3
Modes of Support and Issues of Balance

In this chapter, each of the major modes of support employed by the National Science Foundation (NSF) that now contributes specifically to the atmospheric sciences—that is, grants to individual and multiple principal investigators (PIs), small centers, large national centers, cooperative agreements to support facilities at universities and other locations, NSF-wide initiatives, interagency programs, and field programs—is described and some preliminary analysis of their strengths and limitations are offered. The committee intends to provide a more detailed evaluation of the modes in its final report, in which the questions of which modes are best suited for meeting NSF’s Division of Atmospheric Sciences’ (ATM’s) goals and how to determine an effective balance among the modes will be addressed. In addition, the final report will explore the applicability of modes that are not currently employed by ATM.

GRANTS

ATM supports academic atmospheric research principally through the proposal and peer review process for individual or multiple investigator grants. Table 3-1 shows proposal statistics for ATM as compared to the Geosciences Directorate (GEO) as a whole and to the NSF averages. The bulk of the approximately 300 NSF-funded ATM grants each year are to individual PIs (in many cases with co-investigators), mostly at universities. The number of grants awarded each year has increased slowly over the past two decades (Figure 3-1), but there has been little trend over this time period in the success rate for grant proposals, which has fluctuated around 40 to 50 percent for the division (Jarvis Moyers,



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 39
Strategic Guidance for the National Science Foundation’s Support of the Atmospheric Sciences: An Interim Report 3 Modes of Support and Issues of Balance In this chapter, each of the major modes of support employed by the National Science Foundation (NSF) that now contributes specifically to the atmospheric sciences—that is, grants to individual and multiple principal investigators (PIs), small centers, large national centers, cooperative agreements to support facilities at universities and other locations, NSF-wide initiatives, interagency programs, and field programs—is described and some preliminary analysis of their strengths and limitations are offered. The committee intends to provide a more detailed evaluation of the modes in its final report, in which the questions of which modes are best suited for meeting NSF’s Division of Atmospheric Sciences’ (ATM’s) goals and how to determine an effective balance among the modes will be addressed. In addition, the final report will explore the applicability of modes that are not currently employed by ATM. GRANTS ATM supports academic atmospheric research principally through the proposal and peer review process for individual or multiple investigator grants. Table 3-1 shows proposal statistics for ATM as compared to the Geosciences Directorate (GEO) as a whole and to the NSF averages. The bulk of the approximately 300 NSF-funded ATM grants each year are to individual PIs (in many cases with co-investigators), mostly at universities. The number of grants awarded each year has increased slowly over the past two decades (Figure 3-1), but there has been little trend over this time period in the success rate for grant proposals, which has fluctuated around 40 to 50 percent for the division (Jarvis Moyers,

OCR for page 39
Strategic Guidance for the National Science Foundation’s Support of the Atmospheric Sciences: An Interim Report TABLE 3-1 ATM Research Proposal Statistics for FY 2003   ATM GEO NSF Submitted proposals ~800 ~4,000 ~40,000 Competitive awards ~300 ~1,500 ~11,000 Average annual award (in 1996 dollars) $127,000 ($108,300) $147,000 ($125,350) $136,000 ($116,000) Average duration 3 years 3 years 3 years FIGURE 3-1 Trends in average annual awards (in millions of FY 1996 dollars) and number of grants awarded by ATM since 1985. ATM, personal communication, July 22, 2005). Until recently, most grants were of three-year duration, but this has been changing slowly toward a larger number of four- and five-year grants. The average annual amount of ATM awards to PIs is about $127,000 per year, although actual support to an individual PI may be less if the grant is awarded to multiple investigators or more if allocations of computing or observing facilities are included in the award. For university faculty members, this amount normally includes up to two months of summer salary; support for graduate students, undergraduate students, or both; miscellaneous expenses such as travel, computing, and page charges; and institutionally determined fringe benefits and indirect costs. Over the past 10 years, 570 graduate students, on average, have been supported by ATM research grants each year, constituting a large percent-age of graduate students in atmospheric science departments. The funding is

OCR for page 39
Strategic Guidance for the National Science Foundation’s Support of the Atmospheric Sciences: An Interim Report committed for the duration of the grant, contingent on adequate progress being demonstrated though annual reports. Funding of investigators in nonacademic institutions proceeds similarly. Most grants are unsolicited; scientists with an idea for a research project send in a proposal which is then judged on the basis of scientific excellence and potential broader impacts, such as educational and other societal benefits. A small number of grants of limited scale and duration are awarded as part of the Small Grants for Exploratory Research (SGER) program, which is intended to promote investigation of more radical ideas. NSF and ATM also solicit proposals that address priority research areas or other specific objectives (e.g., Box 3-1). Often, these directed research programs respond to needs identified by the community, thereby alleviating the concern that investigators must shoehorn their proposals to meet research priorities that do not necessarily reflect community goals. This mechanism is used more prominently by the upper atmospheric section. There are several grant programs directed at young faculty and underrepresented groups. For example, the NSF-wide Faculty Early Career Development (CAREER) and the Presidential Early Career Awards for Scientists and Engineers (PECASE) grants target young, tenure-track faculty investigators who have not yet been awarded tenure. The number of these early career grant proposals is relatively small in ATM because of the relatively small number of tenure-track faculty in the field. GEO has grant programs that seek to enhance demographic diversity, including targeted programs for historically black colleges and universities, for tribal colleges and universities, and for improving female and minority representation. While NSF grants from ATM are important for private-sector research companies, they are crucial to the career of university faculty members. The more mission-oriented agencies (e.g., National Aeronautics and Space Administration [NASA], National Oceanic and Atmospheric Administration [NOAA], Department of Energy [DOE], Environmental Protection Agency [EPA], Department of Defense [DoD], and the Federal Aviation Administration [FAA]) support extramural research, but these funds are granted on the basis of mission relevance and scientific merit. Because NSF funding decisions are made primarily on the grounds of scientific excellence, there is a perception that success in obtaining NSF grants is considered more important to academic advancement. Small science and technology oriented businesses can also apply for Small Business Innovation Research (SBIR) and Small Business Technology Transfer (STTR) grants through an NSF-wide solicitation each year (NRC, 2004). STTR projects must involve at least one small business and one not-for-profit research group, usually from an academic institution. SBIR and STTR grants, which receive about 2.7 percent of the NSF’s extramural research budget, have funded the development and demonstration of a number of innovative instruments currently used in atmospheric research. An increasing fraction of NSF grants are for multiple PIs collaborating on a larger-scale project (see Figure 3-2). In particular, multi-PI grants support model-

