross divisional lines also pose significant challenges to agencies and review panels.

The 2019 ROSES solicitation Appendix B.4 for the Open Guest Investigators program was recently amended to allow for ground-based instrumentation associated with the THEMIS (Time History of Events and Macroscale Interactions during Substorms) mission to be used as a primary data source for investigations.²⁴ Note, however, that these particular ground-based instruments were originally funded by NASA

___________________

²³ Information about Plasma 2020 is available on at NASEM, “Decadal Assessment of Plasma Science,” https://sites.nationalacademies.org/BPA/BPA_188502.

²⁴ “April 1, 2019. B.4 Heliophysics Guest Investigator—Open Program has been updated to indicate that All Sky Imagers (ASI) and Ground Magnetometers (GMAG) associated with the THEMIS mission are considered to be part of the Heliophysics System Observatory (HSO). Investigations using these data as their primary data source are permitted.” (From ROSES 2019 solicitation, as amended.)

as part of THEMIS mission development. Currently, NASA only funds the use of other (e.g., NSF-funded) ground-based observations if they are used as supporting data. The importance of coordinated observations is only growing. For example, the combination of ground-based radio and optical data with PSP measurements will be a powerful tool for studying the Sun. Such coordination requires support; the NASA Research Announcement released in September 2019 for HSO data support is a positive step in this direction.

Finding 3.13 Diverse observing platforms continue to produce important scientific results and augment the capabilities of larger facilities. The opportunities for maximizing the use of diverse platforms and combining their measurements have not been fully exploited; further opportunities exist to leverage international collaboration and combine measurements from space-based and ground-based platforms.

The Changing Landscape Related to Integrating Platforms and Strengthening Ties

Heliophysics System Observatory

The decadal survey makes regular reference to the richness of the HSO and its role in major discoveries and progress on the key science goals laid out in the decadal survey. However, as shown in Table 1.2, the HSO is largely populated by missions in their (sometimes much) extended phase of operation. Losing spacecraft will result in the loss of critical measurements necessary to understand the global and system-level picture of the heliosphere. There are several regions within the heliosphere where critical measurements may be lost at any time. Examples include the following:

• SDO is the only satellite providing high-resolution and high-cadence solar magnetograms. These data yield critical information on the solar magnetic field that cannot be obtained in sufficient detail from the ground and enable off-Sun-Earth axis observations that allow reconstruction of three-dimensional features and vector information.
• With the TIMED satellite nearing the end of its life,²⁵ critical information about Earth’s natural thermostat, the nitric oxide 5.3 µm cooling of the thermosphere, will be lost. TIMED also currently provides temperature and constituent measurements at the poles and in the mesosphere that connect Earth’s space environment with Earth’s lower and middle atmosphere.
• With the Van Allen mission at its end, only the Japanese mission Arase (also in extended mission phase) is able to provide measurements across the heart of the radiation belts.
• Since the last decadal, it has become clear that the Voyager spacecraft may be nearing the end of their productive lifetimes. At the same time, both have uncovered new phenomena in the outer heliosphere that cannot be understood with the measurements of their limited payload.

One exception is the specific attention given to the ongoing necessity of L1 solar wind observations and coronagraphs to monitor Earth space weather conditions and to enable space weather–related science. The continuation of those measurements is discussed more in Chapter 4. However, it should be noted that the capabilities of instruments developed primarily for operational use may differ from those developed to satisfy research needs; therefore, there may be some scientific objectives that are not met with the operational measurements.

In addition to its elements aging, the vision of the HSO—to have strategically placed missions that enable the systems-science approach required to understand the Sun and its effects on planets in our solar

___________________

²⁵ The Thermosphere, Ionosphere, Mesosphere, Energetics and Dynamics (TIMED) mission was launched on December 7, 2001. Its nominal design life was 2 years.

system—can only be achieved if the HSO is driven by some strategic planning. Currently, all NASA-selected missions must be stand-alone; a mission’s contributions to the HSO are not considered during the procurement process. Moreover, to fully realize the HSO vision will require integration of ground-based facilities and missions of all sizes into the HSO concept since, more often than not, multiple data sources are used for scientific studies. The decadal survey recognized the HSO as a fully integrated systems-science observatory, but it is not clear that the agencies currently recognize this. The continuation of and enhancements for the HSO are discussed in more detail in Chapter 6.

Finding 3.14 Many elements of the HSO are aging, and there is a risk of losing key capabilities. In order to realize the vision of the HSO, some longer-term strategic planning is required to prioritize the critical support needed at both the mission level and the program level. Moreover, the HSO can be viewed as a national resource that goes beyond NASA missions. Data from small missions, ground-based facilities, and international assets have become increasingly important. An opportunity exists to elevate the HSO concept to better manage and exploit this critical resource for scientific progress.

Crossdisciplinary Science

Some of the most important advances in heliophysics lie in its connections to other disciplinary areas. There are obvious connections with the Earth sciences; for example, the coupling of the ionosphere, thermosphere, and mesosphere above 50 km to the lower atmosphere is studied instensely by the Earth science community. Climate change is another area where heliophysics research overlaps with Earth science; for example, anthropogenic increases of CO₂ are being observed in the thermosphere. Comparative planetology is a growing field within the planetary science community, particularly for Earth-Mars comparisons largely inspired by the MAVEN Mars mission, which also significantly involves the heliophysics community. Applying knowledge from heliophysics research also helps to interpret stellar activity in other systems, while observing other Sun-like stars can teach us about the potential extremes of solar activity. Similarly, heliophysics research contributes to exoplanet science through the applications of concepts and models used for solar system planet–solar wind interactions and space weather influences on atmospheres and surfaces. Understanding planetary evolution and habitability relies in large part on our knowledge of the current solar system environment and solar outputs, as well these conditions in the past and future. Finally, Voyager and IBEX results have transformed our understanding of the interstellar boundaries of astrophysical objects, while observations of other astrospheres provide alternate realizations of heliosphere-like systems with different internal and external properties.

In all of these examples, research that incorporates broader perspectives that go beyond disciplinary boundaries has the potential to open new horizons and raise new questions. Real breakthroughs are often made in cross-disciplinary areas—breakthroughs that benefit heliophysics research as a whole.

Funding structures and review panels currently are not set up for efficient support of the inherently multidisciplinary approach needed to address these science challenges. A few LWS-focused science topics have featured this type of research, but opportunities are limited, and there is no obvious home for such proposals at NSF. The NExSS (Nexus for Exoplanet System Science) Program at NASA attempts to be inclusive in creating virtual institutes from already-selected proposals across all four divisions within SMD to accomplish astrobiology goals in particular. However, heliophysics participation is relatively small, perhaps because this opportunity is not widely known or advertised within heliophysics. The historical lack of support may also be limiting participation in these areas. The rapid development of transiting exoplanet studies warrants particular attention within the context of the decadal survey DRIVE program and based on all four of the survey’s key science goals.

Finding 3.15: Heliophysics has much to contribute to areas of broad interest within NASA SMD, including stellar system and exoplanet research as well as future major exploratory efforts; for example, the Lunar Gateway missions. However, the expertise and knowledge that exists within the heliophysics community is not as widely exploited at SMD as it could be because there are insufficient opportunities to engage across division lines.

