The ocean is a fundamental component of Earth’s biosphere. Because the ocean is so vast and difficult for humans to explore, satellite remote sensing of ocean color is currently the only way to observe and monitor the biological state of the surface ocean globally on time scales of days to decades.
The ocean covers roughly 70 percent of Earth’s surface and plays a pivotal role in the cycling of life’s building blocks such as nitrogen, carbon, oxygen, and sulfur. The ocean also contributes to regulating the climate system. For example, the land and ocean together removed 57 percent of all anthropogenic carbon dioxide (CO2) emissions from 1958 to 2009,1 with the ocean accounting for about half of this. By removing CO2 from the atmosphere, the ocean moderates the rate of human-induced climate change. In addition, the CO2 dissolving in the ocean produces carbonic acid, which is causing the ocean to become more acidic. Moreover, the ocean has absorbed approximately 90 percent of the increased heat associated with climate change (Lyman et al., 2010). As the ocean grows warmer and more acidic, these changes may have adverse effects on whole groups of marine organisms (NRC, 2010). Additional stressors—such as overfishing, nutrient pollution from land runoff, coastal development, and invasive species—further jeopardize the health of the ocean and the vital functions it provides (NRC, 2004a).
Monitoring the health of the ocean and its productivity is critical to understanding and managing the ocean’s essential functions and living resources. Phytoplankton are microscopic organisms responsible for most of the primary production2 in the ocean, are ubiquitous in the surface ocean, and form the base of the marine food web. Tracking changes in phytoplankton in the vast expanse of the ocean requires a perspective that can be gained only from satellite measurements (NRC, 2008a). Ocean color measurements from space have revolutionized our understanding of the ocean on every scale, from local to global and from days to decades.
Ocean color measurements reveal a wealth of ecologically important characteristics including: chlorophyll concentration (a proxy for the biomass of marine plants or phytoplankton), the rate of phytoplankton photosynthesis, sediment transport, dispersion of pollutants, and responses of oceanic biota to long-term climate changes (IOCCG, 2008). Many scientists and operational users, such as managers of coastal resources and fisheries, rely on these measurements for research, ecosystem monitoring, and resource management.
DERIVING OCEAN PROPERTIES FROM OCEAN COLOR RADIANCE
Deriving biological parameters from ocean color measurements is a multi-stage process. Ocean color radiometric sensors measure the upwelling radiance at the top of the atmosphere (LTOA). As illustrated in Figure 1.1, LTOA is the total radiances from three sources: water-leaving radiance (Lw) radiance reflected from the sea surface (surface-reflected radiance), and radiance scattered into the viewing direction by the atmosphere along the path between the sensor and sea surface (atmospheric path radiance).
Of these three radiance sources, the desired measurement is Lw, referred to in this report simply as ocean color. Lw carries information about the biological and chemical constituents in the near-surface waters. To obtain Lw, it is necessary to deduce and remove the contributions of surface reflection and atmospheric path radiance from the measured total, a process known as atmospheric correction. This is difficult because Lw is no more than 10 percent of LTOA, as illustrated in Figure 1.2.
There are four levels of processing of satellite data:
Level 0: Raw data as measured directly from the spacecraft in engineering units (e.g., volts or digital counts).
Level 1: Level 0 data converted to TOA radiance using pre-launch sensor calibration and characterization information adjusted during the life of the mission by vicarious calibration and stability monitoring (for details see Chapter 3). For scientific applications, and in particular to generate
1 See http://www.globalcarbonproject.org/carbonbudget/09/hl-full. htm#naturalSinks; accessed 1/7/2011.
2 Primary production or photosynthesis converts carbon dioxide and water into carbohydrates and oxygen in the presence of light.
FIGURE 1.1 Qualitative illustration of the contributions of water-leaving radiance Lw surface glint, and atmospheric path radiance to the measured TOA radiance.
SOURCE: Adapted from http://www.gps.gov/multimedia/images/.
FIGURE 1.2 Quantitative illustration of the contributions of water-leaving, surface-reflected, and atmospheric path radiance to the measured TOA radiance. The water-leaving radiance—the signal—is at most 10 percent of the TOA radiance (simulations by the HydroLight and Modtran radiative transfer models using typical oceanic and atmospheric properties and 10-nm wavelength resolution).
