Review of the WATERS Network Science Plan (2010)

The WATERS Science Plan (Dozier et al., 2009) presents a high-level vision for interdisciplinary water research that would be enabled by the construction of a network of observatories. The Science Plan envisions an integrated approach involving the natural, engineering, and social sciences to study fundamental processes and activities in the built and natural environments. The overarching science question presented in the plan is: “How can we protect ecosystems and better manage and predict water availability and quality for future generations, given changes to the water cycle caused by human activities and climate trends?” Three “grand challenges” or high-level research questions, which also are at the level of a vision statement rather than a detailed plan, are posed in the Science Plan as natural extensions of the overarching question:

1. How is fresh water availability changing and how can we understand and predict these changes?

2. How can we engineer water infrastructure to be reliable, resilient, and sustainable?

3. How will human behavior, policy design, and institutional decisions affect and be affected by changes in water?

These three questions reflect the focus on the natural, engineering, and social sciences respectively. The committee’s comments are organized around the overarching question and the three grand-challenge questions that flow from it.

The WATERS investigators have identified an important overarching science question, which inextricably links the welfare of humans and ecosystems to the accessibility of high-quality water. The research path forward suggested by the overarching question requires not only additional and better data, but integration of information from various disciplines. In the Science Plan, Dozier et al. (2009) argue that an observatory network is the preferred, and perhaps the only, feasible mechanism for addressing the challenge. The WATERS Network science team further argues that research to support decisions about water-related issues is needed urgently and, given that the envisioned network will require almost a decade for detailed planning and implementation prior to construction and operation, there is a need to act now to initiate the program. As noted in the committee’s July 2009 letter report (see Appendix A), the Science Plan was intended as a broad vision document, and in this light, the document succeeds in communicating a high-level vision for transforming water science and engineering research through establishment of an observatory network. In particular, the overarching question presented in the Science Plan successfully conveys the broad rationale for a major research undertaking.

The WATERS investigators extended the overarching question by posing three “grand challenges” as embodied in the three high-level questions noted earlier. These three questions pose challenges associated with (1) closing the water balance (i.e., independently determining the fluxes and storage of water in both natural and engineered water systems), (2) providing research to support engineered water infrastructure to provide society with safe and reliable water services and protection from hydrologic events (e.g., floods, droughts), and (3) providing a better grasp of the complex interactions and uncertainty between human behavior and variability in the water cycle. In the Science Plan, Dozier and colleagues (2009) argue that the “business as usual” approach to tackling scientific problems drawn upon disciplinary lines will not succeed in answering the challenges as defined. The three grand challenges do, in fact, provide an excellent basis for organizing the plan for the WATERS Network. Furthermore, the challenges embrace the interdisciplinary framework that is a hallmark of the WATERS Science Plan. The com-

mittee judged that the proposed approach to span natural sciences, social sciences, and engineering could lead to WATERS becoming a model for conducting interdisciplinary research within the National Science Foundation (NSF). In particular, the integrated WATERS Network could provide a valuable opportunity to integrate the social sciences and water science.⁴

The Science Plan was conceived as a high-level vision statement and not as a design document. Thus, comments on and critiques of details are not possible. The committee recognizes, however, that careful attention to details will be quite important if the WATERS Network proceeds in the future. Therefore, the committee offers some observations on issues that will need to be addressed in the future to elaborate successfully an approach to meet the three challenges stated in the vision in the Science Plan.

One of the central arguments for hydrologic observatories at various locations or regions of the United States is to enable hydrologists to understand better what happens to water in different types of watersheds. As described in the Science Plan, determining a water balance for a given region involves measurement of fluxes of water (e.g., streamflow, groundwater flow, evapotranspiration) and stocks (e.g., volume of a surface reservoir, amount of water stored in soils). Most often in current practice, one or more of the important fluxes is not measured and must be approximated by calculating the difference (e.g., evapotranspiration as the difference between measured precipitation and measured river runoff) or by some other means. The term “closing the water balance” refers to the notion that appropriate measurements of all important fluxes and stocks be made or estimated independently before determining whether the measurements are internally consistent. Satellite systems and surface-based sensors in the WATERS Network will measure fluxes of water between various stores in the atmosphere, cryosphere, soil, and groundwater (Dozier et al., 2009). Understanding and quantifying the various components of the water cycle are fundamentally important objectives for water science in general and for the broader goals of under-

standing and predicting future changes in water availability within a changing climate. Closing the water balance is critical for Earth System and climate models, which otherwise will not be reliable for long-term predictions of the water cycle or related variables. Thus, these are appropriate long-range goals for the WATERS Network.

