The JPSS-4 VIIRS sensor, the last in a JPSS program that will eventually span some 25+ years of on-orbit data collection, has completed its pre-launch test program. The test program included measurements for characterizing the VIIRS relative spectral response (RSR) in support of the Sensor and Environmental Data Records that will be generated from JPSS-4 VIIRS on-orbit observations after launch. Subject matter experts of the Government Team’s VIIRS DAWG have analyzed the VIIRS spectral measurements and produced the VIIRS spectral characterization, in the form of band averaged and supporting detector level RSR for each VIIRS band. The characterization is based upon the analysis of independent SpMA dual monochromator (all bands) and GSFC GLAMR laser system (reflectance bands only) spectral measurements. The SpMA and GLAMR measurements for reflectance bands (DNBLGS and DNBMGS, I1-I3, M1-M11) were combined to produce a “fused” RSR. For emissive bands (I4, I5, M12-M16), the SpMA measurements provide the entire characterization. The effort has led to the VIIRS Version 2 RSR product, the official at-launch RSR characterization for the JPSS-4 VIIRS mission, which is currently slated to be the next launch (expected Fall 2027) of the JPSS program. As expected, the JPSS-4 RSR are a close match to those of JPSS-3 and JPSS-2 (NOAA-21) VIIRS, while showing some spectral position and shape differences with SNPP and JPSS-1 (NOAA-20) VIIRS. An assessment on compliance with spectral performance metrics finds that VIIRS band-average RSR are compliant on nearly all metrics, with only a single minor exception. The Version 2 RSR will be available under EAR99 restrictions to the science community on the restricted access NASA Sharepoint.
The JPSS-3 VIIRS sensor has completed its pre-launch test program including measurements for characterizing the VIIRS relative spectral response (RSR) in support of the Sensor and Environmental Data Records (SDR and EDR, resp.) that will be generated from VIIRS on-orbit observations. Government team subject matter experts of the VIIRS DAWG have analyzed the VIIRS spectral measurements and produced the VIIRS spectral characterization, in the form of band-average and supporting detector level RSR for each VIIRS band. The characterization is based upon the analysis of independent SpMA dual monochromator (all bands) and GSFC GLAMR laser system (reflectance bands only) spectral measurements. The SpMA and GLAMR measurements for reflectance bands (DNB LGS and MGS, I1-I3, M1-M11) were combined to produce a “fused” RSR. For emissive bands (I4, I5, M12-M16), the SpMA measurements provide the characterization. The effort has led to the VIIRS Version 2 RSR release, the official at-launch RSR characterization for the JPSS-3 VIIRS mission. The JPSS-3 RSR are a close match to those of JPSS-2. An assessment on compliance with spectral performance metrics finds that VIIRS band-average RSR are compliant on nearly all metrics, with only a single minor exception. The Version 2 RSR release is available under EAR99 restrictions to the science community on the restricted access NASA Sharepoint.
KEYWORDS: Tungsten, Sensors, Ceramics, Mirrors, Absorption, Camera shutters, Signal to noise ratio, Long wavelength infrared, Optical filtering, Reflectivity
The JPSS-2 VIIRS instrument is set to be the third VIIRS instrument when it launches in 2022 following S-NPP and NOAA-20 VIIRS which launched in October 2011and November 2017, respectively. To date JPSS-2 VIIRS has undergone extensive pre-launch testing at the instrument system level to determine the radiometric, spatial, and spectral performance. Spectral testing was conducted by the instrument vendor, Raytheon Corporation, at their test facility in mid-to-late 2017 with a test configuration that utilized a double monochrometer with illumination provided by tungsten lamp and ceramic glow bar to cover the full spectral range. The purpose of these measurements was to measure the relative spectral response curve and assess the spectral characteristics necessary to determine compliance with the sensor design requirements. In addition to Raytheon team, the spectral measurements were analyzed by an independent government team with members from NASA, the University of Wisconsin, and Aerospace Corporation. Two RSR curves were released by the government team from this data set: a version 0 release which was the verified RSR as calculated by Raytheon, and version 1 which was the RSR assessment from the government team. The results discussed here are those of the government team (version 1) including the independent assessment of sensor compliance and a comparison of the JPSS-2 VIIRS spectral characteristics with the two previous VIIRS instruments. The version 1 RSR was publically released to the science community in the fall of 2018, and remains available for their use.
