Summary of the Workshop Sessions
The workshop’s breakout sessions were designed to have participants consider the impact of changes to the NPOESS and GOES-R programs from many different perspectives. On day 1 of the workshop, participants considered impacts in terms of their effects on the measurement of essential climate variables (ECVs), as specified by the GCOS Implementation Plan.1 On day 2, impacts were considered in terms of the specific sensors that constituted the original programs’ baselines. The panel recognized that there would be overlap in these discussions, but thought it useful for participants to consider the broad issues of ECV measurement and development of climate data records (CDRs) apart from specific concerns about NPOESS sensors. Day 3 breakout discussions were more loosely organized, to allow for broad discussion of cross-cutting issues, long-term considerations critical to the production of CDRs, and the advance of climate science in general. Indeed, a recurring theme expressed by many participants at the workshop was that ensuring the measurement(s) of a particular climate variable(s) was only a necessary first step toward enabling the creation of time series of measurements of sufficient length, consistency, and continuity to determine climate variability and change, that is, to generate CDRs (see “Panel on Issues Related to CDR Development,” p. 39).
WORKSHOP SUMMARY—DAY 1
The day 1 breakout groups were charged to consider, as a community, the various ECVs that might be affected by the Nunn-McCurdy NPOESS and GOES-R descopes. Participants considered each NPOESS-measured parameter, starting with ones in jeopardy of not meeting Integrated Operational Requirements Document (IORD) specifications, commenting on the relevance of the parameter to climate science and/or long-term climate records, the importance of maintaining the IORD-level value (and potential consequences if it is not met), and noting any additional considerations required to make the NPOESS program’s environmental data records (EDRs) more relevant to GCOS ECV climate parameters and to the climate community as a whole (e.g., additional instrument characterization, calibration, overlap requirements). Participants were also encouraged to suggest mitigation approaches where NPOESS current plans fall short of climate community needs, and to assess whether any of the missions recommended in the Earth science decadal survey2 might enable recovery of the NPOESS climate
The GCOS Implementation Plan (GCOS-107) is available at http://www.wmo.int/pages/prog/gcos/Publications/gcos-107.pdf.
NRC, Earth Science and Applications from Space: National Imperatives for the Next Decade and Beyond, The National Academies Press, Washington, D.C., 2007.
measurements. Participant feedback on each of these areas was captured in real-time in a template,3 and a brief summary of the discussions is provided here.
Consideration of NPOESS and GOES-R Priority Measurements for ECVs— Breakout Sessions
Climate Data Records Related to Observations of the Atmosphere
The atmosphere ECV breakout group was asked to consider 10 ECVs related to observations of the atmosphere: Earth radiation budget (including solar irradiance); aerosol properties; ozone; carbon dioxide, methane, and other greenhouse gases; cloud properties; precipitation; water vapor; surface wind speed and direction; upper-air wind; and upper-air temperature. Recognizing the linkages between the ECVs, the group organized itself into four subgroups:
Radiation budget (Earth radiation budget, aerosol properties),
Ozone and trace gases (ozone; carbon dioxide, methane, and other greenhouse gases),
Clouds and precipitation and water vapor (cloud properties, precipitation, water vapor), and
Winds and temperature (surface wind speed and direction, upper-air wind, upper-air temperature).
A summary of the discussions is provided here, organized according to ECV.
Earth Radiation Budget (Including Solar Irradiance)
Persistent small climate changes are difficult to detect within the diurnal, regional, and seasonal variance of Earth’s reflected (shortwave) and emitted (longwave) energy—hence a continuous long-term (decades) record of Earth’s radiation budget (ERB) is needed to identify subtle long-term shifts related to climate change.4 With the demanifesting of TSIS and ERBS from NPOESS, ERB measurements will end with the last CERES on Aqua (or perhaps NPP, pending addition of CERES FM-5 onto NPP), the TIM record will end with Glory, and the SIM record with SORCE. Planned or proposed international missions and instruments of relevance include EarthCARE, ScaRAB on Megha-Tropiques, and GERB; however, in the view of breakout participants who commented on them, these international missions are insufficient to maintain the ECVs. The Earth science decadal survey recommended that NOAA add CERES to NPP and that NASA develop CLARREO, which would provide spectral ERB measurements. It was noted that ERBS (Earth radiation budget sensor) needs VIIRS cloud imagery, and so flight near NPOESS was desirable. SIM and TIM could be on separate spacecraft from ERBS since they are Sun pointing.
Measurement of aerosol properties is needed to understand the global distribution of aerosols and their impact on Earth’s energy balance, clouds, and precipitation. Aerosol impacts remain a source of major uncertainty in climate prediction in the Intergovernmental Panel on Climate Change (IPCC) 4th Assessment Report (2007).5 Recent and ongoing missions and instruments providing aerosol information include TOMS (1979-), AVHRR (1979-), MODIS (1999-), MISR (1999-), POLDER (2002-), (A)ATSR (1991-), PARASOL (2006-), SCIAMACHY (2003-), CALIPSO (2006-), GLAS (2003-), OMI (2004-), and AIRS (2002-). International missions of relevance include EarthCARE, GCOM-C/SGLI, ADM/Aeolus, and ATLID. The upcoming NASA Glory mission will fly APS, which was originally intended to be followed by subsequent NPOESS flights of APS to provide a continuing data record. With the demanifesting of APS from NPOESS, some aerosol information will be obtained through
The filled-in templates are available at http://www7.nationalacademies.org/ssb/SSB_NPOESS2007_Presentations.html.
See, for example, NRC, Solar Influences on Global Change, National Academy Press, Washington, D.C., 1994.
Intergovernmental Panel on Climate Change, Climate Change 2007, IPCC Fourth Assessment Report, Cambridge University Press, Cambridge, U.K., 2008, available at http://www.ipcc.ch/ipccreports/assessments-reports.htm.
VIIRS, OMPS, and CrIS/ATMS; however, these instruments will not provide polarimetry information. Workshop participants noted that the ACE mission, as described in the Earth science decadal survey, would provide significant advances. Attendees expressed a strong desire to move to a next-generation polarimeter rather than lock in to the technology of APS, as would have been required for accommodation on NPOESS. The 3D-Winds mission recommended in the decadal survey would provide aerosol heights, which would also contribute to measurement of the properties of this ECV.
The ozone ECV is important to monitoring the long-term trends in surface ultraviolet (UV) radiation and recovery of the ozone layer. The ozone ECV is at risk due to the demanifesting of OMPS-Limb by the NPOESS program, although it has recently been restored to the NPP platform. After NPP, no ozone profile measurement is currently planned as part of NPOESS, which after the Nunn-McCurdy action carries only the OMPS-Nadir portion of the original suite. Ongoing missions and instruments of relevance to the ozone ECV include TOMS (1979-), SBUV (1979-), GOME (2006-), MIPAS (2003-), OMI (2003-), SCIAMACHY (2003-), TES (2005-), GOME-II (2006-), MLS (2004-), AIRS (2002-), and IASI (2006-). The decadal survey recommendation for GACM was considered relevant to the ozone ECV, although it was recommended for launch after 2016. In the breakout session, several participants noted that the NPOESS nadir ozone measurement (which is the only ozone measurement to be made by NPOESS) is more than adequately covered by GOME-II on MetOp and that ozone profile measurements would add more value than additional nadir measurements.
Carbon Dioxide, Methane, and Other Greenhouse Gases
Measurements of key greenhouse gases, including CO2 and CH4, are essential parts of a program to understand climate forcings and trends. Indeed, measurements are needed with sufficient quality to detect sources and sinks at regional scales. The NPOESS CrIS instrument will contribute to this ECV, and some breakout participants noted that its value would be increased if all the spectra were downlinked. Ongoing missions and instruments related to the greenhouse gases ECV include IRS (2002-), SCIAMACHY (2003-), MIPAS (2003-), HIRDLS (2004-), MLS (2004-), TES (2004-), GOME-II (2006-), and IASI (2006-). AIRS and IASI both currently produce midtroposphere CO2 data products, although both remain to be validated. NASA’s planned OCO mission (scheduled for launch late in 2008) and the JAXA GOSAT mission will also contribute to the CO2 measurement needs for this ECV. The decadal-survey-recommended ASCENDS mission is also of interest. Some workshop participants noted the desirability of a GIFTS- or HES-like instrument for geostationary measurements (with high temporal resolution) relevant to this ECV.
Ongoing missions and instruments of relevance to the cloud properties ECV include AVHRR/HIRS (1978-), (A)ATSR (1991-), MODIS (2000-), MISR (1999-), AIRS (2002-), SEVIRI (2003-), GOES (1994-), METSAT (2004-), MTSAT-1R (2005-), IASI (2006-), CloudSat (2006-), and CALIPSO (2006-). On NPOESS, contributions include VIIRS (which includes a day and a night imager) and CrIS/ATMS (and, prior to the Nunn-McCurdy action, APS). Planned missions/instruments of relevance include GLM and EarthCARE. The cloud properties ECV can be significantly advanced via the ACE mission recommended by the Earth science decadal survey, which would investigate aerosol-cloud interactions.
The water cycle plays a critical role in climate change. Precipitation measurements are key to understanding and predicting water vapor feedback, water supply, drought, severe storms, and floods. Ongoing missions and instruments of relevance to precipitation measurement include SSM/I (1987-), TMI (1997-), AMSR-E (2002-),
TRMM (1997-), CloudSat, (2006-) MODIS (1999-), and AIRS (2002-) (the last two provide important information on clouds and water vapor). NASA’s upcoming GPM mission is of great relevance to this ECV. International plans for GCOM-W and AMSR F/O (2011) are also of interest. Participants discussed the relevance of the decadal survey’s recommended ACE and PATH missions, which would provide important information on aerosol-cloud interactions and high-temporal-resolution precipitation, respectively. NPOESS CrIS/ATMS measurements will contribute to the precipitation ECV, but questions remain about the still-undefined MIS capability. Some participants expressed concern that a capability for passive microwave precipitation measurements may not emerge in the revised MIS sensor, and they suggested that NPOESS place emphasis on the water cycle (water vapor, liquid water, ice water, and precipitation) when considering MIS requirements, possibly including giant magneto-impedance (GMI) bands.
With measurements available through CrIS/ATMS on NPOESS, IASI on MetOp, and ABI on GOES-R, there was little concern expressed about the water vapor ECV. MIS capabilities, still uncertain, should include total column water vapor information. Several participants suggested that the water vapor channel be added back to VIIRS to further strengthen the water vapor ECV, while also benefiting wind and aerosol measurements. Ongoing missions and instruments of relevance include SSM/I (1987-), SSMIS (2003-), (A)ATSR (1991-), AMSR-E (2002-), MERIS (2002-), HIRS (1979-), AIRS/AMSU (2002-), MODIS (1999-), TMI (1997-), and MLS (2004-). International plans include GCOM-W and AMSR F/O (2011). Decadal survey missions of relevance include GPS/RO and PATH.
Surface Wind Speed and Direction
Measurements of surface wind speed and direction are needed for both climate and operational purposes. For climate, vector winds are required to compute wind stress curl, an essential climate quantity that drives Ekman pumping and suction in the ocean, thereby implying vertical circulations (i.e., upwelling and downwelling). The zonal integral (east to west) of wind stress curl across an ocean basin is proportional to the western boundary current transport (i.e., the transport responsible for the dominant part of the poleward heat flux by the ocean). The climatology of storms (frequency and intensity) depends on vector wind measurement, and measurements are required in all conditions. Several participants noted that the CMIS replacement (MIS) is not expected to meet needs for data on these variables. Several participants also noted that the NPOESS key performance parameter is wind speed only, and so measurement of wind direction is not ensured as trade-offs are explored. Ongoing missions and instruments of relevance to this ECV include QuikSCAT (1999-), ERS (1992-), and WindSat (2003-). The international ASCAT6 measurement and the decadal survey recommendation for XOVWM were also discussed. Participants engaged in a lively debate over the relative merits of passive versus active measurement of surface wind speeds; they also discussed the merits of a future system that would combine the active measurement capabilities of ASCAT with the passive measurements to be provided by MIS. It was the strongly held view of many workshop participants that ASCAT and MIS would be inadequate to meet both operational and climate needs, and that an additional active surface wind speed and wind direction measurement was needed. This ECV was also considered by the oceans breakout group and is further discussed in the summary of its session below.
Three-dimensional upper-air wind, temperature, and moisture profiles with high vertical and temporal resolution are key to improved prediction of hurricane track and intensity. The upper-air wind ECV is at moderate risk due to its partial reliance upon both NPOESS/VIIRS (which lacks the water vapor band needed to continue
MODIS measurements of polar winds) and GOES-R/HES (for continuous full-disk four-dimensional wind vertical profiling, including diurnal coverage). GOES-R/ABI will provide cloud wind tracking and measurements of clear-sky water vapor layer-integrated winds, including diurnal coverage. Ongoing missions and instruments of relevance to the upper-air wind ECV include AVHRR (1979-), MODIS (1999-), (A)ATSR (1991-), GOES (1975-), Meteosat (1978-), GMS (1980s-), Feng Yun (2000s-), and INSAT (2000s-). The international ADM/Aeolus mission is relevant to this ECV, as is the 3D-Winds mission recommended by the Earth science decadal survey.
The upper-air temperature ECV appears to be in good health with the planned flight of CrIS/ATMS on NPOESS and IASI on MetOp, although several participants noted that the inadequate diurnal coverage could be improved by addition of CrIS to the early AM (0530, descending) NPOESS spacecraft. Ongoing missions and instruments of relevance to the upper-air temperature ECV include MSU (1979-), AMSU (1999-), CHAMP (2001-), COSMIC (2006-), GRAS (2006-), HIRS (1979-), and AIRS (2002-). The decadal survey recommendations for GPS/RO, CLARREO, and PATH are also considered relevant to this ECV. Some participants noted that a geosynchronous Earth orbit (GEO) flight of opportunity to fly GIFTS or another Pathfinder could further recover ability to observe and integrate upper-air temperature across the diurnal cycle.