The WATERS objective extends beyond physical measurements and models of water per se. The Science Plan calls for extensive chemical measurements as well, to allow for closure of chemical balances linked to water to complement the water balance. Addressing challenges concerning sensors will be particularly important.

A better understanding of the hydrologic water balance on the watershed scale would provide opportunities for improved management of the surface water and groundwater in those watersheds. Water balance analyses, however, are highly dependent on scale, as judged by space, time, and purpose. As planning for WATERS proceeds, the argument that a network of long-term hydrologic observatories is required to observe processes over a spectrum of scales should be articulated convincingly and clearly and be tied to detailed research descriptions and to implementation plans. In summary, the first grand challenge remains as a critical science need, and the WATERS team will have to elaborate the details required to meet the challenge effectively as the project proceeds.

There are many difficult issues that will have to be engaged to move WATERS ahead to accomplish the stated goals. These include network design, choice and deployment of sensors, integration of NSF-supported efforts with those of other agencies, and cyberinfrastructure. The committee offers some extended comments on networks and linkages with other programs in Chapters 3 and 5. The committee’s interim report (NRC, 2008) contained some discussions related to sensors and cyberinfrastructure. Because the latter topic is of critical importance to the pursuit of all of the science questions outlined, cyberinfrastructure issues are discussed later in this chapter.

A second challenge put forth in the WATERS Science Plan is to understand better how to construct and operate engineered infrastructure to manage water quantity and quality. Sufficient data are lacking on the complex interactions among chemical constituents, pathogenic organisms, and water infrastructure, as is the research infrastructure to test entirely new configurations to optimize water and energy management.

The WATERS Network would address this deficiency by deploying a system of physical and chemical sensors in a variety of water systems and environments to accumulate an empirical base of field data. Another goal as envisaged by the team is the construction of a facility that would be used to test an engineered system’s response to different environmental stresses and novel configurations. Such a facility would also allow researchers to test systems to failure, which cannot be done in existing water and wastewater systems.

The committee finds that this grand challenge is well posed as one of the underpinnings for justifying the WATERS Network. Urban water use, conveyance, and treatment are essential components of the overarching question regarding management of water availability and quality for future generations. There are serious issues, however, that will need to be addressed as the planning process moves forward. These include many of the same details alluded to in the previous comments about the first of the grand challenges (e.g., sensor design and operation, linkages with other programs, cyberinfrastructure). Sensors, in particular, can be costly to purchase, maintain, and implement in a widespread area, and sensor development requires extensive research. Thus, the WATERS Network science team will need to consider what key second-level science questions on water quality have the greatest potential to transform the decisions of water managers and improve ecological integrity and human health and how data collected from sensors contribute toward these objectives.

Although the proposed experimental facility for testing water infrastructure would be an important contribution to science and engineering research, the committee also questioned whether such a facility stretches the concept of an observatory network too far, thereby diluting limited resources for the network as a whole. Future planning efforts should carefully examine the value of these specific test facilities to the WATERS Network. The WATERS team should either make a better case as to how a test facility could increase the knowledge gained across a network of observatory sites or decide that these facilities might more appropriately be of a portable nature that could be deployed to numerous locations for testing which would be integrated into the distributed sensing network.

The third grand challenge posed in the Science Plan seeks to understand how human behavior, policies, and institutional decisions both influence and are influenced by water. A major component of the social science questions facing the WATERS team is to understand and predict water use under a variety of conditions. Major tasks would involve quantifying water use by the human system and using surveys, archival studies of governmental and utility data, and experiments to predict water demand and factors that influence demand (especially in areas where water is scarce). This challenge also requires research to understand the ability of alternative institutional forms to govern water usage, respond to fluctuations in availability and water quality, and balance the costs of developing reliable water supply for human use—be it snowpack, lakes, groundwater, or rivers—with the need to protect these resources and the natural systems that depend on them. One goal of the WATERS Network is to provide scientists, engineers, policy makers, and other stakeholders with the knowledge and tools needed to maintain a reliable and sustainable supply of potable water for the public without damaging watersheds and ecosystems. If properly designed, the WATERS Network could assist water resource managers and stakeholders, in altering human impacts on water use, to adapt to shifts in population and economic dynamics, enhanced knowledge about the human and natural environments, and the consequences of climate change.