3.3.4 DRIVE Venture

Venture forward with science centers and instrument and technology development.

Progress Toward Decadal Survey Recommendations

The decadal survey also made recommendations to push the boundaries in the areas of both theory and technology developments, arguing that transformational progress often comes from collaborations between theorists, modelers, computer scientists, and observers.

The Announcement of Opportunity (AO) for the heliophysics science centers was released in 2019. The selection of the centers will proceed via a two-phase process. Phase 1 proposals proceeded through a standard Step-1 and Step-2 proposal process, with Step 1 due on March 1, 2019 and the Step-2 (full) proposals for phase 1 due on June 20, 2019. The response from the community was significant, with 44 Step-1 proposals ruled to be compliant with the AO. The number of completed Step-2 proposals is unclear, but the community response indicates great enthusiasm for the implementation of this recommendation. It is expected that approximately 6 Phase 1 proposals will be selected and funded at approximately $650,000 per year for 2 years. At the end of this 2 years, the Phase 1 teams will submit Phase 2 proposals for the full implementation on their center concepts.

The 2013 decadal survey recommendation to establish science centers is on its way to being implemented by NASA. While slower to start than anticipated, the ongoing proposal process is nonetheless a very positive step toward ensuring more adequate support for realizing the results of missions, suborbital and ground-based heliophysics observations, and for the basic research that seeds the next heliophysics endeavors. NSF is not currently providing funding for science centers, but it has supported science and technology centers competed across all areas of science and engineering since 1987. CISM was funded through this program from 2002-2013, for example. More recently, NSF has created an Artificial Intelligence Institutes program. NSF is currently contributing to the HSCs by “providing input on best practices for conducting science center operations.”²⁶ However, it is unclear how the different centers and institutes relate to one another. Some coordination between agencies is needed to ensure that the HSCs are effective.

___________________

²⁶ From Geospace Section Head presentation to committee in February 2019.

Finding 3.16: A regular cadence for HSCs is needed. In order for HSCs to be impactful, the next call for Step-1 proposals should be released within a year of the down selection for Step-2 proposals. Moreover, full NSF participation in the HSCs has not been realized.

As mentioned above, ROSES 2013 included support for instrument development through its H-TIDeS program. Figure 3.7 shows the number of Instrument Technology Development (ITD) and LNAPP proposals selected for 2013-2018. An example of the success of the ITD program was the development of terahertz measurement capability for measuring winds. A new sensor technology, called the TeraHertz Limb Sounder, was developed with NASA funding to make these critical wind measurements under a wide range of observational conditions (e.g., day and night, with and without aurora) from LEO.²⁷

As of 2019, flight projects, including small satellite technology demonstrations, have been moved into the new ROSES H-FORT element (Figure 3.8). H-TIDeS retains the ITD line at $4 million per year (in ROSES 2019). This matches the decadal survey goal of consolidating NASA technology funding and exceeds the funding level recommended in the decadal survey (see Table 3.1).

The Changing Landscape Related to Venturing Forward

The growth of the commercial small spacecraft industry and increased launch opportunities are enabling growth in small innovative instrumentation projects. NASA needs to be prepared for a dramatic increase in the number of H-TIDeSand H-FORT proposals over the next few years.

Since publication of the decadal survey, NASA has announced ambitious plans to return to the Moon and to establish a Lunar Gateway. Extending a long-term presence beyond Earth’s protective magnetic shield raises many issues in space weather, both for predictions and for the mitigation of its adverse effects on technological systems and human health (Chapter 4). NASA’s lunar plans also provide potential new opportunities for heliophysics science as NASA extends human flight outside of LEO for extended periods of time. The new NASA Heliophysics Space Weather Science and Applications (SWxSA) program, as discussed in Chapter 4, could explore space weather partnerships with Artemis flight opportunities and in collaboration with the NASA Human Exploration and Operations Mission Directorate.

___________________

²⁷ From NASA Science Mission Directorate Technology Highlights 2016 report.

3.3.5 DRIVE Educate

Educate, empower, and inspire the next generation of space researchers.

Progress Toward Decadal Survey Recommendations

Following a gap in opportunities for NSF FDSS, NSF recently revived the program by selecting six universities (Arizona State University, Georgia State University, University of Hawaii, Montana State University, New Mexico State University, West Virginia University) to hire new faculty in 2019-2020 and to support curriculum development in solar and space physics. As noted in the NSF Geospace Portfolio Review, the FDSS program has been successful. Of the eight faculty members supported by the program through 2014, all but one led to a tenured faculty member. The 2019 selections for the NSF FDSS program meet the goal of this DRIVE recommendation, although a regular cadence is needed to ensure that this program has a positive impact on solar and space physics.

Over the past decade, the NASA Heliophysics Summer School has instructed well over 300 of the most promising students across the variety of research subfields within heliophysics. Each year of the school, a particular theme is selected, enabling the school to continually evolve with its scientific disciplines while always covering the fundamentals of the physics of the local cosmos. In parallel to teaching, the Heliophysics Summer School project has resulted in a series of five books (four published in printed form (Schrijver et al., 2009, 2010a, 2010b, 2016) and a fifth available online at the school’s website²⁸) reviewing the diverse environments and connected processes in the Sun-planet system. These address past, present, and future of the solar system, compare terrestrial and other planets, and look beyond the local cosmos to other planetary systems and their stars. These five books are complemented by recorded lectures and by problem sets and laboratory experiments (largely developed with help from the NASA CCMC), all hosted on the Internet. A condensed textbook based on the extensive heliophysics texts is currently being developed. The continuation of this summer school is a valuable asset in the training of the next generation of heliophysics researchers.

NSF funded the CISM Space Weather Summer School from 2002-2013. Post-CISM, a separate NSF-funded program, called the Boulder Space Weather Summer School (SWSS), administered by the High Altitude Observatory, has been put in place to continue the CISM summer school. This is a 2-week program that targets beginning graduate students and advanced undergraduates who are considering a career in solar, space, atmospheric, or related sciences. It is also open to space weather practitioners in government and industry who are interested in enriching their understanding of the solar-terrestrial system and the causes

___________________

²⁸ See C.J. Schrijver, F. Bagenal, and J.J. Sojka, eds., “Heliophysics V. Space Weather and Society, Early Chapter Collection,” January 5, 2015, https://cpaess.ucar.edu/sites/default/files/heliophysics/documents/HSS5.pdf.

and impacts of space weather events. Admission is open to both U.S. and international students, although the SWSS cannot provide support for international travel. Enrollment is limited to about 30 students each year. The program is funded through 2020 with expected renewal in subsequent years.

In addition, there are student workshops that NSF hosts at the annual CEDAR, GEM, and SHINE workshops. These training opportunities about new data systems and emerging software tools occur both at those workshops, as well as at the Solar Physics Division and American Geophysical Union meetings.

Finding 3.17 NSF and NASA have responded positively to this graduate student training recommendation. The CISM summer school, now the Boulder Space Weather Summer School, has been funded by the NSF. In addition, NASA has continued to fund the Heliophysics Summer School. The former has a focus on beginning students and modeling of space weather, while the latter is more targeted to basic research science for advanced graduate students and post-doctoral researchers. These activities provide an outstanding resource to a community in which heliophysics graduate students in a given department are often few in number and specialized courses in the discipline are not feasible.

