Robert E. Eplee

Calibration of SeaWiFS. II. Vicarious techniques.

We present an overview of the vicarious calibration of the Sea-Viewing Wide Field-of-view Sensor (SeaWiFS). This program has three components: the calibration of the near-infrared bands so that the atmospheric correction algorithm retrieves the optical properties of maritime aerosols in the open ocean; the calibration of the visible bands against in-water measurements from the Marine Optical Buoy (MOBY); and a calibration-verification program that uses comparisons between SeaWiFS retrievals and globally distributed in situ measurements of water-leaving radiances. This paper describes the procedures as implemented for the third reprocessing of the SeaWiFS global mission data set. The uncertainty in the near-infrared vicarious gain is 0.9%. The uncertainties in the visible-band vicarious gains are 0.3%, corresponding to uncertainties in the water-leaving radiances of approximately 3%. The means of the SeaWiFS/in situ matchup ratios for water-leaving radiances are typically within 5% of unity in Case 1 waters, while chlorophyll a ratios are within 1% of unity. SeaWiFS is the first ocean-color mission to use an extensive and ongoing prelaunch and postlaunch calibration program, and the matchup results demonstrate the benefits of a comprehensive approach.

Calibration of SeaWiFS. I. Direct techniques

We present an overview of the calibration of the Sea-viewing Wide Field-of View Sensor (SeaWiFS) from its performance verification at the manufacturers facility to the completion of its third year of on-orbit measurements. These calibration procedures have three principal parts: a prelaunch radiometric calibration that is traceable to the National Institute of Standards and Technology; the Transfer-to-Orbit Experiment, a set of measurements that determine changes in the instruments calibration from its manufacture to the start of on-orbit operations; and measurements of the sun and the moon to determine radiometric changes on orbit. To our knowledge, SeaWiFS is the only instrument that uses routine lunar measurements to determine changes in its radiometric sensitivity. On the basis of these methods, the overall uncertainty in the SeaWiFS top-of-the-atmosphere radiances is estimated to be 4-5%. We also show the results of comparison campaigns with aircraft- and ground-based measurements, plus the results of an experiment, called the Southern Ocean Band 8 Gain Study. These results are used to check the calibration of the SeaWiFS bands. To date, they have not been used to change the instruments prelaunch calibration coefficients. In addition to these procedures, SeaWiFS is a vicariously calibrated instrument for ocean-color measurements. In the vicarious calibration of the SeaWiFS visible bands, the calibration coefficients are modified to force agreement with surface truth measurements from the Marine Optical Buoy, which is moored off the Hawaiian Island of Lanai. This vicarious calibration is described in a companion paper.

On-orbit calibration of the Suomi National Polar-Orbiting Partnership Visible Infrared Imaging Radiometer Suite for ocean color applications

The NASA Ocean Biology Processing Group (OBPG) developed two independent calibrations of the Suomi National Polar-Orbiting Partnership (SNPP) Visible Infrared Imaging Radiometer Suite (VIIRS) moderate resolution reflective solar bands using solar diffuser measurements and lunar observations, and implemented a combined solar- and lunar-based calibration to track temporal changes in radiometric response of the instrument. Differences between the solar and lunar data sets have been used to identify issues and verify improvements in each. Linearization of the counts-to-radiance conversion yields a more consistent calibration at low radiance levels. Correction of a recently identified error in the VIIRS solar unit vector coordinate frame has been incorporated into the solar data and diffuser screen transmission functions. Temporal trends in the solar diffuser stability monitor data have been evaluated and addressed. Fits to the solar calibration time series show mean residuals per band of 0.067%-0.17%. Periodic residuals in the VIIRS lunar data are confirmed to arise from a wavelength-dependent libration effect for the sub-spacecraft point in the output of the U.S. Geological Survey Robotic Lunar Observatory photometric model of the Moon. Temporal variations in the relative spectral responses for each band have been assessed, and significant impact on band M1 (412 nm) lunar data has been identified and rectified. Fits to the lunar calibration time series, incorporating sub-spacecraft point libration corrections, show mean residuals per band of 0.069%-0.20%. Lunar calibrations have been used to adjust the solar-derived radiometric corrections for bands M1, M3, and M4. After all corrections, the relative differences in the solar and lunar calibrations for bands M1-M7 are 0.093%-0.22%. The OBPG has achieved a radiometric stability for the VIIRS on-orbit calibration that is commensurate with those achieved for SeaWiFS and Aqua MODIS, supporting the incorporation of VIIRS data into the long-term NASA ocean color data record.

Cross calibration of SeaWiFS and MODIS using on-orbit observations of the Moon.