The social science vision put forth in the Science Plan outlines the critical research needs in this important area. Key research gaps have been identified in assessing the effectiveness of water policies and management, in understanding the determinants of consumptive water use, and in developing improved water management institutions (NRC, 2001). All of these, as well as others suggested in the Science Plan, will depend critically on integrating work in the natural and social sciences and on linking data from physical observatories with longitudinal archival and survey data that allow social scientists to track changes in the human system over time. For example, archival data on the program and policy decisions of water-related authorities combined with surveys of affected users, user organizations, and the general public in areas of intense hydrologic observations can be used to analyze the interaction between observed fluctuations in the natural system and the human system response. The WATERS Network is an excellent vehicle for achieving the needed research and integration.

The three grand challenges posed in the Science Plan represent the disciplinary perspectives of the three supporting NSF directorates (i.e., Geosciences; Engineering; and Social, Behavioral, and Economic Sciences). Each challenge alone might be substantial enough to support the development of large hydrologic observatories, but the WATERS team argues that all three challenges must be addressed to answer the overarching question: “How can we protect ecosystems and better manage and predict water availability and quality for future generations, given changes to the water cycle caused by human activities and climate trends?”

Dozier et al. (2009, p. 9) state, “As climate and land use change, populations grow and relocate, and our engineered systems age and are taxed with new contaminants, the empirical methods we have traditionally relied on have become inadequate and inaccurate.” The WATERS team clearly recognizes that answers to the questions posed require integration across the natural, engineering, and social sciences, given the coupled nature of natural and human processes within water resources issues. Maintaining a truly interdisciplinary perspective as WATERS moves forward with more detailed planning, however, will be challenging because true integration of social sciences with engineering and hydrologic sciences is currently in its infancy. To nurture this interdisciplinary approach to water research and strengthen future large-scale collaborations between the geosciences, engineering, and social sciences, NSF should consider sponsoring interdisciplinary requests for proposals, jointly issued by the three directorates, that support research projects of sufficient size and duration to enable advancement in this area.

To lend a bit more specificity to what might be accomplished in addressing the three grand challenges, the Science Plan presents a “prototype network” to illustrate how the proposed WATERS Network would allow the combination of models and data to address pressing societal problems through interdisciplinary research. The Science Plan includes three example observatories in this prototype network that leverage prior work from WATERS test-bed projects and elsewhere: the Sierra Nevada for “Snow-Dominated Water Resources in the Mountain West,” the Chesapeake Bay for “Non-Point Source Pollution into Receiving Wa-

ters,” and an engineered system in Pittsburgh for “Integrated and Adaptive Water Cycle Management in Urban Systems.” Note that these examples do not reflect selected observatory sites for the WATERS Network and were only developed for illustrative purposes. The committee appreciated the use of example systems and offers the following comments on them.

Earth’s glaciers and ice caps have been undergoing recession in recent decades, and mountain snowpack in areas such as the intermountain West has been in decline. These changes have significant implications for water resources and ecosystems. The state of the cryosphere has been cited as having “a unique sensitivity to climate change at all spatial and temporal scales” (Slaymaker and Kelly, 2007).

In the United States, snowpack changes in the West are the best documented current hydrologic manifestation of climate change (Barnett et al., 2008; Pierce et al., 2008). About half of the observed decline in snowpack in western mountains (with concomitant changes in the amount and seasonality of river discharge) is clearly linked to a warming climate due to anthropogenic influences. The largest losses in snowpack are occurring in the lower elevations of the Sierra Nevada and Cascade mountains of the Northwest and California, as a result of more rain than snow falling under higher temperature. In climates where the summer growing season is the dry season, as is the case for much of the western United States, this concentration of runoff in the spring season and reduction in later summer runoff stresses the water supply systems and can lead to water shortages in summer (Barnett et al., 2005).

The Science Plan’s mountain snowpack example lays out the societal issues and science challenges well. This example emphasizes the hydrologic science questions (i.e., how is fresh water availability changing and how can we understand and predict these changes?). Some questions, however, are included that address social sciences (e.g., what institutional arrangements would best manage the watershed for ecosystem services?) and the engineered water systems (e.g., what are the effects of changing patterns of water delivery on the provision of drinking water and the management of wastewater and stormwater?). Thus, the mountain snowpack example touches on all three grand challenges, albeit unevenly. The example contains a convincing explanation on how an observatory would provide information that could lead to fundamental new

insights about processes, allow the development and testing of new theory, and lead to much-improved decision making by water managers.