Climate Data Records (CDRs), it is essential to archive Level 0 data, pre-launch calibration and characterization information, and post-launch calibration and stability monitoring data to enable periodic reprocessing of the raw data. Note: CDRs have been defined as “time-series of measurements of sufficient length, consistency, and continuity to determine climate variability and change,” in the NRC report on CDRs from Environmental Satellites (NRC, 2004b).
Level 2: Level 2 data are generated from Level 1 data following atmospheric correction that are in the same satellite viewing coordinates as Level 1 data (i.e., the data have not been mapped to a standard map projection or placed on a grid). Level 2 data include Lw and derived products. Satellite viewing angles and other information are used to map any single Level 2 scene to a standard map projection (see definition of Level 3 data). Lw or ocean color radiance is generated from Level 1 radiance following atmospheric correction. Atmospheric correction for optically deep water3 requires sensor measurements at near and short wave infrared wavelengths, ancillary measurements such as sea-level atmo-
3 Optically deep water refers to water that is deep enough that the bottom reflectance does not contribute to the water-leaving radiance.
spheric pressure and wind speed, and models of atmospheric aerosol properties. The resulting measurement of ocean color radiance is a well-defined geophysical property whose measurement adheres to national and international standards. Ocean color radiance is considered the fundamental product from which all other ocean color products are derived.
Note: Although there is community consensus on the meaning of ocean color radiance, there are multiple approaches and algorithms for generating various derivative products such as measures of chlorophyll or primary production. Thus, it is critical that the path from Level 0 to Level 2 be well understood and documented and that the data to make these conversions be permanently archived. Periodic reprocessing begins with Level 0 data and uses knowledge of how sensor calibration has changed with time, better ancillary information, improved algorithms, and other lessons learned during the mission. Reprocessing is an essential mission requirement for generating quantitative data products, particularly climate data records. Moreover, standard atmospheric correction techniques for Level 2 processing are designed for open ocean waters and might not perform well in turbid coastal water, optically shallow water, and in coastal areas experiencing atmospheric pollutants and dust.
Level 3: Level 3 products are those that have been mapped to a known cartographic projection or placed on a two-dimensional grid at known spatial resolution. Level 0, 1, and 2 products are expressed in satellite coordinates and are not particularly useful to most applications of satellite data. Level 3 data products are often aggregated over time or space. These products are widely disseminated to scientific and operational users.
Level 4: Although gridded satellite data provide far better coverage in space and time than is possible with in situ data, most users want to validate such maps independently for their regions of study through comparisons with in situ data. Results derived from a combination of satellite data and ancillary information, such as ecosystem model output, are called Level 4 products.
New and better algorithms and ocean color products continue to emerge as technology and atmospheric corrections improve. As the scientific understanding advances regarding the relationships between ocean color radiance and the types of particulate and dissolved substances found in water, new and improved algorithms and ocean color products continue to emerge. Primary products (Figure 1.3) are derived with algorithms that rely exclusively on Lw and its relationship to the desired product, such as chlorophyll concentration. Secondary products require knowledge about their relationships to Lw as well as ancillary information obtained from other sensors, in situ observations, or models.
Chlorophyll concentration, the best known and most commonly used ocean color product, is an example of a primary product. The algorithms for determining it are well developed, and satellite-derived chlorophyll values have been validated at various scales, from single images to global composites, and used for a broad array of applications. Nevertheless, assumptions inherent to these algorithms need to be continuously tested and updated in a changing ocean. Particulate organic and inorganic carbon concentrations and Colored Dissolved Organic Matter (CDOM) absorption characteristics can also be derived from Lw spectra. A growing number of new primary products are being developed, such as inherent optical properties (e.g., phytoplankton absorption and backscatter coefficients) and concentrations of other suspended material, including various components of the particulate carbon and dissolved carbon pools in the ocean.
Marine net primary production,4 a secondary product (Figure 1.3), illustrates the utility of ocean color measurements when combined with high-quality in situ data. Estimating net primary production requires ocean color measurements as well as other sources of information such as sea surface temperature or mixed layer depths. The importance of in situ data to enhance ocean color remote sensing will be revisited in Chapter 5. Scientists also use ocean color measurements in combination with other data to learn about the composition of phytoplankton. They accomplish this either by partitioning the total chlorophyll concentration into major size classes (pico-, nano- and micro-phytoplankton) or into major phytoplankton functional groups (diatoms, coccolithophores, blue-green algae, floating sargassum), or by identifying nuisance or harmful algal blooms. However, some methods for retrieving phytoplankton functional types are estimated directly from ocean color radiance (e.g., the diatom discrimination algorithm of Sathyendranath et al., 2004; the algorithm of Alvain et al., 2005).