The breakout group also discussed air quality observation needs, though noted that air quality is not currently a GCOS ECV.
Climate Data Records Related to Observations of the Oceans
The oceans ECV breakout group was tasked to consider six ECVs related to ocean observations: sea level, SST, ocean color, salinity, sea state, and sea ice. Some participants also noted the need for ocean measurement input to several atmospheric ECVs (surface wind speed and direction, precipitation, surface radiation, surface air temperature, and water vapor). A summary of the discussions is provided below, organized according to ECV.
The 15-year record of sea surface height has provided a record of global sea level rise, built on TOPEX and Jason-1 data records. Discussions at the breakout focused on measures to ensure the continuity of this record, a strong desire among most participants. Ongoing missions and instruments of relevance include Jason-1, ENVISAT, and GFO. NASA plans include a Jason follow-on, the Ocean Surface Topography Mission (OSTM)/Jason-2. There are international plans for an accurate altimeter aboard the European Sentinel-3,7 although it will suffer from tidal aliasing due to a Sun-synchronous orbit. The decadal survey recommendation for a NASA advanced altimetry mission called SWOT is also of key interest. Altimeters on NPOESS could help to provide global coverage and measure ocean heat content. However, the removal of the altimeter from NPOESS is not considered a critical issue for climate, as ALT would not have provided a climate-quality sea surface height record due to the NPOESS Sun-synchronous orbit, nor would it have provided information about inland waters and near-coastal areas. For measurements related to the needs of climate researchers, most breakout participants expressed a preference for free-flyer missions that achieve the same quality as Jason, either as a series of Jason follow-on missions such as Jason-3 followed by SWOT, or as a series of SWOT missions, started by advancing the timeline for the first SWOT mission.
Sea Surface Temperature
Remote sensing of sea surface temperature (SST) has a long heritage, dating back to 1980. Climate studies require all-weather SST coverage, involving complementary infrared (IR) and microwave observations. IR obser-
For information on the European Space Agency’s planned Sentinel series, see http://www.esa.int/esaLP/SEMZHM0DU8E_LPgmes_0.html.
vations provide high spatial resolution and radiometric fidelity in clear skies, and microwave observations provide SST measurements in the presence of clouds and aerosols. Ongoing missions and instruments of relevance include AVHRR (1979-), (A)ATSR (1991-), Aqua/AMSR-E (2002-), MSG/SEVIRI, GOES imagers, TRMM/TMI, TRMM/VIRI, and Aqua/Terra MODIS (1999-). International plans include OceanSat-1 and -2, Sentinel-3 series (2013-2020), MetOp (B, C, D), GCOM-C, and GCOM-W/AMSR-2. The decadal survey PATH mission is also of interest. On NPOESS, MIS will replace the canceled CMIS (but currently is not slated for inclusion until the second NPOESS spacecraft launches in 2016). Of particular concern to many workshop participants was the expectation that the certified NPOESS MIS configuration will lack the desired band for passive microwave SST (6.9 GHz), which would create a gap in the SST record. Many participants also suggested the need for sustained daily global coverage of the IR observations. Continuity of both IR and passive microwave SST observations on polar and geostationary platforms was considered by many participants to be essential for an accurate and robust SST CDR, as also noted by the International GHRSST-PP science team.8 Continuity by CMIS/MIS with current AMSR-E observations remains a major concern.
Tracking of trends in ocean productivity via remote sensing of ocean color is an important aspect of ocean climate study. Measurements of water-leaving radiances are needed, and some participants expressed a desire for a more comprehensive approach than observation and monitoring of chlorophyll. Ongoing missions and instruments of relevance include SeaWiFS (1997-), MERIS (2002-), and Aqua/MODIS (2002-). International plans for OceanSat-2, Sentinel-3, and GCOM-C/SGLI are also of interest, as is the ACE mission recommended by the decadal survey. Ocean color measurements were to be provided by NPOESS/VIIRS and GOES-R/HES. The ocean color ECV is considered at risk due to removal of HES from GOES-R. Ocean color scientists noted that the NPOESS platform and its VIIRS sensor will not be satisfactory for ocean color science, in part because NPOESS does not provide for lunar calibration of VIIRS and in part because of VIIRS hardware issues involving increased optical cross-talk.9 Ocean color researchers at the workshop asserted that observations should have band coverage ranging from UV to shortwave, and they suggested modifying the GCOS ECV to include ocean color records beyond chlorophyll. The ocean biology scientists who were present suggested development of a dedicated ocean biology sensor and mission to accommodate the need for lunar calibration, building on the approach taken by the SeaWiFS instrument. In situ calibration with ocean buoys is also an important consideration.
Measurement of sea surface salinity is a new capability. The European Soil Moisture and Ocean Salinity (SMOS) mission and the NASA Aquarius mission will provide the first satellite sensing of sea surface salinity (which will require measurements of surface wind speed and SST as part of the retrieval process). There is as yet. no satellite climate record to evaluate the results of these missions.
As winds over the ocean change in response to climate variability and climate change, there will be changes in sea state. The sea state is important for marine weather and for the safety of life at sea, forecasts and warnings.
For information on the Global High-Resolution Sea Surface Temperature Pilot Project, see http://ghrsst-pp.org.
In remote sensing, optical cross-talk is an important error source that results when a detector responds to impinging light from out-of-channel wavelengths (e.g., due to scattering, internal reflections, or other optical leaks). This out-of-channel component of the detector signal can be difficult or impossible to de-convolve with the in-channel (desired) signal. At the time of the workshop, VIIRS was at risk of not meeting the instrument requirement that limits the level of acceptable optical cross-talk. The optical filter assembly in VIIRS, which separates incoming signal into a number of smaller wavelength channels, is known to be the source of the optical cross-talk problem. Efforts are underway to seal light leaks and reduce scattered light. If the VIIRS optical cross-talk issue is not resolved, ocean color and aerosol products will be adversely affected.
Some participants questioned whether the sea state ECV represented a fundamental measurement. From a climate perspective, the roughness of the sea surface plays a role in air-sea exchanges. It would be ideal to have full wave directional spectral capability, spanning surface gravity wave and surface swell periods. This is not at present a satellite capability.
With MIS delayed until NPOESS C2, there is a need to continue the long (28-year) climate data record of sea ice extent and concentration collected by passive microwave radiometers; continued scatterometer and altimeter measurements are also required. Changes in sea ice and ice coverage are a critical indicator of climate change. Ongoing missions and instruments of interest include SMMR, SSM/I (DMSP), SSMIS, AMSR-E, QuikSCAT, MODIS, and ASCAT. Planned missions include the DMSP missions, F19 and F20, carrying SSMIS; GCOM-W/AMSR-2; GCOM-C/SGLI; RADARSAT-2; and CryoSat-2. The decadal survey recommendations for SCLP, ICESat-II, XOVWM, and DESDynI are also of interest. With MIS delayed, a passive microwave data gap is anticipated. A synthetic aperture radar or equivalent capability is also needed in the production of the sea-ice climate data record for validation of sea ice concentration and edge. This could be provided by the XOVWM scatterometer. To fill the gap, a free-flyer QuikSCAT replacement combined with an AMSR-type instrument would be a backup against DMSP failures.
Surface Wind Speed and Direction
From an oceanographic perspective, there is a need for vector wind measurements, and many participants noted that surface vector winds from passive microwave did not fulfill the need for climate-quality surface vector winds and for observation of extreme weather events. Thus, to these participants, the removal of CMIS from NPOESS was not a major issue. Many of the breakout group’s participants indicated the real need to enhance climate measurement capabilities beyond the QuikSCAT standard in a follow-on, active radar surface vector wind mission. The QuikSCAT mission has provided an 8-year record to date and has exceeded its design lifetime. Follow-on options discussed included relying on ASCAT on MetOp, duplicating QuikSCAT, and flying XOVWM (as recommended by the Earth science decadal survey). The XOVWM option has the advantages that that sensor can measure higher wind speeds than can QuikSCAT, can provide improved vector wind retrievals in rain, and can detect surface rain rate. Higher spatial resolution (~1 km) is also desired. It was also noted that the incremental cost of XOVWM versus a QuikSCAT duplicate would be small, in part because QuikSCAT was designed and developed more than a decade ago.
Precipitation, Surface Radiation, Surface Air Temperature, and Water Vapor
Simultaneous knowledge of the surface forcing of the ocean (heat, water, momentum fluxes from the atmosphere) and ocean-atmosphere exchange is important to monitoring and understanding the ocean’s role in climate. Global ocean remote sensing coverage of rainfall, surface incoming and net shortwave and longwave radiation, and latent and sensible heat fluxes is needed. Latent and sensible heat flux can be parameterized given surface wind, SST, and surface air temperature and humidity. The oceanographic community supports collection of climate-quality surface radiation and rainfall fields. It remains a significant challenge to retrieve surface air temperature and surface humidity from space, and existing data are not considered to be of the quality needed to generate CDRs.
Some participants felt that the requirements to instrument selection process did not sufficiently engage the ocean climate user community, and they expressed a continuing need for this engagement to ensure that the missions flown support collection of climate-quality data records. NASA science teams are one model to ensure such engagement. The science team approach has worked particularly well in terms of federating international activities
for several CDRs, including SST (the GHRSST-PP), ocean color (International Ocean Colour Coordination Group (IOCCG)), and altimetry (Ocean Surface Topography Science Team (OSTST)).
Further, some participants noted that for SST, sea ice, and ocean surface vector winds there is possible synergy and an optimum combination for accuracy, data gap limitation, spatial and temporal resolution, and CDR continuity that should be considered. All three of these CDRs would benefit from sensor collocation. A solution would be to pursue XOVWM and AMSR-type sensors on the same satellite or in formation, and in polar orbit. This approach would entail acceleration of the XOVWM schedule. Another approach would be to modify XOVWM to accommodate passive microwave (6.9 GHz) SST with surface wind speed (required for accurate SST retrievals at 6.9 GHz) together with sea ice monitoring. An XOVWM+SST system in low-inclination orbit would enhance studies of tropical weather and climate.
Climate Data Records Related to Observations of the Land
The land ECV breakout group was asked to consider 10 ECVs related to surface observations: glaciers and ice caps/sheets, snow cover, soil moisture, fire disturbance, lakes, biomass, land cover, surface albedo, fraction of absorbed photosynthetically active radiation (FPAR), and leaf area index (LAI).
The primary NPOESS instrument for land surface climate variables is VIIRS, following the heritage of AVHRR and MODIS sensors. Likewise, for GOES the primary land climate instrument will be the imager (ABI on GOES-R). The first hour or so of the breakout addressed the VIIRS and its known problems, primarily concerning optical cross-talk. The cross-talk as it stands now will affect the aerosol EDRs and the land EDRs, the latter primarily through poor aerosol correction. It is not clear that the cross-talk issue for VIIRS will be fixed in time for its first flight on NPP. Although an improved filter is being constructed and is planned for installation, participants were informed that there remains at least a 30 percent chance that the fixes will not work and that the land EDRs will be out of specification.
Participants considered the importance of land ECVs in terms of scientific impact and the availability of longer-term data sets for comparison and study. The land ECVs were then each evaluated in terms of risk. All risk evaluations in this summary assume that the cross-talk issue for VIIRS will be successfully alleviated.
Glaciers and Ice Caps/Sheets
The glaciers and ice caps/sheets ECV is of importance to climate models and albedo, water balance, sea level, and radiation budget climate studies. Ongoing missions/instruments of relevance include Landsat (1984-), SPOT, ASTER (2000-), GRACE (2002-), ICESat, MODIS (1999-), and MISR (2000-). The international Cryosat-2 mission, currently in its implementation phase, and the ICESat-II, GRACE-II, DESDynI, and SCLP missions recommended by the Earth science decadal survey are also relevant. NPOESS’ VIIRS is expected to contribute to this ECV; however, there is some risk to the ECV associated with the lack of ALT data required to estimate mass balance, although other altimeter measurements (if secured) can meet the need.
Measurement of snow cover is a high priority because of snow cover’s role in radiation budget and water cycle studies. Ongoing missions and instruments of relevance include AVHRR (originally VHRR; 1972-), MODIS (1999-), (A)ATSR (1991-), Landsat, SPOT, and SSM/I. NPOESS will contribute via VIIRS and ATMS; however, planned contributions by the CMIS replacement, MIS, are now uncertain. The snow cover ECV is also affected because VIIRS data can be used to map areal extent through time but a height/depth-related measure, which is required to make key calculations of mass, is missing. The decadal survey SCLP mission is relevant to this ECV, as it would provide passive and active microwave measurements of snow water equivalent. GOES-R ABI measurements are also of relevance, as are international plans for Sentinel-3.
The soil moisture ECV is important to climate science due to its impact on biogeochemical cycling, mesoscale climate models, vegetation dynamics, albedo, and surface roughness. Ongoing missions and instruments of relevance include AMSR-E (2002-), ALOS (2006-), Landsat, MODIS (1999-), and ASCAT (2006-). The planned NASA LDCM mission and international SMOS missions are also of interest. The NPOESS VIIRS and CMIS instruments are relevant to soil moisture; however, the soil moisture ECV is considered at high risk due to the CMIS descope, which effectively eliminates any possibility of retrieving this measurement. Even with CMIS, soil moisture measurements would have been limited to bare or very sparsely vegetated soils. Recommended by the Earth science decadal survey, SMAP, an active and passive L-band mission to directly measure soil moisture, would provide direct global soil moisture measurements with greater penetration depth.
The fire disturbance record has climate science implications in terms of understanding biogeochemical cycling, disturbance, and disasters. Ongoing missions and instruments of relevance include AVHRR (1982-), (A)ATSR (1991-), SPOT (1998-), Landsat, ASTER, MODIS (1999-), and MERIS (2002-). International plans for GCOM-C/SGLI (2012-2025) and Sentinel-2 are also of interest. VIIRS on NPOESS and ABI on GOES-R are expected to contribute to this ECV; however, there is a moderate risk to the ECV due to the low saturation level of the VIIRS instrument and the lack of VIIRS in a midmorning orbit. The saturation issue prevents the retrieval of fire radiative power,10 which is an important component of this ECV, and the loss of the midmorning orbit reduces the measurement of fire diurnal cycles.