Cultural eutrophication is a serious and ubiquitous global water quality problem in lacustrine, estuarine, and coastal aquatic environments. It is normally the result of inputs of nutrients—nitrogen and phosphorus—resulting from a variety of human activities (Vollenweider, 1971). Estuarine and coastal waters along all coasts of the United States are severely impacted. Eutrophication is often accompanied by massive algal blooms, reduced transparency, development of littoral mats of filamentous algae in near-shore areas, disappearance of submerged aquatic vegetation, speciation changes, development of low-oxygen conditions, decline of fisheries, obnoxious tastes and odors, and the production and release of toxins that impair human and animal health.

The link between increased nutrients and increased primary production is well established. Difficulties in understanding the problems associated with eutrophication are related to the changes in fluxes of materials from the watershed and airshed, their speciation and quantity, and the sources. Point sources of pollution are more readily identifiable than nonpoint and diffuse sources. An additional difficulty in defining impaired water quality related to nutrient increases and its negative impacts are the multiple stressors impinging on a water body, such as habitat loss, changes in fishing pressures, hydrologic modifications, dredging, and the presence of other chemical contaminants such as metals and pesticides. Water quality management related to nutrients is severely hampered by the complexity of the aquatic ecosystem, its human inhabitants, and the multiple sources of nutrient inputs.

The Chesapeake Bay is a sensible example for the illustrative purposes of the Science Plan. The overarching societal questions are well developed, and the example emphasizes the hydrologic and social science challenges in the Chesapeake Bay. The discussion is not as effective as that for the Sierra example, however, in part because the Science Plan does not clearly explain how the WATERS Network would complement the extensive long-term research, monitoring, and management efforts in the Chesapeake Bay to create significant advances in understanding. Moreover, the monitoring strategies discussed in the Science Plan appear to focus on physical parameter measurements (e.g., flow, temperature) to elucidate the physical causes of the problem (e.g., strati-

fication, turbidity) while eutrophication is dominantly a biogeochemical phenomenon. Linkages to the grand challenge on engineered infrastructure involve management practices in the watershed to control nutrient loadings, although the associated scientific challenges are not explored in the same depth as the hydrologic and social sciences challenges.

The ubiquity of the problem across the nation does indicate that the overall choice of impacts of nutrient additions to the nation’s waters is an excellent vehicle to express the advantages of the WATERS Network approach. The Science Plan example would be more compelling if it included an expanded description of the problem and a broader vision of the network to extend the discussion to the nation as a whole.

The third example observatory in the Science Plan examines how society can better sustain the engineered water cycle despite changing land use and population growth. The Science Plan describes the compelling science challenges associated with the urban water cycle. It uses the example of the Pittsburgh watershed to outline ways that the WATERS Network could help address these challenges in an urban setting and at the interface between the engineered and natural environments.

The WATERS team emphasized the importance of designing the engineered observatory correctly because it is nested within a larger watershed observatory. The hypothetical observatory could provide valuable information regarding the state of the existing built water infrastructure and insights into how future upgrades can be designed in a smarter, more adaptive, and sustainable manner, while examining issues of water quality and availability in the broader watershed. The linkages between the three grand challenges—water infrastructure, changing availability of freshwater (focused primarily on water quality in this example), and human and institutional decisions—are particularly compelling.

The WATERS team also proposed the use of test facilities to overcome the challenges of research in an existing built environment, including the inability to test changes to treatment processes (e.g., the use of a different disinfectant) or alterations in water quality (e.g., chemical spills, introduction of pathogens) or to test an existing system to failure. Their vision of the proposed experimental facilities for the built environment effectively describes the research advantages associated with such facilities. However, as noted previously, the committee questions

whether this experimental facility is consistent with the concept and resources of an observatory network.

Each of the example observatories in the prototype network, discussed in the previous sections, addresses the overarching interdisciplinary question posed in the Science Plan. The examples also all touch on each of the three grand challenges, albeit sometimes unevenly. Important synergies are evident from addressing social science questions concurrently with engineering and hydrologic science questions. The linkages between the example observatories, however, are not clear, nor are they explained. Are there compelling reasons why simultaneously answering the questions posed, in common locations, across a network of sites, is important and provides important synergies over answering them separately? To justify a national network, the WATERS team should make a stronger case that these science questions require intersite comparison to answer or identify other integrated questions and hypotheses that are posed across sites.

