B
Workshop Discussion and Participants

WORKSHOP ELEMENTS

Context

When completed, the National Polar-orbiting Operational Environmental Satellite System (NPOESS) will provide operational support for a wide variety of Earth remote sensing measurements. With the longer planned lifetime of its satellites and the more stringent performance and stability requirements (relative to Polar Orbiting Environmental Satellites (POES) and the Defense Meteorological Satellite Program (DMSP), the current operational weather systems NPOESS is to replace), the NPOESS system offers an opportunity to begin developing an operational component of an integrated satellite observing system for climate research and climate monitoring. This was the subject of a workshop hosted by the committee on July 26-27, 1999. This appendix summarizes the committee’s discussions at the workshop along with the committee’s perspective on the views of workshop participants.

Differences in agency history and culture and the distinctions that characterize the operating philosophy of a “research” agency such as the National Aeronautics and Space Administration (NASA) and an “operational” agency such as the National Oceanic and Atmospheric Administration (NOAA) formed the backdrop for many of the workshop’s discussions. In general, NOAA approaches climate research as an outgrowth of its primary mission, which is weather prediction and warning, while NASA’s Earth Science Enterprise (ESE) carries out its climate-related research in connection with a program to answer specific science questions. Despite these differences, both agencies frequently require access to long-term, systematic observations to support their missions. However, while the particular observations may have considerable overlap, there are often significant differences in the measurement requirements. These similarities and differences underlie the challenge in crafting a strategy to meet the needs of climate researchers. They also account for the belief among members of the Committee on Earth Studies that long-term (or systematic) observations in support of climate research require a comprehensive strategy, not simply modified requirements.

Because climate variability occurs on a wide range of time and space scales and involves complex interactions between atmosphere, ocean, and land, researchers can completely specify with certainty neither the types of variables that need to be measured nor the appropriate sampling strategy. However, according to workshop



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 53
Issues in the Integration of Research and Operational Satellite Systems for Climate Research: II. Implementation B Workshop Discussion and Participants WORKSHOP ELEMENTS Context When completed, the National Polar-orbiting Operational Environmental Satellite System (NPOESS) will provide operational support for a wide variety of Earth remote sensing measurements. With the longer planned lifetime of its satellites and the more stringent performance and stability requirements (relative to Polar Orbiting Environmental Satellites (POES) and the Defense Meteorological Satellite Program (DMSP), the current operational weather systems NPOESS is to replace), the NPOESS system offers an opportunity to begin developing an operational component of an integrated satellite observing system for climate research and climate monitoring. This was the subject of a workshop hosted by the committee on July 26-27, 1999. This appendix summarizes the committee’s discussions at the workshop along with the committee’s perspective on the views of workshop participants. Differences in agency history and culture and the distinctions that characterize the operating philosophy of a “research” agency such as the National Aeronautics and Space Administration (NASA) and an “operational” agency such as the National Oceanic and Atmospheric Administration (NOAA) formed the backdrop for many of the workshop’s discussions. In general, NOAA approaches climate research as an outgrowth of its primary mission, which is weather prediction and warning, while NASA’s Earth Science Enterprise (ESE) carries out its climate-related research in connection with a program to answer specific science questions. Despite these differences, both agencies frequently require access to long-term, systematic observations to support their missions. However, while the particular observations may have considerable overlap, there are often significant differences in the measurement requirements. These similarities and differences underlie the challenge in crafting a strategy to meet the needs of climate researchers. They also account for the belief among members of the Committee on Earth Studies that long-term (or systematic) observations in support of climate research require a comprehensive strategy, not simply modified requirements. Because climate variability occurs on a wide range of time and space scales and involves complex interactions between atmosphere, ocean, and land, researchers can completely specify with certainty neither the types of variables that need to be measured nor the appropriate sampling strategy. However, according to workshop

OCR for page 53
Issues in the Integration of Research and Operational Satellite Systems for Climate Research: II. Implementation participants, it is possible to define the key elements of a strategy that would facilitate climate research. They suggested the following elements: Scientific insight into sensor requirements and implementation; Calibration and validation of the sensor and its data products as well as comprehensive sensor characterization; Data analysis to assess data quality and develop new data products; Reprocessing of data to incorporate new knowledge; Ground networks for validation; Multiple approaches to variables to increase confidence; Assessment of temporal and spatial sampling strategies; and Technology development to lower costs and improve performance. Workshop participants said that execution of this strategy was beyond the capabilities of any single agency. They also noted that the strategy required more than just observations—its components also include data analysis and technology development. Workshop participants viewed NPOESS as a critical element in executing a climate research strategy as it provides a platform for the long-term, continuous observations that will be required for the study of many climate-related variables. Programs sponsored by NASA ESE were also thought to be a critical element of a climate research technology development strategy. Currently, the ESE plans to rely heavily on NPOESS for the systematic observations that are required to meet its climate research objectives. Workshop participants saw the challenge for developing a NASA-NOAA climate research and monitoring program as being one of integrating the operational stability of NPOESS with the research flexibility of NASA/ESE. At present, no agency is charged, nor is there a formal process in place, to assess climate requirements and the overall approach and balance of a climate observational system. In advance of the workshop, the committee distributed materials, including the following questions, for discussion: How can research-quality data sets be obtained that are suitable for study of decadal-scale processes? How can agencies implement the required program of long-term measurements to support climate research in a constrained fiscal environment? A number of workshop participants concluded that the answers to these questions included both operational satellite platforms and research satellite platforms. Moreover, they believed it would involve a new category of missions in which both research and operations work together on data sets of mutual interest. Such preoperational missions could be used to bridge the needs of research and operational users as well as forge links between the observational and satellite sensor communities. An integrated approach to climate research is needed, according to workshop participants, because of the critical need for continuous high-quality data, the size of the national investment involved, and the need to demonstrate scientific progress upon which to base sound national policies regarding climate. Goals As noted, participants at the workshop were asked to consider opportunities in the near term to make incremental investments that would improve the suitability of operational missions for climate research. Further, participants were asked to be sensitive to the need for climate research to be carried out in a way that blends the flexibility provided by research missions with the long-term stability provided by operational missions. The workshop focused on the following two questions: In the near-term (the next 2 to 3 years), what marginal investments can be made by the Integrated Program Office (IPO) to improve the climate capabilities of the NPOESS Preparatory Program (NPP) and NPOESS? What