OCR for page 39
Strategic Guidance for the National Science Foundation’s Support of the Atmospheric Sciences: An Interim Report BOX 3-1 Focused Programs That Are Community-Driven Ongoing Programs with an Annual Competition for Funding: Coupling, Energetics, and Dynamics of Atmospheric Regions (CEDAR) is a broad-based upper-atmospheric research program with the goal of understanding the behavior of atmospheric regions from the middle atmosphere upward through the thermosphere and ionosphere into the exosphere in terms of coupling, energetics, chemistry, and dynamics on regional and global scales. The Geospace Environment Modeling (GEM) program supports basic research into the dynamical and structural properties of the magnetosphere. One of the objectives is the construction of a global geospace general circulation model with predictive capability. Solar and Heliospheric Interaction (SHINE) research focuses on the connections between eruptive events and magnetic phenomena on the Sun and the corresponding solar wind structures in the inner heliosphere. The goal of SHINE research is to enhance both our physical understanding and predictive capabilities for solar-driven geoeffective events. Earth System History (ESH) is a cross-divisional research program, which is managed by ATM’s Paleoclimate Program Director. The program seeks to provide better understanding of Earth’s paleoenvironmental system and its evolution over geologic time by (a) documenting the past temporal and spatial variability of the Earth system, (b) assessing the rates of change associated with this variability, and (c) determining the sensitivity of the Earth system to variations in climate-forcing factors. The Geoscience Education program aims at initiating or encouraging innovative geoscience education activities. It specifically seeks projects that are informed by results of current education-related research or that conduct educational research with a geoscience education venue. The Opportunities for Enhancing Diversity in the Geosciences program supports activities that will increase the number of members of underrepresented groups that (a) are involved in formal precollege geoscience education programs; (b) pursue bachelor, master, and doctoral degrees in the geosciences; (c) enter geoscience careers; and (d) participate in informal geoscience education programs. Recent Solicitations for Proposals on Targeted Topics: The Pilot Climate Process and Modeling Teams (CPT) program was cosponsored by NOAA and NSF. The goal was to further the development of global coupled climate models by enhancing collaborations between theoreticians, field observationalists, process modelers, and the large modeling centers. The Water Cycle Research initiative was intended to enhance innovative basic research contributing to the understanding of the water cycle and its function as a transport agent for energy and mass (water and biologically/geochemically reactive substances).

OCR for page 39
Strategic Guidance for the National Science Foundation’s Support of the Atmospheric Sciences: An Interim Report FIGURE 3-2 Percent of grants (top panel) and funding (bottom panel) awarded to single PIs (white) and multiple PIs (grey). ing and measurement efforts. Atmospheric scientists have long recognized the value of collaboration (NAS/NRC, 1958) and are increasingly seeing the need to form teams that can access the multiple skills, tools, and facilities that are frequently required to plow new scientific ground. The demand on ATM for multi-investigator project funding is likely to continue to grow. An issue that arises as the scale grows is the ability for agencies to work together, and for agencies to coordinate with international partners, in the fostering and support of such programs.

OCR for page 39
Strategic Guidance for the National Science Foundation’s Support of the Atmospheric Sciences: An Interim Report Increasingly, advances in modeling capabilities rest on critical collaborations and shared infrastructure. Likewise, the increasing complexity and frequent multidisciplinary nature of atmospheric science measurements—including laboratory experiments, ground-based and airborne field measurements, and advanced research instrument development and testing—often require collaboration of two or more research groups to be addressed effectively. Atmospheric field measurements often need to be performed at one or more remote sites, may require complex logistics involving site access or mobile measurement platforms, usually require the simultaneous measurement of multiple physical and/or chemical parameters, and normally require significant modeling capabilities for proper analysis. All of these factors push the requirement for multiple-PI projects. There is a synergy between ATM PI grants and National Center for Atmospheric Research (NCAR) programs for both individual and multiple PIs. Many NSF grantees use research tools developed and maintained at NCAR. These include numerical models, equipment, and computing. Also, there is a great deal of science collaboration between NCAR scientists, who are frequently unfunded co-PIs on grants, and PIs from universities or the private sector in the conduct of their research, including field programs. This mode of core grant support has benefited the atmospheric sciences in several ways. First, it has enabled lots of good science. For example, grants to individual and multiple PIs have enabled the development of theory, analysis of observation and model results, process studies, provision of data to a broad suite of users, and development and acquisition of instruments by universities. Second, it has provided multiple options and flexibility in the ways ATM supports PIs, including unsolicited proposals, solicitation for new money that came in via various NSF-wide initiatives, ATM-initiated solicitations, and solicitations for field programs. This flexibility allows ATM to both encourage submission of proposals around specific themes and to encourage good ideas for proposals to be submitted at any time. The NSF approach to reviewing and selecting research activities to support generally ensures that good science is funded and poor or mediocre science is not. A challenge to this approach is making sure to fund some science that is particularly innovative, high risk, and may have large potential payoffs. Such research efforts are more likely to fail, but also may lead to transformative discoveries. Whereas other federal agencies typically fund research directly related to their mission, NSF is the primary place where scientists turn for support of research that has no obvious applications or even a guarantee of success. Identifying proposals that fall into this category and ensuring adequate support for them has presented challenges for NSF as a whole, despite encouragement from NSF leadership to pursue innovation and risk taking (NAPA, 2004). Aside from those grants awarded through the SGER program, most proposals that might be considered high risk undergo the regular merit review process; thus it is unknown how much research of this sort is supported. Furthermore, because peer reviewers tend