The JPSS-2 VIIRS sensor has completed its pre-launch test program and is now awaiting launch in the 2022 timeframe. The VIIRS spectral characterization, in the form of band averaged and supporting detector level relative spectral response (RSR) for each VIIRS band, was completed in 2019 and is based upon independent SpMA dual monochromator (all bands) and GSFC GLAMR laser system (reflectance bands only) spectral measurements, including first time measurements of the VIIRS SWIR bands by a laser system. The measurements and subsequent analysis effort by subject matter experts of the VIIRS DAWG has led to the July 2019 VIIRS Version 2 RSR release, the official at-launch RSR characterization for the JPSS-2 VIIRS mission. Version 2 replaces and improves upon the August 2018 Version 1 release by incorporating the GLAMR measurements into the analysis to produce an updated “fused” RSR for reflective solar bands (M1- M10, I1-I3, DNBLGS, DNBMGS) and by applying a CO2 absorption correction to the SpMA measurements for thermal band M13. For all other bands (M11, M12, M14-M16, I4, I5), the Version 1 characterization, based entirely upon the SpMA measurements, is carried forward into the Version 2 release. An assessment on compliance with spectral performance metrics finds that VIIRS is compliant on nearly all metrics, with a few minor exceptions. The version 2 RSR release includes band average (over all detectors and subsamples) RSR plus supporting RSR for each detector and subsample, and is available under EAR99 restrictions to the science community at a restricted access NASA eRoom site.
KEYWORDS: Calibration, Sensors, Infrared radiation, Black bodies, Temperature metrology, Reflectivity, Long wavelength infrared, Thermal modeling, Error analysis, Space operations
The National Oceanic and Atmospheric Administration 20 (NOAA-20) operational satellite, also known as the Joint Polar Satellite System 1 (JPSS-1), is the follow-on to the Suomi-National Polar-orbiting Partnership (S-NPP) with launch dates of November 2017 and October 2011, respectively. S-NPP and NOAA-20 provide critical weather and global climate products to the user community. The Visible-Infrared Imaging Radiometer Suite (VIIRS), a primary sensor on both SNPP and NOAA-20, has 22 bands covering a spectral range of 0.412-12.0μm with spatial resolutions of 750m and 375m for moderate and imaging bands, respectively. VIIRS provides calibrated Earth observations within the Sensor Data Records (SDRs) using on-orbit calibration sources such as the Solar Diffuser (SD) for the Reflective Solar Bands (RSBs) and an On-Board Calibrator BlackBody (OBCBB) for the Thermal Emissive Bands (TEBs), combined with pre-launch characterization information. Both the on-orbit calibration sources and pre-launch measurements contain calibration errors that propagate into the SDR radiance retrievals and degrade the performance of the Environmental Data Records (EDRs). This paper will focus on the TEB SDR calibration products and investigate the sources of the on-orbit calibration errors observed. This includes looking at gain drifts during the OBCBB warm-up and cool-down, along-scan temperature biases, and thermal model errors used in the estimation of the sensor’s background thermal emission. The pre-launch errors from the Response Versus Scan angle (RVS), calibration coefficients, and Ground Source Equipment (GSE) will also be included in the discussion. Finally, this paper will compare the differences in calibration errors between the S-NPP and NOAA-20 sensors and how they impact the SDR products in unique ways.
Optical and electronic cross-talk effects are present in the Terra MODIS sensor. Those effects are reviewed and the physical (engineering) characteristics that give rise to the effects are described when they are known. The potential for performance degradation also is assessed for each effect. The long-term consequence of these effects is to give rise to Terra MODIS to Aqua MODIS performance trend differences and researchers are cautioned to use care in interpretation of these trend differences as potential diurnal environmental effects.
The Cross-track Infrared Sounder (CrIS) onboard Suomi NPP (SNPP) and JPSS series has high radiometric accuracy, which can be used for validating some infrared bands of Visible Infrared Imaging Radiometer Suite (VIIRS) onboard the same platform. The collocated CrIS and VIIRS sensor data record (SDR) along with the VIIRS cloud mask product from 19 to 21 September 2016 (during a period of blackbody warm-up cool-down, or WUCD) are used for inter-comparisons. This study addresses the questions on how the sub-pixel cloud presence and the local zenith angle impact the radiometric differences between CrIS and VIIRS. Both VIIRS brightness temperature (BT) bias and standard deviation for I5, M13, M15 and M15 whose spectral response functions (SRFs) have the full coverages over the CrIS spectral regions, are analyzed over the clear and cloudy skies, respectively. Results show good agreement between VIIRS and CrIS, cloud presence has substantial impact on STD, and also impact on BIAS, local zenith angle has also substantial impact on STD, but impact on bias is small. Both bias and STD are large in DCC (deep convective cloud) areas. The study clearly shows the VIIRS scene temperature bias during WUCD, as well as the bias removal after reprocessing the M15 and M16 with the improved calibration bias correction algorithm. The methodology can be applied to monitor and validate the imager with advanced infrared (IR) sounder onboard the same platform, such as CrIS for VIIRS (SNPP, JPSS), IASI for AVHRR (Metop), and GIIRS for AGRI (FengYun-4).