The lakes ECV is of relevance to biogeochemical cycling, eutrophication, mesoscale climate models, human impact, vegetation dynamics, water cycle, and radiation budget climate studies. Ongoing missions and instruments of relevance include ERS-2/AATSR (1995-), MERIS (2002-), SeaWiFS (1997-), Jason-1 (2001-), Landsat (Landsat-7, 1999-), SPOT (SPOT-5, 2002-), and AVHRR (on NOAA POES). NASA plans for OSTM/Jason-2 and LDCM, international plans for Sentinel-3 and GCOM-C/SGLI, and the decadal survey recommendation for SWOT are also of interest. NPOESS/VIIRS can address the surface area of lakes; however, there remains a lack of three-dimensional measurement capability.
Measurements of biomass are important to studies of biogeochemical cycling, modeling, mesoscale climate models, human impact, vegetation dynamics, and surface roughness. Ongoing missions and instruments of relevance include ALOS/PALSAR (2006-), ENVISAT/ASAR, Landsat, MODIS (1999-), MERIS (2002-), ICESat, and ASTER. NASA plans for LDCM, international plans for Cryosat-2, ALOS, and ESA-BIOMASS, and the decadal survey recommendations for DESDynI and ICESat-II are also of interest. NPOESS/VIIRS is expected to contribute to this ECV; however, there remains a lack of three-dimensional measurement capability (e.g., from lidar or radar).
Land Cover, Surface Albedo, Fraction of Absorbed Photosynthetically Active Radiation, and Leaf Area Index
The above ECVs are important to climate studies due to their role in biogeochemical cycling, modeling, mesoscale climate models, human impact, vegetation dynamics, albedo, and surface roughness. Ongoing missions and instruments of relevance include AVHRR, MODIS (1999-), (A)ATSR (1991-), Landsat, SPOT, MERIS (2002-), GLI, ASTER, MISR (2000-), GOES, MSG, and POLDER. The NASA-planned LDCM mission, the international plans for Sentinel-3 and GCOM-C/SGLI, and the decadal survey recommendation for HyspIRI are also of interest. These ECVs are considered to be at low risk because they can be adequately addressed by VIIRS (assuming cross-talk is mitigated). If the VIIRS cross-talk issue is not resolved, there will be moderate risk to these ECVs.
WORKSHOP SUMMARY—DAY 2
The breakout groups on day 2 focused on the impacts of NPOESS and GOES-R descopes sensor by sensor. Participants were asked to comment on the various mitigation options suggested by NASA and NOAA presenters on day 1 and to suggest other mitigations to recover lost capabilities of importance to the climate community. Where appropriate, participants also considered whether missions in the Earth science decadal survey mission set might enable the recovery of the NPOESS climate measurements.11 As on day 1, templates were filled in during the breakout sessions, and they are available online.12 After the workshop a short background section was added to each breakout session summary to provide context for the discussions. It is important for the reader to recognize that the mitigation options presented below do not include all that might be considered and that both the options and the analysis are necessarily the subjective and not always disinterested views of presenters and participants.
Radiation Sensor Measurements
TSIS, ERBS, and OMPS-Limb measure, respectively, the incoming solar energy, the energy reflected and emitted by Earth, and the height-dependent concentration of atmospheric ozone that modulates these energy fluxes. Since the balance of incoming and outgoing radiation (Figure 2.1) determines Earth’s global temperature, these quantities are critical physical components of climate variability and change.
The 28-year-plus time series of total solar irradiance, total ozone, and outgoing longwave radiation allows researchers to address unique aspects of climate change, climate sensitivity, and cloud feedbacks; however, questions remain. Termination of the solar irradiance, energy budget, and ozone profile time series will leave unanswered crucial questions concerning the Sun’s impact on climate, both from direct surface heating and indirectly through its modulation of ozone and the stratosphere; the recovery (or not) of the ozone layer from chlorofluorocarbon reductions; the climatic impacts of a changing stratosphere; and the high-precision monitoring of clouds, aerosols, and ocean heat storage over the globe.
Total and Spectral Solar Irradiance
The TSIS instrument that would have flown on NPOESS comprises the Total Irradiance Monitor (TIM) and Spectral Irradiance Monitor (SIM) components, copies of which are currently operating successfully on the NASA SORCE (Solar Radiation and Climate Experiment) free-flying spacecraft (launched in 2003).
The decadal survey missions represent a set of community consensus priorities spanning Earth science including, but not limited to, climate science. Participants were asked to consider whether missions in the decadal survey mission set might enable recovery of NPOESS climate measurements to determine whether there are opportunities for synergism between NPOESS climate measurement recovery strategies and implementation of the community consensus decadal survey plan. Mitigation strategies were considered entirely within the context of climate measurement recovery and are not to be construed as a review of decadal survey mission priorities. The notion of synergy versus competition with the decadal survey is further discussed in Chapter 3, “Cross-Cutting Issues.”
The SORCE TIM sensor provides improved absolute accuracy and long-term stability relative to the radiometers flown on the Nimbus-7, Solar Maximum Mission, Upper Atmosphere Research Satellite (UARS), ACRIMSAT, and SOHO spacecraft. ACRIMSAT (launched in 1999) and SOHO (launched in 1995) are still operating. The SORCE SIM instrument is the first to measure the visible and near-infrared spectral irradiances, and it continues the monitoring of the middle UV spectrum, done earlier by UARS.
A TIM instrument is scheduled to fly on the Glory mission (launch in late 2008, 3-year mission design lifetime, 5-year goal), after which there are no current plans to ensure continuation of the 35-year record of total solar irradiance. The end of the SORCE mission in 2011 (assuming a 4-year extension of the core 5-year mission) will terminate a 9-year record of solar visible and infrared spectral irradiance and a 20-year record of solar ultraviolet spectral irradiance. Solar irradiance measurements from 1978 to 2013 will have sampled only three 11-year irradiance cycles, which alone is insufficient time to determine whether long-term irradiance trends occur or to quantify the broad range of irradiance changes possible in activity cycles of varying strength.
Earth Radiation Budget
Earth’s radiation budget parameters have, like solar irradiance, been measured since 1978 via instruments onboard seven different spacecraft. Each CERES instrument contains three scanning thermistor bolometer radiometers to monitor the longwave and visible components of Earth’s radiative energy budget. CERES achieves high radiometric measurement precision and accuracy, and it measures comprehensive Earth radiation budget parameters at higher accuracy than did its predecessors. CERES instruments on TRMM (launched 1997), Terra (launched 1999), and Aqua (launched 2002) have significantly enhanced capability relative to that of the initial sensors flown on Nimbus 7, ERBS, NOAA-9, and NOAA-10.
The paired CERES on Terra and on Aqua provide both of those missions with the possibility of coincident fixed azimuth plane scanning from one and rotating azimuth plane scanning from the other CERES, enhancing the quality of the final products. The CERES Terra and Aqua biaxial scan mode permits observations of the angular radiation fields in order to greatly improve the accuracy of the final fluxes of solar and thermal energy used to derive Earth’s radiation balance. These biaxial observations allow future missions in the same 10:30 or 13:30 orbits to fly a single CERES instrument while achieving the same accuracy as Terra and Aqua. The demanifesting of ERBS, which was to have had the same performance specifications as CERES but updated components, means that Earth radiation budget measurements will terminate with the CERES measurements on Aqua. While the CERES instruments are the most accurate broadband instruments yet flown, they are still not accurate enough to observe the subtle but critical decadal climate change signals unless the instruments are overlapped for at least 6 months in orbit according to the GCOS climate-monitoring principles.13 For this reason it is crucial that measurement record gaps are avoided. Both on-orbit CERES instruments have already exceeded their 5-year mission design life.
The total column and the vertical profile of ozone have been measured from space since 1978, primarily by the TOMS and SBUV instruments, respectively. The NPOESS OMPS-Nadir sensor is a combined TOMS/SBUV sensor. Although the SBUV is capable of measuring the ozone profile, its spatial resolution is poor (250 × 250 km), and the observations extend only above the peak ozone concentration. Therefore, the original OMPS design also included a limb sensor (OMPS-Limb) to achieve much higher spatial resolution and, in addition, measure the entire ozone vertical profile, including in the troposphere, below the stratospheric peak. Elimination of OMPS-Limb from NPOESS means that measurements of the complete ozone profile will end upon completion of the Aura mission (launched in 2004 with a 5-year mission design lifetime). The OMPS-Nadir sensor on NPOESS will continue only the total column ozone record.
Summary of Breakout Group Discussions
Participants in the breakout discussion considered various mitigation options for each demanifested sensor. A common theme throughout the session was the general preference for free flyers rather than a remanifesting of sensors on NPOESS, although the advantages of assimilation onto an operational platform in terms of data continuity were also noted. Should free flyers play a role in NPOESS mitigation, some participants indicated that there would be requirements for formation flight with the NPOESS platforms that might present a requirement for station keeping for NPOESS itself. The ability of the Integrated Program Office to accommodate such a requirement is uncertain.
TSIS. Although “absolute calibration” has been a goal, expected accuracy has yet to be demonstrated, and so the overlap requirement remains. Ensuring the continuity of the solar irradiance record requires the flight of TSIS indefinitely, overlapping with the current observations. With the demanifesting of TSIS from NPOESS, the sensor can be flown only if provided to the program as government-furnished equipment. In the near term, TIM on Glory
will overlap with TIM on SORCE. However, with the earliest flight of a remanifested TSIS on C2 in 2016, the likelihood of a measurement gap is high. Some participants noted that assimilation of total solar irradiance (TSI) and spectral solar irradiance (SSI) observations into the NPOESS operational environment would ensure eventual continuity of the measurements in the longer term, but with an increased risk of gaps in the near term.
The participants considered several mitigation scenarios, which are summarized below.
Mitigation Scenario 1. In scenario 1, the TIM instrument flies on Glory, as planned, in 2009 (continuing the record of total solar irradiance) and NASA builds two additional TSIS (containing TIM and SIM for total and spectrally resolved irradiance measurements) instruments for NPOESS C2 (2016) and C4 (2022). Most participants felt that this option, involving eventual restoration of the TSIS instrument to the NPOESS platform, provided for the eventual continuity of total and spectral irradiance observations in the longer term. However, the potential risk is high for creating gaps in total solar irradiance and SSI records. It was also noted that waiting for an NPOESS C2 launch would very likely create a gap, avoidable only in the (unlikely) event that SORCE continues beyond 2016, a mission life of over 13 years.
Mitigation Scenario 2. Scenario 2 includes all the provisions of scenario 1 but adds TSIS to the LDCM in 2011. This scenario would provide the opportunity to avoid an otherwise-likely data gap, but a solar pointing platform or mechanism would have to be provided to accommodate TSIS. In this scenario, a gap in TSI observations will likely be avoided, provided that there is sufficient overlap of SORCE, Glory, LDCM, and NPOESS C2. The probability of a gap in SSI is also reduced, since SORCE SSI measurements need only continue beyond 2011 (instead of beyond 2016). LDCM is a high-priority mission that reduces the probability of launch delays that could create a gap in the irradiance data.
Mitigation Scenario 3. Scenario 3, which was preferred by most of the breakout session participants, involves flying TSIS on LDCM and then on subsequent free flyers in 2014 and 2020. Having a dedicated mission is considered desirable in order to reduce the higher integration costs presumed for a multisensor Earth-pointed platform and to allow flexibility in planning and launches. During discussions, a participant noted that free flyers can be canceled more easily than can a multiple-sensor mission; he considered this a potential drawback. However, in the short time available for discussion, the potential trade-offs involved in free flyers versus alternatives could not be explored in detail.
Other Mitigation Options. Participants also discussed other options for securing the TSIS data record, including acceleration of the decadal survey recommendation for CLARREO, flying a series of dedicated spacecraft, and accommodation of TSIS on already-planned missions as an instrument of opportunity (e.g., on GOES-R or DSCOVR). The drawbacks of these options include the risk of relying on unapproved missions, the perceived higher risk of cancellation of single-instrument missions, and physical14 and programmatic instrument accommodation challenges, respectively.
ERBS. Ensuring a long-term record of Earth’s radiation budget requires the flight of ERBS-type sensors indefinitely, overlapping with Aqua in the near term. CERES is currently manifested on C1 and the NPOESS ERBS was canceled, making an ERB gap likely. To avoid a gap with Aqua, many participants strongly suggested that CERES FM-5 should fly on NPP rather than C1 (2013).
Participants considered several mitigation scenarios, which are summarized below. Some also offered several suggestions for CERES upgrades or improvements, including changes to the mirror attenuated mosaic (MAM)15
to facilitate solar calibration, switching the 8-12 µm window channel with the ERBE longwave channel to improve determination of longwave and shortwave flux components, and changing materials and instrument operation to avoid UV degradation of the solar channel.
Mitigation Scenario 1. Scenario 1 involves flying CERES on NPP in 2009 rather than on NPOESS C1 to avoid a gap with Aqua, while developing ERBS or a CERES-II for NPOESS C1 and C3. This scenario ensures continuation of ERB measurements on an operational platform and reduces the risk of a gap by a factor of three for putting ERBS on C1 (based on an engineering model of instrument and spacecraft failure rates). A downside of choosing a CERES-II approach is that the original CERES instrument team has been disbanded and the technology is old, so costs, risks, and available capability for building the instruments are unknown.
Mitigation Scenario 2. Mitigation scenario 2 for providing the needed measurement during the NPOESS program span is to fly the existing CERES on NPP and develop ERBS for launch on two subsequent free flyers. Because generation of the Earth radiation budget CDR requires inputs from other sensors on NPOESS, the free flyers would have to fly in formation with the NPOESS 13:30 spacecraft (within 5 minutes of VIIRS coverage). Some participants again noted the advantages of dedicated missions, which allow for flexibility in mission planning and launch dates, but acknowledged the increased risk of cancellation of individual free flyers and of thus jeopardizing measurement continuity.