OCR for page 39
Strategic Guidance for the National Science Foundation’s Support of the Atmospheric Sciences: An Interim Report to be risk averse, particularly innovative proposals may not fare well when competing against regular proposals. NAPA (2004) found that NSF’s support for high-risk research could be enhanced by better communicating opportunities for such support to the scientific community, perhaps through specialized calls for proposals; by modifying the review criteria used to evaluate proposals to place more weight on innovation; or by subjecting high-risk proposals to a specially designated review process. Currently, ATM does not set aside any funds specifically for high-risk research, but program officers are encouraged to be receptive to such proposals that come in through the regular grant process. In some cases, awards are made despite the lack of reviewer endorsement, shorter-duration proof-of-concept awards are made, or ATM or GEO reserves are used to fund such activities. One example of such an action by an ATM program director took place in the early 1980s when Dr. Ronald Taylor put funding into the newly emerging area of the MST (Mesosphere-Stratosphere-Troposphere) radar. This action accelerated progress in this field so that we now have many such radars around the world collecting valuable data. ATM does not track how many grants are awarded for high-risk proposals, either through the regular grant process or through the discretion of the program directors, or the outcomes of the high-risk research that is funded. Some high-risk projects that are of limited duration and of modest cost are supported through the SGER program. No more than 5 percent of any NSF program can be used for SGER awards; in ATM, typically 1 to 2 percent of each program’s funds are applied to SGER. It is not entirely clear to investigators what funding mechanisms are available for support of high-risk projects that are larger in scope than that which an individual program director could fund. It is essential to preserve opportunities for high-risk, potentially transformative research. Finding: Among federal science agencies, NSF is a leader in its commitment to support high-risk, potentially transformative research (excluding satellite instrument development). This type of research is instrumental in making major advances in the field, as well as in sustaining the nation’s economic development and global competitiveness. Currently, program directors have discretion to use 5 percent of their budgets for SGER projects, though typically about 1 to 2 percent of each program’s funds are applied this way. In addition, program directors can choose to support other high-risk work through regular grant mechanisms as they see fit. However, it is unknown to what extent this flexibility to support exploratory research is utilized. Furthermore, there may be some research questions of this type that require a bigger investment than what typically can be made by a program director. One option to be more effective is to pool some of the funding for exploratory research from all ATM programs and run an internal competition to which program directors can submit promising, high-risk ideas for consideration.

OCR for page 39
Strategic Guidance for the National Science Foundation’s Support of the Atmospheric Sciences: An Interim Report Recommendation: ATM should support high-risk, potentially transformative research at the current rate or a greater one, seeking new mechanisms to enhance opportunities for investigators, such as pooling some of the existing funding. The success of this effort should be evaluated every five years. SMALL CENTERS Over the past two decades, NSF has begun to employ a small-center mode of funding. This mode was initiated by the Engineering Directorate, which introduced Engineering Research Centers (ERCs) in the early 1980s. Subsequently, the Office of Integrated Activities created Science and Technology Centers (STCs), which are designed to enable innovative research and education projects of national importance that require a center to achieve significant research, education, and knowledge-transfer goals shared by the partners. ERCs and STCs are funded at the level of $2 million to $5 million per year. In addition, there are centers supported under the NSF-wide Information Technology Research (ITR) program and ATM supports some centers from core funds. Box 3-2 lists atmospheric science centers established over the past 15 years along with the science problems they are addressing. Because these centers are supported primarily by other parts of NSF, they provide an opportunity to expand the overall NSF level of support for atmospheric sciences. The NSF Office of Integrative Activities supports 11 STCs. Competition for the class of 2005 is underway, with six to eight proposals to be selected. Two atmospheric-sciences-related STCs were awarded in the early years: the Center for the Analysis and Prediction of Storms (CAPS) housed at the University of Oklahoma and the Center for Clouds, Chemistry and Climate (C4) at Scripps Institution of Oceanography. Although CAPS and C4 have been sunsetted as STCs, support for the research initiated at these centers has continued because of successful competition for ATM core funding. The Division of Engineering Education and Centers supports 23 current ERCs, with 4 new centers expected to be funded in 2006. There have been a total of 41 centers since the program started in 1985, and the last competition for new centers was in 2003, with 3 funded. Currently, ATM is represented by one STC, the Center for Integrated Space Weather Modeling (CISM), and one ERC, the Center for Collaborative Adaptive Sensing of the Atmosphere (CASA). The STC and ERC programs provide participating investigators with long-term, relatively stable funding of sufficient size to tackle large problems. They involve the creation of large, interdisciplinary research efforts with targeted goals. Such a goal-oriented research focus, with milestones and metrics, is a different environment than the work of the individual PI. Stable funding benefits graduate students and postdoctoral fellows, and allows researchers to focus on key science issues that extend beyond the regular grant cycle for single and multiple PIs. With a recent trend of three to five years for grants to individual investigators, this advantage of centers may become less important.

OCR for page 39
Strategic Guidance for the National Science Foundation’s Support of the Atmospheric Sciences: An Interim Report BOX 3-2 Small Atmospheric Centers Supported by NSF Center for Analysis and Prediction of Storms (CAPS) was an STC at the University of Oklahoma from 1989 to 2000, funded at a rate of $0.9 million to $1.5 million per year. The CAPS mission was the development of techniques for the computer-based prediction of high-impact local weather with operational Doppler radars serving as key data sources. Center for Clouds, Chemistry and Climate (C4) was an STC spearheaded by Scripps Institution of Oceanography from 1991 to 2001, funded at a rate of $1.5 million per year. The goal of C4 was to develop theoretical, observational, and modeling bases required to understand and predict Earth’s changing climate as affected by clouds, radiation, and atmospheric chemistry and their interactions. Center for Integrated Space Weather Modeling (CISM) is an STC coordinated by Boston University, starting in 2002, funded at a rate of $4 million per year for up to 10 years. CISM consists of research groups at eight universities and several government and private nonprofit research organizations and commercial firms. The center’s mandate is to construct a comprehensive physics-based numerical simulation model that describes the space environment from the Sun to the Earth, thus enabling reliable prediction of space weather events at least two days in advance. Center for Collaborative Adaptive Sensing of the Atmosphere (CASA) is an ERC led by the University of Massachusetts-Amherst, funded at a rate of $1.5 million to $2 million per year for up to 10 years. Established in late 2003, the center brings together a multidisciplinary group of engineers, computer scientists, meteorologists, sociologists, and industry and government representatives to conduct fundamental research, develop enabling technology, and deploy prototype engineering systems based on a new paradigm: distributed collaborative adaptive sensing. These networks are deployed to overcome fundamental limitations of current tropospheric observational approaches by using large numbers of appropriately spaced sensors capable of high spatial and temporal resolution. Linked Environments for Atmospheric Discovery (LEAD) is an ITR program led by the University of Oklahoma and established in 2003. It is funded at a rate of $11.25 million for five years. The transforming element of LEAD is dynamic workflow orchestration and data management, which will allow use of analysis tools, forecast models, and data repositories as dynamically adaptive, on-demand systems. Global Multi-Scale Kinetic Simulations of the Earth’s Magnetosphere Using Parallel Discrete Event Simulation is an ITR project at the Georgia Institute of Technology to develop scalable, parallel, numerical models for the simulation of space plasmas and the dynamics of the Earth’s magnetosphere, based on discrete event simulation (DES). The investigators will develop DES methods with situation-dependent physics, suitable for space physics problems, and then develop the algorithms required to execute these efficiently on massively parallel computer systems. Tree Ring Reconstruction of Asian Monsoon Climate Dynamics is a new five-year collaborative project at Columbia University. The project will use the science of dendrochronology to examine the relationship between the Asian monsoon and the large-scale coupled processes that drive much of its variability.