Observations in the Terra MODIS PVLWIR bands 27 – 30 are known to be influenced by electronic crosstalk from those bands as senders and into those same bands as receivers. The magnitude of this crosstalk affecting L1B radiances has been steadily increasing throughout the mission lifetime, and has resulted in several detectors within these bands to be unusable for making L2 and L3 science products. In recent years, the crosstalk contamination has been recognized as compromising the climate quality status of several MODIS L2 and L3 science products that depend on the PVLWIR bands. In response, the MODIS Characterization Support Team (MCST) has undertaken an effort to generate a crosstalk correction algorithm in the operational L1B radiance algorithm. The correction algorithm has been tested and established and crosstalk corrected L1B radiances have been tested in several Terra MODIS L2 science product algorithms, including MOD35 (Cloud Mask), MOD06 (Cloud Fraction, Cloud Particle Phase, Cloud Top Properties), and MOD07 (Water Vapor Profiles). Comparisons of Terra MODIS to Aqua MODIS and Terra MODIS to MetOp-A IASI show that long-term trends in Collection 6 L1B radiances and the associated L2 and L3 science products are greatly improved by the crosstalk correction. The crosstalk correction is slated for implementation into Collect 6.1 of MODIS processing.
The Joint Polar Satellite System 2 (JPSS-2) Visible Infrared Imaging Radiometer Suite (VIIRS) includes one spectral band centered in a strong atmospheric absorption region. As much of the pre-launch calibration is performed under laboratory ambient conditions, accurately accounting for the absorption, and thereby ensuring the transfer of the sensor calibration to on-orbit operations, is necessary to generate science quality data products. This work is focused on the response versus scan angle (RVS) measurements, which characterize the relative scan angle dependent reflectance of the JPSS-2 VIIRS instrument optics. The spectral band of interest, centered around 1378 nm, is within a spectral region strongly effected by water vapor absorption. The methodology used to model the absolute humidity and the atmospheric transmittance under the laboratory conditions is detailed. The application of this transmittance to the RVS determination is then described including an uncertainty estimate; a comparison to the pre-launch measurements from earlier sensor builds is also performed.
The relative spectral response (RSR) characterization of the JPSS-1 VIIRS spectral bands has achieved “at launch” status in the VIIRS Data Analysis Working Group February 2016 Version 2 RSR release. The Version 2 release improves upon the June 2015 Version 1 release by including December 2014 NIST TSIRCUS spectral measurements of VIIRS VisNIR bands in the analysis plus correcting CO2 influence on the band M13 RSR. The T-SIRCUS based characterization is merged with the summer 2014 SpMA based characterization of VisNIR bands (Version 1 release) to yield a “fused” RSR for these bands, combining the strengths of the T-SIRCUS and the SpMA measurement systems. The M13 RSR is updated by applying a model-based correction to mitigate CO2 attenuation of the SpMA source signal that occurred during M13 spectral measurements. The Version 2 release carries forward the Version 1 RSR for those bands that were not updated (M8-M12, M14-M16A/B, I3-I5, DNBMGS). The Version 2 release includes band average (over all detectors and subsamples) RSR plus supporting RSR for each detector and subsample. The at-launch band average RSR have been used to populate Look-Up Tables supporting the sensor data record and environmental data record at-launch science products. Spectral performance metrics show that JPSS-1 VIIRS RSR are compliant on specifications with a few minor exceptions. The Version 2 release, which replaces the Version 1 release, is currently available on the password-protected NASA JPSS-1 eRooms under EAR99 control.
KEYWORDS: Sensors, Tungsten, Ceramics, Signal detection, Signal to noise ratio, Optical filters, Long wavelength infrared, Absorption, Optical filtering, Calibration
The JPSS-1 VIIRS instrument completed sensor level testing, including spectral characterization, at the Raytheon El Segundo facility in 2014. Spectral subject matter experts (SMEs) of the VIIRS DAWG have reviewed and analyzed the spectral measurements leading to a Version 1 release of JPSS-1 VIIRS relative spectral response (RSR) data in June 2015. The analysis demonstrates that all bands are well characterized with minor performance specification non-compliances in a few bands similar to those seen on S-NPP VIIRS. A major reduction in the out-of-band response (compared to that seen for S-NPP) has been realized through the redesign of the JPSS-1 VIIRS integrated filter assembly for the warm focal plane bands. An EAR99 restricted DAWG Version 1 release consisting of detector and band average (over all detectors) in-band + out-of-band RSR for all VIIRS bands is available on the NASA JPSS-1 eRoom limited access site and replaces the previous Version 0 Beta release.