OCR for page 53
Issues in the Integration of Research and Operational Satellite Systems for Climate Research: II. Implementation marginal investments can be made by NASA and NOAA to improve the climate capabilities of NPP and NPOESS, as well as other missions? What activities are needed on a longer time scale to increase the likelihood that NPOESS will have the capability to be responsive to both currently anticipated needs and those needs that might arise in response to new scientific or technical developments? Workshop participants developed their opinions on what can be done now to ensure a reasonable record of critical climate variables for the study of key climate processes. The objective was not to develop the “best” system but to develop one that might make significant contributions to climate research within the fiscal, technical, and programmatic constraints of existing and planned programs. Approach A variety of intra- and interagency actions appear necessary to address the issues outlined above. Many participants saw the need for a national policy regarding climate and long-term environmental variability, particularly to direct interagency efforts. Others focused on issues regarding the integration of NASA and NOAA research and operational observing systems. Looking in more detail, some participants commented on the need for national policy objectives regarding the collection and analysis of decadal-scale observations. Finally, some participants focused on comparatively narrow issues regarding the development of NPOESS and the NASA/ESE missions. At the workshop, attention was focused on making interim progress on specific issues of integration of the NPOESS and EOS programs. It was anticipated that this approach would also inform policymaking discussions that might occur at higher levels of government. The workshop was organized as follows: Prior to the meeting, the committee selected a small set of critical satellite-based data sets for discussion. These data sets, chosen in part based on previous Intergovernmental Panel on Climate Change (IPCC), National Research Council (NRC), and NASA reports, were thought to be representative of those needed to study decadal-scale climate variability. The workshop addressed the structures (including policy, agency, and observing systems) that might be needed to obtain these data sets with sufficient overall quality to meet climate research requirements. These needs were then examined in light of the plans of the NPOESS IPO and NASA/ESE. Workshop participants were asked to focus their discussions on approaches to meet climate observational priorities within the technical, fiscal, schedule, and programmatic constraints of agency programs. Specific attention was paid to those variables where realistic changes could be made within existing NPOESS plans. This implied, for example, a consideration of improved sensor performance or better sensor characterization versus consideration of a sensor that was not already in the baseline plan. Systematic measurements that had been identified by NASA/ESE as candidates for continued measurement by NPOESS were examined in detail. Workshop participants also discussed a strategy to acquire or continue measurements of high-priority climate variables that are not part of the NPOESS framework. Elements of the strategy included measurements by NASA/ ESE, bridging missions between research and operational programs, technology development, or international partnerships. Participants recognized that scientific understanding and technology will evolve, and that the needs of policymakers may change. Therefore, they also considered ways agencies might formalize a process to reassess their observation strategies. Climate Variables Discussed The variables chosen for discussion at the workshop are listed below: Atmospheric and forcing variables Atmospheric temperature and moisture profiles Cloud properties Stratospheric ozone and aerosols

OCR for page 53
Issues in the Integration of Research and Operational Satellite Systems for Climate Research: II. Implementation Precipitation Earth radiation budget Solar irradiance Ocean variables Ocean topography Ocean winds Sea surface temperature (SST) Sea ice Phytoplankton biomass Land variables Land cover. While all of these variables have a demonstrated or anticipated importance to climate research, it should be recognized that only a subset have a long heritage in satellite remote sensing. They are atmospheric temperature and moisture, stratospheric ozone, solar irradiance, SST, sea ice, Earth radiation budget, and land cover. To be useful in climate research, these variables require measurement nearly continuously over multidecadal time scales (essentially indefinitely), that is, data acquisition in an operational manner. The committee recognized that there were other variables that needed such long-term records; however, budgetary and programmatic constraints at NASA and NOAA suggested that the initial focus of the workshop should be on a comparatively smaller set of variables. The connection between these variables, the key scientific questions identified in the Draft NASA Earth Science Implementation Plan (which in turn draws on the “Pathways” report (NRC, 1998)), and a measurement strategy is shown below. Workshop discussions did not consider all of the critical measurements that are necessary to address these questions. Instead, participants deliberately selected a small number of measurements where long-term continuity was thought to be essential and where expectations of marginal investments in the present measurement strategy were thought to have a reasonable chance to yield data records suitable for climate research. Biology and Biogeochemistry of Ecosystems and the Global Carbon Cycle Key Questions How do ecosystems respond to and affect global environmental change? How are land cover and land use changing? What are the causes and consequences? Required Measurements Systematic measurements at moderate spatial resolution of land cover Systematic measurements at moderate spatial resolution of phytoplankton biomass Global Water and Energy Cycle Key Questions Is the global water cycle accelerating? Can weather systems, precipitation, and the hydrologic processes that control water resources be related to observed or predicted large-scale climate anomalies? Can the integrated effect of fast atmospheric and surface processes be accurately represented in large-scale model predictions of climate change?

OCR for page 53
Issues in the Integration of Research and Operational Satellite Systems for Climate Research: II. Implementation Required Measurements Systematic measurements of global temperature and water vapor Systematic measurements of global precipitation Systematic measurements of the components of Earth’s radiation budget Systematic measurements of cloud properties Climate Variability and Prediction Key Questions Is climate varying in ways that we can understand and predict? What causal relationships can be established between observed climate changes and specific forcing factors? Required Measurements Systematic measurements of solar irradiance Systematic measurements of the components of Earth’s radiation budget Systematic measurements of global stratospheric aerosols Systematic measurements of global ocean topography Systematic measurements of ocean surface vector winds Systematic measurements of sea surface temperature Systematic measurements of sea ice Atmospheric Chemistry Key Questions Is the Montreal Protocol working as expected to stop stratospheric ozone depletion by industrially produced chemicals? How are meteorological and chemical processes in the atmosphere affecting the distribution of trace constituents? How much will industrial and urban pollution expand globally and with what consequences? Required Measurements Systematic measurements of stratospheric ozone and aerosols Emerging Requirements for Climate Data Products The requirements for climate observations to meet the needs of scientific research were recently considered in connection with the decadal review of the U.S. Global Change Research Program (USGCRP) (NRC, 1999). In addition, national assessments1 reflect the growing interest in climate changes on both the interannual and decadal scales by business, governmental planners, and the general public. Interest in the economic impact of climate change is evident, for example, in business decisions regarding investment in the energy sector of the economy, an area sensitive to both climate trends and anomalies. Further, the impact of climate change on the energy sector is 1   Information on the USGCRP-initiated “U.S. National Assessment of the Potential Consequences of Climate Variability and Change for the Nation” is available online at <http://www.nacc.usgcrp.gov/>.

OCR for page 53
Issues in the Integration of Research and Operational Satellite Systems for Climate Research: II. Implementation expected to grow as the nation expands its energy use during the coming years. Planning to meet that growing demand requires climate projections concerning decadal time scales, whereas generating scenarios for dealing with seasonal peaks requires interannual climate products. At the moment, important climate products are being prepared on an operational basis. The newly developed market for weather futures is placing demands on timeliness, which differ from the needs of the research community. The potential for increasing these observing capabilities with the advent of NPOESS is great. Realizing that potential, however, is neither an easy nor a well-defined process. Strategies To date, the climate research community has had only a limited involvement in the process that determines the NPOESS requirements (NOAA, 1997). Further, there is substantial concern that the NPOESS data products will be only marginally useful for climate purposes. As noted earlier in this report, the needs of the climate research community were not a driving part of the process for determining the NPOESS environmental data records (EDRs), which reflect the joint operational needs of NOAA and the Department of Defense (DOD). Workshop participants highlighted key issues and identified several actions for consideration by the IPO that they believed would greatly improve the suitability of NPOESS data for climate studies. At the moment, for example, there is no provision for archiving the metadata concerning the sensor calibrations, housekeeping information, or prelaunch characterizations of the sensors, which are critical to the ability to construct climate applications based on the NPOESS data, especially when attempting to identify trends. Another concern of workshop participants is related to continuity of the NPOESS data. Data continuity might not appear to be an issue as the NPOESS satellites have a 7-year life expectancy and are planned for launch on a 5-year cycle. However, it is expected that the launch cycle will be driven by cost considerations and launches of replacement systems will be made on a “demand” basis, that is, when systems critical to operational weather services begin to fail. There is a 9 to 12 month period between the time the decision is made to make a replacement launch and the actual operational service of the new satellite. A gap in the record of that magnitude, where there is no overlap with the then-current satellite, cannot be bridged for the investigation of climatic trends. (An example is the interannual variability associated with El Niño/Southern Oscillation (ENSO) events.) Workshop participants also had concerns related to the actual production of climate data products. For example, the IPO does not have a requirement to provide a data archive and retrieval capability. That responsibility is currently left to the operational agencies, namely NOAA. There has been only preliminary planning within NOAA to address this responsibility. If climate applications require affordable and relatively easy access to raw radiances, the storage capacity needs to be larger than present NOAA capabilities. Other concerns relate to data formats, access procedures, media obsolescence, and provision for reprocessing when errors are detected in the original data sets. In fact, workshop participants discussed the need to design a climate processing system that would operate in parallel with the operational one. The impact of tight budgets on agency capabilities to serve the climate community was another concern of workshop participants. Present plans call for transition from the research-demonstration mode of NPP to the operational mode of NPOESS. However, the budget implications of a complete operational end-to-end NPOESS that are sustainable are greater than the totality of all other operational observing programs in NOAA. The disparity appears to be so large that it will not be possible to make traditional trade-offs between one or the other operational observing system to pay for the infrastructure to support the operational NPOESS. Eliminating the entire rawinsonde program, for example, would net only some $6 million per year. Supporting the operational NPOESS would require significantly more funding. In summary, workshop participants believed that the process of realizing major climate benefits from NPOESS was incomplete—scientific understanding needs to be enhanced and the infrastructure for embedding NPOESS into existing operational programs needs to be defined and established along with the organizational structures for making that process work. As the actions required are very complex, participants suggested that agencies consider testing concepts in “real time” to gain experience and inform decisions on how the operational program should function. Important parts of such a process already exist and workshop participants thought they could be