OCR for page 39
Strategic Guidance for the National Science Foundation’s Support of the Atmospheric Sciences: An Interim Report In addition to their research objectives, STCs and ERCs have mandates to conduct education activities and to develop applications and knowledge transfer. The STCs and ERCs are required to spend approximately 20 percent of their resources on education and diversity programs, well beyond the requirements of other grants and agency requirements. Thus, the centers significantly broaden education resources. For example, CISM holds a two-week summer school that provides a broad-based exposure to space weather in the entire Sun-Earth system, which has proved to be very successful (Simpson, 2004). ERCs are specifically mandated to include minority-serving institutions in the team. STCs and ERCs also have to devote considerable resources to knowledge transfer—making the products of the research useful to users in the real world. For ATM, this has meant moving atmospheric or space weather predictive capability from research into operations (NRC, 2000). LARGE NATIONAL CENTER One of the mechanisms used by NSF for support of research is a large national center. Typically designated as federally funded research and development centers (FFRDCs), they provide for a larger aggregation of research capability than that which could ordinarily be expected to occur at an individual university department. The largest of NSF’s FFRDC is NCAR, located in Boulder, Colorado. The University Corporation for Atmospheric Research (UCAR), a nonprofit consortium of 68 North American universities with graduate programs in atmospheric sciences, has managed NCAR since its founding in 1960 through a cooperative agreement with ATM. This structure was designed to foster interactions and joint management between NCAR and the university community. The specific objectives for NCAR were laid out in the 1959 “Blue Book” authored by the University Committee on Atmospheric Research (“UCAR”; see Box 3-3). The critical mass of resources that NCAR brings to bear on the atmospheric sciences includes computational resources, aircraft resources, observational capabilities, laboratories, and machine shops. An additional objective was to provide a personnel base that could support large-scale research, including interdisciplinary research. The center would have sufficient support personnel to enhance the research environment. The initial planning for NCAR called for half the staff to be from the atmospheric sciences with the remainder being from disciplines such as physics, mathematics, chemistry, and engineering. This disciplinary composition has evolved since 1959 as demanded by new research avenues in the atmospheric sciences. Today, NCAR has about 220 scientists, 100 associate scientists, and 620 support personnel (which encompasses everything from software engineers to administrative assistants) who conduct research in the atmospheric and ocean sciences and in solar and space physics, and participate in a suite of activities that support the broad community. As shown in Box 3-4, NCAR and its scientists

OCR for page 39
Strategic Guidance for the National Science Foundation’s Support of the Atmospheric Sciences: An Interim Report BOX 3-3 Four Compelling Reasons for Establishing a National Institute for Atmospheric Research Identified in the “Blue Book” (“UCAR,” 1959): The need to mount an attack on the fundamental atmospheric problems on a scale commensurate with their global nature and importance. The fact that the extent of such an attack requires facilities and technological assistance beyond those that can properly be made available at individual universities. The fact that the difficulties of the problems are such that they require the best talents from various disciplines to be applied to them in a coordinated fashion, on a scale not feasible in a university department. The fact that such an Institute offers the possibility of preserving the natural alliance of research and education without unbalancing the university programs. BOX 3-4 Overview of NCAR Organization, Activities, and Facilities NCAR Organization: Computational Information and Systems Laboratory, which houses the Institute for Mathematical Applications in the Geosciences and the Scientific Computing Division. Earth Observing Laboratory (EOL), which includes the Atmospheric Technology Division (ATD) and the High-performance Instrumented Airborne Platform for Environmental Research (HIAPER). EOL maintains and deploys observational facilities for the lower-atmosphere research community. Earth and Sun Systems Laboratory (ESSL), which houses much of NCAR’s scientific research as well as its community models. ESSL includes: Atmospheric Chemistry Division Climate and Global Dynamics Division High Altitude Observatory (HAO) Mesoscale and Microscale Meteorology Division The Institute for Multidisciplinary Earth Studies The NCAR library Research Applications Laboratory, which includes the Research Applications Programs, is involved in a spectrum of activities relating to technology transfer and application of new knowledge to practical use. Societal and Environmental Research and Education Laboratory, including the Advanced Study Program, which offers postdoctoral positions that enable