Other Mitigation Options. Flying ERBS on the decadal survey’s recommended CLARREO mission was considered; however, the orbits were found to be incompatible, as the CLARREO mission concept (as it is currently defined) requires precessing orbits, whereas the ERBS continuation of CERES requires a 13:30 Sun-synchronous orbit. The CLARREO and ERBS observations could be directly compared, however, during orbit crossings of CLARREO with NPP and/or the NPOESS 13:30 orbit.
OMPS Limb Subsystem. OMPS-Limb was removed from the NPOESS manifest as part of Nunn-McCurdy certification. Omitting OMPS-Limb will result in the complete loss of precise information about the ozone-height profile after 2014, because OMPS-Limb was the only instrument planned to fly after Aura that would be capable of determining ozone profiles below the peak concentration in the stratosphere.
Some participants noted that even though a descoped OMPS on NPOESS will continue total-column ozone measurements, the OMPS-Nadir sensor lacks the state-of-the-art capability for measuring other trace species and for high spatial resolution, both of which are essential for advancing atmospheric research in the future. Furthermore, OMPS-Nadir measurements are duplicated by GOME-II on MetOp, which has a smaller footprint (~40 km × 40 km). The GOME-II instrument also measures aerosols, NO2, SO2, BrO, and OClO. With the availability of the higher-resolution OMI data, the science community has realized that OMPS-Nadir and GOME-II have inadequate spatial resolution—thus, there is a desire for higher resolution and more capable sensors than OMPS-Nadir.
For near-term mitigation, most participants would have the 2010-2014 NPP mission fly both OMPS-Nadir and OMPS-Limb, since OMPS-Limb is already built (NASA and NOAA have indicated that OMPS-Limb will indeed be flown on NPP16).
Mitigation Scenario 1. The most basic mitigation scenario involves remanifesting of OMPS-Limb onto all NPOESS satellites flying OMPS-Nadir. Some participants suggested that because OMPS-Limb and OMPS-Nadir were designed as an integrated package and thus share common electronics, reintegration of OMPS-Limb would present a low risk and should be low in cost. The expected launch date of C3, however, presents a measurement gap risk beyond Aura and NPP.
Press Release: NOAA, NASA Restore Climate Sensor to Upcoming NPP Satellite, April 11, 2007, available at http://www.nasa.gov/home/hqnews/2007/apr/HQ_07085_NOAA_NASA_instrument.html.
Mitigation Scenario 2. Scenario 2 involves flight of the OMPS suite (nadir and limb) as above, but replacing the C3 flight with a free flyer. Many participants again noted the advantages of dedicated missions, which allow for flexibility in mission planning and launch dates; however, they also acknowledged the increased risk of cancellation of individual free flyers, which jeopardizes measurement continuity.
Mitigation Scenario 3. Participants discussed a scenario involving flight of solar occultation instruments (e.g., SAGE or Canadian ACE) on free flyers in inclined, precessing orbits to ensure continuity of measurements of stratospheric ozone and pertinent trace-gas profiles. Many participants again noted the advantages and risks associated with free flyers.
Other Mitigation Scenarios. Participants also discussed the relevance of the GACM mission recommended by the Earth science decadal survey. Although it was noted that GACM would provide higher resolution than OMPS, its anticipated launch date is too far in the future for GACM to be relied on as a mitigation option. The flight of an OMI follow-on instrument would preserve the continuity of Aura’s higher-resolution ozone data, but only at nadir—meaning that a limb capability would still be needed. GOME-II was also discussed as a possible source of some desired trace gas information, but the spatial resolution is relatively low, and the MetOp 9:30 orbit would present difficulty in merging the data into the current data record.
Visible and Infrared Imager and Sounder Measurements
Nunn-McCurdy NPOESS certification resulted in the demanifesting of APS and reduced the coverage of CrIS/ATMS. The VIIRS sensor has experienced hardware challenges that might impair the sensor’s ability to meet certain IORD objectives. A brief background on each of the sensors is presented below. Mitigation options were explored for APS (the demanifested sensor), and participants made suggestions and comments regarding VIIRS and CrIS/ATMS.
Operational 2010+ low-Earth orbit (LEO) environmental monitoring will be provided by the NPOESS VIIRS. VIIRS combines and dramatically improves upon the POES AVHRR and the DMSP Operational Line Scanner (OLS). Combining AVHRR and OLS capabilities into a single sensor will provide advantages of simultaneity along with dramatic improvements in spatial resolution and radiometry for vegetation index, SST, cloud top temperature, and day-night cross-terminator cloud imaging for DOD applications.
Moreover, to satisfy the VIIRS EDRs prescribed by the NPOESS IORD, VIIRS will also provide many of the scientific remote-sensing features of the Earth Observing System (EOS) MODIS and SeaWiFS instruments. VIIRS offers most MODIS and SeaWiFS capabilities except near-IR and microwave/IR water vapor bands, IR sounding bands, and near-IR fluorescence radiometry not required to meet the prescribed VIIRS EDRs. VIIRS will also dramatically improve on MODIS and SeaWiFS spatial resolution (via a patented OLS-like17 detector aggregation technique) and global coverage (via a 40 percent wider imaging swath), while offering comparable absolute radiometry and sensitivity as well as the long-term stability required by the IORD to support CDRs. Indeed, most of the 23 VIIRS EDRs are also ECVs.
VIIRS is manifested on the NPP and is planned for a 13:30 Sun-synchronous orbit as part of the EOS “A-Train,” to augment the EOS Aqua spacecraft carrying MODIS, and to complement the NOAA N′ and DMSP midafternoon spacecraft. Following NPP, the NPOESS C1 spacecraft carrying a VIIRS will operate in the terminator orbit to replace the DMSP F16. Finally, the NPP will be replaced by NPOESS C3; the operational replacement for the DMSP, NOAA N′, and EOS Aqua satellites, all operating in midafternoon orbits.
The pre-Nunn-McCurdy NPOESS constellation was also to include NPOESS C2 for the midmorning orbit with a VIIRS to replace the EOS Terra MODIS as well as the NOAA N′ and DMSP midmorning AVHRR and OLS, respectively. Post-Nunn-McCurdy, the midmorning orbit has been deleted, and the ESA/EUMETSAT’s MetOp-A, which became operational in late 2006, substitutes AVHRR for the NPOESS midmorning VIIRS, offering no replacement for the midmorning OLS or MODIS.
Along with the MetOp AVHRR in the 2130 orbit (9:30 pm local ascending node—“midmorning” refers to the 9:30 am local descending node), the NPP/MetOp pair will provide continuity of civil environmental imaging, but the deletion of the NPOESS 2130 orbit results in reduced capability, given that the AVHRR on MetOp will only address (and not meet) a fraction of the VIIRS EDRs. In particular, the requirements for VIIRS EDR long-term stability were specified to assist CDR production. The AVHRR is not specified to meet these requirements even for the limited set of VIIRS EDRs it does address.
Aerosol information available from current operational and research satellite observations is primarily in the form of aerosol optical depth, with additional coarse information about particle size provided in the form of a coarse/fine mode discrimination of optical depth or in the form of an aerosol index. Much of this information is restricted to over-ocean observations, given the complexity that land surface adds due to variable surface reflections. The information currently available is far short of what is needed—quantified aerosol absorption is needed to apportion the aerosol forcing contributions between atmosphere and surface—to monitor aerosol forcings of climate.18 APS offers limited ability for determining the absorbing properties of aerosols, which is nevertheless a significant step forward from existing capabilities.
APS on Glory is scheduled for launch in 2008 and is expected to operate into 2013. With the removal of the APS instrument from NPOESS, VIIRS will by necessity become the principal sensor for deriving aerosol parameters needed for estimation of aerosol climate forcing post-2013. Without a remanifesting of the APS, the monitoring of aerosol forcing beyond 2013 will likely decline to pre-2013 capability, particularly given the uncertain performance of VIIRS.
The power of hyperspectral sounding has been amply demonstrated by the NASA EOS Atmospheric Infrared Sounder (AIRS) flying on the Aqua mission in a 13:30 orbit19 in terms of improved retrieval uncertainty and a significant positive impact on forecast skill.20 Operational LEO atmospheric temperature and moisture sounding capability in the 2010+ time frame will be provided by two instrument pairs (three during the transition from the current system). The NPOESS program will fly an operational hyperspectral infrared sounder—the Cross-track Infrared Sounder (CrIS). The CrIS instrument will be accompanied by the Advanced Technology Microwave Sounder (ATMS).
The CrIS and ATMS instrument pair is currently manifested on the NPP flight, which is planned for a 13:30 Sun-synchronous orbit as a part of the EOS “A-Train.” The NPP will subsequently be replaced with the NPOESS flights C1 and C3; these are the operational replacements for NOAA N′. The NOAA M (midmorning), N, and N′ spacecraft carry the current-generation multispectral HIRS, along with the AMSU. In the midmorning orbit, the multispectral sounding capability of NOAA-M is being replaced by the Infrared Atmospheric Sounding Interferometer (IASI) carried on ESA/EUMETSAT’s MetOp-A, which became operational in late 2006. The MetOp
NRC, Radiative Forcing of Climate Change: Expanding the Concept and Addressing Uncertainties, The National Academies Press, Washington, D.C., 2005.
The Aqua orbit is controlled to maintain an ascending node equatorial crossing time of 13:30 local time.
J. Le Marshall, “The Use of Global AIRS Hyperspectral Observations in Numerical Weather Prediction,” 11th Symposium on Integrated Observing and Assimilation Systems for the Atmosphere, Oceans, and Land Surface, 87th American Meteorological Society Annual Meeting, San Antonio, Texas, January 15-18, 2007, available at http://ams.confex.com/ams/pdfpapers/119660.pdf.
series carries additional profiling capability via the Microwave Humidity Sounder, Advanced Microwave Sounding Units (AMSU-A1 and AMSU-A2), and High-resolution Infrared Radiation Sounder (HIRS/4).
Retrieved variables from CrIS and ATMS include temperature, moisture, and pressure profiles, surface emissivity and temperature, total-column ozone, and additional possible data products such as trace gases (CO, N2O, CH4, and CO2). In particular, upper-air temperature and water vapor are considered to be global ECVs.
The demanifesting of CrIS/ATMS from the NPOESS 17:30 orbit results in reduced coverage, because the CrIS/ATMS ±48.3° cross-track scans and 2,250 km swaths do not provide global contiguous coverage. The reduction from three to two orbit planes for atmospheric moisture and temperature profiling represents a loss in diurnal sampling (from 4- to 6-hour refresh) compared to the pre-Nunn-McCurdy NPOESS baseline, which will reduce the quality of diurnally averaged climate analyses. It should be noted, however, that the current operational satellite architecture of NOAA POES, DMSP, and MetOp does not include a 17:30 infrared sounder, and so coverage will not worsen, but rather fail to improve over that provided by the current system. This reduction in diurnal coverage is compounded by the recent NOAA decision to suspend taking operational geosynchronous upper-air temperature and water vapor profile measurements after the current GOES-N/O/P series until approximately 2025.
Summary of Breakout Group Discussions
A number of mitigation options and instrument improvements were considered for APS, VIIRS, and CrIS/ ATMS. An idea that received particular attention was that on all subsequent flight builds there would be extensive preflight characterization and improved documentation to increase climate science utility (to date this is not currently planned); these preflight characterizations would ensure that the sensors are stable, as nearly identical as possible from sensor to sensor, and thus climate relevant.
Two of the mitigation options discussed below were identified by some participants as involving small to moderate changes to existing instruments that might be accomplished with minimal additional investment and could yield high returns to the climate science community. Specifically:
The VIIRS fire product (the VIIRS active fire EDR) can be improved by adding an M15 saturation flag. Participants familiar with the instrument design suggested that this might be possible to implement early in the program (as soon as NPP).
CrIS/ATMS data can be downlinked at full spectral resolution to enable production of additional climate data products without changes to the hardware. Increased preflight testing and documentation would also be necessary to produce climate-quality greenhouse gas measurements from the instruments.
VIIRS. A comparison of MODIS and VIIRS was presented. It was noted that the MODIS functional architecture is a flat “paddle” scan-mirror favored for midmorning and afternoon orbits while the VIIRS functional architecture is a rotating telescope required for terminator orbits. VIIRS will provide improved imagery (with more constant field of view than MODIS), but VIIRS has no IR channels sensitive to atmospheric H2O (or CO2). Regarding EDR performance, VIIRS is expected to meet all requirements, and in tandem with CrIS improves on most.21
VIIRS is currently progressing through vacuum tests. While emphasizing the importance of not disrupting these tests so as to maintain schedule, several participants noted a number of highly desirable improvements. In particular, the VIIRS fire product can be improved by mitigating the aggregation of saturated pixels with nonsaturated pixels, or at least providing a flag. In the future, a higher saturation level in the shortwave infrared window should be considered; this could be accomplished with a dual-gain sensor and likely not affect SST determinations. The inclusion of water-vapor-sensitive measurements that enable estimation of winds over the poles day and night is planned for the C3 VIIRS and should be pursued. It was noted that synergy with the visible/near-infrared (NIR) channels on ABI has been suggested and is planned; this synergy provides in-flight calibration opportunities for the
geostationary ABI sensor (which lacks on-board visible/NIR calibration) leveraged from the LEO sensor (VIIRS with onboard calibration).
Finally, to achieve comparable imaging capabilities in the midmorning orbit, participants advised that the Integrated Program Office work with EUMETSAT to fly a VIIRS imager on subsequent MetOp platforms so that an imager more capable than the AVHRR is flying in the midmorning (MetOp) orbit as soon as possible.
One participant noted that MODIS displays problems with saturation that could be mitigated for VIIRS by incorporating dual gains especially for the 746 nm channel, and further suggested that signal-to-noise improvements by a factor of two in the 1,240 and 1,610 nm bands would enhance the ocean-sensing capabilities of VIIRS significantly. This participant noted that VIIRS could be very helpful to the ocean CDR (even more so with the above-mentioned improvements).