NASA also recently established the ECIP at a level of $1.5 million per year in FY 2019 and FY 2020 (expected). The program supports early career professionals within 10 years of receiving their Ph.D. In response to its first offering, the program received broad interest with 101 Step-1 proposals submitted, 50 step-2 proposals reviewed, and 11 proposals selected for funding.²⁹ While the program addresses the challenges faced by early career professionals in heliophysics to establish secure funding, it is obviously heavily oversubscribed.

No progress has been made on this recommendation. The 2016 NSF Portfolio Review reinforced the decadal survey recommendation,³⁰ pointing out that students who apply for NSF Graduate Research Fellowships in geospace science may be at a disadvantage due to the absence of solar and space physics as a category.

The Changing Landscape Related to Education

As described in Section 3.3.2, significant developments in data analysis methods and tools have occurred both within and beyond the science community. Moreover, many of these software developments are “open source”—facilitating further collaborative development. The National Academies’ report on opensource software (OSS) states, “the fact that coding is becoming as essential as calculus to scientists could motivate secondary schools and colleges to include software development best practices in their curricula for all science, technology, engineering, and mathematics (STEM)-bound students. OSS provides a way to educate and train new talent” (NASEM, 2018, p. 55).

___________________

²⁹ Data taken from NASA NSPIRES.

³⁰ Recommendation 4.10 in the report: The GS should work with the NSF office that maintains “Survey of Earned Doctorates” to implement immediately the category “Solar and Space Physics” (or another name to be determined) into the survey.

Finding 3.18: Advances in the capability of open-source software and the related heliophysics tool sets are not often covered in undergraduate and graduate education. Training the next generation in software best practices enables robust and maintainable code.

Since the publication of the 2013 decadal survey, citizen science (public participation in scientific research) has become more prominent in solar and space physics. An example of the scientific benefits of citizen science, the discovery of STEVE, was discussed in Chapter 2. Another example is the Aurorasaurus project,³¹ a citizen science website where participants report sightings and details of the aurora. The data have been used to improve models for auroral forecasts. Citizen science allows the research community to leverage a large volunteer workforce that can provide a unique set of measurements—for example, those distributed around the globe in the case of Aurorasaurus. In addition, citizen science provides an important outreach tool. It has the ability to engage many thousands of volunteers in scientific research and the potential to inspire new generations of heliophysics researchers.

3.3.6 DRIVE Recommendations for the Next Four Years

The 2013 decadal survey made specific actionable recommendations under the DRIVE initiative, and these have been largely addressed by NASA and NSF. The DRIVE elements have led to increased funding of suborbital and CubeSat missions, a boost to R&A programs, and the imminent selection of the first DRIVE science centers. The spirit of DRIVE is to continue to innovate and look for new ways to maximize scientific progress. Thus DRIVE should be viewed as a means for organizing the R&A programs in a way that can respond and adapt to new opportunities.

Finding 3.19 DRIVE is an organizational framework that encourages innovation and balance across NASA and NSF R&A programs, thus maximizing the science return of agency investments. In the future, DRIVE may include new elements or augmentations that go beyond the limited number of recommendations made in the decadal survey. It is essential to continue tracking and making visible the elements of DRIVE.

Recommendation 3.1: NASA and NSF should continue to use the DRIVE framework within their Research and Analysis programs. As the program elements that are part of DRIVE continue to evolve, they should remain visible and continue to be tracked in a transparent manner.

Finding 3.20 NASA and NSF have made progress on most of their DRIVE elements, although some of the DRIVE elements were implemented only recently. Funding constraints imposed by the decadal survey requirement to complete the current program are a contributing factor.

Finding 3.21 Some elements of DRIVE for NSF have not been fully implemented. These include ensuring funding for science areas that fall between divisions such as outer heliosphere research, full participation in HSCs, and recognition of solar and space physics as a subdiscipline in the annual survey of earned doctorates.

In addition to evaluating the progress on decadal survey recommendations, the committee identified new opportunities that have emerged since the decadal survey was published. The findings outlined in the

___________________

³¹ See the AuroraSaurus website at http://www.aurorasaurus.org/.

sections above lead to a recommendation for building on recent progress and taking advantage of new opportunities through the rest of the decade.

3.4 ACCELERATE AND EXPAND THE HELIOPHYSICS EXPLORERS PROGRAM

Progress Toward Decadal Survey Recommendations

The third recommendation of the decadal survey was to accelerate and expand the highly successful Heliophysics Explorers program, enabling a Medium-Classs Explorer (MIDEX) line and frequent MoOs.

In April 2013, shortly after the decadal survey was released, NASA selected the ICON mission, along with the GOLD MoO, for development. GOLD was launched in January 2018 into a geostationary orbit onboard a commercial telecommunications satellite. GOLD makes images of the thermosphere and ionosphere, providing atmospheric composition and temperature and the density and structure of the ionosphere. The first data from GOLD have been released and reveal, for example, dramatic plasma instabilities in the ionosphere during and after sunset that appear far more common than anticipated (Eastes et al., 2019). The ICON spacecraft was delivered on schedule in late 2017; however, its launch was delayed repeatedly due to issues with the Pegasus launch vehicle.³² ICON finally launched on October 10, 2019; all systems are currently operating nominally, and an initial return of science data is expected November 2019. ICON will provide coordinated observations of the neutral atmosphere and ionosphere at low latitude, aimed at understanding the interaction between the gas and plasma. Given their complementary views from geostationary and low earth orbits, overlap between the GOLD and ICON observations are expected to enable significant discoveries that would not have been possible with either mission alone.

Between 2013 and 2015, no Explorer AOs were released in heliophysics. In 2016, the Heliophysics SMEX AO was released which included a Stand-alone MoOs (SALMON) element. Five SMEX missions and three MoOs were selected for Phase A concept studies. In 2019, the Atmospheric Waves Experiment (AWE) MoO was selected for implementation and is expected to launch to the ISS in 2022. AWE is focused on studying atmospheric waves in the mesopause region, where such waves often become large in amplitude and non-linear effects increase dramatically. The SunRISE mission to study solar radio bursts was also provided with additional funding for an extended Phase A study. In June 2019, two SMEX missions were selected to continue into Phase B. PUNCH (Polarimeter to Unify the Corona and Heliosphere) will focus on the Sun’s outer atmosphere, the corona, and how it generates the solar wind. TRACERS (Tandem Reconnection and Cusp Electrodynamics Reconnaissance Satellites) will study magnetic reconnection in the cusp region of Earth’s magnetosphere. In July 2019, a MIDEX AO was released, with proposals due in September 2019; thus a cadence of 3 years was achieved between the SMEX and MIDEX AOs.

The 2018 SALMON opportunity solicited several different types of MoO. Two technology demonstrations and two science missions were selected for Phase A study as potential rideshares for the IMAP mission.³³ It is anticipated that one from each category will be selected to launch with IMAP. Rideshare opportunities at NASA are discussed in more detail below. In September 2019, three stand-alone science MoOs were also selected for Phase A study.³⁴

The Changing Landscape Related to Explorers

Small missions, from CubeSats up to MIDEX, are critical elements of the toolset needed to advance the science of heliophysics. These relatively small to mid-sized missions, if effectively implemented along with the large strategic missions, should enable us (1) to fill gaps in observables, particularly in the current environment of infrequent large missions and aging on-orbit resources, (2) to efficiently implement the use of innovative technologies, and (3) to motivate and involve a larger and younger segment of the research

___________________

³² See NASA OIG (2019), pp. 17-18 and Table 5.