OCR for page 39
Strategic Guidance for the National Science Foundation’s Support of the Atmospheric Sciences: An Interim Report and Climate (COSMIC). AMISR has just been initiated; the grant for building it was awarded to SRI International. COSMIC is being operated by UCAR. A number of issues arise in making choices about which observing facilities to support and how to implement them. One must consider the balance in needs for observational platforms across the disciplines (i.e., climate, mesoscale convection, space weather, etc.) and in needs for different types of platforms (i.e., aircraft, radars, etc.). It is not clear whether the appropriate distribution should be determined by the number of researchers, the cost, or other priorities. Another dimension of balance to consider is the extent to which small or large centers, universities, or private-sector entities should support development and maintenance of observational platforms. Similarly, NSF must determine an appropriate balance for maintaining and keeping existing facilities up to date, retiring facilities as appropriate, and developing new facilities. Since some facilities are very expensive to operate and maintain, it is important that NSF frequently and carefully continue to determine which facilities are essential for research and which facilities might best not be supported any more. How best to utilize partnerships of NSF with other agencies that support observational facilities is another area of consideration. NSF has collaborated with other agencies to develop observing facilities, as it is currently doing in the case of COSMIC, and to deploy observing facilities for large field programs, such as the Indian Ocean Experiment (INDOEX) campaign. There may be further opportunities to build such collaborations. How ATM’s support for facilities should evolve will be considered in more detail in the committee’s final report. NSF-WIDE INITIATIVES ATM participates in a number of NSF-wide, interagency, and international programs, which in some cases require different approaches to providing support. The NSF-wide emphasis areas result from national initiatives spearheaded by Congress or the President, or else are activities such as the STCs that NSF leadership chooses as a priority. They can bring new funds into the Foundation, which are then distributed to relevant divisions. Since 2000, ATM has received additional funds toward five NSF priority areas, as well as from the STC and ERC programs described previously (Table 3-3). Typically, these funds are distributed as grants to individual and multiple PIs who respond to specialized calls for proposals. INTERAGENCY PROGRAMS Several government agencies support extramural research in the atmospheric sciences—including NASA, NOAA, EPA, DOE, DoD, and FAA—in part because atmospheric science is directly relevant to the missions of these agencies. Effective coordination of ATM with other agencies is important for meeting ATM’s

OCR for page 39
Strategic Guidance for the National Science Foundation’s Support of the Atmospheric Sciences: An Interim Report TABLE 3-3 Investments in ATM Research from NSF-wide Priority Areas (millions of dollars for each fiscal year) Priority Area 2000 2001 2002 2003 2004 Biocomplexity in the Environment: improve environmental forecasting capabilities; enhance decision-making tools; and integrate human, social, and ecological factors into investigations of the physical environment and environmental engineering. 0.00 7.50 7.40 7.40 12.00 Information Technology Research: deepen fundamental research on large-scale networks and create new integrative software and advanced architectures for high-end computing. 0.00 3.40 3.40 4.60 5.00 Nanoscale Science and Engineering: develop and strengthen promising fields (including nanobiotechnology, manufacturing at the nanoscale) and establish the science and engineering infrastructure and workforce needed to exploit new capabilities in systematic organization, manipulation, and control of matter at atomic, molecular, and supramolecular levels. NSF activities are part of the larger, cross-agency National Nanotechnology Initiative. 0.00 0.00 0.50 0.50 0.60 Mathematical Sciences: deepen support for fundamental research in the mathematical sciences and statistics and integrate mathematical and statistical research and education across the full range of science and engineering disciplines. 0.00 0.00 0.00 1.50 2.40 Human and Social Dynamics: draw on recent convergence of research in biology, engineering, information technology, and cognitive science to investigate the causes and ramifications of change and its complex consequences—cultural, economic, individual, political, and social. 0.00 0.00 0.00 0.00 0.50 goals for several reasons. First, many essential resources for atmospheric sciences research are created and supported by other agencies. These include space-based observational platforms, long-term monitoring efforts, and data archiving. Pooling resources supported by multiple agencies is an important component of many field programs. Second, whereas NSF’s funding has remained fairly stable in recent decades, these other agencies have had more volatility. Thus, scientists supported by the other agencies turn to NSF for support when those agencies have downswings in funding, placing a larger demand on the NSF support for the atmospheric sciences. Third, because ATM is the one source for federal funding

OCR for page 39
Strategic Guidance for the National Science Foundation’s Support of the Atmospheric Sciences: An Interim Report that aspires to address research needs spanning all of atmospheric science, the division has additional responsibility to consider supporting critical areas of the science not addressed by other agencies for programmatic reasons. ATM participates in three major interagency programs that include atmospheric components (see Box 3-6): the U.S. Climate Change Science Program (CCSP), the U.S. Weather Research Program (USWRP), and the National Space Weather Program (NSWP). In addition, ATM supports the Center for Ocean, Land, and Atmosphere (COLA), a not-for-profit research institution in Calverton, Maryland, with interagency support that has some of the characteristics of the small centers discussed earlier (see Box 3-6). The division contributes to these efforts by supporting scientists who are doing research on related topics and in some cases providing funds for central coordination of the programs. ATM’s involvement in the CCSP, USWRP, NSWP, and COLA commits the division to ongoing support of research that addresses the goals of these programs. A possible concern has been that these targeted initiatives would constrain the community to follow certain lines of inquiry, possibly channeling emphasis away from other important research areas. However, this has not proved to be the case in the initiatives listed in Box 3-6. In fact, these initiatives have all brought new funds into ATM, thus supporting more investigators and resulting in excellent science. Many of these funds have been distributed through PI grants, and significant funds within CCSP have gone to NCAR, helping to support climate system modeling. Interagency activities in operational meteorology and supporting research have been coordinated by the federally mandated Office of the Federal Coordinator for Meteorology (OFCM) since 1964. Fifteen federal departments and agencies currently participate in OFCM’s coordination infrastructure, which includes program councils, committees, working groups, and joint action groups staffed and populated by representatives from the federal agencies. OFCM focuses on coordinating operational weather observing and forecasting requirements. In addition, it produces annual reports on federal investments in weather-related activities and research and, as needed, holds workshops and produces reports on specific issues. Like the other interagency coordination efforts, OFCM has had varied effectiveness over its tenure. Interagency coordination is a longstanding challenge for federally funded research in the atmospheric sciences, as recognized in many previous reports (e.g., NRC, 1997b, 1998, 2003), and requires the commitment of other agencies along with NSF. Yet it is essential to ensure that the critical science issues identified by the programs in Box 3-6, as well as other issues that require interagency coordination, are adequately addressed. Over the past decades, there have been mixed levels of success in these programs and in other efforts at interagency coordination, such as the Committee on Environment and Natural Resources, Subcommittee on Air Quality Research. The success depends in part on the leadership of each program, the willingness of the participating agencies to work