OCR for page 53
Issues in the Integration of Research and Operational Satellite Systems for Climate Research: II. Implementation modified at modest cost to provide a basis for generating the capability to use NPOESS effectively for climate applications. In particular, workshop participants focused on the NPP as a key element in their suggested plans. Workshop participants saw the NPP as an excellent vehicle to test more than the functioning of the satellite systems and their application to weather forecasting. Participants expected the NPP to provide hard information on the operational costs likely to be involved with NPOESS. In addition, they believed experience with the NPP would provide a basis for identifying the support required from the in-situ observing system, as well as the efforts required to transition that system into the NPOESS era. Finally, they expected the data stream from the NPP to provide the basis for designing an effective archival and retrieval system. Another existing resource of interest to participants was the North American Atmospheric Observing System (NAOS).2 NAOS is a cooperative program supported by governmental organizations and universities in Canada, Mexico, and the United States. Its purpose is to make recommendations on the configuration of the upper air observing program, in those three countries and adjacent water areas, that would meet societal needs in the coming few decades. The main thrusts are (1) a scientific evaluation program focused on, but not limited to, the value of various combinations of observing systems to numerical weather prediction, and (2) an assessment of the operational, financial, and organizational implications of configurations based on the results of the scientific evaluations. NAOS supports network design and scientific studies of the impact of current observing systems, including the potential impact of future systems on the ability of the weather services to provide both weather and climate services. A fundamental concern of NAOS members is the mix of satellite and surface-based observations. This established activity provides a vehicle for building a consensus of the relative roles of satellites and other observations in the future observing system. Organizational Issues Any climate observing system will be based on a complex combination of elements that are often perceived to be in opposition. These include short-term, focused satellite missions versus long-term, continuous missions; research-driven measurements versus operation-driven measurements; systematic versus process studies; frequent technology insertion versus tried and true systems; “facility-class” instruments versus those developed in “principal investigator mode”; and in situ versus satellite remote sensing. All of these elements are needed because climate research requires both an understanding of complex, interrelated processes and systematic measurements to detect subtle changes in the Earth system. Many workshop participants thought the present organizational structure within the federal government was not adequate to manage this complicated arsenal of observing systems in a coherent manner. Responsibilities cross agency boundaries, and many organizations are tasked with specific components without sufficient infrastructure or expertise to manage these components. There is no forum in which to balance the responsibilities or to set scientific or policy priorities. Thus, the present U.S. strategy for climate observations relies on a set of largely ad hoc agreements as well as on missions whose primary purpose is not climate research and monitoring. NPOESS will be an enormous asset for observing Earth. Its stable orbit and commitment to the continuous observation of a broad range of climate-related geophysical variables are its most notable features from a climate research and monitoring perspective. However, NPOESS is organized primarily to deliver data products for use in short-term environmental prediction. Its mission does not include further exploitation of these products, although it is expected that they will be used by both civilian and military forecast systems. For NPOESS to realize its full potential for climate observation, agencies with different missions and cultures will need to work together effectively. Workshop participants were concerned about the adequacy of existing organizational structures to meet this challenge. Participants at the workshop also detailed the functional requirements for an organization that would support a climate observing system: 2   Information about NAOS is available online at <http://is1715.nws.noaa.gov/naos/>.

OCR for page 53
Issues in the Integration of Research and Operational Satellite Systems for Climate Research: II. Implementation Promoting the necessary insight into sensor characterization and performance. This includes involvement in the development of the sensor and the prelaunch tests. The goal is not to direct these activities but rather to gain insight into the process. Additional testing might be suggested to enhance the suitability of the sensors in climate research. Promoting access to sensor data, including operational data. The climate research community requires easy, affordable access to low-level sensor data to develop long-term consistent data records, as well as new data products. In most cases, climate data sets will be based on specific processing algorithms rather than on the operational algorithms necessary to meet the NPOESS objectives. Metadata describing the sensor and its operational history are also required. Promoting continuing data analysis and product validation. Subtle errors in algorithms or sensors often appear only in long time series. A vigorous data analysis program is necessary to ensure the long-term viability of the data records. Moreover, data product validation means more than just comparisons with ground truth for algorithm tuning; it also involves continuing assessments of temporal and spatial errors in the data. Supporting data reprocessing and long-term archiving. Data should be reprocessed to ensure a consistent data record in light of improved knowledge of sensor performance, as well as improved understanding of the physics underlying the algorithms. To study decadal-scale trends requires that the data and metadata be preserved and made accessible far into the future. Supporting the development of algorithms for climate data records (CDRs). NPOESS will be delivering data products primarily for use in near-real time. However, CDRs will be based on extensive analysis as well as ancillary data that may not be available until long after the data are collected. Such algorithm development is an essential component of climate research to ensure that the data records remain viable for studies of long-term trends. This also includes aggregation of data in time and in space at scales that are not part of the EDR specifications. Such development requires long lead times, perhaps as long as 10 years. Developing long-term data records suitable for climate research based on data from existing operational data streams, research missions, NPP, and NPOESS. For many critical data sets, there exist multiyear to decadal-scale records from operational satellites such as POES and DMSP as research systems such as TOPEX/Poseidon, SeaWiFS, UARS, and EOS. Specific attention should be given to bridging between these data records and NPOESS (including the various pre-NPOESS missions such as NPP and Windsat). Participants noted that the suggestions above would result in only a small percentage increase in the overall cost of NPOESS; however, they recognized that in absolute dollar terms the increase would be significant. Workshop participants considered the creation of three organizational units they thought would assist the NPOESS program in meeting the needs of climate researchers: climate science teams, a climate research working group, and a joint steering council. These ideas would require further consideration if the NPOESS agencies decided to pursue them. Climate Science Teams NASA and NOAA would select climate science teams through a peer-reviewed, competitive process. They would provide support for these teams to develop selected CDRs based on NPOESS measurements (including the pre-NPOESS missions) rather than simply on specific sensors. The composition and focus of the teams would change over time as the satellite systems move from definition to launch to postlaunch. The teams would provide advice to NASA, NOAA, and the IPO on sensor characterization, data product validation, and algorithm development and would also conduct research. In the near term, attention would be paid to the transition from existing satellite-based data records to data from the new satellite systems. The teams would be formed soon after the contractors are selected for the NPOESS sensors to ensure that scientific insight is provided into the sensor development and testing process.