³³ The two science rideshare missions are the Spatial/Spectral Imaging of Heliospheric Lyman Alpha (SIHLA) mission to study the heliosphere boundary with the interstellar medium and the Global Lyman-alpha Imagers of the Dynamic Exosphere (GLIDE) to study Earth’s exosphere, consisting mostly of hydrogen. The two tech-demo rideshare missions selected are the Science-Enabling Technologies for Heliophysics (SETH) mission to demonstrate higher data rates from deep space with optical communication and the Solar Cruiser mission to demonstrate solar sail technology. A downselect to one science mission and one tech-demo mission is expected after Phase A studies are completed in 2020, and both will be launched in October 2024 as rideshare missions with IMAP.

³⁴ Extreme Ultraviolet High-Throughput Spectroscopic Telescope (EUVST) Epsilon Mission, Aeronomy at Earth: Tools for Heliophysics Exploration and Research (AETHER), and Electrojet Zeeman Imaging Explorer (EZIE).

and engineering communities. Increasing proposal costs and mission budgets, the burden of undesirably high standards for risk mitigation, and the growing AO-to-launch intervals all need to be addressed to optimize the role that small-sats up to MIDEX can play in advancing solar and space physics.

Launch costs continue to be a major component of the Explorer budget. NASA has recently committed to including an ESPA³⁵ ring on every SMD launch. This has the potential to benefit the Explorers program, in particular by reducing launch costs and providing launch opportunities to orbits that are not easily accessible otherwise. For example, NASA has taken an innovative approach in its planning for the IMAP mission by offering five ESPA-ring slots for small satellites to share the ride to the Lagrange L1 point. For one of the slots, NOAA has partnered with NASA to launch its Space Weather Follow-On mission to L1.³⁶ As discussed above, in 2018, NASA released a call for both a science and a technology demonstration MoO to fly as rideshares with the IMAP mission. While rideshares can contribute to reduced cost and increased flight opportunities, the potential impact of delays imposed on the major mission by the minor mission has to be managed.

The Explorers Office website³⁷ states, “The mission of the Explorers Program is to provide frequent flight opportunities for world-class scientific investigations from space utilizing innovative, streamlined and efficient management approaches within the heliophysics and astrophysics science areas.” However, the most recent SMEX selection took 3 years from AO to selection (this timeframe included AO, review, Phase A, review, and selection). Reducing the number of requirements for the Phase A Concept Study Report might help shorten the Phase A duration. If future SMEX missions continue to have a long review and down-select time and high cost, there will be adverse impacts on the rate of scientific progress and innovation. There is also a potential negative impact on workforce development and retention. This topic is further discussed in Chapter 5.

The commercial sector is developing new ways to manufacture small satellites using technologies and processes learned from aircraft manufacturing. One example, among several, is OneWeb, which has partnered with Airbus to produce 900 satellites with a mass of 150 kg each—similar to a SMEX—at a rate of three per day and for less than $1 million per satellite (Iannotta, 2019). There is an opportunity for the science community and agencies to learn from and leverage these developments in order to reduce the costs of small missions, to enable more frequent access to space, and to support constellations of small satellites.

In December 2017, NASA Associate Administrator Thomas Zurbuchen released the document “Class D Tailoring/Streamlining Decision Memorandum”³⁸ announcing a new streamlined process for implementing Class-D missions with costs below $150 million (not including launch cost), which includes the Explorers SMEX and MoOs.³⁹ The impact of this memo on the recent MoO and SMEX proposals and the development of the selected missions requires tracking and evaluation. This topic is further discussed in Chapter 6.

___________________

³⁵ ESPA stands for EELV Secondary Payload Adapter. It was originally developed by the Air Force to facilitate launch of secondary payloads on large launch vehicles.

³⁶ For more on SWFO, see, E. Talaat, “Accelerating Progress Toward NOAA’s Next Generation Architecture,” 2019 Goddard Memorial Symposium, March 21, 2019, https://astronautical.org/dev/wp-content/uploads/2019/04/RHG_Thu_0800_Talaat.pdf.

³⁷ See NASA Goddard Space Flight Center’s Explorer Program website at https://explorers.gsfc.nasa.gov.

³⁸ See NASA, “NASA Science Mission Directorate (SMD) Class-D Tailoring/Streamlining Decision Memorandum,” signed December 7, 2017, https://essp.nasa.gov/essp/files/2018/05/SMD-Class-D-Policy.pdf.

³⁹ Information about risk classification for NASA missions is further described in a December 15, 2014, slide presentation by Chief Safety and Mission Assurance Engineer Dr. Jesse Leitner, NASA Goddard Space Flight Center, “Risk Classification and Risk-based Safety and Mission Assurance,” which can be found at https://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/20150001352.pdf.

Findings on the Explorer Program

3.5 RESTRUCTURE STP AS A MODERATE-SCALE, PI-LED LINE

Implement IMAP, DYNAMIC, MEDICI-Like Missions

___________________

⁴⁰ See NASA Goddard Space Flight Center’s Explorer Program website at https://explorers.gsfc.nasa.gov.

ensure coordination with NASA Voyager missions. The mission implementation also requires measurements of the critical solar wind inputs to the terrestrial system.
R3.2: The second STP science target is to provide a comprehensive understanding of the variability in space weather driven by lower-atmosphere weather on Earth. This target is illustrated by the reference mission Dynamical Neutral Atmosphere-Ionosphere Coupling (DYNAMIC).
R3.3: The third STP science target is to determine how the magnetosphere-ionosphere-thermosphere system is coupled and how it responds to solar and magnetospheric forcing. This target is illustrated by the reference mission Magnetosphere Energetics, Dynamics, and Ionospheric Coupling Investigation (MEDICI).

NASA has transitioned the STP program to be a moderate-scale, competed, PI-led mission line. Figure 3.9 shows the decadal survey–recommended budget and 4-year cadence for the next STP missions. The first of these, IMAP, was prioritized to take advantage of the overlap with the historic Voyager missions and would study the outer heliosphere and its interaction with the interstellar medium. The launch date for IMAP anticipated in the decadal survey was 2021 and had already been shifted to 2022 by the time NASA issued its survey implementation plan in the 2014 Roadmap. The second STP science target would study the variability in space weather driven by lower-atmosphere weather on Earth. This notional mission was called DYNAMIC with a decadal survey anticipated launch date of 2025. The 2014 Roadmap estimated DYNAMIC to start in 2020 and launch in 2025. The third and final STP science target recommended for this decade focused on how the magnetosphere-ionosphere-thermosphere system is coupled and how it responds to solar and magnetospheric forcing. This science target was illustrated by the reference mission called MEDICI with a decadal survey anticipated launch date of 2029. The decadal survey already

anticipated that MEDICI would not start before the next decadal interval; thus, this midterm assessment focuses on IMAP and DYNAMIC.