Over the past three decades, the oceanographic community has witnessed astounding growth in the capabilities of ocean color remote sensing. The Sea-viewing Wide Field-of-view Sensor-Moderate Resolution Imaging Spectroradiometer (SeaWiFS-MODIS) era from 1997 to present has provided scientists with a high-quality, well-calibrated Lw time-series from which to estimate chlorophyll concentration and primary production. As a result, for the first time, a climate-quality5 data record can be compiled to demonstrate the strong link between interannual climate variability and the marine biosphere during the El Niño to La Niña transition on ocean-basin scales. Many of these recent discoveries and accomplishments in biological oceanography have been described in Earth Observations from Space: The First 50
4 Net primary production quantifies the net conversion of carbon dioxide and water into carbohydrates and oxygen in the presence of light and represents the energy supply to the base of marine food webs.
5 To demonstrate long-term trends in a time-series with large natural interannual variability, the data record requires very high accuracy. For the ocean color climate record, accuracy requirements are discussed in Chapters 3-5.
FIGURE 1.3 Ocean color radiance is used to derive products directly or indirectly. Secondary products are based on the primary products and ancillary data. These products are then used to address scientific and societal questions. Some satellite missions apply the vicarious calibration when processing Level 2 data. (CDOM: Colored Dissolved Organic Matter; PAR: Photosynthetically Available Radiance; PIC: Particulate Inorganic Carbon; POC: Particulate Organic Carbon; K490: diffuse attenuation coefficient at 490 nm; HAB: Harmful Algal Bloom).
Years of Scientific Achievements (NRC, 2008a) and a recent International Ocean Colour Coordinating Group report (IOCCG, 2008).
To sustain and build on these achievements, the climate research community requires access to uninterrupted climate-quality data records for the marine biosphere. Such records are central to validating new, more sophisticated climate models that incorporate biogeochemical processes, such as primary production, and validating and improving the accuracy of products in a changing ocean.
Moreover, as detailed in the research community’s plan entitled Advanced Plan for the Ocean Biology and Biogeochemistry Program (NASA, 2007); continued support for ocean color remote sensing is needed to improve computer-based modeling of ecosystem dynamics. Ocean color remote sensing data are necessary to build accurate and useful models related to climate change, which will increase our understanding of variability of in situ organic and inorganic carbon constituents; continental shelf ecosystem dynamics and variability of the mixed-layer thickness; and variability of particulates and aerosols in the ocean and atmosphere. Furthermore, as the length of the climate-quality ocean color
data record grows, its intrinsic value for recording long-term changes in the marine ecosystem also increases.
It is important to note that, with the availability of routine measurements and with free and easy access to ocean color data, the user community has expanded dramatically to include state and federal coastal and fisheries resource managers who depend on the data for ecosystem monitoring (e.g., coral ecosystems and harmful algal blooms). Therefore, any gap in the time-series—regardless of how short—would be detrimental not only to the ocean color and climate research community but also to resource managers.
To assess the risk for losing access to high-quality Lw data and to identify mitigation options, the National Oceanic and Atmospheric Administration (NOAA), NASA, the National Science Foundation (NSF), and the Office of Naval Research (ONR) asked the National Academy of Sciences (NAS) to convene an ad hoc committee. The committee was asked to assess the requirements to sustain global ocean color research and operational applications (see Box 1.1 for the statement of task).
Since the task statement was written, significant changes related to ocean color remote sensing have occurred that have shifted the baseline for this study. Most significantly, in February 2010, the White House ordered the restructuring of the National Polar-orbiting Operational Environmental Satellite System (NPOESS) program and a separation of the civilian programs from the Department of Defense (DOD) program. The civilian portion of the NPOESS program has become the Joint Polar Satellite System (JPSS). In addition, the latest Visible Infrared Imager Radiometer Suite (VIIRS) characterization yielded positive results and cautious optimism about the sensor’s performance (see Chapter 3 for a detailed discussion). In more good news, planning has started at NASA for a new ocean color mission, the Pre-AerosolClouds-Ecosystem (PACE) mission, with a launch date of 2019 or later.