Observations of the Moon provide a primary technique for the on-orbit cross calibration of Earth remote sensing instruments. Monthly lunar observations are major components of the on-orbit calibration strategies of SeaWiFS and MODIS. SeaWiFS has collected more than 132 low phase angle and 59 high phase angle lunar observations over 12 years, Terra MODIS has collected more than 82 scheduled and 297 unscheduled lunar observations over nine years, and Aqua MODIS has collected more than 61 scheduled and 171 unscheduled lunar observations over seven years. The NASA Ocean Biology Processing Group Calibration and Validation Team and the NASA MODIS Characterization Support Team use the USGS RObotic Lunar Observatory (ROLO) photometric model of the Moon to compare these time series of lunar observations over time and varying observing geometries. The cross-calibration results show that Terra MODIS and Aqua MODIS agree, band to band, at the 1%-3% level, while SeaWiFS and either MODIS instrument agree at the 3%-8% level. The combined uncertainties of these comparisons are 1.3% for Terra and Aqua MODIS, 1.4% for SeaWiFS and Terra MODIS, and 1.3% for SeaWiFS and Aqua MODIS. Any residual phase dependence in the ROLO model, based on these observations, is less than 1.7% over the phase angle range of -80° to -6° and +5° to +82°. The lunar cross calibration of SeaWiFS, Terra MODIS, and Aqua MODIS is consistent with the vicarious calibration of ocean color products for these instruments, with the vicarious gains mitigating the calibration biases for the ocean color bands.

Comparison of SeaWiFS measurements of the Moon with the U.S. Geological Survey lunar model

The Sea-Viewing Wide-Field-of-View Sensor (SeaWiFS) has made monthly observations of the Moon since 1997. Using 66 monthly measurements, the SeaWiFS calibration team has developed a correction for the instruments on-orbit response changes. Concurrently, a lunar irradiance model has been developed by the U.S. Geological Survey (USGS) from extensive Earth-based observations of the Moon. The lunar irradiances measured by SeaWiFS are compared with the USGS model. The comparison shows essentially identical response histories for SeaWiFS, with differences from the model of less than 0.05% per thousand days in the long-term trends. From the SeaWiFS experience we have learned that it is important to view the entire lunar image at a constant phase angle from measurement to measurement and to understand, as best as possible, the size of each lunar image. However, a constant phase angle is not required for using the USGS model. With a long-term satellite lunar data set it is possible to determine instrument changes at a quality level approximating that from the USGS lunar model. However, early in a mission, when the dependence on factors such as phase and libration cannot be adequately determined from satellite measurements alone, the USGS model is critical to an understanding of trends in instruments that use the Moon for calibration. This is the case for SeaWiFS.

SeaWiFS Lunar Calibration Methodology after Six Years on Orbit

The SeaWiFS Project uses monthly lunar calibrations to monitor the on-orbit radiometric stability of SeaWiFS over the course of its mission. Ongoing analyses of the steadily increasing lunar calibration data set have led to improvements in the calibration methodology over time. The lunar measurements must be normalized to a common viewing geometry for the calibration time series to track the radiometric stability of the instrument. Corrections computed from the time and geometry of the observations include Sun-Moon and instrument-Moon distances, oversampling of the lunar image, and variations in the lunar phase angles. The Project has recently implemented a correction for lunar libration that is computed from regressions of the libration angles of the observations against the lunar radiances. Decaying exponential functions of time are fit to the geometry-corrected calibration time series. The observations for bands 1,2,and 5-8 are fit to two simultaneous exponential functions of time, while bands 3 and 4 are fit to single exponential functions of time. The corrections to the radiometric response of the instrument over time are the inverses of these fits. The lunar calibration methodology provides top-of-the-atmosphere radiances for SeaWiFS that are stable to better than 0.07% over the course of the mission, with residual time drifts that are smaller than -0.004% per thousand days. The resulting water-leaving radiances are stable to better than 0.7%, allowing the Project to implement a vicarious calibration of the water-leaving radiances that is independent of time. The calibration methodology presented here will be used to generate the calibration table for the fifth reprocessing of the SeaWiFS global ocean data set.

On-orbit calibration of SeaWiFS

Ocean color climate data records (CDRs) require water-leaving radiances with 5% absolute and 1% relative accuracies as input. Because of the amplification of any sensor calibration errors by the atmospheric correction, the 1% relative accuracy requirement translates into a 0.1% long-term radiometric stability requirement for top-of-the-atmosphere (TOA) radiances. The rigorous prelaunch and on-orbit calibration program developed and implemented for Sea-viewing Wide Field-of-view Sensor (SeaWiFS) by the NASA Ocean Biology Processing Group (OBPG) has led to the incorporation of significant changes into the on-orbit calibration methodology over the 13-year lifetime of the instrument. Evolving instrument performance and ongoing algorithm refinement have resulted in updates to approaches for the lunar, solar, and vicarious calibration of SeaWiFS. The uncertainties in the calibrated TOA radiances are addressed in terms of accuracy (biases in the measurements), precision (scatter in the measurements), and stability (repeatability of the measurements). The biases are 2%-3% from lunar calibration and 1%-2% from vicarious calibration. The precision is 0.16% from solar signal-to-noise ratios, 0.13% from lunar residuals, and 0.10% from vicarious gains. The long-term stability of the TOA radiances, derived from the lunar time series, is 0.13%. The stability of the vicariously calibrated TOA radiances, incorporating the uncertainties of the in situ measurements and the atmospheric correction, is 0.30%. This stability of the radiometric calibration of SeaWiFS over its 13-year on-orbit lifetime has allowed the OBPG to produce CDRs from the ocean color data set.