APS. The APS instrument scientist for Glory, Brian Cairns, delivered a presentation regarding APS and APS-MODIS/VIIRS synergy. Dr. Cairns noted that aerosols come in various sizes and shapes; the key requirement is to determine the type of aerosol that is present. APS is intended to assist in measuring particle composition and size and shape. There are two cloud data products and an experimental product that are thought to be able to infer cloud base height. Instantaneous field of view cloud screening of APS at 6 km is accomplished using VIIRS/MODIS. Summarizing the APS and MODIS/VIIRS synergy, Cairns remarked, “APS with MODIS/VIIRS tells you a lot, but alone APS tells you nothing.”
Mitigation options considered by the participants are summarized below.
Mitigation Scenario 1. In scenario 1, APS will fly on Glory as a demonstration, and if successful will be integrated onto NPOESS C3. Many participants believed APS on Glory would begin a valuable record; however, it was also noted that NPOESS C3 does not have a needed lunar calibration capability.
Mitigation Scenario 2. Scenario 2 includes the elements of scenario 1 but adds a climate free flyer between Glory and C3. The added value of this scenario is the continuation of the aerosol data record, with the ability to lunar-calibrate APS on the free flyer. A variation on this option considers another free flyer in place of reintegration onto C3. This approach would avoid the concern about lack of lunar calibration associated with C3, although likely at a higher cost.
Mitigation Scenario 3. Another scenario involves proceeding with the ACE mission recommended in the Earth science decadal survey. This mission calls for cross-track polarimetric coverage, which is an advance over the single-pixel APS; however, the technology readiness of such an instrument was questioned by some participants. The perceived low technology readiness level22 of the polarimeter was also considered a risk of this scenario.
CrIS/ATMS. The role of AIRS/IASI/CrIS-ATMS in climate research was discussed. It was noted that the requirements for AIRS and CrIS are similar and that hyperspectral infrared measurements have been demonstrated to improve weather forecasting, largely accounting for the initial decision to include CrIS on NPOESS. AIRS has demonstrated a positive impact in weather forecasting, but the hyperspectral IR also helps in climate observations. AIRS radiances are accurate and traceable to National Institute of Standards and Technology (NIST) standards. Further, AIRS is stable as verified by ground truth. With cross-calibrations, hyperspectral IR measurements have been used for quality control for other sensors (including MODIS). Some participants stated that the advent of spectrally resolved NIST-traceable infrared measurements will assist climate science appreciably.
In subsequent discussion, breakout participants considered having CrIS/ATMS restored to the early morning orbit so that the diurnal cycle would be measured adequately. A breakout participant suggested that data with the
Technology readiness levels are defined in J.C. Mankins, “Technology Readiness Levels: A White Paper,” NASA Advanced Concepts Office, April 6, 1995, available at http://www.hq.nasa.gov/office/codeq/trl/trl.pdf.
full spectral resolution measured by CrIS be downlinked so that more accurate trace gas measurements could become available. Some participants also discussed their desire for additional improvements to hyperspectral sounding capability, including a dedicated sounder free flyer.
Microwave Sensor Measurements
The NPOESS altimeter, ALT, was demanifested as a result of the Nunn-McCurdy action, and CMIS is being restructured as a (still largely undefined) MIS instrument with reduced capability. In light of these changes, the microwave sensor breakout session divided its presentations and discussion into three subsessions: altimetry, radiometry, and scatterometry. While scatterometry was not considered as part of the NPOESS baseline, some participants felt that the pressing need for continuation of operational active ocean vector wind measurements warranted further discussion, particularly in light of the CMIS descope. Further, some participants asserted that passive microwave vector wind measurements did not constitute a climate data product, whereas the value for climate studies of scatterometry-derived wind measurements has been demonstrated.
A 15-year CDR of global sea level rise and interannual variability has been established by TOPEX/Poseidon (1992-2002) and Jason (2002-present).23 The duration of this data record is just beginning to provide insight into decadal variability. Altimeter data are used extensively in observationally based studies of ocean climate variability on seasonal and longer time scales. These data are also assimilated into many ocean circulation models. The 15-year sea level data record has established a unique record of the effects of global warming. As the ocean absorbs more than 80 percent of the heat from global warming, the information on the state of ocean circulation revealed from altimetry is also important for understanding climate change. The altimetry sea level record is crucial for checking the validity of the assessment of the extent of global warming and future projections and for monitoring the effects of global warming. The continuation of a precise sea level record is thus of unique and critical importance.
There are a number of other altimetry missions planned for the next decade: the France/India AltiKa/SARAL mission, the ESA Sentinel-3 mission, and the Chinese HY-2 mission. These missions will certainly complement precision altimetry missions but cannot be relied on as alternate approaches to the continuation of the sea level record because of their non-optimal orbits for resolving ocean tides, less accurate orbit determination, and the lack of an associated well-balanced science program focused on sea level and ocean circulation.
Although the present record will be continued by the follow-on mission to Jason—the OSTM/Jason-2 to be launched in 2008 as a joint mission of NASA, NOAA, Centre National d’Etudes Spatiales (CNES), and EUMETSAT—the next mission after OSTM/Jason-2 is not yet confirmed. The certified NPOESS program does not include an altimeter.
CMIS represented the state of the art in satellite microwave radiometers and was intended to continue, with a higher degree of accuracy and resolution, the time series of many fundamental climate variables, including SST and wind, sea ice and snow coverage, soil moisture, and atmospheric moisture (vapor, clouds, and rain). The ability of CMIS to measure surface characteristics through cloud cover made it a unique and essential sensor for climate.
CMIS had a number of advanced capabilities that are not available from the current operational microwave imaging radiometers SSM/I and SSMIS. These included:
Low-frequency channels at 6.9 and 10.7 GHz,
Higher spatial resolution (a factor of three better than SSM/I and SSMIS), and
Better spatial/temporal coverage: three orbit times as compared to two.
CMIS also had polarimetric channels capable of inferring wind direction, which is addressed elsewhere (see “Scatterometry” below).
The capabilities of the MIS instrument that is to replace CMIS are still largely undefined; however, there were indications that certain low-frequency channels were likely to be lost. The loss of the low-frequency channels, particularly at 6.9 GHz, would mainly impact the measurement of SST and soil moisture, although it also would degrade the accuracy of some other retrievals such as measurements of wind speed. Loss of high spatial resolution would have a detrimental effect on measurements of all parameters, including sea ice, snow cover, and precipitation. Reduced spatial/temporal coverage, due to the deletion of MIS from the midmorning orbit, will significantly limit the ability to characterize the climate’s diurnal cycle, especially with respect to global precipitation. The impact on climate monitoring and research of losing these advanced capabilities is substantial.
Microwave “through-cloud” SST measurements have proven to be a boon for climate research and oceanography. Unlike IR measurements, which are limited to cloud-free areas, microwave retrievals provide a largely uninterrupted view of the surface temperature over the world’s oceans. The importance of SST to climate research is hard to overstate. SST is a key parameter in determining how the water and energy fluxes at the air-sea interface affect the hydrologic cycle and the surface radiation balance. The intensity, frequency, and location of hurricanes are in part determined by where the necessary oceanic heat is available to sustain, encourage, or dissipate these storms. Climate oscillations such as the El Niño Southern Oscillation, North Atlantic Oscillation, and Pacific Decadal Oscillation all have distinctive SST signatures that characterize the relevant forcings. The endemic cloud cover at high latitudes prevents monitoring of ocean temperatures by IR radiometers, and microwave radiometers provide the only way to continually measure SST in these vital Arctic regions, which are now experiencing rapid climate change. Tropical convergence zones are also prime examples of persistently cloudy regions where SST detection by AVHRR is problematic. Microwave measurements in the 5-7 GHz band are required to retrieve SST over the full range of global temperature (−3°C to 35°C).
Soil moisture is a key determinant of the interaction between the land and the atmosphere. In many respects, it plays a role similar to that of SST in the case of air-sea interactions. Soil moisture controls the relationship between actual and potential evapotranspiration and hence is a key determinant of the recycling of moisture from the land surface to the atmosphere. Notwithstanding that 6.9 GHz is limited to sensing soil moisture in the top few centimeters of the soil only in areas of sparse vegetation, the portion of the globe so covered is substantial. Furthermore, the areas where construction of a CDR would be feasible include substantial areas (e.g., of the African continent) where hydrologic extremes have great consequences both economically and in terms of loss of human life. Given the potential for acceleration of the hydrologic cycle associated with global warming, construction of a long-term CDR for soil moisture would have significant scientific and societal value. Furthermore, planned soil moisture missions (ESA/SMOS, NASA/SMAP) at the L-band, while emphasizing a product technically superior to the product that could be derived from a 6.9 GHz channel, are experimental in nature and are not alone intended to produce long-term, multidecadal CDRs. These planned L-band missions would, however, have great value in terms of refining and characterizing the temporal and spatial variability of the 6.9 GHz retrievals.
Sea ice plays a key role in global climate change by regulating ocean-atmosphere transfers of energy and water and helping to control ocean surface salinity. Sea ice albedo feedbacks amplify climate impacts in the polar regions. Variables such as ice extent, concentration, and type are important for navigation as well as for marine habitat assessment. The passive microwave satellite record of sea ice concentration and extent extends from 1979 to the present. Documented decreases of Arctic sea ice extent currently exceed 8 percent per decade and appear to be accelerating. Snow cover in the Northern Hemisphere has also been declining at a rate of about 3 to 5 percent per decade during spring and summer. This decline in snow cover is significant because, compared with other land cover types, snow has a very high albedo and climate feedbacks are felt on local, regional, and even hemispheric scales. Moreover, snowmelt runoff is a key component in the hydrologic cycle and the primary source of fresh
water for many millions of people. At a time when Arctic sea ice and snow cover are changing most rapidly, the loss of the all-weather monitoring capability of CMIS represents a major setback.
Global measurements of precipitation will be adversely affected by all three lost capabilities. Accurate measurements of heavy rain require the 11 GHz channels. Higher spatial resolution is essential to discriminate convective versus stratoform features and to measure the intense rain that often comes from small rain cells. Finally, better spatial/temporal coverage is a main prerequisite for improving current knowledge of global rainfall over the complete diurnal cycle. The advanced capabilities of CMIS will be dearly missed by the precipitation community.
The cancellation of CMIS leaves JAXA’s AMSR-E and the U.S. Navy’s WindSat as the only low-frequency, high-spatial-resolution microwave radiometers in space.
Data derived from ocean scatterometers is vital to scientists in their studies of air-sea interaction and ocean circulation, and their effects on weather patterns and global climate. These data are also useful in the study of unusual weather phenomena such as El Niño, the long-term effects of deforestation on our rain forests, and changes in the sea-ice masses around the polar regions. These all play a central role in regulating global climate. An 8-year CDR of ocean surface vector winds has been established by QuikSCAT (1999-present). This data set has been crucial in advancing scientific research into marine meteorology, wind-driven upper-ocean circulation, and air-sea interaction processes from local to basin-wide scales. The QuikSCAT measurements have revealed energetic small-scale structure in the surface wind field that was not previously known to exist. The Ekman upwelling from the wind stress curl associated with these structures plays an important role in ocean circulation theory, as well as in ocean biology from upwelling of nutrients from the deep water into the upper ocean where they can be utilized by phytoplankton. The QuikSCAT data record is approaching the 10-year duration that is considered the baseline minimum for use in numerical simulations of wind-forced ocean circulation.
QuikSCAT is also heavily used in operational severe weather forecasting. The QuikSCAT measurements have had a major impact on tropical cyclone forecasting, especially for cyclones outside the range of aircraft reconnaissance. QuikSCAT data have helped in the estimation of the intensity of tropical storms, in determining the radial extent of winds of tropical storm force in tropical storms and hurricanes, and in locating circulation centers for tropical depressions and tropical storms. QuikSCAT occasionally provides earlier detection of surface circulations in developing tropical cyclones, and some studies have indicated a positive impact on hurricane track forecasts by numerical models, especially over the open-ocean regions that are not accessible by aircraft.
The high resolution of QuikSCAT measurements has improved forecasting, warnings of localized wind events, and ability to locate frontal systems over the ocean. In midlatitudes, QuikSCAT revolutionized wind warning categories by enabling the introduction of hurricane-force wind warnings in 2000. Hurricane-force winds were rarely forecast outside the tropics prior to the availability of QuikSCAT data. During the months of September 2006 through May 2007, forecasters at the NOAA Ocean Prediction Center used QuikSCAT wind measurements to identify 114 individual extratropical cyclones (64 in the North Atlantic and 50 in the North Pacific) containing extreme hurricane-force wind conditions.
In the original configuration of NPOESS, the ocean surface vector wind data record established by QuikSCAT was to be replaced by passive microwave measurements of wind speed and direction by the polarimetric CMIS radiometer. From the beginning, there were serious concerns within the scientific community (both research and operational) about the viability of passive microwave measurements of ocean surface vector winds, especially in storms and in other areas of rain and large amounts of cloud liquid water.
In preparation for CMIS, the U.S. Navy launched WindSat in January 2003 as a “risk reduction demonstration project.”24 WindSat is similar but not identical to CMIS, allowing insight into the accuracy of vector wind
Windsat is a joint IPO/DOD/NASA risk reduction demonstration project intended to measure ocean surface wind speed and wind direction from space using a polarimetric radiometer. It was launched in January 2006. See http://www.ipo.NOAA.gov/Projects/Windsat.html.
retrievals that could be expected from CMIS. WindSat results thus far have not allayed scientists’ concerns about passive microwave measurement of ocean vector winds.25
Summary of Breakout Group Discussions
ALT/Altimetry. Workshop participants considered currently operating and planned altimetry missions and their adequacy to meet climate measurement needs. Since the Sun-synchronous orbit of the NPOESS platforms is not acceptable for measuring global sea level change with the required precision, the loss of the NPOESS altimeter has little impact on continuation of this CDR.