In December 2010, the SeaWiFS mission ceased operation. Thus the ocean color community lost the sensor that had become the gold standard for ocean color remote sensing.
Lastly, the Deep Water Horizon oil spill that began in April 2010 was a stark reminder of coastal communities’ dependence on healthy marine ecosystems. The spill reinforced the ways in which human activities can jeopardize those ecosystems and the communities that rely on them for their livelihoods and survival. Remote sensing from both planes and satellites was critical in monitoring and projecting the evolution of the oil slick and highlighted the importance of ocean color remote sensing to an oil spill response.
To address the task, this report identifies in Chapter 2 the research and operational applications for ocean color products and the data specifications to generate them. Chapter 3 evaluates lessons from past and current sensors and missions. Based on these lessons learned, the committee establishes the minimum requirements for sustaining the capability to obtain remotely sensed ocean color data. Chapter 4 assesses the gaps in meeting the requirements, evaluates current and future capabilities of U.S. and foreign missions, and provides options to minimize the risk of a data gap in the near term. Chapter 5 provides a long-term view; it describes challenges in meeting all research and operational requirements and lists many existing opportunities for building on lessons learned and advancing current capabilities.
Statement of Task
Continuity of satellite ocean color data and associated climate research products are presently at significant risk for the U.S. ocean color community. Temporal, radiometric, spectral, and geometric performance of future global ocean color observing systems must be considered in the context of the full range of research and operational/application user needs. This study aims to identify the ocean color data needs for a broad range of end users, develop a consensus for the minimum requirements, and outline options to meet these needs on a sustained basis.
An ad hoc committee will assess lessons learned in global ocean color remote sensing from the SeaWiFS/MODIS era to guide planning for acquisition of future global ocean color radiance data to support U.S. research and operational needs. In particular, the committee will assess the sensor and system requirements necessary to produce high-quality global ocean color climate data records that are consistent with those from SeaWiFS/MODIS. The committee will also review the operational and research objectives, such as described in the Ocean Research Priorities Plan and Implementation Strategy, for the next generation of global ocean color satellite sensors and provide guidance on how to ensure both operational and research goals of the oceanographic community are met. In particular the study will address the following:
1. Identify research and operational needs, and the associated global ocean color sensor and system high-level requirements for a sustained, systematic capability to observe ocean color radiance (OCR) from space;
2. Review the capability, to the extent possible based on available information, of current and planned national and international sensors in meeting these requirements (including but not limited to: VIIRS on NPP and subsequent JPSS spacecrafts; MERIS on ENVISAT and subsequent sensors on ESA’s Sentinel-3; S-GLI on JAXA’s GCOM-C; OCM-2 on ISRO’s Oceansat-2; COCTS on SOA’s HY-1; and MERSI on CMA’s FY-3);
3. Identify and assess the observational gaps and options for filling these gaps between the current and planned sensor capabilities and timelines; define the minimum observational requirements for future ocean color sensors based on future oceanographic research and operational needs across a spectrum of scales from basin-scale synoptic to local process study, such as expected system launch dates, lifetimes, and data accessibility;
4. Identify and describe requirements for a sustained, rigorous on-board and vicarious calibration and data validation program, which incorporates a mix of measurement platforms (e.g., satellites, aircraft, and in situ platforms such as ships and buoys) using a layered approach through an assessment of needs for multiple data user communities; and
5. Identify minimum requirements for a sustained, long-term global ocean color program within the United States for the maintenance and improvement of associated ocean biological, ecological, and biogeochemical records, which ensures continuity and overlap among sensors, including plans for sustained rigorous on-orbit sensor inter-calibration and data validation; algorithm development and evaluation; data processing, re-processing, distribution, and archiving; as well as recommended funding levels for research and operational use of the data.
The review will also evaluate the minimum observational research requirements in the context of relevant missions outlined in previous NRC reports, such as the NRC “Decadal Survey” of Earth Science and Applications from Space. The committee will build on the Advance Plan developed by NASA’s Ocean Biology and Biogeochemistry program and comment on future ocean color remote sensing support of oceanographic research goals that have evolved since the publication of that report. Also included in the review will be an evaluation of ongoing national and international planning efforts related to ocean color measurements from geostationary platforms.