OCR for page 39
Strategic Guidance for the National Science Foundation’s Support of the Atmospheric Sciences: An Interim Report BOX 3-6 Major Interagency Programs The U.S. Climate Change Science Program (CCSP) is an interagency effort to better understand how climate, climate variability, and potential human-induced changes in climate, affect the environment, natural resources, infrastructure, and the economy in our nation and the world. The guiding vision for CCSP is “a nation and the global community empowered with the science-based knowledge to manage the risks and opportunities of change in the climate and related environmental systems.” The U.S. Weather Research Program (USWRP) has the goal of improving the delivery and use of weather information. NSF’s role is to provide leadership and support for all aspects of the fundamental science components—experimental, theoretical, and numerical. The current three priority thrust areas are quantitative precipitation forecasting and estimation, hurricane landfall, optimal mix of observing systems. The overarching goal of the National Space Weather Program (NSWP) is to achieve an active, synergistic, interagency system to provide timely, accurate, and reliable space environment observations, specifications, and forecasts. The program includes contributions from the user community, operational forecasters, researchers, modelers, and experts in instruments, communications, and data processing and analysis. It is a partnership between NSF, NASA, DoD, NOAA, DOE, the Department of the Interior, academia, and industry. NSF provides support to advance state-of-the-art instruments and data gathering techniques, to understand the physical processes, to develop predictive models, and to perform detailed analysis of data associated with past events that have caused significant impacts to space systems. The Center for Ocean, Land, and Atmosphere (COLA) is devoted to understanding the predictability of Earth’s current climate fluctuations on seasonal to decadal timescales using state-of-the-art, comprehensive models of the global atmosphere, world oceans, and land surface. COLA activities include (a) independently evaluating the climate variability characteristics of the nation’s climate change models, (b) providing leadership on prediction of climate variability on seasonal-to-interannual timescales, (c) characterizing the impact of long-term climate change on climate variability, and (d) providing information technology infrastructure for efficient exchange of climate model and observational data. COLA is supported by NSF, NOAA, and NASA. toward mutual objectives, and the extent to which opportunities for coordination are clearly communicated to the research community. Typically, these interagency programs have not asserted control over the budgets of individual agencies, but instead facilitate coordination by defining shared research agendas to which each agency contributes.

OCR for page 39
Strategic Guidance for the National Science Foundation’s Support of the Atmospheric Sciences: An Interim Report ATM is to be commended for its participation in the large interagency efforts described in Box 3-6. Furthermore, ATM program directors have been proactive about working with their colleagues from other agencies to support cross-agency research efforts, in particular, field programs (see Table 2-2). The committee is concerned, however, that ATM does not appear to have a strategic approach to its interagency activities. Thus, it is not clear to the research community exactly how ATM intends to contribute to large interagency programs, and interactions between program directors from NSF and other agencies appear to have an ad hoc nature. A more strategic approach is especially important for addressing large research problems that span the research investments of multiple agencies, such as climate or air quality, and for research avenues that have significant potential applications for operational capabilities, such as weather, for which coordination with mission-oriented agencies such as the National Weather Service is critical. A more strategic approach is needed to facilitate interagency coordination. Finding: Despite compelling motivations for interagency coordination, ATM does not always have clear mechanisms for effectively facilitating such interactions. Some interagency coordination takes place through formalized interagency programs (e.g., CCSP, NSWP), interagency working groups, community-driven initiatives (e.g., Climate Variability and Change [CLIVAR]), and ad hoc interactions between program directors. A strategic plan would both increase the transparency and decrease the ad hoc nature of NSF’s approach to these interagency collaborations. Another way to address this problem would be to facilitate the establishment of an interagency Federal Coordinator for Atmospheric Research. This individual would be supported by all relevant agencies, with duties and responsibilities similar to the role of OFCM, but with a focus on sustaining the overall health of basic research in atmospheric science by maintaining liaisons with all relevant agencies and identifying their contributions to atmospheric research. Other options for fostering interagency coordination could also be effective. Recommendation: ATM should be even more proactive in developing clear mechanisms for interagency collaborations. FIELD PROGRAMS Taking observations of the atmosphere in organized field programs to study specific processes continues to be integral to atmospheric research. Major field programs supported by ATM during the past decade are described in Table 2-2. Field programs are supported through a combination of modes, usually including grants to individuals or groups, NCAR, or university facilities, and often involve other agencies or countries. ATM supports smaller field programs through individual investigator grants and the facilities deployment pool. However, ATM

OCR for page 39
Strategic Guidance for the National Science Foundation’s Support of the Atmospheric Sciences: An Interim Report supports large field programs in a variety of ways: as the lead agency (e.g., Bow Echo and MCV Experiment [BAMEX], IHOP_2002), as a major partner in an international effort (e.g., Tropical Ocean and Global Atmosphere Coupled Ocean Atmosphere Response Experiment [TOGA COARE]), as a supporting agency for field programs sponsored by other agencies (e.g., Boreal Ecosystem-Atmosphere Study [BOREAS], led by NASA), and, on occasion, supplying NSF facilities for which other agencies pay. Individual NSF-funded PIs can also participate in field campaigns sponsored by other agencies through individual grants. ATM indirectly supports field programs by supporting investigators to develop research capabilities that are then employed in campaigns funded by other agencies. In the case of INDOEX, the C4 STC was instrumental in initiating and carrying out the field program. To facilitate the planning of field programs, ATM requires those interested in using facilities from the NSF deployment pool to submit requests as much as two years in advance. The procedures for reviewing field programs were updated in February 2005 (NSF, 2005). As the atmospheric sciences have become more complex, conducting field programs has presented new challenges for ATM in determining how to support these efforts, including: Increased demand for facilities. Particularly for the large and diverse lower-atmosphere community, there is significant demand for facilities that often leads to conflicts in scheduling. Carefully developed protocols for requesting facilities years in advance, negotiation with NSF program officers and facility providers, and input from the Observing Facility Advisory Panel have often, but not always, resolved conflicts. The problem is exacerbated by the fact that scheduling is often driven by probable weather and the scheduling of other facilities belonging to other agencies and countries (e.g., University-National Oceanographic Laboratory System) or the schedules of cooperating institutions. Data archiving and development of data analysis tools. Currently, there are varied destinations for data archival, including NCAR, Web sites set up by universities, and data archives established by other government agencies (e.g., National Climatic Data Center). For lower-atmospheric field campaigns back to the early 1990s, UCAR/JOSS has served as a center for data archiving for observational data and model simulations, or as a clearinghouse for PI-supported datasets archived elsewhere. Likewise, HAO maintains data archives from its solar instruments. Other government agencies, such as NASA, NOAA, and DOE, also have made efforts to establish data archives for data from field programs, satellite instruments, and monitoring networks. However, it is becoming increasingly difficult to access older observational or model datasets: changing technology and analysis packages make these datasets more difficult to analyze, and supporting metadata are often absent for the historical datasets. There is not always a clear responsibility for providing archived data for researchers for both large, multi-investigator field experiments and small field experiments. Thus,