The Jason altimeter is expected to continue operating at least long enough to overlap its successor, Jason-2 (also known as OSTM), which is expected to launch in June 2008. Jason-2 is essentially equivalent to the currently operating Jason altimeter. The overlap of TOPEX/Poseidon and Jason enabled the identification of a 14 cm bias between the two altimeters. It is likely that a similar bias will exist between Jason and Jason-2; therefore, an overlap of Jason and Jason-2 is highly desirable in order to cross-calibrate the two altimeters and ensure accurate continuation of the sea level CDR. If there is no overlap, tide gauge data will provide a viable alternative to cross-calibration, as long as the gap between Jason and Jason-2 is not long. While Jason-2 may continue to operate for more than its nominal 5-year lifetime, it is critical that a successor to Jason-2 be launched by 2013 to ensure continuation of a sea level CDR that is indispensable for monitoring the state of the global ocean and its role in future climate variability.
Because of its Sun-synchronous orbit, the currently operating ENVISAT altimeter and its successor Sentinel-3 are not viable mitigation strategies for continuation of the sea level CDR beyond Jason-2. Three mitigation scenarios were discussed. All three consist of a sequence of two successors to Jason 2, referred to here as Jason-3 and Jason-4.
Mitigation Scenario 1. In the first scenario, which was the scenario most preferred by participants, Jason-3 consists of a Jason-2-type altimeter to be launched by NOAA and EUMETSAT, and Jason-4 consists of a wide-swath altimeter, referred to in the Earth science decadal survey as the SWOT mission, to be developed and launched by NASA and CNES. To allow for precise intercalibration, the preferred orbit for Jason-3 is the same as that of TOPEX/Poseidon, Jason, and Jason-2. The orbit for SWOT would have to be changed to a higher inclination and longer repeat period in order to satisfy the sampling requirements for the terrestrial water (lakes and rivers) applications. In addition to broadening applications to include measurements of terrestrial water, the synthetic aperture radar-interferometric technology of SWOT will provide much higher resolution measurements for studies of ocean eddies and measurements very near land for coastal applications.
An advantage of this scenario is that Jason-3 would be a clone of Jason-2, in terms of both hardware and being a jointly funded project with EUMETSAT and other European partners. Many components have already been manufactured as spares for Jason-2, including a spare Proteus bus. If partnerships could be secured, the United States would only be responsible for approximately half of the cost of the mission. A potential disadvantage of this two-mission scenario is that the launch of Jason-3 could jeopardize a subsequent launch of the decadal survey’s recommended SWOT mission if sufficient funding is not provided for both missions sequentially.
Intercalibration issues between Jason-3 and a SWOT altimeter for Jason-4 would be unavoidable because of the need to change to a different orbit for SWOT. An overlap between Jason-3 and Jason-4/SWOT is therefore highly desirable, although the tide gauge network could also be a viable method for intercalibration. The 10-day repeat orbit for Jason-3 in this scenario would not satisfy Navy requirements. SWOT’s higher-inclination orbit provides a wider swath and a repeat period that would satisfy Navy requirements.
Mitigation Scenario 2. In the second scenario, Jason-3 and Jason-4 are both Jason-2-type altimeters in the same orbit that has been used for TOPEX/Poseidon and Jason, which is also to be used for Jason-2. This scenario would
See, for example, M. Brennan, R. Knabb, P. Chang, J. Sienkiewicz, Z. Jelenak, and K. Schrab, “The Operational Impact of and Future Requirements for Satellite Ocean Surface Vector Winds in Tropical Cyclone Analysis,” 61st Interdepartmental Hurricane Conference, March 6, 2007, available at http://www.ofcm.gov/ihc07/Presentations/s4-04brennan.ppt.
eliminate any issues with cross-calibration and would thus ensure continuation of the CDR for sea level rise. The primary disadvantage of this scenario is the delay in the launch of a SWOT altimeter, thus postponing the capabilities to measure the full spectrum of eddy variability in the ocean, to measure sea surface height near land, and to measure terrestrial water. Another issue for this scenario is that there are no spare satellite buses available for Jason-4.
Mitigation Scenario 3. Jason-3 is a SWOT-type altimeter. The advantage of scenario 3 is the near-term broadening of applications of satellite altimetry to include studies of ocean eddies, near-coastal sea level variability, and terrestrial water. A potential disadvantage is the possibility of a gap occurring in the sea level CDR due to limitations in how quickly SWOT could be built, tested, and launched. Since the orbit of SWOT would be different from that of TOPEX/Poseidon, Jason, and Jason-2, potential problems with cross-calibration for continuity of the sea level CDR would be an issue. An overlap between Jason-2 and Jason-3/SWOT is therefore highly desirable, although tide gauge data could also be a viable method for intercalibration.
CMIS/MIS/Radiometry. Participants in the radiometry breakout session focused on the likely loss of capability of the CMIS instrument, which was canceled and is to be re-competed as a simpler, less capable instrument launching no earlier than 2016 on NPOESS C2. This descope and delay were of most concern for applications requiring the 6.9 GHz band, which is of prime importance for measurement of global SST and soil moisture. As noted earlier, many participants were less concerned about the potential loss of ocean vector winds measurements from CMIS, because this CMIS data product was considered inadequate even prior to the descoping; ocean vector wind measurement is addressed further in the section “Scatterometry” below.
Two presentations were given on the importance of microwave SST retrievals to climate studies. One talk stressed the strong synergism that is obtained when microwave SST retrievals are combined with IR SST retrievals; both are necessary for doing climate research. The other presentation focused on the detrimental impact associated with the cancellation of CMIS and then suggested several possible mitigation strategies. Global SST retrievals require a channel near 6.9 GHz, and currently only AMSR-E and WindSat have this low-frequency channel. CMIS also had a 6.9 GHz channel, but with its cancellation there is a very high risk that a break will occur in the microwave SST climate record when AMSR-E and WindSat cease to operate. Both sensors are past their mission design lifetimes, and AMSR-E is experiencing some torque anomalies. The soil moisture CDR, which also requires the 6.9 GHz channels, will suffer a break as well.
Without any mitigation measures, the future for low-frequency, high-resolution microwave radiometry looks austere. A follow-on AMSR-2 is scheduled to fly on JAXA’s GCOM-W platform, but not until 2012, and no follow-on is planned for WindSat. NASA’s GMI radiometer is scheduled for launch in 2013, but it does not have the 6.9 GHz channels or the high spatial resolution of CMIS and AMSR. In addition, GMI will not view the high latitudes due to its low-inclination orbit. In 2016, assuming no more delays, NPOESS will launch MIS, a descoped version of CMIS with capabilities yet to be defined. One participant noted that this “thin thread of current and future microwave missions is completely inadequate for climate monitoring and research.” It was pointed out that a significant launch delay of MIS past 2016 could be disastrous. The DMSP F-series of satellites comes to an end at the end of the next decade. The microwave imagers SSM/I and SSMIS on these DMSP satellites have provided the research community with extremely important CDRs, including sea ice coverage, water vapor, wind speed, rainfall, and cloud water. A break in any of these time series due to a delay or aborted launch of MIS would be devastating to climate monitoring.
With respect to descoping CMIS to MIS, there was strong support for maintaining the low-frequency channels, particularly 6.9 GHz, and also for maintaining the high spatial resolution of AMSR-E and WindSat. Most participants considered these capabilities more important than maintaining the polarimetric channels for wind direction retrievals; the preferred approach for obtaining wind direction was via scatterometry.
Several mitigation strategies were discussed.
Mitigation Scenario 1. The first scenario involves making the most of what is possible now with AMSR-2 and MIS by advising NASA and NOAA to establish a memorandum of understanding with JAXA that would
make the AMSR-2 data and the supporting documentation that is required to develop CDRs freely available to the research community. The workshop discussion stressed the need for proper documentation for each satellite data stream to be freely available to the user community as an aid to application of the data within the CDR. In addition, science teams need to be funded to utilize the AMSR-2 data for climate research. However, there were some concerns expressed about relying too much on AMSR-2 because of past problems with platform stability and longevity, and some additional mitigation was thought to be highly desirable.
Mitigation Scenario 2. The second mitigation scenario is to add a 6.9 GHz channel to GMI. Currently the lowest channel on GMI is 11 GHz, and it is feasible that a 6.9 GHz channel could share the same feedhorn as the 11 GHz channel. It is also possible that the size of the GMI antenna could be increased. However, the GMI project has already undergone several delays, and it is not clear if these new modifications would be possible considering the current schedule. Another drawback is that SST in polar areas will not be observed by GMI.
Mitigation Scenario 3. The third mitigation scenario, most intriguing to many participants, is to enhance the microwave radiometer onboard the planned (but not yet funded) XOVWM, which has a suggested launch date around 2012. The synergy of an active scatterometer and a passive radiometer on the same platform is significant and would improve both the scatterometer vector wind retrievals and the radiometer SST retrievals. As currently planned the XOVWM radiometer has channels at 6.9 and 14 GHz. It also has a very large antenna that will provide higher spatial resolution than would AMSR-E. To obtain accurate SST retrievals, at least one higher-frequency channel would be required and the onboard calibration system would have to be improved. The feasibility of these enhancements needs to be investigated.
Mitigation Scenario 4. A final mitigation strategy is a free-flyer radiometer with AMSR-type capabilities. Existing radiometers such as GMI (with a 6.9 GHz channel), JAXA AMSR-2, or WindSat are all possibilities. However, this would be a costly scenario in that it would require an entirely new mission.
Other Breakout Group Discussions. In addition to mitigation strategies, a few other matters were discussed, including the idea of reinstating microwave sounding channels on the morning NPOESS platform. For this purpose, ATMS is preferable to sounding channels on MIS. Interest in this approach comes from the need to continue the MSU/AMSU tropospheric and stratospheric temperature CDRs without any spurious discontinuities. These temperature time series have been based on a combination of morning and afternoon orbits for the last 28 years and represent one of the most important CDRs coming from satellite remote sensing.
Scatterometry. Breakout group participants discussed the CDR that exists thus far for ocean vector winds, based primarily on 8 years of QuikSCAT measurements. Other platforms’ contributions were discussed, including those of ASCAT and WindSat. These discussions are briefly described here, although the discussion was extensive.
Some participants noted that the currently operating ASCAT scatterometer on MetOp will not maintain the CDR established by QuikSCAT, primarily because of sampling inadequacy; the combined coverage of the two parallel measurement swaths of ASCAT is only approximately 55 percent that of QuikSCAT. The 720 km gap between the two ASCAT swaths exacerbates these sampling problems. In addition, the spatial resolution of ASCAT is half that of QuikSCAT, which limits ASCAT’s usefulness in coastal applications to those that are about 50 km or farther from land, and in the resolution of small-scale features in the wind field such as hurricane structure, fronts, and jets. ASCAT also has a different wind directional ambiguity structure that results in larger potential errors in the interpretation of vector wind fields. Further, because of the reduced sensitivity of vertically polarized radar returns to high winds compared with horizontal polarization and the fact that ASCAT is a single-channel vertically polarized radar, the performance of ASCAT in high-wind conditions remains to be demonstrated.
Some participants also remarked on the difficulty of assessing the accuracy of WindSat estimates of wind speed and direction due to frequent updates of the wind retrieval algorithms under development by the Navy, although the evolving nature of these algorithms was not considered surprising in view of the newness of the passive microwave
technology for measurements of ocean surface vector winds. In presentations to the participants, WindSat wind retrievals (based on 4 years of data) were compared with QuikSCAT observations. Based on analyses of these comparisons, the following observations were made:
There is significantly larger wind direction uncertainty in WindSat retrievals at low-to-moderate wind speeds;
Depending on the version of the algorithm, WindSat wind retrievals can be biased either high or low in high-wind-speed conditions such as hurricanes and extratropical cyclones;
WindSat retrievals of wind vectors are more susceptible to error in cloudy and rainy conditions, which are often associated with extreme weather events; this susceptibility may affect the use of WindSat data in forecast systems and for wind warnings and the development of accurate climatologies of such events;
The spatial resolution of WindSat is less than half that of QuikSCAT;
The coverage of the WindSat measurement swath is only approximately 55 percent that of QuikSCAT; and
Passive measurements are much more subject to contamination by land in the antenna sidelobes; as a result, WindSat’s retrievals are not possible within approximately 75 km of land.
While some of these issues are being addressed by ongoing improvements in the WindSat retrieval algorithms, several participants expressed the strongly held view that passive microwave measurements would never be comparable in accuracy, coverage, or resolution to the measurements from a radar scatterometer. Passive microwave measurements would be especially problematic in cloudy and rainy conditions and for measurement of winds near land.
In the certified NPOESS program, CMIS has been descoped to MIS, which has not yet been defined in detail. Participants frequently commented that CMIS was adopted with no input from the scientific user community and with limited evidence of the capabilities of passive microwave for estimation of ocean surface vector winds. Regardless of whether MIS includes the polarimetric measurements required to estimate wind direction, it would result in a degradation of the accuracy, coverage, and resolution of ocean vector winds provided by QuikSCAT, especially in rainy conditions. Moreover, MIS would worsen the sampling of the wind field near land compared with QuikSCAT. MIS is therefore not a viable mitigation strategy for maintaining the ocean surface vector winds CDR.
India and China plan to launch scatterometers in 2008 and 2010, respectively. The instrument designs for these scatterometers are unknown and data availability remains uncertain for both missions. Neither of these scatterometers can therefore be considered viable mitigation strategies for continuation of the ocean surface vector winds CDR.
While QuikSCAT has provided many benefits and has established a baseline CDR for ocean surface vector winds, there are important limitations to the QuikSCAT data. For example, the Ku-band QuikSCAT radar cannot measure extreme winds or winds in heavy rain (although it can measure wind speeds of up to about 90 kt, if those winds occur outside of rain and are not confined to a very small area, both of which are the case in most hurricanes). QuikSCAT measurements are also limited to a spatial resolution of 12.5 km and are not routinely made closer than about 30 km from land.26 Many in the microwave breakout group argued that high priority should be given to a sustained, more capable, next-generation scatterometer program that can meet these requirements while at the same time continuing the ocean surface vector winds CDR established by QuikSCAT.