OCR for page 53
Issues in the Integration of Research and Operational Satellite Systems for Climate Research: II. Implementation Climate Research Working Group The working group would consist of representatives from each of the climate science teams as well as additional representatives from NASA, NOAA, and the IPO. The participation of still other scientists might be sought to ensure that all areas of climate research (especially modeling and interdisciplinary research) are represented. The working group would focus on broad issues of importance for all of the climate science teams (e.g., data systems and archiving) and would advise agencies on these issues, as well as in areas where there are competing demands for limited funds, such as the reprocessing of data sets, overlap strategies for operational missions, and the definition of requirements for system replenishment from a climate research perspective. Issues to be studied might range from continuity of current research and operational missions to those that are still in the planning and development stage (such as NPOESS). The NPP could chart the way through the definition of agency responsibilities and relationships, the setting of priorities, and the development of an advice and decision-making process. The NPOESS total system performance requirements (TSPR) contractor would have to ensure that pre-NPOESS missions such as NPP and Windsat are a part of its activities. Joint Steering Council The joint steering council would be composed of agency managers and researchers from government and academia. It would define and evaluate the government’s strategy for long-term observations for climate research and monitoring. Relying on various national and international assessments, the council would identify key uncertainties and develop priorities for the observing strategy necessary to reduce these uncertainties. The joint steering council would work closely with NAOS to ensure that its recommendations were compatible with operational plans. The council would report to NASA and NOAA, as well as to other USGCRP agencies. It would consider both the satellite and nonsatellite components of the observing systems. Summary Although there are significant technical challenges in developing a climate observing system, workshop participants thought that the organizational issues were the most difficult because of the (1) substantial cultural differences from one agency to another and even within the agencies, (2) complexities of a multiagency budget process, (3) balance of civil and military interests that must be achieved and maintained within NPOESS, and (4) need for climate research to balance resources to enable the continuation of long-term data sets (operational programs) versus embarking on new scientific paths or employing new observational techniques (research programs). NASA and NOAA have both acknowledged these issues in letters (unpublished) to Neal Lane, the head of the Office of Science and Technology Policy. Workshop participants generally believed that interagency issues related to the roles and responsibilities for climate research would be resolved only with direction from higher levels within the executive branch. This belief was reinforced by the perception that agencies are already taxed in responding to changing budgets and programmatic directions. Workshop participants stressed the importance of having the climate research community both in the government and in academia play an active role in the design, implementation, operation, and evolution of the climate observing strategy. Although missions such as NPOESS initially were focused on requirements other than climate, workshop participants believed that climate requirements could also be included without disrupting the primary mission requirements. In this regard, they believed that pre-NPOESS missions such as NPP could play an important role in addressing both the science and programmatic issues associated with the transition (or, more accurately, the integration) of research systems to operational missions. Although NPOESS and its precursors are not ideal platforms for collecting every critical data set for climate research and monitoring, in many respects they are an adequate and, in some cases, a significant step in building a climate observing system. Stable orbit crossing times, data stability requirements (for some of the variables), and longer sensor/platform lifetimes are some of the NPOESS capabilities that represent a significant improvement

OCR for page 53
Issues in the Integration of Research and Operational Satellite Systems for Climate Research: II. Implementation from the present generation of operational satellite missions in terms of climate research needs. Workshop participants thought that an organizational structure that provided the necessary oversight and leadership should be able to meet the primary objective of the climate research community—developing viable data sets for research and monitoring. Workshop participants thought the three agencies involved with NPOESS—NASA, DOD, and NOAA—could create the infrastructure needed to use NPP as a platform for designing, testing, and evaluating an end-to-end capability that would allow using the NPOESS data stream for climate applications. Participants acknowledged the likely need to fine-tune existing NPOESS requirements but did not believe substantive revisions would be necessary to achieve much enhanced utility to climate researchers. Indeed, most agreed that the emphasis should be on establishing appropriate organizational relationships and responsibilities, setting in place processes for determining priorities, creating and using scientific bases for making decisions, and developing a processing system attuned to climate needs and the generation of climate products. SYNOPSIS OF DISCIPLINE GROUP DISCUSSIONS3 Report of the Atmospheres Discipline Group General Issues Following are summary comments of the atmospheres discipline group as compiled by its chair, Dennis Hartmann: The NPOESS program, if properly executed, has the potential to provide critical data for climate monitoring and research. It is perhaps the only program that can provide long time series of calibrated measurements for climate purposes. Every EDR to be used for climate purposes should specify stability in its definition. In the 1996 IORD (IPO, 1996), stability is included for many variables. In the recent update that the discipline group received from the IPO, long-term stability had too often been deleted. Accuracy and precision were sometimes replaced with a variable called uncertainty. The discussion of the specific EDRs below suggests stability requirements that are thought to be both useful and achievable. A data system is needed for taking the observations from the NPOESS data stream and putting them into an efficient, accessible, affordable archive and retrieval system that will allow for science investigations and reprocessing. This system should be capable of retaining the information necessary to reproduce the EDRs and to understand the calibration, validation, and processing. The required information is as follows: The raw radiances (RDR), Metadata on the NPOESS data, All calibration algorithms and data, The data processing algorithms, and Validation and calibration data. A data validation plan is needed for each variable of importance. Overlap between each succeeding instrument in each orbit is critical. It would be best to have 1 year of overlap to adequately sample EDR differences through all the seasons, although during periods of high solar activity, it might be necessary to have 2 years for measurements of solar irradiance. Insufficient overlap will lead 3   Three discipline groups met during the course of the workshop to consider the particular data needs for climate researchers studying processes occurring within and among the atmosphere, the oceans, and land. Synopses of the discipline groups’ views were prepared by their chairs for consideration by the Committee on Earth Studies. The committee is including these summaries in its report because they contain valuable ideas and suggestions. However, these summaries did not undergo the NRC review process; therefore, they should be interpreted only as the opinions of individual participants at the workshop.

OCR for page 53
Issues in the Integration of Research and Operational Satellite Systems for Climate Research: II. Implementation to a loss of data continuity needed for measuring variability on time scales ranging from interannual (for instance, for ENSO monitoring and prediction) to decadal (for global change studies). The stable orbits (i.e., fixed altitude and fixed equator crossing time) will be of great benefit to climate monitoring and research. Active research programs intimately related to the data production process are needed, particularly programs that focus on topics that will lead to the discovery and documentation of systematic errors in the NPOESS data stream. A thorough, well-documented instrument characterization should be performed before launch and retained for later use by researchers. If it can be accomplished without compromising data quality, some NPOESS channels should overlap the channels from heritage instruments that NPOESS will replace, such as AVHRR, SSMI, and MSU. Such an overlap would be helpful in resolving continuity-of-data issues. NPP General Issues A CERES-like instrument is needed on NPP to provide a continuous set of broadband Earth radiation budget measurements across the gap between EOS-PM and planned broadband measurements on NPOESS. The broadband measurement data set begins in 1978 with the Nimbus-7 ERB. There could be a gap between solar irradiance measurements from the the Solar Radiation and Climate Experiment (SORCE) mission and the NPOESS mission. SORCE is scheduled for 2002 to 2007 and NPOESS from 2009 onward. A total solar irradiance (TSI) data set has been continued since 1978 through a strategy of overlapping measurements and multiple redundant instruments in orbit at one time. Solar irradiance is a central climate forcing parameter, and it needs to be monitored precisely for climate detection and attribution purposes. Solar irradiance needs to be measured in the NPP/NPOESS time frame, which might be accomplished on an independent small satellite. From the perspective of atmospheric measurements for weather prediction and climate, it would be better for a number of reasons if the NPP were in a 1330 local time orbit rather than a 1030 orbit:4 NPP will provide a new, more capable interferometer sounding instrument. For sampling longitude and the diurnal cycle, it would be better to have the 0930 METOP interferometer and a 1330 NPP interferometer rather than a 0930 and 1030 configuration. In a 1330 orbit the NPP instruments would measure at the same local time and within 45 minutes of the heritage instruments on POES and EOS-PM and would also overlap the first NPOESS 1330 orbiter, providing a link between NPOESS and its predecessor instruments. This would allow NPP to provide inter-instrument calibration for the infrared sounders HIRS → AIRS → CrIS and the microwave sounders MSU → AMSU → ATMS, AVHRR → MODIS → VIIRS and also tie together broadband radiation measurements that may be taken from EOS-PM, NPP, and NPOESS. The POES and EOS instruments will not be available in a morning orbit at the time NPP is planned to be in orbit (2006 to 2011). Environmental Data Records The EDRs to be provided by NPOESS could be extremely important for climate monitoring and research. Because its operational missions require it to be up and operating continuously, NPOESS is particularly important for data continuity. The stable orbits are good for climate monitoring, and the three-satellite system gives adequate diurnal sampling of key variables. 4   The land-surface community would use the relatively low resolution global data from VIIRS to extend high-resolution data from other sources. They prefer the morning because cloud coverage over land is less in the morning than in the afternoon. Perhaps the planned Japanese instrument Global Imager might be used for this purpose.