The MODerate-resolution Imaging Spectroradiometer (MODIS) on NASA’s Earth Observing System Terra satellite, launched into orbit on 18 December 1999, will have a “first light” 15th anniversary on 24 February 2015. For nearly 15 years the MODIS instrument has provided radiances in all spectral bands. Though some detectors have fallen below SNR thresholds, the vast majority of spectral bands continue to provide high quality L1B measurements for use in L2 science algorithms supporting global climate research. Radiometric accuracy of the Terra MODIS thermal emissive bands (TEBs) in the C6 L1B product has been assessed using various approaches over the nearly 15 year Terra MODIS data record, including comparisons with instruments on the ground, in aircraft under-flights, and on other satellites. All of these approaches contribute to the understanding of the Terra MODIS radiometric L1B performance. Early in the lifetime of Terra, ground-based measurements and NASA ER-2 aircraft under-flights revealed that TEBs in the infrared window (“window” bands) are well calibrated and performing within accuracy specifications. The ER-2 under-flights also suggested that many atmospheric bands may be performing outside of specification, especially LWIR CO2 sensitive bands that are subject to optical crosstalk, although analysis uncertainties are larger for atmospheric bands. Beginning in 2007, MetOp-A IASI observations were used to evaluate Terra MODIS TEB performance through Simultaneous Nadir Overpass (SNO) comparisons. These inter-satellite comparisons largely affirm the early aircraft and ground-based evaluations, showing that all Terra MODIS window bands have small biases, minimal trending, and minor detector and mirror side striping over the 2007-2013 timeframe. Most atmospheric bands are performing satisfactorily near to specification; however, biases, striping and trending are large and significantly out of specification in the water vapor sensitive band 27 and ozone sensitive band 30 while the CO2 sensitive band 36 bias significantly exceeds specification. The investigation has found that an effective spectral shift significantly reduces biases and scene temperature dependence (but not trends) in most atmospheric bands, bringing them closer to, if not within, specification.
The Suomi National Polar-orbiting Partnership (S-NPP) satellite, carrying the first Visible Infrared Imager Radiometer Suite (VIIRS) was successfully launched on October 28, 2011 with first light on November 21, 2011. The passive cryo-radiator cooler doors were opened on January 18, 2012 allowing the cold focal planes (S/MWIR and LWIR) to cool to the nominal operating temperature of 80K. After an early on-orbit functional checkout period, an intensive Cal/Val (ICV) phase has been underway. During the ICV, the VIIRS SDR performance for thermal emissive bands (TEB) has been under evaluation using on-orbit comparisons between VIIRS and the CrIS instrument on S-NPP, as well as VIIRS and the IASI instrument on MetOp-A. CrIS has spectral coverage of VIIRS bands M13, M15, M16, and I5 while IASI covers all VIIRS TEB. These comparisons largely verify that VIIRS TEB SDR are performing within or nearly within pre-launch requirements across the full dynamic range of these VIIRS bands, with the possible exception of warm scenes (<280 K) in band M12 as suggested by VIIRS-IASI comparisons. The comparisons with CrIS also indicate that the VIIRS Half Angle Mirror (HAM) reflectance versus scan (RVS) is well-characterized by virtue that the VIIRS-CrIS differences show little or no dependence on scan angle. The VIIRS-IASI and VIIRS-CrIS findings closely agree for bands M13, M15, and M16 for warm scenes but small offsets exist at cold scenes for M15, M16, and particularly M13. IASI comparisons also show that spectral out-of-band influence on the VIIRS SDR is <0.05 K for all bands across the full dynamic range with the exception of very cold scenes in Band M13 where the OOB influence reaches 0.10 K. TEB performance, outside of small adjustments to the SDR algorithm and supporting look-up tables, has been very stable through 18 months on-orbit. Preliminary analysis from an S-NPP underflight using a NASA ER-2 aircraft with the SHIS instrument (NIST-traceable source) confirms TEB SDR accuracy as compliant for a typical warm earth scene (285-290 K).