In 2017, an AO was released for a mission addressing IMAP science goals. The AO outlined a PI-led mission cost capped at $492 million (in GFY 2017 dollars). In response to this AO, two proposals were submitted, and one of these was selected in 2018. The selected IMAP mission with 10 instruments is planned for a launch to Lagrange point L1 in October 2024. The IMAP mission will study the solar wind and energetic particles from the Sun that modulate the boundary of the heliosphere and the harmful cosmic radiation that penetrates deep into the heliosphere to Earth.

Finding 3.26 Formulation of the first of three recommended STP missions has begun, but IMAP comes 3 years later than anticipated in the decadal survey, and the next STP mission (DYNAMIC) has not started. As anticipated in the decadal survey, the MEDICI mission is not expected to start until the next decade.

In June 2019, a community workshop was held at the CEDAR meeting to discuss the science goals for the notional DYNAMIC constellation mission in light of the selection of two MoOs (single instruments; GOLD and AWE) and one Explorer (ICON) since the decadal survey was published. DYNAMIC’s main science goals are (1) to resolve lower atmosphere influences on the atmosphere-ionosphere-magnetosphere (AIM) system by measuring the height evolution of the wave spectrum in the thermosphere that produces spatio-temporal variability within the system and (2) to provide the much needed day and nighttime wind measurements throughout the whole thermosphere to study ion-neutral interactions and dynamo processes.

The AWE instrument, to be installed on the ISS in late 2022, will remotely sense airglow emission from the lower boundary of the AIM system to extract part of the gravity wave spectrum impinging on this boundary from lower-atmosphere sources. The GOLD instrument, launched in January 2018, offers a unique geosynchronous vantage point, viewing one-third of the globe with a resolution sufficient to resolve large- and small-scale thermosphere structures in temperature and composition. As such, both missions address parts of Atmosphere-Ionosphere-Magnetosphere Interactions (AIMI) panel science goals 1 (GOLD) and 3 (GOLD and AWE) specified in the decadal survey. The ICON mission will advance our understanding of day-to-day ionospheric variability and monthly mean global-scale neutral atmospheric wave-ionospheric interactions at low-to-midlatitudes and partly contributes to AIMI science goal 2.

The newly selected GOLD and AWE MoOs and ICON Explorer mission will answer important targeted science questions that contribute to our understanding of lower-atmospheric impacts on the AIM system. However, these missions do not adequately observe Earth’s poles and thus will not fully address the decadal survey top-level Research Recommendation 3.2 “to provide a comprehensive understanding of the variability in space weather driven by lower-atmosphere weather on Earth.” As discussed in Chapter 2, progress made since the decadal survey highlights an increasing community need for a DYNAMIC-like whole atmosphere mission, particularly to resolve day-to-day wave and mean state variability, and to obtain the highly coveted day and nighttime wind measurements throughout the whole thermosphere that drive many processes in the AIM system. GOLD, AWE, and ICON can only partially address science questions that would be answered by DYNAMIC. These missions do not provide needed coverage in high latitudes, and their instruments lack the nighttime capability to measure winds in a portion of the thermosphere where the transition into diffusive equilibrium occurs, waves dissipate, and ion-neutral and dynamo interactions take place. The daily local time resolution needed to resolve day-to-day wave variability cannot be provided by single satellites or instruments. Single platforms are unable to resolve the tidal and planetary wave fields from the lower atmosphere due to aliasing caused by incomplete sampling. A constellation of satellites covering all latitudes and multiple local times is required to remove aliasing issues and address the influence of planetary-scale wave sources on the AIM system and dynamo processes. Day and night-

time wind measurements throughout the thermosphere are needed to study ion-neutral interactions and dynamo processes in the AIM system.

Finding 3.27 The DYNAMIC science goals remain compelling and of high priority for the heliophysics community. The targeted science goals and measurement capabilities of GOLD, AWE, and ICON do not address several key objectives in the top-level decadal survey science challenge posed by DYNAMIC.

Recommendation 3.4: NASA should take the steps necessary to prepare for the release an Annoucement of Opportunity for a DYNAMIC-like mission.⁴¹

3.6 IMPLEMENT A LARGE LWS GDC-LIKE MISSION

GDC was outlined as a notional LWS mission concept in the decadal survey. The mission would consist of six identical satellites in LEO, providing simultaneous, global observations of the coupled AIM system. GDC will address crucial scientific questions pertaining to the dynamic processes active in Earth’s upper atmosphere; their local, regional, and global structure; their response to magnetospheric drivers; and their role in modifying magnetospheric activity. It will be the first mission to address these questions on a global scale due to its use of a constellation of spacecraft that permit simultaneous multi-point observations. This investigation is central to understanding the basic physics and chemistry of the upper atmosphere and its interaction with Earth’s magnetosphere, but also will produce insights into space weather processes. GDC science continues to be of high priority to the heliophysics community, specifically in the field of AIM interactions. If successful, GDC would revolutionize scientific understanding of the dynamics within the ionosphere-thermosphere system.

The decadal survey’s recommended scientific investigation and design reference mission for GDC was refined and updated in 2019 by a community-based Science and Technology Definition Team (STDT)⁴² established as a subcommittee of the Heliophysics Advisory Committee (HPAC), an advisory committee established under the Federal Advisory Committee Act (FACA). The STDT assessed the science rationale for the mission and the provision of science and mission parameters, including the optimal number of spacecraft required to address the science, and any other scientific aspects needed. At the end of its work, the STDT submitted a final report on October 2, 2019, to HPAC that contained potential mission implementation scenarios and the scientific trade-off among the studied scenarios (NASA, 2019).

The STDT defined various mission implementation scenarios that are feasible, effective, allow for the evolution of the ionosphere-thermosphere system to be tracked across a range of temporal and spatial

___________________

⁴¹ Per the decadal survey, such a mission would begin as the next STP mission after IMAP. Steps in preparation for the AO could include a new study of its mission goals and objectives.

⁴² The STDT’s membership consisted of 17 experts from the Heliophysics community that covered relevant scientific and technical expertise. The committee met in person 3 times, with additional teleconferences, and solicited community input through a NASA Request for Information (RFI); 56 responses were transferred to the STDT by NASA.

scales. The options in the STDT fully address the requirements specified by NASA in the STDT charter while also ensuring alignment with the recommendations of the 2013 decadal survey.⁴³

The GDC STDT (NASA, 2019) was an important first step in reviewing and refining the science goals for the reference mission outlined in the decadal survey. However, in contrast to previous STDTs for similar missions, a number of topics related to mission formulation and implementation were excluded from consideration.44 The report examines four possible mission architectures and assesses the science closure that would be achieved with each of these. No recommendation on the preferred architecture was made, although the report suggests that CubeSats may offer some advantages. In order to fully realize the mission goals for a GDC-like mission, it will be necessary to determine the best implementation of a satellite constellation.

Finding 3.28 The GDC STDT, per their charge, was not permitted by FACA regulations to select a particular mission architecture to meet GDC science objectives. (Finding 3.28)

Recommendation 3.5: In order to proceed toward meeting the top-level decadal survey Living With a Star mission recommendation, NASA should take the steps necessary to define a specific mission architecture formulation and implementation scheme for the Geospace Dynamics Constellation within the next 3 years.