In addition to the benefits noted above and also in the Science Plan, the WATERS project has the potential to bring several widespread benefits to the global scientific community. The nation’s applied research and operational programs can benefit from the modernization and transformation that the WATERS Network would bring. For example, the WATERS Network would bring new technologies, such as novel remote sensing capabilities, into common use in water science. Although data collection has been headed in this direction for several years, the WATERS Network could accelerate the development of technologies that are less dependent on the collection of data by humans and save funding, and perhaps lives, while enabling faster, more accurate monitoring.

The WATERS Network could also make significant contributions to areas such as water and human health, water–energy linkages, and water economics. For example, societal benefits could be realized from the redesign of coupled built and natural water systems to maximize the availability of clean water. Also, the WATERS Network could spur the development of sensors to detect drinking water contaminants such as

harmful microorganisms, pesticides and herbicides, and emerging pollutants such as pharmaceuticals and personal care products that enter potable water supplies. In the area of water and energy, the WATERS Network could examine the societal trade-offs between increasing domestic energy production and the resulting impacts on water resources, both quality and quantity. As an example, increased use of irrigated crops for the production of biofuels could have significant impacts on water supplies and cause much greater water quality problems, including increased nitrogen and phosphorus loads (NRC, 2008). As fresh water becomes scarcer in certain parts of the nation, economics will play an increasing role in determining water policy and promoting conservation. One of the strengths of the WATERS Science Plan is that social sciences is an integral component of the network and can address some of these economic considerations.

The inclusion of social science data not only strengthens timely policy-related analyses, but can also encourage transformative research that is not currently possible. For example, research on collaborative governance institutions for watershed planning now uses only expert judgments to assess the performance of alternative institutional design. An integrated database that includes both institutional and hydrologic measures of performance would allow more meaningful analysis. Similarly, long-term measurement series focused on changes relating to climate and water quality can be linked to the responses of water users as well as water managers and political overseers to study the adaptive capacity of the system.

Another important benefit of the WATERS program is the potential for affecting water science in the international arena, particularly in developing countries where water supply and water quality problems are typically much greater than in the United States. For example, climate change is a global issue, yet many countries cannot provide adequate safe drinking water much less the resources to examine the impacts of climate change on their water resources. For example, WATERS could make major contributions on behalf of the United States to the Global Water System Project, which is addressing human interactions with the hydrologic cycle in terms of natural and built environments; the Group on Earth Observations, which is developing the data and information components of the Global Earth Observation System of Systems; and the Global Energy and Water Cycle Experiment, which is focusing on hydrologic prediction systems. The WATERS research findings also could become an important U.S. contribution to international develop-

ment programs, particularly for those countries where water issues are critical.

As discussed previously in this chapter, cyberinfrastructure is a critical element of the WATERS Network to link the local observatories and to enable multiscale and networkwide analyses by a wide array of researchers. Considering the lack of cyberinfrastructure details in the Science Plan, the most that the committee can offer in terms of a critique is to note several general areas that deserve attention. Once WATERS moves toward more detailed planning, the design team will derive cyberinfrastructure requirements that will inform their design decisions and architecture. This could lead to a conceptual design with sufficient detail to support a rigorous review. In the absence of a conceptual design, the committee can only speculate on what the needs will be.

1. Scope of services: It will be important to clarify the scope of the cyberinfrastructure services that the WATERS Network will provide. The Science Plan mentions data products, workflows, models, and collaboration services. Mention is made of “real-time” observational data, including synthesized “real-time” observations. Observations, forecasts, and decision support are listed among the network services that will be enabled by cyberinfrastructure. Other observational Major Research and Facilities Construction (MREFC) projects (e.g., National Ecological Observatory Network, USArray) have a more limited focus, namely standardized data product publication. It will be important that the detailed plans for WATERS cyberinfrastructure be evaluated within the context of technical feasibility and budgetary reality to determine the appropriate scope of services.