OCR for page 53
Issues in the Integration of Research and Operational Satellite Systems for Climate Research: II. Implementation V40.4.1 Cloud Base Height Data from future missions using radar and lidar should help with this important variable. Inferences of cloud base height from microwave or other passive measurements involve a large amount of modeling and, therefore, are of limited use for climate monitoring or climate model validation. V40.4.3 Cloud Effective Particle Size Meets climate requirements; the 2 percent stability requirement is good. V40.4.6 Cloud Optical Depth The range should be at least 0 to 50, not 0 to 10. The variation from 10 to 50 is observable and important. Add a long-term stability requirement of 2 percent. Justification: Cloud optical depth is an important climate forcing parameter, and stable long-term measurements of it are needed. V40.4.7 Cloud Top Height Satisfies important climate requirements. V40.4.8 Cloud Top Pressure Satisfies important climate requirements. V40.4.9 Cloud Top Temperature Satisfies important climate requirements. 4.1.6.4.3 Cloud Ice Water Path The EDR described in the IORD satisfies important climate requirements. C40.3.4 Precipitation Change range to 0 to 75 mm/h rather than 0-50 mm/h. Add a stability requirement of the greater of 0.5 mm/h or 2 percent. Change the accuracy to threshold: 2 mm/h or 10 percent; objective: 1 mm/h or 5 percent. Change the precision to threshold: 1 mm/h or 5 percent; objective: 0.5 mm/h or 2 percent. Justification: Precipitation is a critical constraint on the hydrological and energy budgets of Earth and it needs to be known to these levels of precision, accuracy, and stability. A precipitation rate of 0.5 mm/h corresponds to a latent heat release rate of 350 Wm–2. 4.1.6.3.5 Cloud Liquid Water In the IORD the accuracy specifications are too large, and land and ocean are reversed. The table in the NPOESS climate workshop report (NOAA, 1997) is good. Change the accuracy to the greater of 0.025 mm or 10 percent over ocean; greater of 0.05 mm or 20 percent over land. Add a stability requirement of 2 percent.

OCR for page 53
Issues in the Integration of Research and Operational Satellite Systems for Climate Research: II. Implementation SRD3.2.1.1 Stratospheric Ozone The EDR described in the IPO draft satisfies important climate requirements. Tropospheric ozone is extremely important for health and climate. The discipline group suggests that a precision requirement of 10 percent be added. V40.3.1 Stratospheric Aerosols This category should be relabeled “Total Aerosol Burden.” Since it is derived from VIIRS, it will likely be dominated by tropospheric aerosols in most cases, but it will be difficult to distinguish tropospheric and stratospheric aerosols from solar reflection measurements. For measurement accuracy over land, the discipline group recommends changing the formula given in row V40.3.1.1-12 to a straight 0.2. This should be more realistic and easier to achieve. V40.3.1.2 Aerosol Size The EDR described in the IPO draft satisfies important climate requirements. The requirements should be relaxed over land, where the measurement is more difficult. 4.1.6.4.1 Albedo Surface Visible The product is useful, but for climate purposes a broadband albedo would also be useful. 4.1.6.4.2. Downward Longwave Radiation at the Surface The precision and accuracy thresholds should be relaxed to 10 Wm–2 and 15 Wm–2, respectively; the discipline group believes that values given in the IORD are unachievable. 4.1.6.4.3. Downward Shortwave Radiation at the Surface The precision and accuracy thresholds should be relaxed to 10 Wm–2. This is a modeled parameter and cannot be measured directly from space. 4.1.6.4.6 Outgoing Longwave Radiation Thresholds for broadband measurements should be precision, 5 Wm–2; accuracy, 2 Wm–2; and stability, 1 Wm–2. Justification: These are useful levels and are what the CERES team says it can produce. 4.1.6.4.6 Absorbed Solar Radiation Since these are given as instantaneous measurements within a range of 0 to 1400 Wm–2, the limits should be given in percentages. Thresholds for broadband measurements should be precision, 5 percent; accuracy, 1 percent; and stability 0.5 percent. Justification: These are useful levels and are what the CERES team says it can produce.

OCR for page 53
Issues in the Integration of Research and Operational Satellite Systems for Climate Research: II. Implementation 4.1.6.4.5 Solar Irradiance Solar irradiance is the radiated power incident on a radiometer aperture whose surface is orthogonal to the radiometer’s optical axis, which is aligned to the line of sight to the Sun from the aperture’s center. The total irradiance (i.e., the integral over all wavelengths) and the spectral irradiance from 0.2 to 2 μm are to be reported. This EDR supports monitoring of the total and spectral irradiance for determining solar influences on global change. These influences involve the entire radiation spectrum via varying mechanisms and are known to have strong wavelength dependencies. Solar variability can influence global surface temperatures directly because the visible and infrared radiation (longward of 0.3 μm) that constitutes the bulk of the total irradiance penetrates to Earth’s lower atmosphere and heats the surface, atmosphere, and upper ocean layers. Proper attribution of climate change thus requires reliable characterization of direct solar forcing by solar visible and near-infrared radiation. Variations in solar ultraviolet irradiance in the 0.2 to 0.3 μm range, which is absorbed by ozone in Earth’s atmosphere, is necessary to assess long-term ozone variations and for reliable detection of ozone depletion and recovery. As well, the measured total irradiance variability must be adjusted to account for variability of this radiation, which varies by an order of magnitude more than does the visible spectrum but does not reach Earth’s surface. Because of this, monitoring the ultraviolet spectral interval (0.2 to 0.3 μm) is a high priority for spectral irradiance monitoring. Also of high priority is monitoring the near-infrared band centered near 1.6 microns. This radiation emerges from the deepest layers of the Sun’s atmosphere, is thought to be the least variable part of the spectrum, and provides a measure of sunspot modulation of irradiance for interpreting the solar irradiance measurements. The Total Solar Irradiance Monitor is the sensor for this EDR (Table B.1). TABLE B.1 Solar Irradiance Systems Capability Threshold Objective Measurement range Total 1310-1420 Wm–2 1310-1420 Wm–2 Spectral (0.2-2 μm, bandpass 2-20 nm) 0-100 Wm–2 0-100 Wm–2 Measurement precision Total 0.002% (20 ppm) 0.001% (10 ppm) Spectral (0.2-2 μm) 0.02% 0.01% Measurement Accuracy Total 1.5 Wm–2 (0.1%) 0.15 Wm–2 (0.01%) Spectral (0.2-2 μm) 1% 0.1% Refresh 20 min stare, each orbit 50 min stare, each orbit   2 satellites 3 satellites Long-term stability Total 0.002%/yr (20 ppm/yr) 0.0005%/yr (5 ppm/yr) Spectral (0.2-2 μm) 0.02%/yr 0.01%/yr