3.7 REFERENCES

___________________

⁴³ The GDC mission will unravel complex mysteries in the combined and interacting ionized and neutral gases of the IT (ionosphere-thermosphere) system by using an array of satellites. GDC is anticipated to be capable of measuring, for the first time, both the large-scale and localized dynamics of the interaction between the upper atmosphere and the near-Earth space plasma environment. The GDC mission will address two overarching science goals with specifically actionable objectives: (1) understand how the high latitude ionosphere-thermosphere system responds to variable solar wind/magnetosphere forcing; (2) understand how internal processes in the global ionosphere-thermosphere system redistribute mass, momentum, and energy.

⁴⁴ These are listed in the STDT report (NASA, 2019) as follows:

1. Particular instrument types, instrument builds, non-spacecraft capabilities (e.g. models, ground-based observatories). While some measurement requirements have generally been met by particular instruments, the STDT shall not recommend those particular instruments to the exclusion of other instruments (or combinations thereof) that could meet the requirement of measuring particular physical parameters.
2. The method, structure, content, or target of any mission formulation activity. This includes the direction, competition (e.g. AO, RFP), or invited contribution (e.g. from international partners) of mission components (e.g., spacecraft, instruments, inter-mission collaboration).
3. Any procurement activity in support of the mission formulation activity. In instances where a need or opportunity outside of the mission concept is recognized, the STDT shall identify for NASA to address.
4. Mission development costs or mission budget targets, either projected or recommended. All needed budgetary constraints will be provided by NASA.
5. Any provider-specific bus or bus type.
6. Any specific launch vehicle or launch strategy. In instances where a design reference mission requires the launch or deploying of multiple spacecraft, NASA will provide launch constraints.
7. Any potential NASA collaborations with specific US or non-US organizations.
8. Any space weather operational goals or requirements.

Astropy Collaboration, A.M. Price-Whelan, B.M. Sipocz, H.M. Günther, P.L. Lim, S.M. Crawford, S. Conseil, et al. 2018. The Astropy Project: Building an open-science project and status of the v2.0 core package. Astrophysical Journal 156:123. doi:10.3847/15383881/aabc4f.

Baker, M. 2016. 1,500 scientists lift the lid on reproducibility. Nature 533:452. doi:10.1038/533452a.

Bauer, A.E., E.C. Bellm, A.S. Bolton, S. Chaudhuri, A.J. Connolly, K.L. Cruz, V. Desai, et al. 2019. “Petabytes to Science.” https://arxiv.org/abs/1905.05116.

Berukoff, S., K. Reardon, T. Hays, and D. Spiess. 2015. “The DKIST Data Center: Challenges and Opportunities Ahead.” Paper 402.01 presented at AAS/AGU Triennial Earth-Sun Summit.

BiteSizeBio. 2017. “Does Your h-index Measure Up?” April 2. https://bitesizebio.com/13614/does-your-h-index-measure-up/.

Bobra, M.G., and S. Couvidat. 2015. Solar flare prediction using SDO/HMI vector magnetic field data with a machine-learning algorithm. Astrophysical Journal 798:135. doi:10.1088/0004-637X/798/2/135.

Bortnik, J., W. Li, R.M. Thorne, and V. Angelopoulos. A unified approach to inner magnetospheric state prediction. 2016. Journal of Geophysical Research: Space Physics 121:2423. doi:10.1002/2015JA021733.

Buckheit, J.B., and D.L. Donoho. 1995. Improved linear discrimination using time-frequency dictionaries. Proceedings of SPIE 2569:540-551. doi:10.1117/12.217608.

Burrell, A., A. Halford, J. Klenzing, et al. 2018. Snakes on a spaceship—An overview of Python in heliophysics. Journal of Geophysical Research: Space Physics 123:10. doi:10.1029/2018JA025877.

Camporeale, E., A. Carè, and J.E. Borovsky. 2017. Classification of solar wind with machine learning. Journal of Geophysical Research (Space Physics) 122:10910-10920. doi:10.1002/2017JA024383.

Claerbout, J.F., and Karrenbach, M. 2005. Electronic documents give reproducible research a new meaning. Pp. 601-604 in SEG Technical Program Expanded Abstracts 1992. Stanford University Press, Calif. doi:10.1190/1.1822162.

Dubois-Felsmann, G., Economou, F., Lim, K.-T., et al. 2019. “Large Synoptic Survey Telescope (LSST) Data Management Science Platform Design.” https://ldm-542.lsst.io/.

Eastes, R.W., S.C. Solomon, R.E. Daniell, D.N. Anderson, A.G. Burns, S.L. England, C.R. Martinis, and W.E. McClintock. 2019. Global-scale observations of the equatorial ionization anomaly. Geophysical Research Letters 46:9318. doi:10.1029/2019GL084199.

Eghbal, N. 2016. Roads and Bridges: The Unseen Labor Behind Our Digital Infrastructure. Ford Foundation. https://www.fordfoundation.org/about/library/reports-and-studies/ roads-and-bridges-the-unseen-labor-behind-our-digital-infrastructure/.

Elmore, D.F., T. Rimmele, R. Casini, S. Hegwer, J. Kuhn, H. Lin, J.P. McMullin, K. Reardon, W. Schmidt, A. Tritschler, and F. Wöger. 2014. The Daniel K. Inouye Solar Telescope first light instruments and critical science plan. Proceedings of SPIE 914707. doi:10.1117/12.2057038.

Faris, J., E. Kolker, A. Szalay, L. Bradlow, E. Deelman, W. Feng, J. Qiu, D. Russell, E. Stewart, and E. Kolker. 2011. Communication and data-intensive science in the beginning of the 21st century. OMICS: A Journal of Integrative Biology 15:213. doi:10.1089/omi.2011.0008.

Fitzpatrick, M.J., K. Olsen, F. Economou, E.B. Stobie, T.C. Beers, M. Dickinson, P. Norris, et al. 2014. The NOAO Data Laboratory: A conceptual overview. Proceedings of SPIE 91491T. https://doi.org/10.1117/12.2057445.

Geiger, R.S., C. Cabasse, C.Y. Cullens, L. Norén, B. Fiore-Gartland, D. Das, and H. Brady. 2018. “Career Paths and Prospects in Academic Data Science: Report of the Moore-Sloan Data Science Environments Survey.” SocArXiv. doi:10.31235/osf.io/xe823.

GitHub. 2019. “A Gallery of Interesting Jupyter Notebooks: Reproducible Academic Publications.” https://github.com/jupyter/jupyter/wiki/A-gallery-of-interesting-Jupyter-Notebooks#reproducible-academic-publications.

Grebowsky, J., K. Fast, E. Talaat, M. Combi, F. Crary, S. England, Y. Ma, M. Mendillo, P. Rosenblatt, K. Seki, M. Stevens, and P. Withers. 2015. Science enhancements by the MAVEN participating scientists. Space Science Reviews 195:319. doi:10.1007/s11214-014-0080-4.

Hunter, J.D. 2007. Matplotlib: A 2D graphics environment. Computing in Science and Engineering 9:90-95. doi:10.1109/MCSE.2007.55.