KEYWORDS: Calibration, Sensors, Infrared radiation, Reflectivity, Thermography, Space operations, Long wavelength infrared, Infrared sensors, Black bodies, Temperature metrology
The Visible Infrared Imager Radiometer Suite (VIIRS), a primary sensor on-board the Suomi-National Polar-orbiting Partnership (SNPP) spacecraft, was launched October 28, 2011. It has 22 bands: 7 thermal emissive bands (TEBs), 14 reflective solar bands (RSBs) and a Day Night Band (DNB). The TEBs cover the spectral wavelengths between 3.7 to 12 μm and have two 371 m and five 742 m spatial resolution bands. A VIIRS Key Performance Parameter (KPP) is the sea surface temperature (SST) which uses bands M12 (3.7 μm), M15 (10.8 μm) and M16’s (12.0 μm) calibrated Science Data Records (SDRs). The TEB SDRs rely on pre-launch calibration coefficients used in a quadratic algorithm to convert the detector’s response to calibrated radiance. This paper will evaluate the performance of these prelaunch calibration coefficients using vicarious calibration information from the Cross-track Infrared Sounder (CrIS) also onboard the SNPP spacecraft and the Infrared Atmospheric Sounding Interferometer (IASI) on-board the Meteorological Operational (MetOp) satellite. Changes to the pre-launch calibration coefficients’ offset term c0 to improve the SDR’s performance at cold scene temperatures will also be discussed.
KEYWORDS: Space operations, Sensors, Aerospace engineering, Quality measurement, Polarization, Aerosols, Infrared radiation, Long wavelength infrared, Clouds, Signal to noise ratio
The Suomi National Polar-orbiting Partnership (S-NPP) satellite was successfully launched on October 28,
2011, beginning the on-orbit era of the Visible Infrared Imager Radiometer Suite (VIIRS). In support of atlaunch
readiness, VIIRS underwent a rigorous pre-launch test program to characterize its spatial, radiometric,
and spectral performance. Spectral measurements, the subject of this paper, were collected during instrument
level testing at Raytheon Corp. (summer 2009), and then again in a special spectral test for VisNIR bands
during spacecraft level testing at Ball Aerospace and Technologies Corp. (spring 2010). These spectral
performance measurements were analyzed by industry (Northrop Grumman, NG) and by the Relative Spectral
Response (RSR) subgroup of the Government team, (NASA, Aerospace Corp., MIT/Lincoln Lab, Univ.
Wisconsin) leading to releases of the S-NPP VIIRS RSR characterization by both NG and the Government
team. The NG RSR analysis was planned to populate the Look-Up-Tables (LUTs) that support the various
VIIRS operational products, while the Government team analysis was initially intended as a verification of
the NG RSR product as well as an early release RSR characterization for the science community’s pre-launch
application. While the Government team deemed the NG December 2010 RSR release as acceptable for the
“at-launch” RSR characterization during the pre-launch phase, the Government team has now (post-launch
checkout phase) recommended for using the NG October 2011 RSR release as an update for the LUTs used in
VIIRS SDR and EDR operational processing. Meanwhile the Government team RSR releases remain
available to the community for their investigative interests, and may evolve if new understanding of VIIRS
spectral performance is revealed in the S-NPP post-launch era.
The new NASA Enhanced MODIS Airborne Simulator (eMAS) is based on the legacy MAS system,
which has been used extensively in support of the NASA Earth Observing System program since
1995. eMAS consists of two separate instruments designed to fly together on the NASA ER-2 and
Global Hawk high altitude aircraft.
The eMAS-IR instrument is an upgraded version of the legacy MAS line-scanning spectrometer,
with 38 spectral bands in the wavelength range from 0.47 to 14.1 μm. The original LN2-cooled
MAS MWIR and LWIR spectrometers are replaced with a single vacuum-sealed, Stirling-cooled
assembly, having a single MWIR and twelve LWIR bands. This spectrometer module contains a
cold optical bench where both dispersive optics and detector arrays are maintained at cryogenic
temperatures to reduce infrared background noise, and ensure spectral stability during high altitude
airborne operations.
The EMAS-HS instrument is a stand-alone push-broom imaging spectrometer, with 202 contiguous
spectral bands in the wavelength range from 0.38 to 2.40 μm. It consists of two Offner
spectrometers, mated to a 4-mirror anastigmatic telescope. The system has a single slit, and uses a
dichroic beam-splitter to divide the incoming energy between VNIR and SWIR focal plane arrays.
It will be synchronized and bore-sighted with the IR line-scanner, and includes an active source for
monitoring calibration stability.
eMAS is intended to support future satellite missions including the Hyperspectral Infrared Imager (
HyspIRI,) the National Polar-orbiting Operational Environmental Satellite System (NPOESS)
Preparatory Project (NPP,) and the follow-on Joint Polar Satellite System (JPSS.)
KEYWORDS: Sensors, Signal to noise ratio, MODIS, Camera shutters, Aerospace engineering, Optical filters, Absorption, Optical design, Polarization, Long wavelength infrared
The VIIRS Flight 1 (F1) instrument completed sensor level testing, including relative spectral response
(RSR) characterization in 2009 and is moving forward towards a launch on the NPP platform late in 2011.