Since QuikSCAT is already 3 years past its designed instrument lifetime, it was a widely held view that continuation of the ocean surface vector wind CDR is in serious jeopardy. None of the currently operating or future planned instruments can continue the ocean surface vector winds CDR. Two mitigation scenarios were discussed. Both consist of a dedicated free-flyer scatterometer mission at the nearest possible opportunity in order to avoid, or at least minimize, a gap in the ocean surface vector winds CDR. This mission is envisioned as the first in a sequence of such missions.
See “Oceans Community Letter,” April 6, 2006, available at http://cioss.coas.oregonstate.edu/CIOSS/Documents/Oceans_Community_Letter.pdf.
Mitigation Scenario 1. The first scenario involves a QuikSCAT clone, which is the minimal solution for continuing the accuracy, resolution, and coverage of the 8-year ocean surface vector winds CDR established by QuikSCAT. The advantage is that a QuikSCAT clone is preliminarily estimated by NASA to be approximately 10 percent less expensive and could be readied 6 months sooner than the advanced scatterometer considered in the second scenario. The small percentage cost differential is because QuikSCAT is based on 1980s technology that would have to be updated to currently available electronic components. This updating would lead to a redesign of major instrument subsystems, thereby losing many of the cost advantages of a true “build-to-print” duplication of the QuikSCAT instrument. The disadvantage of a QuikSCAT clone is that some of the most important NOAA operational requirements established at the June 2006 NOAA Operational Ocean Surface Vector Winds Requirements Workshop27 would not be met (e.g., measurements of extreme winds, higher spatial resolution, and reduced contamination from rain and land).
Mitigation Scenario 2. The second scenario, preferred by many participants, consists of a next-generation synthetic-aperture-radar-based scatterometer mission referred to in the Earth science decadal survey as XOVWM. XOVWM would include a dual-frequency Ku-band and C-band radar and an X-band radiometer, which would allow measurements in rainy conditions, as well as measurements of the extreme winds in hurricanes and extratropical cyclones. The next-generation system would provide measurements with a resolution of better than 5 km and to within 1-3 km of land. XOVWM would thus satisfy most of the NOAA operational requirements, while at the same time maintaining the ocean surface vector winds CDR established by QuikSCAT and beginning a more accurate record of strong storms at sea, including hurricanes. The relatively minor disadvantages of XOVWM over a QuikSCAT clone are an approximate 10 percent cost increase (based on preliminary NASA estimates) and a 6-month longer delay to launch. The minor cost increase for XOVWM versus a QuikSCAT clone reflects the reality that even an attempt to duplicate the existing QuikSCAT would incur many of the nonrecurring costs of XOVWM, in part because of the long delay since QuikSCAT’s initial development and the obsolescence or unavailability of the hardware components used. XOVWM is a mission recommended in the decadal survey; several workshop participants argued that the proposed schedule for launch of this mission—2013-2016—be accelerated. Finally, while discussing this mitigation scenario, some participants indicated the desirability of an enhanced XOVWM+SST mission, a point that was also made during day 1 discussions.
Geostationary Hyperspectral Measurements
GOES-R is being developed as NOAA’s next generation of geostationary weather satellites. In late 2006, following large increases in estimates for completion of the program, NOAA canceled plans to incorporate a key instrument on the spacecraft—HES. HES was planned to provide both an advanced sounding capability for measurements of atmospheric temperature and moisture content and an imager for monitoring coastal water quality and assessing coastal hazards. Background on the HES instrument, along with a summary of the breakout participant discussions, is provided below.
Geostationary sounders provide unique, rapidly updated moisture profile measurements. In 1980, through the Operational Satellite Improvement Program (OSIP), NASA and NOAA partnered to fly a critical demonstration mission—the Visible and Infrared Spin Scan Radiometer (VISSR) Atmospheric Sounder (VAS). VAS was the first atmospheric temperature and moisture profiler flown in GEO. Subsequent three-axis-stabilized operational GOES-I-class sounders significantly improved upon VAS’s precision and have collected long-term records of atmospheric variables and diurnal cycles over the Western Hemisphere through the present time. These measure-
The NOAA Operational Ocean Surface Vector Winds Requirements Workshop, held June 5-7, 2006, at the National Hurricane Center in Miami, Florida, was sponsored by the Office of the Federal Coordinator for Meteorology. The final report of the workshop is available at http://www.ofcm.gov/tcr/reference/Ocean%20Surface%20Vector%20Winds_workshop_report_final.pdf.
ments will continue through the flight of the GOES-N/O/P series. With the termination of the GOES-R sounder, these long-term records will end.
The value of sounding from GEO, however, goes beyond maintenance of a long-term record. The ability to sense water vapor in the atmosphere is crucial for monitoring and predicting hazardous weather conditions. Large variations in atmospheric water vapor occur over fine scales of 10 km in the horizontal and 1 km in the vertical, and over tens of minutes; therefore, high-temporal-resolution monitoring is essential. The current GOES-N-class sounder temperature and moisture profiles provide relatively coarse temporal and spatial coverage, which is informative for indicating the synoptic-scale severe weather threat to areas, but insufficient for “nowcasting” cell development on the mesoscale or adequately resolving boundary-layer structures critical for nowcasts of severe thunderstorms.
Summary of Breakout Group Discussion
The GOES-R/HES breakout group session focused on mitigation options to restore the high-vertical-resolution temperature and water vapor sounding products and associated derived products planned for the HES payload on the GOES-R series. The breakout group did not address the coastal water imager because the ocean color community was not sufficiently represented. As noted above, the reader is advised that the options presented do not include all that might be considered, and that both the options and the analysis are necessarily the subjective and not always disinterested views of presenters and participants.
The breakout group heard a presentation regarding the importance that high-temporal-resolution hyperspectral observations of key atmospheric state variables and their trends have for climate data records. Such measurements are not easily made except from a geostationary orbit. The role of geostationary hyperspectral measurements in characterizing diurnal variations, identifying the sources, sinks, and transport of pollutants and greenhouse gases, and a potential key role in sensor intercalibration,28 were also discussed.
The case was then presented for advanced geostationary sounding capabilities as a contribution to GEOSS societal benefit areas, atmospheric ECVs, Numerical Weather Prediction capabilities improved by four-dimensional data assimilation, nowcasting capability, and sensor intercalibration.29 The value of nonclimate applications of such measurements was emphasized.
A presenter then reviewed the NESDIS Office of Systems Development Analysis of Alternatives (AoA) study,30 which considered a broad array of advanced geosynchronous sounder alternatives and trade-offs. The AoA study’s conclusions were discussed, particularly the need for an advanced sounder and space-based technology demonstration as early as feasible. It was suggested that previous ground system cost estimates were driven up by the inclusion of the coastal waters imager and that a recent proposal by NESDIS/STAR,31 considering only the advanced sounder in a demonstration mode, reduced the cost estimates significantly from the original estimates. In addition, the presenter noted the similarities between the AoA and Earth science decadal survey recommendations, which endorse the need for (at reasonable cost and risk) an operational advanced imaging sounder for GOES and an early demonstration. GIFTS was then introduced as a potentially viable option to get a demonstration instrument into GEO as early as possible. The presenter suggested that if launch services could be identified, such a mission could be done for approximately $150 million. This proposed track would not interfere with the GOES-R schedule but would retain the timing necessary to influence the design of the operational version for GOES-T. Concurrently, the presenter argued, reduced-capability advanced sounders should be developed for the GOES series.
For example, geostationary hyperspectral sounders are identified as a key component of a Global Space-Based Inter-calibration (GSICS) system. See http://www.star.nesdis.noaa.gov/smcd/spb/calibration/icvs/GSICS/index.html.
P. Ardanuy, B. Bergen, A. Huang, G. Kratz, J. Puschell, C. Schueler, and J. Walker, “Simultaneous Overpass Off Nadir (SOON): A method for unified calibration/validation across IEOS and GEOSS system of systems,” in Atmospheric and Environmental Remote Sensing Data Processing and Utilization II: Perspective on Calibration/Validation Initiatives and Strategies (A.H.L. Huang and H.J. Bloom, eds.), Proceedings of SPIE, Volume 6301, 2006.
NESDIS and OSD, Analysis of Alternatives, 2007. Participants in the AoA study included NOAA/NESDIS offices, university/cooperative institutes, contractors, DOD, and NASA.
NESDIS/STAR (Center for Satellite Applications and Research) is the new name for the former Office of Research and Applications.
Some attendees at the breakout group argued forcefully that an advanced sounder with HES-like capabilities would revolutionize short-term prediction, most notably of severe weather. Some workshop participants also referenced a NOAA/NESDIS-commissioned analysis of the potential economic benefits of the GOES-R ABI and HES instruments,32 which supported the economic justification for a HES-like capability. Advocates for including HES-like capabilities on GOES-R, which in this self-selected breakout group seemed to be most of the attendees, were very displeased by the indication during a plenary presentation by a NOAA official that an advanced hyperspectral sounder was “off the table” for GOES-R/S, and would most likely be next considered as a demonstration instrument on GOES-T. Some participants suggested that NASA and NOAA partner to achieve earlier GEO hyperspectral sounder capability, taking advantage of the inherent strengths of both agencies (and reinvigorating the OSIP).
Mitigation Scenario 1. Scenario 1 involves use of simulated sounder products taking advantage of only ABI observations. Many participants considered this option to be generally undesirable, as ABI lacks spectral, and therefore vertical, resolution and would be unable to provide the many products expected from HES.
Mitigation Scenario 2. Scenario 2 involves adding CrIS/ATMS back to the early morning (05:30) NPOESS orbit platforms. This remanifesting would add a useful additional pair of diurnal observations that would provide hyperspectral information. It would not, however, approach the temporal refresh available from geostationary orbit.
Mitigation Scenario 3. Participants suggested a scenario involving an opportunity for an early demonstration of GEO hyperspectral capabilities by launching GIFTS on a near-term flight of opportunity (i.e., free flyer or international partnership) to advance user readiness and allow algorithm development. It was noted that savings in nonrecurring engineering would be lost with this approach, as the demonstration unit (i.e., GIFTS) would not be the same as subsequent units, requiring subsequent demonstrations. Flight of an engineering model (rather than GIFTS) as a demonstration was seen as a way to save on nonrecurring engineering costs. However, there were differences of opinion among the group on the question of whether it would be less expensive or more desirable to launch GIFTS, build a different early demonstration model, or build the first flight model of the desired sounder.
Mitigation Scenario 4. Another potential approach to retaining (and advancing) the sounder capabilities on GOES was presented by a representative of ITT Space Systems who argued that the ITT “ABX” sounder is a simpler approach that could bridge the gap between the GOES-N legacy sounder and a full hyperspectral sounder on GOES-T. For GOES-R, the ABX would involve 18 sounding channels by reducing the ABI scan rate to improve the signal-to-noise ratio. This could “evolve” into a full hyperspectral capability by GOES-T using the preplanned product improvement (P3I) track. This option would allow retention and enhancement of existing capabilities, provision of GIFTS-like bands, and the potential for extensive reuse for subsequent flights. The perceived negative aspect of this solution is that a full hyperspectral demonstration may be delayed until GOES-T. Other proposed GOES-R series sounder options and paths have been considered by industry; given the competitive nature of such options, however, the representatives at the workshop indicated that they were not at liberty to share the specifics.
Other Discussions. It was stated that much of the cost of HES was attributable to the ground system requirements of NPOESS, which are driven by latency requirements. However, according to participants at the breakout session, latency is not a large concern of the hyperspectral community. Thus, most participants also argued that the cost savings that could result from a relaxation of the latency requirement should be pursued. Indeed, the demonstration mode referred to by presenters largely implies relaxation of latency as a cost-savings strategy.
Due to session time limitations, the HES breakout group was not able to consider the merit of a HES Observing System Simulation Experiment (OSSE).33 However, an expert on OSSEs provided a background handout for the group
Centrec Consulting Group, LLC, An Investigation of the Economic and Social Value of Selected NOAA Data and Products for Geostationary Operational Environmental Satellites (GOES). GOES-R Sounder and Imager Cost/Benefit Analysis, NOAA/NESDIS, 2007. The economic analysis suggested that the inclusion of hyperspectral sounding capability in addition to ABI would nearly double the socioeconomic benefit of GOES-R from $2.4 billion to $4.3 billion.
For details on OSSEs, see http://www.emc.ncep.noaa.gov/research/osse/.
and suggested to the chair of the session that a mesoscale OSSE for the HES instrument could be extremely valuable if done correctly. However, it would require considerable development and a great deal of caution for the conclusions of such a study to be deemed credible. Such a mesoscale OSSE has, to the workshop participants’ knowledge, never been done. Additional comments on the OSSE topic by European experts during the international videoconference session on day 3 suggested that the HES OSSE would be very difficult and likely not possible in a timely manner.
WORKSHOP SUMMARY—DAY 3
Plenary Session on International Considerations
On Thursday morning, the workshop held a joint international session, through videoconference, with participants at the World Meteorological Organization (WMO) “Workshop on the Re-design and Optimization of the Space-based Global Observing System” that was underway in Geneva, Switzerland. WMO workshop participants included high-level representatives of operational and research and development space agencies, the Committee on Earth Observations Satellites (CEOS), Global Climate Observing System (GCOS), the WMO Space Programme, the WMO Open Programme Area Group/Integrated Observing System (OPAG/IOS), and the Expert Team on Evolution of the Global Observing System (ET-EGOS). That workshop is expected to result in recommendations for both weather and climate monitoring from space being forwarded to the appropriate levels of WMO, the Coordination Group for Meteorological Satellites (CGMS), and CEOS. Anthony Hollingsworth (European Centre for Medium-Range Weather Forecasting; ECMWF) also participated in the videoconference from Reading, England.