OCR for page 39
Strategic Guidance for the National Science Foundation’s Support of the Atmospheric Sciences: An Interim Report data archival formats, quality control, and metadata are not necessarily standardized. Supporting data analysis. Inadequate time and resources for analysis of data collected in the field has been a problem for decades. LeMone (1983) reported that it took six years to reach the peak in publications from GATE data. There was a time lag of five to six years between Cooperative Convective Precipitation Experiment (1981) and the peak in resulting publications, and the peak in Free Atmosphere Carbon Experiment publications was four years after the experiment. Some scientists analyzing TOGA COARE data ran out of funding before they completed analysis and publication; some even ran out of funding before they obtained all their data. A 2- to 10-year post-analysis phase is recognized in the lifetime of a generic large NSF field program, discussed in the recently released document, “Field Program Support at UCAR” (available at http://www.ucar.edu/fps/fps.pdf). NSF’s new procedures for reviewing field programs (NSF, 2005) emphasize advance notice more than the post-field phase. Providing adequate time for careful analysis and synthesis of field data, which today typically involves complementary numerical simulations, increases the probability of significant payoff. Although the recent trend toward five-year grants allows more time for data analysis, the closer spacing of field programs exacerbates this problem. Spacing of field programs. In addition to increased demand for facilities and the need for additional time and money for data analysis and synthesis, there are other factors that must be considered. The large infrastructure maintained to operate the facilities requires a certain level of use, not only to justify its existence, but to test instruments and maintain proficiency of the personnel, a requirement for airplane pilots. Furthermore, field programs are effective ways to inspire and recruit new students and to stimulate new questions. A need for longer-term sustained intensive measurements. While ATM has a distinguished record in supporting long-term measurements of the upper atmosphere (Table 3-2) and the Sun (Box 3-4), current ATM policies and procedures for lower-atmosphere field programs are consistent with instrument deployments on the order of a few months. However, many problems related to weather and climate—for example, the interaction between the atmosphere and Earth’s surface in the context of heat, moisture, or biogeochemical cycles—require sustained, specially designed, and focused measurements for a complete annual cycle or even several years. There are examples where ATM supported longer-term measurement goals by supporting field programs on an episodic basis (e.g., First ISLSCP [International Satellite Land Surface Climatology Project] Field Experiment in the 1980s), but sustained measurements are often needed. There are also efforts within other divisions of NSF to develop capabilities for long-term observations over the ocean (e.g., Ocean Research Interactive Observatory Networks Ocean Observing Initiative [ORION OOI]) and the land surface (e.g., Consortium of Universities for the Advancement of Hydrologic Science, Inc.

OCR for page 39
Strategic Guidance for the National Science Foundation’s Support of the Atmospheric Sciences: An Interim Report [CUASHI] Hydrological Observatories). Operational weather- and climate-monitoring networks provide observations over the longer term, but often not at the intensive level needed for process studies. Adapting to a changing international scene. Historically, the United States usually has been the leader or at least a major partner in international field efforts. In the past few years, however, the major leadership in field programs has started to come from other nations. For example, the African Monsoon Multiscale Analysis (AMMA) field program is a large international field program supported by the European Union and led by France. Development of innovative observing techniques and methods. For the U.S. atmospheric science community to remain at the cutting edge of field research, innovative techniques and methods need to be developed in order to obtain the observations needed to test hypotheses, better resolve the variability and structure of the atmosphere, and understand the coupling of the atmosphere to the land, ocean, and space. Once developed and proven, these new methods need to be transferred to facilities that can make them available to the broader community. Longer-term field programs have not received sufficient support. Finding: ATM has well-established mechanisms for supporting short-duration field programs. However, ATM has not yet clearly articulated mechanisms for supporting field programs that require continuous, longer-term (i.e., up to multiyear) deployment and observations not available from operational monitoring networks. This type of observation protocol is generally ill-suited to the existing funding opportunities, in part because they were prohibitively expensive until recently. Three factors motivate the need and appropriateness of this approach today: (1) these types of observations are especially critical to understanding the interaction between the atmosphere and Earth’s surface, which are growing areas of research and concern; (2) many instruments that would be used are less expensive, making it reasonable to deploy them in the field for longer durations; and (3) there are existing observational programs developed by other NSF divisions and agencies (e.g., Long Term Ecological Research, ORION OOI, the proposed CUASHI Hydrological Observatories), which can be leveraged with additional investments to conduct atmospheric research. Recommendation: ATM, in coordination with other NSF divisions and federal agencies, should develop the explicit capability to support longer-term (i.e., up to multiyear) lower-atmosphere field programs to study atmospheric processes that are important on these timescales. Support for field data archives, visualization tool development, and analysis is not commensurate with the investment in obtaining the measurements. Finding: A longstanding challenge in the atmospheric sciences is providing sufficient support for scientists to analyze data obtained during field programs and from observational networks. Because analysis comes at the end of a field