As part of its mandate to produce analyses of F1 performance essentials, the VIIRS Government Team,
consisting of NASA, Aerospace Corp., and MIT/Lincoln Lab elements, has produced an independent (from
that of industry) analysis of F1 RSR. The test data used to derive RSR for all VIIRS spectral bands was
collected in the TVAC environment using the Spectral Measurement Assembly (SpMA), a dual
monochromator system with tungsten and ceramic glow bar sources. These spectrally contiguous
measurements were analyzed by the Government Team to produce a complete in-band + out-of-band RSR
for 21 of the 22 VIIRS bands (exception of the Day-Night Band). The analysis shows that VIIRS RSR was
well measured in the pre-launch test program for all bands, although the measurement noise floor is high on
the thermal imager band I5. The RSR contain expected detector to detector variation resulting from the
VIIRS non-telecentric optical design, and out-of-band features are present in some bands; non-compliances
on the integrated out-of-band spectral performance metric are noted in M15 and M16A,B bands and also for
several VisNIR bands, though the VisNIR non-compliances were expected due to known scattering in the
VisNIR integrated filter assembly. The Government Team "best" RSR have been released into the public
domain for use by the science community in preparation for the post-launch era of VIIRS F1.
KEYWORDS: MODIS, Temperature metrology, Calibration, Sensors, Long wavelength infrared, Mid-IR, Infrared radiation, Space operations, Mirrors, Black bodies
MODIS is a major instrument for the NASA EOS Terra and Aqua missions, launched in December 1999 and May 2002
respectively. MODIS has 16 thermal emissive bands and they are calibrated using an onboard blackbody (BB) based on
a nonlinear second order relationship. While the gains of the MODIS thermal bands are calibrated on a scan-by-scan
basis, the offset and non-linear terms are determined either from prelaunch or on-orbit measurements during scheduled
BB warm-up and cool-down cycles. A major concern on determination of the offset and non-linear terms from on-orbit
BB measurements is that the controlled BB temperature range is relative small compared to the temperature range used
in prelaunch tests, which could have impacts on the retrieval of brightness temperatures (BT) well outside the calibration
range. In this study, the stability of offset and non-linear terms obtained from BB warm-up/cool-down cycles is
presented. Several approaches to derive the on-orbit offset and non-linear terms are used and their impacts on the Earth
scene BT estimates are examined. By comparison with BT derived using prelaunch offset and nonlinear terms under the
same electronic configuration, it is shown that the current approach of deriving on-orbit offset and nonlinear terms
applied in L1B radiance products causes positive BT biases of exceeding 1K at low temperatures for middle- to longwave
IR bands. Comparison of MODIS and AIRS (The Atmospheric Infrared Sounder), both on-board Aqua spacecraft,
for cold temperature scenes at Antarctica for two long-wave IR bands also indicates that there are temperature-dependent
positive BT biases for about the same magnitudes. Results of this study have a significant impact on improving the current approach of setting a0 and a2 used to produce MODIS L1B data products.
During the Tahoe 2006 field effort, the NASA ER-2 aircraft flew 2 nighttime science missions (September 29
and October 13, 2006) over California and the nearby Pacific Ocean. Because of its high cruising altitude
(above ~95% of the atmosphere), the ER-2 platform closely simulates satellite-based observations of the
earth-atmosphere system. Each Tahoe 2006 mission included overpasses of the Lake Tahoe Validation
Network and an underflight of the Aqua satellite. An 8+ minute ER-2 flight segment of clear sky data over
the Pacific Ocean on the October 13 flight has been used to assess the MODIS thermal band (bands 20-36)
radiometric performance. For the assessment MODIS radiances are simulated using the MAS high spatial
resolution and SHIS high spectral resolution radiances (the calibration backbone of the MODIS assessment),
and compared to the co-located MODIS observations. The assessment (286 matchups) shows that Aqua
MODIS thermal bands continue to perform within or very nearly within their 1% radiometric specification
(0.5% for window bands 31, 32; 0.75% for window band 20) with the exception of band 30 (ozone) and band
36 (CO2). There is low confidence in the ozone band 30 assessment due to lack of information on the ozone
profile above the ER-2 flight level; band 36 however, appears to be consistently about 0.7 K warmer than
expected. These results are consistent with a previous Aqua MODIS comparison to SHIS and MAS in 2002
and with previously reported1 comparisons of Aqua AIRS and MODIS observations.