2. Integration with existing facilities: The proposed integration of WATERS cyberinfrastructure with existing facilities and other agencies raises numerous issues in cyberinfrastructure development and operations. Although the intention to leverage existing systems is commendable, and probably necessary, the actual execution of this idea will be a challenge. It is relatively straightforward to exchange data products, but it is significantly more difficult to develop and manage complex cyberinfrastructure systems across autonomous agencies. There are technical and operational challenges, such as agreements on common interface standards, integration of heterogeneous and legacy software systems, and

agreement on system tasking, operations, and maintenance. Resolving these issues will be a challenge that should be addressed in the conceptual design document.

1. Metadata standards: The success of the WATERS Network will depend, in part, on the metadata specifications. It could be useful for the WATERS design team to detail the approach to creating and getting consensus on appropriate metadata standards for both deployment details (e.g., sensors, locations, calibration histories) and data products. In addition to the standards, it would be useful to create detailed example scenarios to demonstrate the operational roles of various metadata tagging and processing activities.

2. Acquisition of cyberinfrastructure components: An approach to cyberinfrastructure acquisition should be described in the next level of design documents. What approach will the WATERS design team take in identifying and acquiring the cyberinfrastructure? The WATERS team will need a plan for surveying the current portfolio of NSF-sponsored cyberinfrastructure and identifying elements that can be adapted to their needs. They should avoid the temptation to design and build everything from scratch. For adapting cyberinfrastructure from the WATERS prototypes, they will need a plan that transitions the current test-bed or prototype activities into a continental-scale production facility. Practices and policies that are adequate for small-scale independent systems will not suffice for an integrated system of the scale envisioned for WATERS. They should also engage with the other observational MREFCs to explore options for cyberinfrastructure acquisition.

3. Operations and maintenance: A detailed projection of the anticipated costs for operations and maintenance should be performed as soon as possible. The number of fielded sensors and their geographical distribution will require considerable maintenance for deployment, cleaning, calibration, rotation, etc. Similarly, the need to maintain decision theaters and modeling clusters will also require administrators, support staff, and managers. An early estimate of the costs of operations and maintenance could perhaps provide useful feedback to determine the scope of cyberinfrastructure services that the WATERS Network will provide.

4. Phased deployment: Given the broad distribution of the WATERS cyberinfrastructure and the rapid rate of technology evolution, the WATERS team should consider various deployment schedules. Also, given the loosely coupled nature of many of the cyberinfrastructure components, it will likely be possible to mitigate some development risks by strategically testing and hardening system components on a relatively

small scale (e.g., a watershed). In addition to a phased system deployment, it will be important to have a plan for periodic technology refresh. Given the intended lifespan of the WATERS Network, the cyberinfrastructure will most likely need to support several generations of sensors and instruments.

1. Management: The committee recognizes that WATERS is a large and complex system. Although there is a sizeable and qualified project team, it could be useful to hire a full-time systems design engineer who has experience with complex large-scale systems as the WATERS Network moves forward.

As noted in the committee’s letter report (NRC, 2009), the Science Plan was intended as a broad vision document, and in this light the document succeeds in communicating a high-level vision for transforming water science and engineering research through establishing an observatory network. The Plan outlines the opportunity to collect, analyze, and integrate hydrologic, environmental science and engineering, and social sciences data at a level that has not previously been possible. Opportunities for cutting-edge research are envisioned through the collection of data from sensors distributed at sites along gradients or at nested sites that will enable the improvement of the understanding of the complete water balance within a research site. The proposed network also could support research on how better to design and build engineering systems for water management and the collection of social science data from individual sites and on a national scale to better understand human–water resource interactions and impacts. Overall, the committee finds that the presentation of the overarching science question and the three grand challenges in hydrology, engineering, and social sciences provide compelling arguments in support of the WATERS Network. The hypothetical observatory examples illustrate the potential interdisciplinary research that could be undertaken using the WATERS Network, although the linkages between the example observatories that would lead to a national network based on intersite comparison remain poorly defined. The integration of social sciences with engineering and hydrology is a key benefit of the WATERS Network. The committee commends the WATERS team for its efforts to bring together this community of researchers, and encourages the team to continue to nurture the integration of multiple disciplines.

Although the committee finds the high-level vision for science to be well done in the Science Plan, as the WATERS Network moves ahead through the conceptual design phase, a much more detailed “science plan” will need to be developed in parallel with the design to make sure that the necessary coordination between the desired science and the feasibility of network construction is accomplished. That is, the natural progression from high-level vision to detailed description of scientific objectives will have to occur.