SeaWiFS long-term solar diffuser reflectance and sensor noise analyses

The NASA Ocean Biology Processing Groups Calibration and Validation (Cal/Val) team has undertaken an analysis of the mission-long Sea-Viewing Wide Field-of-View Sensor (SeaWiFS) solar calibration time series to assess the long-term degradation of the solar diffuser reflectance over 9 years on orbit. The SeaWiFS diffuser is an aluminum plate coated with YB71 paint. The bidirectional reflectance distribution function of the diffuser was not fully characterized before launch, so the Cal/Val team has implemented a regression of the solar incidence angles and the drift in the node of the satellites orbit against the diffuser time series to correct for solar incidence angle effects. An exponential function with a time constant of 200 days yields the best fit to the diffuser time series. The decrease in diffuser reflectance over the mission is wavelength dependent, ranging from 9% in the blue (412 nm) to 5% in the red and near infrared (670-865 nm). The Cal/Val team has developed a methodology for computing the signal-to-noise ratio (SNR) for SeaWiFS on orbit from the diffuser time series corrected for both the varying solar incidence angles and the diffuser reflectance degradation. A sensor noise model is used to compare on-orbit SNRs computed for radiances reflected from the diffuser with prelaunch SNRs measured at typical radiances specified for the instrument. To within the uncertainties in the measurements, the SNRs for SeaWiFS have not changed over the mission. The on-orbit performance of the SeaWiFS solar diffuser should offer insight into the long-term on-orbit performance of solar diffusers on other instruments, such as the Moderate-Resolution Imaging Spectrometer [currently flying on the Earth Observing System (EOS) Terra and Aqua satellites], the Visible and Infrared Radiometer Suite [scheduled to fly on the NASA National Polar-orbiting Operational Environmental Satellite System (NPOESS) and NPOESS Preparatory Project (NPP) satellites] and the Advanced Baseline Imager [scheduled to fly on the National Oceanic and Atmospheric Administration Geostationary Environmental Operational Satellite Series R (GOES-R) satellites].

Use of the moon as a calibration reference for NPP VIIRS

The Moon has served as a reference for several satellite instruments including SeaWiFS and MODIS, both of which provide design innovations for NPP VIIRS. However, as yet, the Moon is not a formal part of the calibration baseline for NPP VIIRS. In particular, the lunar measurements by the MODIS instruments require on-orbit maneuvers (spacecraft rolls of up to 20 degrees) to maintain a constant lunar phase angle. Here, we use a simulated set of NPP VIIRS lunar measurements to demonstrate the quality of the Moon as a reference for long-term measurements by VIIRS. With nine lunar comparisons (1 year of VIIRS measurements), it is possible to detect linear changes over time in the calibration of the VIIRS reflective solar bands at the 0.1% per year level or better. In addition, the surface of the Moon does not change over periods of a million years or more. As a result, the Moon can act as a cross-calibration reference for NPP VIIRS and the Terra MODIS instrument that precedes it, even with a time gap between the operation of the two sensors. The quality of this cross-comparison reference is estimated to be significantly better than 1%. However, to accomplish both of these functions, NPP VIIRS must make measurements at the same lunar phase angle as Terra MODIS, that is, at 55 degrees after full phase. This requires periodic spacecraft maneuvers.

SeaWiFS transfer-to-orbit experiment

We present the results of an experiment designed to measure the changes in the radiometric calibration of the Sea-viewing Wide Field-of-view Sensor (SeaWiFS) from the time of its manufacture to the time of the start of on-orbit operations. The experiment uses measurements of the Sun at the manufacturers facility to predict the instrument outputs during solar measurements immediately after launch. Because an onboard diffuser plate is required for these measurements, the experiment measures changes in the instrument-diffuser system. There is no mechanism in this experiment to separate changes in the diffuser from changes in the instrument. For the eight SeaWiFS bands, the initial instrument outputs on orbit averaged 0.8% higher than predicted with a standard deviation of 0.9%. The greatest difference was 2.1% (actual output higher than predicted) for band 3. The estimated uncertainty for the experiment is 3%. Thus the transfer-to-orbit experiment shows no changes in the radiometric sensitivities of the SeaWiFS bands--at the 3% level--from the completion of the instruments manufacture to its insertion into orbit.