Five cases using NASA ER-2 aircraft based SHIS and MAS radiances have been used to assess the L1B radiometric performance of Terra and Aqua MODIS Collection 5 radiances for LWIR bands 31-36. The composite results of these cases show that the split window bands 31 (11 μm) and 32 (12 μm) have performed well within the 0.5% radiometric specification over their lifetime. This is in agreement with results from other ground based and satellite based comparisons that are discussed in the paper. However, the LWIR CO2-sensitive bands 34-36 appear to be performing outside of their 1% accuracy specification, especially for Terra MODIS. This is also observed in global Aqua AIRS-MODIS comparisons. Possible causes for this behavior are under investigation, with the most likely contributors being spectral characterization error, OOB influences due to spectral filter leaks, or possibly scan mirror characterization. It seems that an optical leak from Terra MODIS band 31 into bands 32-36 is probably not a significant contributor to the large residuals of bands 34-36, owing to an effective radiometric correction. Calibration coefficient error is probably only a small contributor since, after adjustments in 2002, the on-orbit calibration now closely follows that of the pre-launch calibration.
Exploiting the redundancy in high spectral resolution observations, dependent set Principle Component Analysis (PCA)
is a simple yet very powerful tool not only for noise filtering and lossy compression, but also for the characterization of
sensor noise and other variable artifacts using Earth scene data. Our approach for dependent set PCA of AIRS Earth
scene data is presented. Aspects of the analyses include 1) estimation of NEDT using PCA and comparisons to values
derived from on-board blackbodies, 2) estimation of the scene dependence of NEDN, 3) estimation of the spectrally
correlated component of NEDT and comparison to pre-launch analyses using blackbody views, 4) investigation of non-
Gaussian noise behavior, and 5) inspection of individual PCs. The results of the PCA analyses are generally consistent
with results obtained pre-launch and on-orbit using blackbody and/or space view data. Specific findings include: 1)
PCA estimates of AIRS spectrally random and spectrally correlated NEDN compare well with estimates computed from
on-board blackbody and space views, 2) the signal dependence of AIRS NEDN is accurately parameterized in terms of
the scene radiance, 3) examination of the reconstruction error allows non-Gaussian phenomenon such as popping to be
characterized, and 4) inspection of the PCs and individual PC filtered radiance spectra is a powerful technique for
diagnosing low level artifacts in hyperspectral data.
The ability to accurately validate high spectral resolution infrared radiance measurements from space using comparisons with aircraft spectrometer observations has been successfully demonstrated. The demonstration is based on an under-flight of the Atmospheric Infrared Sounder (AIRS) on the NASA Aqua spacecraft by the Scanning High resolution Interferometer Sounder (S-HIS) on the NASA ER-2 high altitude aircraft on 21 November 2002 and resulted in brightness temperature differences approaching 0.1K for most of the spectrum. This paper presents the details of this AIRS/S-HIS validation case and also presents comparisons of Aqua AIRS and Moderate Resolution Imaging Spectroradiometer (MODIS) radiance observations. Aircraft comparisons of this type provide a mechanism for periodically testing the absolute calibration of spacecraft instruments with instrumentation for which the calibration can be carefully maintained on the ground. This capability is especially valuable for assuring the long-term consistency and accuracy of climate observations. It is expected that aircraft flights of the S-HIS and its close cousin the National Polar Orbiting Environmental Satellite System (NPOESS) Atmospheric Sounder Testbed (NAST) will be used to check the long-term stability of the NASA EOS spacecrafts (Terra, Aqua and Aura) and the follow-on complement of operational instruments, including the Cross-track Infrared Sounder (CrIS).
During the Terra-Aqua experiment -- 2002 (TX-2002), a NASA ER-2 was used to underfly the EOS Aqua satellite over the Gulf of Mexico for the purpose of gaining insight on the accuracy of MODIS and AIRS thermal infrared (TIR) band radiances. The ER-2 payload included the MODIS Airborne Simulator (MAS) and the Scanning High resolution Interferometer Sounder (SHIS); these instruments have flown previously on the ER-2 for assessing Terra MODIS TIR band radiances. On November 21, 2002, the ER-2 flew directly under the Aqua satellite, with MODIS and AIRS, as it swept over a clear sky region of the Gulf of Mexico. The MAS and SHIS observations were used to simulate the MODIS thermal IR band radiances for the warm (~ 295 K) Gulf of Mexico scene. The results of comparing the simulated MODIS radiances with the MODIS observations show Aqua MODIS TIR bands are performing well. The residuals (MAS - MODIS) in most bands are within or very near specification. The split window 11 and 12 μm band residuals are small and very close to one another at -0.15°C and -0.13°C, respectively. The comparisons suggest that MODIS LWIR CO2 sensitive bands 35 (13.9 μm) and 36 (14.2μm) may be calibrated slightly warm by about 1°C. Early direct comparisons between MODIS and AIRS on Aqua also suggest that the MODIS bands 35 and 36 may be calibrated slightly warm.