Iannotta, B. 2019. “Mass Producer.” Aerospace America. https://aerospaceamerica.aiaa.org/departments/mass-producer/.

Ivezic, Z., A.J. Connelly, J.T. VanderPlas, and A. Gray. 2014. Statistics, Data Mining, and Machine Learning in Astronomy. Princeton University Press. https://press.princeton.edu/books/hardcover/9780691151687/statistics-data-mining-and-machine-learning-inastronomy.

Jones, E., T. Oliphant, P. Peterson. 2001. “SciPy: Open Source Scientific Tools for Python.” http://www.scipy.org.

Katz, L. 2019. Evaluation of the Moore-Sloan Data Science Environments Final Report. http://msdse.org/files/MSDSE Eval Final Report Feb 2019 v2.pdf.

Lecun, Y., Y. Bengio, and G. Hinton. 2015. Deep learning. Nature 521:436. doi:10.1038/nature14539.

Li, X., R. Selesnick, Q. Schiller, K. Zhang, H. Zhao, D.N. Baker, and M.A. Temerin. 2017. Measurement of electrons from albedo neutron decay and neutron density in near-Earth space. Nature 552:382-385. doi:10.1038/nature24642.

McComas, D.J., E.R. Christian, and N.A. Schwadron. 2018. Interstellar Mapping and Acceleration Probe (IMAP): A new NASA mission. Space Science Reviews 214:116. doi:10.1007/s11214-018-0550-1.

McGranaghan, R.M., A.J. Mannucci, B. Wilson, C.A. Mattmann, and R. Chadwick. 2018. New capabilities for prediction of high-latitude ionospheric scintillation: A novel approach with machine learning. Space Weather 16:1817. doi:10.1029/2018SW002018.

McKinney, W. 2010. “Data Structures for Statistical Computing in Python.” Pp. 51-56 in 9th Python in Science Conference. doi:10.25080/Majora-92bf1922-00a.

Millan, R.M., R. von Steiger, M. Ariel, S. Bartalev, M. Borgeaud, S. Campagnola, J.C. Castillo-Rogez, et al. 2019. Small satellites for space science: A COSPAR scientific roadmap. Advances in Space Research 64:1466. doi:10.1016/j.asr.2019.07.035.

Momcheva, I., and E. Tollerud. 2015. “Software Use in Astronomy: An Informal Survey.” https://ui.adsabs.harvard.edu/abs/2015arXiv150703989M.

Morley, S., J. Koller, D. Welling, B. Larsen, and J. Niehof. 2014. “SpacePy: Python-Based Tools for the Space Science Community.” http://ascl.net/1401.002.

Muna, D., M. Alexander, A. Allen, R. Ashley, D. Asmus, R. Azzollini, M. Bannister, et al. 2016. “The Astropy Problem.” https://arxiv.org/abs/1610.03159.

NASA OIG (Office of the Inspector General). 2019. NASA’s Heliophysics Portfolio. IG-19-018. May 7. NASA OIG, Office of Audits, Washington, DC. https://oig.nasa.gov/docs/IG-19-018.pdf.

NASA. 2010. “Announcement of Opportunity: Explorer 2011, NNH11ZDA002O.” Science Mission Directorate. https://nspires.nasaprs.com/external/viewrepositorydocument/cmdocumentid=247426/solicitationId=%7BBEB74D4A-7AC0-5590-C8C4-90E7C54B4425%7D/viewSolicitationDocument=1/EX2011 AO corrected2.pdf.

NASA. 2014. Heliophysics Science and Technology Roadmap for 2014-2033. NP-2014-12-226-GSFC. Heliophysics Roadmap Team. NASA Goddard Space Flight Center, Greenbelt, MD.

NASA. 2013. “Research Opportunities in Space and Earth Sciences 2013 (ROSES-2013). https://nspires.nasaprs.com/external.

NASA. 2018. “Research Opportunities in Space and Earth Sciences 2018 (ROSES-2018).” NNH18ZDA001N. http://solicitation.nasaprs.com/ROSES2018.

NASA. 2019. NASA Science and Technology Definition Team for the Geospace Dynamics Constellation: Final Report. Living With a Star Program, Heliophysics Division. October 4. https://science.nasa.gov/heliophysics/resources/stdts/geospace-dynamicsconstellation/documents-and-reports.

NASA NSPIRES. 2019. “Outer Heliosphere Guest Investigators,” Appendix B.13 in “Research Opportunities in Space and Earth Sciences 2019 (ROSES-2019).” https://nspires.nasaprs.com/external.

NASA NSPIRES. 2019. Heliophysics Technology and Instrument Development for Science (HTIDeS), Selected Proposals. https://tinyurl.com/yxcgsn5h.

NASEM (National Academies of Sciences, Engineering, and Medicine). 2016. Achieving Science with CubeSats: Thinking Inside the Box. The National Academies Press, Washington, DC. doi:10.17226/23503.

NASEM. 2018. Open Source Software Policy Options for NASA Earth and Space Sciences. The National Academies Press, Washington, DC. doi:10.17226/25217.

NASEM. 2019. Reproducibility and Replicability in Science. The National Academies Press, Washington, DC. doi:10.17226/25303.

NRC (National Research Council). 2013. Solar and Space Physics: A Science for a Technological Society. The National Academies Press, Washington, DC. doi:10.17226/13060.

NSB (National Science Board). 2018a. Bridging the Gap: Building a Sustained Approach to Mid-scale Research Infrastructure and Cyberinfrastructure at NSF. NSB-2018-40. National Science Foundation. https://www.nsf.gov/nsb/publications/2018/NSB2018-40-Midscale-Research-Infrastructure-Report-to-Congress-Oct2018.pdf.

NSB. 2018b. Study of Operations and Maintenance Costs for NSF Facilities. NSB-2018-17. https://www.nsf.gov/nsb/publications/2018/NSB-2018-17-Operations-and-Maintenance-Report-to-Congress.pdf.

NSF (National Science Foundation). 2016. Investments in Critical Capabilities for Geospace Science 2016 to 2025: A Portfolio Review of the Geospace Section of the Division of Atmospheric and Geospace Science. Submitted on February 5. https://www.nsf.gov/geo/adgeo/geospace-review/geospace-portfolio-review-final-rpt-2016.pdf.

NSF. 2019a. “Ideas Lab: Cross-cutting Initiative in CubeSat Innovations.” https://www.nsf.gov/funding/pgm summ.jsp?pims id=505605.

NSF. 2019b. “FY 2019 NSF Budget Request to Congress.” https://www.nsf.gov/about/budget/fy2019/pdf/40t fy2019.pdf.

NSF. 2019c. “Harnessing the Data Revolution (HDR) at NSF.” https://www.nsf.gov/cise/harnessingdata/.

Nishizuka, N., K. Sugiura, Y. Kubo, M. Den, S.-i. Watari, and M. Ishii. 2018. “Solar Flare Prediction Using Machine Learning with Multiwavelength Observations.” Pp. 310-313 in Space Weather of the Heliosphere: Processes and Forecasts (C. Foullon and O.E. Malandraki, eds.). doi:10.1017/S1743921317007293.

NumFocus Team. 2017. NumFocus Annual Report 2017. https://numfocus.org/wp-content/uploads/2018/03/numfocus-annual-report2017-WEB.pdf.