WMO coordinates efforts for meeting the needs for climate information, such as for climate monitoring, climate-data management, climate-change detection, seasonal-to-interannual climate predictions, and assessments of the impacts of climate change. In the view of WMO representatives, measurements of the climate system should be considered as an operational requirement, and climate monitoring and climate measurements should be given equally high priority within the Global Observing System (GOS). In the WMO Rolling Review of Requirements process, climate requirements are represented by the GCOS Implementation Plan. The WMO presentation noted that taken as a whole, there has not been a concerted strategy for sustained climate observations from space. Instead, the climate community has relied on suboptimal sensors to create a climate record, resulting in significant challenges in terms of handling bias differences, orbit drift, data gaps, and spectral differences between follow-on instruments when reprocessing multiple-satellite data—often at considerable cost.
The CEOS presentation provided valuable insight into how various thematic issues could be addressed on a global basis utilizing the CEOS constellation concept, which considers virtual constellations of research and operational satellites to meet observational needs. Study teams have been established and international cooperation among space agencies has been stimulated to explore four representative Constellation prototypes, including atmospheric composition, global precipitation, land surface imaging, and ocean surface topography. It was noted by several speakers that the impact of NPOESS descoping was immediately significant in terms of GOS/GCOS planning and the quality of the CDRs for several variables.
The Global Monitoring for Environmental Security (GMES) and climate modeling presentation addressed key uncertainties identified by the IPCC Fourth Assessment report,34 global satellite provisions for atmospheric composition in the 2003-2019 time frame, European launch plans for 2007-2015, the GMES Sentinel program, and progress on the Global and regional Earth-system (Atmosphere) Monitoring using satellite and in situ data (GEMS) atmosphere project at ECMWF.
The need for hyperspectral observations from geostationary satellites was also addressed, including a discussion of their potential role in calibration of the space-based observing system (within those spectral ranges); monitoring of the diurnal cycle; and provision of spectrally resolved radiances (hyperspectral visible/near-IR and IR) as a climate reference.
Barbara Ryan, U.S. Geological Survey, reminded the teams that CEOS was strongly promoting an integrated observing system that included in situ data for ongoing verification and validation of satellite observations. In
Intergovernmental Panel on Climate Change, Climate Change 2007, IPCC Fourth Assessment Report, Cambridge University Press, Cambridge, U.K., 2008, available at http://www.ipcc.ch/ipccreports/assessments-reports.htm.
situ data are essential and complementary to the space segment data streams, enabling long-term monitoring of satellite data quality and as an independent component of the long-term climate record.
A number of other important considerations were brought forth during the videoconference. The importance of sustaining climate-quality climate data from space was addressed, along with the need to keep valuable space assets in operation after they have passed their design lifetime (e.g., Terra, Aqua, and Aura, which provide data for a variety of applications). There was recognition of the importance of determining how to preserve the heritage of past and current instruments with the natural evolution to advanced future instruments for extending climate records. It was further recognized that with limited financial and human resources, a response to GCOS requirements can be achieved only through enhanced international cooperation. Such cooperation should involve global planning with international contributions, in such a way that implementation problems encountered by an individual agency do not dramatically affect the global system. It was recognized that a number of missions planned in Europe will be of great value for climate analysis and that there is an acute need for better international collaboration and awareness spanning the full spectrum of activities from high-level data access agreements to pragmatic documentation exchange.
Concerning the NPOESS configuration, many participants supported:
Remanifesting hyperspectral IR and microwave sounders in the early morning orbit—both for operational purposes and for reanalysis and climate-related activity.
Maintaining continuity of microwave SST measurements at 6.9 GHz (AMSR-E type). With the loss of CMIS on C1 and increasing concern regarding the health of the AMSR-E on-board Aqua (indications of a failing antenna bearing), there is a significant risk of a microwave SST data gap prior to the launch of the Japanese GCOM mission; this could be addressed by the future MIS.
Maintaining a high-precision Jason-type altimeter in non-Sun-synchronous orbit (to mitigate the impact of tidal aliasing on sea level measurements) complemented by at least two other altimeter missions (Sentinel-3 will be one) in a Sun-synchronous orbit. This was stated as an urgent need by many participants.
Flying a CERES-class instrument for continuity of Earth radiation budget measurements.
Accelerating development of an active vector wind mission to replace QuikSCAT.
Finally, during the closing plenary session, there was discussion again of the requirements for constructing, managing, and maintaining CDRs. As in previous sessions, participants discussed the intellectual and resource challenges in developing CDRs, which require attention and adequate budgets in the space segment, ground segment, and CDR production units themselves. It was noted that at present, the last is often limited in resources so that problems with satellite data are only discovered following dedicated ad hoc CDR processing projects. Some participants stated that considerable cost benefits would almost certainly be realized if CDR processing could be sustained in an operational near-real-time-style environment.
A general theme of the videoconference echoed the need for organizations to work together with synergies among international satellite programs and the importance of multilateral agreements in addressing climate monitoring. In the future, it is through effective international cooperation and global partnerships that useful climate monitoring from space will be realized. A frequently expressed sentiment was that the joint Geneva-Washington session was extremely important in terms of bringing the international satellite climate community together and that such communication should be encouraged through future meetings.
The breakout groups on day 3 were loosely organized to enable participants to offer comments. Two panels were given specific topics, namely, to assess the NASA-NOAA suggested mitigation options and to further explore the intricacies of CDR development. These two breakout sessions are summarized here. A third breakout session allowed participants to comment on any topic within the scope of the workshop, and key points have been integrated into this report where relevant (many are covered in Chapter 3) and will be considered further during a follow-on study.
Panel to Assess NASA-NOAA Mitigation Options
The breakout panel reviewed a summary (see Appendix C) of the draft NASA-NOAA white paper titled “Mitigation Approaches to Address Impacts of NPOESS Nunn-McCurdy Certification on Joint NASA-NOAA Goals.”35 The Office of Science and Technology Policy (OSTP) had asked NOAA and NASA to provide this analysis of possible options for mitigation of the climate research impacts of the NPOESS Nunn-McCurdy certification through 2026, along with an assessment of the potential costs of these options, with the primary goal of ensuring the continuity of long-term climate records. The primary goal of the NASA-NOAA white paper was to identify means for ensuring the continuity of long-term climate records.
NASA noted that the white paper was based on a single sentence from the June 5, 2006, Nunn-McCurdy Acquisition Decision Memorandum: “[The restructured program] does not include funding for the following sensors: APS, TSIS, OMPS-Limb, ERBS, ALT, SuS, and the full SESS; however, the program will plan and fund for integration of these sensors onto the satellite buses, if the sensors are provided from outside the program.”36
The options presented in the draft white paper represent a departure from the traditional NPOESS/EOS/MetOp big-platform approach. They are a combination of NPOESS operational flights, accommodations of opportunity, and “climate free flyers.” These focused missions would be dedicated to a limited number of specialized sensors; simpler instruments could have dedicated functions (e.g., to separate reflective from emissive bands). The apparent intent is to use a constellation approach to obtain as many complementary measurements as possible through formation flight.
The panel was encouraged by the imagination shown by the NASA-NOAA team and was extremely supportive of their ideas for implementation flexibility—specifically including flights on diverse platforms, including formation flight with NPOESS. However, the white paper options focused on only five instruments: TSIS, ERBS/CERES, ALT, OMPS, and APS. NASA noted that the white paper does not consider mitigation options for VIIRS, CrIS/ATMS, CMIS/MIS, and SESS. Some workshop participants commented that the lack of attention to the other instruments should not be construed as a de facto lower prioritization of their suitability as options for mitigation of lost capabilities. NASA and NOAA will expand the white paper options to consider the other sensors that will fly, revising the white paper based on comments from this workshop. They plan to deliver a revised draft to OSTP by late summer.
Panel on Issues Related to CDR Development
Underemphasized during certain sessions of the workshop, but recognized as fundamental for ensuring the climate record from space, is the technical issue of generating the needed CDRs from the operational EDRs. Crucial issues include the accommodation of ancillary observations critical for CDRs but absent from the current and planned satellite systems, and the ability to adequately develop and maintain CDRs. The breakout session considered requirements for CDRs (particularly in contrast to EDR retrievals) and the adequacy of current (post-Nunn-McCurdy) plans for prelaunch instrument calibration and characterization; on-orbit calibration and validation; measurement overlap and replenishment requirements; and data storage, archiving, distribution, reprocessing, analysis, and interpretation concerns.
Presenters and many participants at the breakout session echoed a concern that the fundamental definitions of EDRs and CDRs and the requirements for CDR generation and maintenance are not adequately understood by the operational and research community. Proper communication of requirements for CDRs requires that these distinctions be clearly understood. According to presenters from the NOAA National Climatic Data Center, even though the sensor signals used to generate EDRs are also used for CDRs, the EDRs themselves are frequently of little use for climate research. EDRs are (in general) poorly calibrated, quick-turnaround products that lack long-term repeatability, whereas CDRs are fully calibrated time series having high precision (repeatability) and accuracy, often requiring reprocessing of entire data sets as algorithms are improved (Box 2.1).
Generation of Climate Data Records
The instruments and data system for NPOESS are designed to produce a number of operational geophysical products, which are called environmental data records (EDRs). EDRs are generally produced by applying an appropriate set of algorithms to raw data records. Although NPP- and NPOESS-derived EDRs may have considerable scientific value, climate data records (CDRs)a are far more than a time series of EDRs. Participants at the workshop emphasized the fundamental differences between products that are generated to meet short-term needs (EDRs) and those for which consistency of processing and reprocessing over years to decades is an essential requirement.
Climate research and monitoring often require the detection of very small changes against a naturally noisy background. For example, sea surface temperatures can vary by several degrees between daytime and nighttime, or from year to year, whereas the climate signal of interest may change by only 0.1 K over a decade. Moreover, changes in sensor performance or data-processing algorithms often introduce artificial noise that may be greater than the climate signal. In addition to natural and artificial noise, spatial and temporal biases in the measurements confound climate researchers. A CDR suitable for studying interannual to decadal climate variability and trends includes a time series produced with stable, high-quality data, and error characteristics that have been quantified by accounting for all of the above sources of error and noise. The production of a CDR requires considerable refinement of the raw data and the blending of multiple data streams. These streams may come from multiple copies of the same sensor, or they may be ancillary data fields that are used in synergy with the primary data stream.b Thorough analysis of sensor performance and improved processing algorithms are also required, as are quantitative estimates of spatial and temporal errors. Figure 2.1.1 illustrates the notional pathways that result in generation of an EDR and a CDR.c
The past experience of the climate research community with the Microwave Sounding Unit (MSU) and Advanced Microwave Sounding Unit (AMSU) provides a constructive case study in the challenges associated with constructing CDRs with satellite data. Starting in late 1978, nine polar-orbiting satellites carried identical copies of the MSU to measure atmospheric temperatures. In a 2000 National Research Council report,d it was noted that the last MSU occupied the afternoon orbit slot (NOAA-14), while the morning slot was monitored by the AMSU on NOAA-15.e Constructing CDRs from MSU instruments revealed that even though the prelaunch instruments were essentially identical, postlaunch differences among them were as large as the climate signal being sought. Once in space, each was found to have a unique response to variations in direct solar heating. Others experienced shifts in responses to onboard calibration targets. Another was found, after launch, to have been improperly calibrated in the laboratory. A final complication was due to the fact that the frequencies monitored with the new AMSU were slightly different from those monitored with the legacy MSUs. Scientists who were interested in stable, long-term temperature records from the MSU were required to commit considerable resources to discover the aforementioned problems and to test adjustments.
A similar example is seen in the generation of sea surface temperature CDRs. Sea surface temperature (SST) CDRs were improved through several joint agency efforts (e.g., NOAA-NASA Pathfinder program) and,and, more recently, merging of complementary infrared and passive microwave satellite data having global daily coverage together with in situ observations as part of the international Global High Resolution SST Pilot Project (GHRSST-PP).f The GHRSST-PP is also pioneering the development of a high-resolution SST CDR within a dedicated reanalysis project, led by the NOAA’s National Oceanic Data Center, for the satellite era (1981-present).
Calibration and validation in the context of CDRs can be considered a process that encompasses the entire system, from sensor to data product. The objective is to develop a quantitative understanding and characterization of the measurement system and its biases in time and space, which involves a wide range of strategies that depend on the type of sensor and data product.
Many participants noted that CDR science teams are crucial for maintaining the CDRs over many years (climate change time scales are long compared with those for weather), a task that is expected to require additional research, analysis, and validation of the observations (and thus funding, well beyond that applied to the EDRs). Prelaunch calibration and characterization that meet EDR requirements do not always (typically) meet the more exhaustive requirements for CDR accuracy and stability. Data-handling requirements are also completely different from those for EDRs and will likely require an independent CDR system.
Whereas functionally the EDRs are short-lived operational products, the CDRs must be permanently stored and continuously accessible for considerable additional ongoing research and analysis if they are to be of use in climate change policy making and societal applications. Given that data requirements for CDRs can exceed those for EDRs, a list of missing data should be developed and considered as part of the mitigation option analysis. Participants also noted that where CDRs are particularly affected by the demanifesting of a sensor (e.g., APS), restoring the sensor without the capability for long-term CDR generation and maintenance is of little benefit.
The workshop breakout session discussion specifically avoided defining agency roles and responsibilities, consistent with the workshop’s overall focus on identifying various options, but not their funding source. Further, the session participants suggested that the forthcoming National Research Council study on a strategy to mitigate the climate impacts that resulted from the NPOESS restructuring also avoid any attempt to assess costs or agency responsibilities, noting that these efforts should be initiated by the government in response to general study findings and recommendations regarding CDR generation requirements.
Personnel training and maintenance of scientific capability over the long term were cited as essential elements of successful CDR development. It was noted that although operational programs also require skill continuity, the types and levels of skills required for CDRs are substantially more demanding and therefore more expensive to maintain than those for EDRs.