OCR for page 53
Issues in the Integration of Research and Operational Satellite Systems for Climate Research: II. Implementation Report of the Ocean Discipline Group Following are summary comments by the ocean discipline group as compiled by its chair, Frank Wentz. Data System Archive All of the RDRs should be placed in a long-term archive with sufficient metadata so that future investigators can fully understand the characteristics of the data sets. The archive should be easily accessible at no cost (or minimal cost) for scientific research. Basic archival features, such as data subsetting, should be implemented. Climate Data System The requirements for high-quality research and climate products differ greatly from those for operational data products. Climate-specific algorithms may require ancillary data not available in real time. Furthermore, algorithm evolution and satellite intercalibation require continual reprocessing and refinement. Accordingly, a data system designed specifically for producing climate data records (CDRs) needs to be developed. In designing this data system, the NASA ESSIPs model should be considered as one possible prototype. Integrated Data Processing System The integrated data processing system (IDPS) for NPOESS should be sized so that it has the processing power to handle data from nonoperational spacecraft that may still have functioning sensors providing valuable data. These nonoperational data should not be discarded, but should be processed to RDRs for archiving. In this case, the IDPS only needs to produce the RDRs. Science Teams The NPOESS data products will be of a considerable benefit to the research and climate communities. To fully realize the benefit of these observations, science teams need to be formed. These teams will have responsibility for producing the CDRs, calibration and validation, and subsequent research. They should be selected via an open competition with peer review, similar to that associated with a NASA research announcement. Calibration and Validation Prelaunch There should be more scientific involvement in the prelaunch calibration and characterization of the NPOESS sensors. Oversight by the science teams of these critical measurements (antenna pattern and thermal-vacuum) will benefit both the NPOESS program and the research community. All engineering data from these prelaunch tests should be archived for future reference. Postlaunch Calibration Discussions should be held with the IPO about the possibility of doing on-orbit maneuvers so that the sensors can be calibrated to cold space and to the Moon. The stability of the Moon’s emissivity may allow calibration of out-instrument drift, thereby providing reliable long-term trends on climate variability. On the other hand, calibration maneuvers may negatively affect orbit stability, and their impact with respect to ocean altimetry measurements should be studied. The science teams should be responsible for intersatellite calibration after launch.

OCR for page 53
Issues in the Integration of Research and Operational Satellite Systems for Climate Research: II. Implementation TABLE B.2 Stability Requirements for Climate Data Records Parameter Trend Accuracy Spatial Resolution (km) Intersatellite Overlap (yr) Sea surface temperature 0.1 °C/decade 100 1 Ocean color (443 nm) 0.5 mW/cm2sr μm/decade 100 1 Ocean color (555 nm) 0.25 mW/cm2sr μm/decade 100 1 Ocean color (865 nm) 0.08 mW/cm2sr μm/decade 100 1 Ocean topography 5 mm/decade 100 See altimetry Ocean wind speed 0.5 m/s/decade 50 1 Ocean wind stress curl 10–9 N/m3/decade 50 1 Sea ice concentration 1%/decade 50 1 Stability Requirements for Climate Data Records Table B.2 gives the stability requirements for CDRs. The trend accuracy refers to the slope of a least-squares regression to the particular CDR over a long time baseline (> 5 years) at the specified spatial resolution. The intersatellite overlap is the minimum overlap time between successive satellites required to remove intersatellite biases. Altimetry The NPOESS Sun-synchronous orbit presents serious problems for doing climate research with altimetry. The fixed local time for the equator crossings will alias the tidal signal into the time series. Perhaps more serious, the NPOESS ground track is not the same as the TOPEX/Jason ground track, effectively breaking the 20-year time series for sea-level height. In addition, it is not clear if the NPOESS altimeter will meet the climate requirements of 3 cm radial orbit accuracy with a 1 km repeat cycle. For these reasons, the option of flying the altimeter for NPOESS on a separate, dedicated spacecraft, following the TOPEX/Jason ground track, should remain open until these technical and tidal aliasing issues are settled. Also, NASA should continue to sponsor technology development for the altimeter, as well as the work of the science teams. The requirement for an overlap between successive satellites is of particular concern for the altimeter. The Jason-2 mission will end in 2010, and the first NPOESS altimeter is not scheduled until 2011. In the discipline group’s view, a disconnect at 2010 would have devastating consequences for the long-term monitoring of sea-level variability. Furthermore, if the NPOESS altimeter does not repeat the TOPEX/Jason ground track, then a 3 to 5 year overlap would be required to intercalibrate the two altimeters. An overlap of only 0.5 years would be required if the NPOESS altimeter has the same ground track as TOPEX/Jason. In the discipline group’s view, these serious questions on the continuity of the altimeter climate mission need to be addressed jointly by NASA and the NPOESS IPO. Advantages of the NPOESS Preparatory Project (NPP) and NPOESS Continuity of Data Sets Many critical ocean data records that were begun as research missions will be continued with NPP and NPOESS. This includes phytoplankton chlorophyll (started with SeaWiFS and MODIS and to be continued with VIIRS on NPP) and combined microwave/infrared measurements of SST (begun on TRMM and to be continued with CMIS and VIIRS on NPOESS). These geophysical variables are characterized by low-frequency variability; multiyear gaps, which would likely occur if NPP and NPOESS were not to fly, would greatly inhibit the ability to study these climate-related processes.

OCR for page 53
Issues in the Integration of Research and Operational Satellite Systems for Climate Research: II. Implementation Improved Sensors CMIS is likely to be a significant improvement over the present operational passive microwave radiometer, SSM/I, which is on board the DMSP series. With its additional frequencies and improved sensor performance, CMIS will probably provide higher-quality data products. Similarly, VIIRS will likely have better capabilities than the AVHRR. However, all of these improvements are based on the discipline group’s assessment of the EDRs derived from these sensors. Until the details of the sensor designs are known, it will not be possible to make definitive statements. Equator Crossing Time Both the NPOESS and NPP platforms will maintain a constant equator crossing time. This capability greatly simplifies sensor calibration and improves long-term data continuity. The present generation of POES satellites does not maintain crossing times, and the orbits drift over the life of the platform such that missions that began with morning crossing times eventually have afternoon crossing times. Many apparent long-term trends in geophysical data sets derived from the POES sensors can be attributed to orbit drift. Colocation of Sensors Simultaneity has sometimes been used to justify the need for large platforms, such as the early plan for the Earth Observing System. However, except for rapidly changing phenomena such as clouds, there are few requirements for strict simultaneity of observations. Climate research does require contemporaneous observations, and because the processes and trends are often subtle, the combination of different measurement approaches to observe the same geophysical variable is advantageous. For example, the combination of passive microwave and infrared measurements improves SST retrievals. Moreover, climate research often focuses on the coupling of Earth system components. Thus, measurements of winds, SST, and ocean color are essential for understanding ocean productivity and its relationship to ocean mixing. The facility approach of NPP (and eventually NPOESS) to platform configuration means that many critical sensors will be in orbit at the same time. This not only simplifies program and platform management, but also ensures that critical data sets will be collected together to study coupled Earth system processes. Early Test of Programmatic Relationships NPOESS will eventually become an important element of the climate observing system. However, there is much to be done to define the roles and responsibilities of the various agencies if the integration of operational and research missions is to be successful. NPP can serve as a testbed for these activities, as attempts are made to develop climate-quality data products in an operational setting. This goal will require coordination among multiple agencies, and NPP could provide an early test. In addition to its observing elements, NPP should also serve as a testbed for ground data processing, archiving, and distribution. While the focus of the IPO is on an early test of the ground system capabilities necessary to meet its primary mission of short-term forecasting, NASA and NOAA should use NPP to develop methods to integrate climate science requirements into the ground system. This may mean the development of a parallel system that branches off from the main IPO system, which is focused on short-term forecasts. Report of the Land Discipline Group Following are summary comments of the land discipline group as compiled by its chair, Chris Justice. The NPOESS Preparatory Project mission is an important step forward in the transition from satellite missions whose primary purpose is scientific research on the evolution of Earth processes and the climate system and missions whose primary purpose is to monitor parameters necessary for short-term numerical weather prediction.