OCR for page 39
Strategic Guidance for the National Science Foundation’s Support of the Atmospheric Sciences: An Interim Report program and competes against the start of other new field programs, it is at times subject to reduction in support. Thus, the full benefit from the investment in a field program often is not realized. Maximum benefit from many NSF-supported studies also would be facilitated by easy access to data from operational observational and monitoring networks (including surface, upper air, radar, and satellite) in addition to easy access to field-program data, historical data, and numerical model data. All these datasets should be archived and provided to the community in a manner consistent with common standards, along with the necessary analysis and visualization tools. In enhancing these capabilities, there are opportunities for NSF to work with other federal agencies who have faced similar challenges, particularly in terms of data archiving. Recommendation: ATM should maximize the benefit of field data by ensuring that archiving, visualization, and analysis activities are well supported and continue for many years after the completion of field campaigns. ATM is encouraged to work with the community by sponsoring a series of workshops on development of standards for metadata, data archival, and software tools and by providing support for the implementation of the recommendations of the workshops. ENSURING A BALANCED PROGRAM The committee’s preliminary analysis of the modes of support employed by ATM leads to the conclusion that each of the modes is serving an important function. In particular, the complementary roles of a large national center and grants to PIs have been a constructive component of the atmospheric science enterprise. The diversity of available modes has facilitated several different ways to tackle the scientific questions in the atmospheric sciences. Indeed, it appears that many of the newer modes arose out of emerging needs of the research community. The current balance among the modes is serving the community well, but may need to shift in coming years to respond to a changing research environment. For example, domestic budget constraints at NSF and other federal agencies that support atmospheric research, increasing sophistication and investments in the international research community, and changing societal expectations of research may make it necessary to rely more on some modes of support or to introduce new modes to the ATM portfolio. In its final report, the committee will consider the extent to which the balance should be modified. Having diverse modes of support available has benefited the atmospheric sciences. Finding: The committee finds that the diversity of activities and modes of support is a strength of the program and of our nation’s scientific system. The approach and vision outlined in NAS/NRC (1958) and the Blue Book (“UCAR,” 1959), which together mapped out the complementary roles of a large national

OCR for page 39
Strategic Guidance for the National Science Foundation’s Support of the Atmospheric Sciences: An Interim Report center and the individual investigator university grants program, has served the atmospheric science community well and is the envy of many other scientific communities. The newer modes of support (i.e., multi-investigator awards, cooperative agreements, and centers sited at universities) reflect the maturation and increasing interdisciplinary nature of atmospheric sciences. The community input received to date supports this multifaceted approach. The present balance is approximately right and reflects the current needs of the community. Recommendation: ATM should continue to utilize the current mix of modes of support for a diverse portfolio of activities (i.e., research, observations and facilities, technology development, education, outreach, and applications). ATM has not published a strategic plan to guide its activities in the coming years. Some indication of strategic directions for the division can be found in a major, long-term planning effort undertaken by GEO and culminating in a report titled NSF Geosciences Beyond 2000: Understanding and Predicting Earth’s Environment and Habitability (GEO, 2000). A community-based strategic planning effort could provide a means by which ATM can advance on many of the issues identified in this report, thereby bringing the division’s activities even closer to the guiding principles laid out in Chapter 1. A clear strategic vision would help guide choices among different priorities and help facilitate interdisciplinary, interagency, and international collaborations. Likewise, a strategic plan would help schedule multiyear commitments of facilities, especially to ensure an approach to field programs that would balance many competing demands. Although there is some concern that a strategic plan could impose constraints on funding opportunities, the committee feels that the benefits of transparency in making difficult choices among competing demands outweigh this concern. Strategic plans can take many different forms, ranging from describing a mission and fairly high-level goals for a program to providing more details about implementation. At a minimum the strategic plan recommended below should clearly articulate ATM’s mission and goals in the context of the multidisciplinary, multiagency, and multinational environment of atmospheric research. However, the committee envisions ATM’s strategic plan going beyond providing a set of goals to include actions on how to attain the goals, although not prescribing in great detail the specifics of implementation. Rather, it should address practical implementation challenges, such as interagency relations, international relations, and university relations with NCAR. Further, the plan should put flexible structures in place that will give ATM a means for making decisions about prioritization, for example, in response to pressures resulting from an evolving budgetary environment, competing international initiatives, and multiple demands for facilities. Strategic planning should provide a broad framework to address key long-term scientific needs such as those related to climate change. Having a strategic plan in place may call for a reorganization of ATM to direct staff and

OCR for page 39
Strategic Guidance for the National Science Foundation’s Support of the Atmospheric Sciences: An Interim Report resources in a way that may better address emerging challenges. Furthermore, the balance of modes should evolve in the future in a manner that is consistent with strategic planning efforts. The committee believes that the strategic plan itself will be useful to ATM, but the process of producing it may prove even more valuable, particularly if it is conducted with ample and transparent community engagement. The committee envisions the strategic planning process as providing a mechanism for the community as a whole to participate in an active conversation about the direction of the field and where best to use resources, while remaining sensitive to the societal expectations of that research. Thus, the strategic plan must be flexible and responsive, and developed by the science community in collaboration with ATM management. Ideally, the process of developing the strategic plan would be simple, revisited at regular intervals, and eventually ingrained in the ATM culture. A strategic plan will be essential to maintain a balanced, effective portfolio in an evolving programmatic environment. Finding: We are now in a phase of rapid change in graduate education demographics, the role of the United States in the global atmospheric science community, potentially the role of NSF in national atmospheric science funding, and the maturation and interdisciplinary growth of atmospheric science, as well as a likely period of constrained budgets. GEO (2000) represents a broad strategic plan for NSF GEO and reflects the considerable evolution of the geophysical scientific enterprise. Yet, ATM has not developed its own strategic plan. Given the changing programmatic environment, ATM should take a more proactive approach to strategic planning. A flexible strategic plan developed with ample community input will enable determination of the appropriate balance of activities and modes of support in the ATM portfolio; help plan for large or long-term investments; facilitate appropriate allocation of resources to interdisciplinary, interagency, and international research efforts; and ensure that the United States will continue to be a leader in atmospheric research. In addition, a strategic planning effort that effectively engages the community will enhance the transparency of the rationale behind ATM’s decisions. Recommendation: ATM should engage the community in the development of a strategic plan, to be revisited at regular intervals, and should rethink its programmatic organization in light of this plan.