NumFocus Team. 2018. NumFocus Annual Report 2018. https://numfocus.org/wp-content/uploads/2019/04/NumFOCUS-AnnualReport-2018-FINAL.pdf.

Open Source Initiative. 2007. “The Open Source Definition.” https://opensource.org/osd.

OpenHub. 2019. “SunPy Estimated Cost.” https://www.openhub.net/p/sunpy/estimated_cost.

Panos, B., L. Kleint, C. Huwyler, S. Krucker, M. Melchior, D. Ullmann, and S. Voloshynovskiy. 2018. Identifying typical Mg_II flare spectra using machine learning. Astrophysical Journal 861:62. doi:10.3847/1538-4357/aac779.

Parente, P. 2019. “Estimate of Public Jupyter Notebooks on GitHub.” https://nbviewer.jupyter.org/github/parente/nbestimate/blob/master/estimate.ipynb.

Pedregosa, F., G. Varoquaux, A. Gramfort, V. Michel, B. Thirion, O. Grisel, M. Blondel, et al. 2011. Scikit-learn: Machine learning in Python. Journal of Machine Learning Research 12:2825. http://www.jmlr.org/papers/volume12/pedregosa11a/pedregosa11a.pdf.

Perez, F., and B.E. Granger. 2007. IPython: A system for interactive scientific computing. Computing in Science and Engineering 9:21. doi:10.1109/MCSE.2007.53.

Pesnell, W.D., B.J. Thompson, and P.C. Chamberlin. 2012. The Solar Dynamics Observatory (SDO). Solar Physics 275:3-15. doi:10.1007/s11207-011-9841-3.

PlasmaPy Community, N.A. Murphy, A.J. Leonard, et al. 2018. “PlasmaPy: An Open Source Community-Developed Python Package for Plasma Physics.” Zenodo. doi:10.5281/zenodo.1238132.

Radtke, J., C. Kebschull, and E. Stoll. 2017. Interactions of the space debris environment with mega constellations—Using the example of the OneWeb constellation. Acta Astronautica 131:55. doi:10.1016/j.actaastro.2016.11.021.

Rimmele, T.R. 2019. “Status of the Daniel K. Inouye Solar Telescope.” Abstract 226.01 in American Astronomical Society Meeting Abstracts, Vol. 234. Bulletin of the American Astronomical Society, Vol. 51, No. 4.

Russell, C.T., B.J. Anderson, W. Baumjohann, K.R. Bromund, D. Dearborn, D. Fischer, G. Le, et al. 2016. The Magnetospheric Multiscale magnetometers. Space Science Reviews 199:189. doi:10.1007/s11214-014-0057-3.

Schrijver, C., F. Bagenal, and J. Sojka, eds. 2016a. Heliophysics IV: Active Stars, their Astrospheres, and Impacts on Planetary Environments. Cambridge University Press. doi:10.1017/CBO9781316106778.

Schrijver, C., F. Bagenal, and J. Sojka, eds. 2016b. Heliophysics V: Space Weather and Society. Cambridge University Press. https://cpaess.ucar.edu/sites/default/files/heliophysics/documents/HSS5.pdf.

Schrijver, C., and G. Siscoe, eds. 2009. Heliophysics I: Plasma Physics of the Local Cosmos. Cambridge University Press. doi:10.1017/CBO9781107340657.

Schrijver, C., and G. Siscoe, eds. 2010a. Heliophysics II: Space Storms and Radiation: Causes and Effects. Cambridge University Press. doi:10.1017/CBO9781139194532.

Schrijver, C., and G. Siscoe, eds. 2010b. Heliophysics III: Evolving Solar Activity and the Climates of Space and Earth. Cambridge University Press. doi:10.1017/CBO9780511760358.

Seablom, M.S. 2017. “NASA 2016 Science Mission Directorate Technology Highlights.” https://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/20170003801.pdf.

Sloan Digital Sky Survey. 2018. https://www.sdss.org/surveys/.

SunPy Community, S.J. Mumford, S. Christe, D. Pérez-Suárez, J. Ireland, A.Y Shih, A.R Inglis, et al. 2015. SunPy—Python for solar physics. Computational Science and Discovery 8:014009. doi:10.1088/1749-4699/8/1/014009.

Talaat, E. 2019. “Accelerating Progress Toward NOAA’s Next Generation Architecture.” https://astronautical.org/dev/wp-content/uploads/2019/04/RHG_Thu_0800_Talaat.pdf.

van der Walt, S., S.C. Colbert, and G. Varoquaux. 2011. The NumPy Array: A structure for efficient numerical computation. Computing in Science and Engineering 13:22. doi:10.1109/MCSE.2011.37.

van der Walt, S., J.L. Schönberger, J. Nunez-Iglesias, F. Boulogne, J.D. Warner, N. Yager, E. Gouillart, T. Yu, and the scikit-image contributors. 2014. Scikit-image: Image processing in Python. PeerJ 2:e453. doi:10.7717/peerj.453.

VanderPlas, J. 2013. “The Big Data Brain Drain: Why Science Is in Trouble.” https://jakevdp.github.io/blog/2013/10/26/big-databrain-drain/.

VanderPlas, J. 2014. “Hacking Academia: Data Science and the University.” https://jakevdp.github.io/blog/2014/08/22/hackingacademia/.

VanderPlas, J., A.J. Connolly, Z. Ivezic, and A. Gray. 2012. “Introduction to astroML: Machine Learning for Astrophysics.” Pp. 47-54 in Proceedings of Conference on Intelligent Data Understanding (CIDU). doi:10.1109/CIDU.2012.6382200.

Walkowicz, L., A. Miller, and V. Kalogera. 2016. “The LSSTC Data Science Fellowship Program.” https://astrodatascience.org/.

Winter, L.M., and K. Ledbetter. 2015. Type II and type III radio bursts and their correlation with solar energetic proton events. Astrophysical Journal 809:105. doi:10.1088/0004-637X/809/1/105.

Woodger, L.A., A.J. Halford, R.M. Millan, M.P. McCarthy, D.M. Smith, G.S. Bowers, J.G. Sample, B.R. Anderson, and X. Liang. 2015. A summary of the BARREL campaigns: Technique for studying electron precipitation. Journal of Geophysical Research (Space Physics) 120:4922. doi:10.1002/2014JA020874.

Woods, T.N., A. Caspi, P.C. Chamberlin, A. Jones, R. Kohnert, J.P. Mason, C.S. Moore, et al. 2017. New solar irradiance measurements from the Miniature X-Ray Solar Spectrometer CubeSat. Astrophysical Journal 835:122. doi:10.3847/1538-4357/835/2/122.

Woolston, C. 2015. Graduate survey: Uncertain futures. Nature 526:597.

Zhelavskaya, I.S., M. Spasojevic, Y.Y. Shprits, and W.S. Kurth. 2016. Automated determination of electron density from electric field measurements on the Van Allen Probes spacecraft. Journal of Geophysical Research (Space Physics) 121:4611. doi:10.1002/2015JA022132.

Zooniverse. 2019. “Galaxy Zoo.” https://www.zooniverse.org/projects/zookeeper/galaxy-zoo/.