OCR for page 53
Issues in the Integration of Research and Operational Satellite Systems for Climate Research: II. Implementation The NPP should provide both scientifically interesting and useful data sets, as well as a base of experience for the operation of NPOESS that will serve to reduce risk in the final NPOESS configuration. Like NOAA and DOD, NASA has an important role to play in this transition, particularly in the identification and provision of systematic measurements for understanding the state and dynamics of climate. The three agencies, but perhaps especially NASA, need to create a management philosophy that is robust in the face of the inevitable peregrinations of their individual budgets, that can balance the sometimes competing demands of operational and climate users, and that can deliver and operate a series of missions that address long-term scientific needs without compromising the primary operational needs. NPP is the first mission in which these management philosophies, as well as the instruments and spacecraft themselves, can be tested. The VIIRS instrument will be the culmination of the sequence, AVHRR→MODIS→VIIRS (NPP)→VIIRS (NPOESS). These multispectral instruments have the general characteristics of relatively coarse spatial resolution, broad swath widths, and relatively fine temporal resolution. Careful calibration and periods of overlap of these instruments will be required for climate science. Measurements with these characteristics have been identified several times, most recently by the NASA science community in the Easton process,5 as crucial for understanding biological productivity and its underlying processes on both land and ocean. However, two other types of measurements have also been consistently identified as either crucial or important for terrestrial processes: a Landsat-class imager (but not an Enhanced Thematic Mapper (ETM+)), and a hyperspatial resolution sampler. The latter instrument, which has the potential to be important scientifically, clearly is being pursued by commercial vendors, but the former, while scientifically critical, does not appear for consideration in the NPOESS suite of instruments. These issues are examined in more detail below. Landsat and Instrument Synergies The Landsat class of polar-orbiting multispectral measurements is crucial for climate and global change research, especially that part related to the carbon budget, land-use change, and ecosystem impacts. Many of the changes in land cover occur on spatial scales that are small relative to the coarse pixel sizes of AVHRR, MODIS, and VIIRS, and, while those instruments should be able to identify where areas of rapid change are taking place, only Landsat-class imagers will be able to quantify the changes. These changes are important for climate: approximately 20 percent of the net addition of carbon to the atmosphere comes from land-use change, particularly deforestation in the tropics, and the putative “missing sink” is thought to be in part a land-use issue as well. The discipline group found strong arguments for making Landsat-class measurements more nearly operational. There exists a good data record dating back to 1972, and there is increasing awareness of the importance of capturing interannual variability in land-cover changes. Such an instrument could logically be included in the NPOESS suite for providing systematic environmental measurements. No agency has made a long-term commitment to providing Landsat-class measurements since Landsat-7 (however, NASA has identified the Landsat measurement as a high priority). The discipline group noted with interest the potential to fly “instruments of opportunity” on NPP and NPOESS utilizing spacecraft volume, weight, and power margins. From a terrestrial science perspective, this may be an opportunity to fly a lightweight, multispectral resolution imager with spatial resolution of about 30 m, that is, a Landsat-class imager. Colocation of a Landsat-class instrument with a moderate resolution imager like VIRRS is scientifically desirable but not crucial. Orbit Crossing Time AVHRR has flown in both 0730 and 1430 orbits. The land data record (1981 to the present) uses the 1430 overpass time based on availability of the data, rather than on what was scientifically optimal for the measurements themselves. The 1430 overpass will continue through NPOESS C1. 5   The Easton process refers to NASA’s post-2002 mission planning exercise, which culminated in an August 1998 meeting at Easton, Maryland. A link to the full report from the Easton Workshop is available online at <http://eospso.gsfc.nasa.gov/eos_homepage/misc_html/intro_kennel.html>.

OCR for page 53
Issues in the Integration of Research and Operational Satellite Systems for Climate Research: II. Implementation MODIS is slated to fly in both 1030 orbits on EOS-AM-1 (Terra) and a 1330 orbit on EOS-PM (Aqua). The land research community prefers the earlier crossing time for reduced cloud cover and more usable data sets. NPP is currently carrying a nominal 1030 crossing time, which would be very good for the land community interested in VIIRS. However, the NPOESS VIIRS is slated for 0530 and 1330 crossing times, neither of which is optimal for terrestrial observations. While METOP is slated for the 0930 crossing time in the final configuration, clarification is needed on its instrument capability. At the time of the workshop (July 1999), the discipline group understood that METOP would not have a VIIRS but would carry an AVHRR or similar instrument until 2018. Because the AVHRR is a much less capable instrument for terrestrial observations than the VIIRS is designed to be, this would clearly be a problem for the land community. Environmental Data Records There is an urgent need for the climate science community to evaluate the actual instrument specifications and proposed design. This need arises because the environmental data records for the VIIRS, as they are presented now, do not allow an understanding of the quality of the actual measurements. Needed for a more complete scientific understanding are comprehensive specifications for radiances (brightness temperatures). In addition, there need to be strict specifications for geolocation and band-to-band registration and for long-term stability of the measurements. While the discipline group recognizes that the 500 m spatial resolution threshold for land products is an improvement over earlier versions of the VIIRS documentation, the discipline group would have preferred a 250 m spatial resolution threshold and a specification for repeat cycle that nominally provides daily coverage of Earth at 1 km resolution or better. With respect to particular EDRs, the discipline group emphasizes that the fire detection capability that is part of the MODIS suite of measurements is necessary for climate science and should be part of the VIIRS suite. Improvements are needed for the proposed vegetation index and land-cover specifications from the standpoint of climate science and the long-term scientific record. While it is clear that the climate science community will want to make use of the operational EDRs to the maximum extent possible, other land products are also needed. Among these are leaf area index, albedo, percent green vegetation, and percent tree cover. These are unlikely to be of immediate operational use in the NPP time frame but are considered essential scientifically by the land discipline group. The IPO organizes EDRs by instrument. The discipline group found that this organization made it difficult to ascertain whether there are significant gaps or overlaps (from the standpoint of climate science) within the current complement of instruments. The discipline group suggests that NASA and NOAA examine the EDRs by cross-cutting scientific issues (e.g. carbon cycle, land-atmosphere interactions, clouds, aerosols, and radiation) to ensure that there are no major inconsistencies. Calibration and Characterization One of the thorniest problems inherent in the NPOESS mission is the degree of instrument characterization and calibration that will be required for climate science. The discipline group recognizes that a fair degree of instrument characterization and calibration will be done, but what is required is a full plan for each activity from the standpoint of the needs for long-term data sets required for climate science. For example, discipline group members thought it essential that there be stability specifications for radiances from VIIRS. It would be very desirable to obtain periodically a lunar view or deep-space view for calibration purposes, and it was unclear from the briefings and available material whether this is in the plans for NPP and ultimately for NPOESS. A calibration plan should include the means of doing on-board calibration utilizing a solar diffuser and vicarious calibration through ground programs and field campaigns. Geolocation and postlaunch geolocation validation are critical elements. A science team could provide valuable advice to NASA, NOAA, and the IPO on these elements.