The calibration accuracy of the Moderate resolution Imaging Spectro-radiometer (MODIS) on Terra near its one year anniversary of first light has been assessed using ER-2 aircraft underflights during the Terra eXperiment (TX-2001) in the spring, 2001. The ER-2, equipped with the MAS and SHIS instruments, underflew Terra several times viewing clear sky earth scenes of the Gulf of Mexico. MAS and SHIS form a powerful tandem, combining high spatial resolution imaging with high spectral resolution sampling in the midwave to longwave infrared region. The assessment is based on co-located MODIS and MAS fields of view with matching viewing geometry and removes spatial and spectral dependencies. The MAS L1B calibration accuracy is improved by transferring the SHIS calibration accuracy (conservatively 0.5 K) to MAS. The early results of two days from TX-2001 indicate that MODIS bands are performing well, but not optimally. The MODIS MWIR window bands appear to be close to the 0.75 - 1% radiometric accuracy specification for the uniform warm, low reflectance scenes assessed, perhaps suggesting that known electronic crosstalk in MODIS SWIR and MWIR bands is small for such scenes. MODIS LWIR window bands show residuals of about 0.5 K to 0.6 K, larger than the 0.5% radiometric accuracy specification. However with the 0.5 K (window bands) to 1 K (atmospheric bands) uncertainties associated with the current assessment, it is not possible to definitively state whether these MODIS bands are or are not within specification. MODIS LWIR atmospheric CO2 bands appear to perform near the 1% accuracy specification with the exception of bands 35 and 36, the upper tropospheric CO2 bands at 13.9micrometers and 14.1micrometers . Different MODIS viewing geometry on the two days seems to suggest that scan mirror reflectance dependence on mirror angles may be influencing the MODIS L1B calibration for some bands, most notably the 8.6micrometers and LWIR CO2 bands; however this assessment is dependent upon the accuracy of the spectral correction (a function of atmospheric conditions), which will be further investigated in coming months. It was surprising to find large MODIS residuals for several bands when the mirror angle to the earth scene closely matched that of when MODIS views its onboard blackbody.
Several MODIS cloud product algorithms are being developed at the University of Wisconsin for the generation of day-1 products after the launch of MODIS. MODIS Airborne Simulator (MAS) radiometric data collected form NASA's ER-2 platform is being used to simulate MODIS spectral bands for testing and refinement of the cloud product algorithms. Spectral characterization is an important component of the MAS calibration. MAS LWIR bands are spectrally characterized in ambient conditions using a monochromator and are corrected for source spectral shape and atmospheric attenuation. An atmospheric correction based on LBLRTM forward model transmittances demonstrates that strong spectral absorption features, such as Q-branch CO2 absorption near 13.9 micrometers , are effectively removed from the spectral measurements with the aid of a small spectral position correction. Comparisons of MAS in-flight data to well- calibrated HIS instrument data indicate that MAS LWIR spectral calibration drift over time is less than 5 percent of FWHM. The MODIS CO2 cloud top height retrieval shows small dependence on the spectral characterization, with retrieved cloud top height changing by less than 0.5 km in response to a 5 percent spectral position change. This is within the tolerance of other error sources in the cloud top properties algorithm.
KEYWORDS: Black bodies, Calibration, Long wavelength infrared, Temperature metrology, MODIS, Short wave infrared radiation, Carbon dioxide, Absorption, Clouds, Spectral resolution
The impact of non-unit calibration blackbody emissivity on MODIS airborne simulator (MAS) absolute thermal calibration accuracy is investigated. Estimates of blackbody effective emissivity were produced for MAS infrared channels using laboratory observations of a thermally controlled external source in a stable ambient environment. Results are consistent for spectrally close atmospheric window channels. SWIR channels show an effective emissivity of about 0.98; LWIR channels show an effective emissivity of about 0.94. Using non-unit blackbody effective emissivity reduces MAS warm scene brightness temperatures by about 1 degree Celsius and increases cold scene brightness temperatures by more than 5 degrees Celsius as compared to those inferred from assuming a unit emissivity blackbody. To test the MAS non- unit effective emissivity calibration, MAS and high- resolution interferometer sounder (HIS) LWIR data from a January 1995 ER-2 flight over the Gulf of Mexico were compared. Results show that including MAS blackbody effective emissivity decreases LWIR absolute calibration biases between the instruments to less than 0.5 degrees Celsius for all scene temperatures, and removes scene temperature dependence from the bias.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.