OCR for page 53
Issues in the Integration of Research and Operational Satellite Systems for Climate Research: II. Implementation Instrument characterization is equally important. The experience with the ETM+ on Landsat-7 and with MODIS on the NASA Terra mission is that the full participation of a science team in instrument characterization is crucial for ensuring that the measurements ultimately derived are maximally useful for scientific purposes. With respect to VIIRS, the land discipline group stresses the importance of allowing sufficient time to correct problems that may be identified after the analysis and interpretation of test results. Climate Data System The land discipline group finds the data system envisioned by the IPO deficient in several respects in terms of providing the services needed for climate science. In particular, the discipline group believes that the lack of plans for a long-term archive or the capability to reprocess data as algorithms mature would impede the ability of NPP or NPOESS to serve the needs of climate science. Discussions focused on a separate data stream for climate products that would be linked closely to a science team responsible for analysis. Such a data stream would originate at level 1a/1b of the existing data stream, so it would be necessary for these data to be of climate quality. In addition to the land surface measurements themselves, the climate data stream would need to be combined with comprehensive data on the instrument itself in order for anomalies due to variations in instrument and spacecraft characteristics to be identified. The discipline group believes that the design for such a data system should be initiated as soon as possible and involve both NASA and NOAA at a minimum. The discipline group does not envision a data system as ambitious as that originally designed for EOS and believes that design of the data system would benefit from study of earlier experiences with EOSDIS and the Earth System Science Information Partnerships, as well as from study of the planned NewDIS. Validation Validation of data products is as important to the climate science community as the calibration of the VIIRS instrument itself. There is an enormous opportunity for NPP/VIIRS to build on existing EOS validation activities and also on activities that are under way elsewhere in NASA, NOAA, and DOD. In addition, the calibration/ validation working group of the Committee on Earth Observation Satellites (CEOS)6 provides a crucial forum for international cooperation. The discipline group notes that part of the reason that cooperation on validation activities is so important is that they can be, and arguably should be, a large budget item. The discipline group suggests that opportunities to leverage both government and private-sector efforts be investigated to support validation efforts for NPP/VIIRS. Need for a Climate Science Team In the discipline group’s view, the development of VIIRS would clearly benefit from having a science team. Such a team could stand alone or be part of an overall science team for the NPP mission. Roles for the science team include the following: Providing guidance for instrument characterization and calibration efforts; Providing a climate science review of level 1 data; Assessing the EDRs for climate use and suggesting improvements; Providing algorithms for additional products that are necessary for climate; Undertaking climate product validation; and Assisting in the planning and implementation of the climate data system. 6   CEOS was created in 1984 as a result of the international Economic Summit of Industrialized Nations and serves as the focal point for international coordination of space-related Earth observation activities. Information about CEOS is available online at <http://www2.ncdc.noaa.gov/CEOS/ceospg1.html>.

OCR for page 53
Issues in the Integration of Research and Operational Satellite Systems for Climate Research: II. Implementation References Integrated Program Office (IPO), National Polar-orbiting Operational Environmental Satellite System (NPOESS). 1996. Integrated Operational Requirements Document (First Version) (IORD-1) 1996. Issued by Office of Primary Responsibility: Joint Agency Requirements Group (JARG) Administrators, March 28. The updated IORD and other documents related to NPOESS are available online at <http://npoesslib.ipo.noaa.gov/ElectLib.htm>. National Oceanic and Atmospheric Administration (NOAA). 1997. Climate Measurement Requirements for the National Polar-orbiting Operational Environmental Satellite System (NPOESS), Workshop Report. Herbert Jacobowitz (ed.), Office of Research and Applications, NESDIS-NOAA, Washington, D.C. National Research Council (NRC). 1998. Overview, Global Environmental Change: Research Pathways for the Next Decade. National Academy Press, Washington, D.C. National Research Council (NRC). 1999. Adequacy of Climate Observing Systems. National Academy Press, Washington, D.C. WORKSHOP PARTICIPANTS Mark Abbott, Oregon State University/Chair, CES Ina B. Alterman, National Research Council/Space Studies Board Peter Backlund, Office of Science and Technology Policy Tony Busalacchi, NASA/Goddard Space Flight Center Art Charo, National Research Council/Space Studies Board John Christy, University of Alabama at Huntsville/CES Bob Corell, National Science Foundation H. Lee Dantzler, NOAA/NESDIS Michael H. Freilich, Oregon State University Joe Friday, National Research Council/Board on Atmospheric Sciences and Climate Jim Giraytys, NAOS/Consultant Arnold Gruber, NOAA/NESDIS Mike Haas, NOAA/Integrated Program Office/Aerospace Robert Harriss, National Center for Atmospheric Research Dennis Hartmann, University of Washington Craig Herbold, National Research Council/Space Studies Board Sarah Horrigan, Office of Management and Budget Tony Janetos, World Resources Institute John Janowiak, NOAA/National Weather Service Chris Justice, University of Virginia/CES Tom Karl, NOAA/National Climatic Data Center Jack Kaye, NASA Headquarters Michael King, NASA/Goddard Space Flight Center Judith Lean, Naval Research Laboratory Jerry Mahlman, NOAA/Geophysical Fluid Dynamics Laboratory Martha Maiden, NASA Headquarters Jon Malay, Ball Aerospace Stephen Mango, NPOESS/Integrated Program Office Bruce Marcus, TRW (retired)/CES Alvin (Jim) Miller, NOAA/National Weather Service Ralph Milliff, National Center for Atmospheric Research/CES Gary Mitchum, University of South Florida Berrien Moore, University of New Hampshire Robert E. Murphy, NASA/Goddard Space Flight Center Craig Nelson, NOAA/NPOESS/Integrated Program Office Dallas Peck, U.S. Geological Survey (retired)/CES Walt Planet, NOAA/NESDIS

OCR for page 53
Issues in the Integration of Research and Operational Satellite Systems for Climate Research: II. Implementation V. Ramanathan, Scripps Institution of Oceanography/UCSD P. Krishna Rao, NOAA/NESDIS Eugene Rasmusson, University of Maryland Dick Reynolds, NOAA/National Climatic Data Center Gary Rottman, University of Colorado Stan Schneider, NPOESS/Integrated Program Office/NASA Larry Scholz, Lockheed Martin (retired)/CES Phil Schwartz, Naval Research Laboratory Roy Spencer, NASA/Manned Space Flight Center Detlef Stammer, Scripps Institution of Oceanography/UCSD Larry Stowe, NOAA/NESDIS Dan Tarpley, NOAA/NESDIS Ray Taylor, NASA/Goddard Space Flight Center Susan Ustin, University of California at Davis/CES Frank Wentz, Remote Sensing Systems/CES Bruce Wielicki, NASA/Langley Dave Wilkinson, Stirling Strategic Services Greg Williams, NASA Headquarters Richard Willson, Columbia University Jim Yoder, University of Rhode Island