The NASA GSFC Filter Radiometer Monitoring System (FRMS) was used to compare lamp-based and detector-based spectral radiance calibration of an integrating sphere. The FRMS is a telecentric, filter radiometer employing two apertures, a filter wheel, and a detector. The FRMS uses nine filters at specific wavelengths from 360 to 2400 nm. The lamp-based calibration used a National Institute of Standards and Technology (NIST) calibrated irradiance standard lamp to calibrate the irradiance responsivity of a scanning spectroradiometer. The spectroradiometer was then used to transfer its irradiance calibration to an integrating sphere. The lamp-based spectral radiance calibration of the sphere was calculated using the sphere irradiance, the sizes of the sphere exit and spectroradiometer entrance apertures, and the distance between those apertures. The detector-based calibration of the sphere used NIST calibrated absolute radiance Si photodiode detector to determine the absolute spectral radiance responsivity of the FRMS with the NASA GSFC Automated Laser Tuned Advanced Radiometry (ALTAR) laser system as the source. The absolute spectral radiance responsivity of the FRMS was measured at the following channels: 380, 410, 640, and 840, nm. The FRMS measured the integrating sphere to make a direct determination of its absolute radiance at those channels. Analysis of lamp-based and detector-based radiance measurements of the integrating sphere at four wavelength bands will be presented.

Two scatterometers at NASA GSFC Diffuser Calibration Lab (DCL) are used to realize the Bidirectional Reflectance Distribution Function (BRDF) scale transfer from a small NIST traceable BRDF lab standard sample to large-area solar diffuser panels in support of the calibration of remote sensing instruments. The capability of measuring the BRDF of large area solar diffuser panels has been realized by developing a large-area collimated uniform light source and setting the desired fields-of view (FOVs) of instrument detectors. The BRDF measurements for three types of diffuse samples using two different scatterometer measurement equations are compared and consistent results are obtained. The clocking effect of BRDF measured from those diffuse samples is also investigated, and the uncertainty budget of BRDF measurements for the large-area diffuse panel is discussed.

Continuing the successful on-orbit operation of the VIIRS instruments currently on-board the Suomi-NPP and NOAA-20 spacecraft, additional VIIRS instruments are in development to be launched on future JPSS missions. Pre-launch testing of VIIRS, before integration with the spacecraft, is an important step in verifying the performance and operation of the instrument. As part of that testing, investigation into the near-field response (NFR) for each detector is required to assess the detector performance and assure there is no interference due to scattered light that could influence the radiometric measurements. A key element of this test is measuring the detector response of a bright target, followed by viewing the same target through neutral density filters. The measurements are stitched together to quantify the detector response over the entire dynamic range and fit using a Harvey-Shack scatter model. We present our findings for JPSS-4 VIIRS NFR performance evaluated during instrument ambient testing. The results include performance of the VIIRS detectors as compared against their specifications and comparisons against previous flight models.

The Day Night Band (DNB) has been featured on the first two VIIRS instruments aboard the Suomi NPP and NOAA-20 satellites that are both currently in service, and to date prelaunch sensor level testing has been completed for the next two instruments in the series JPSS-2 and JPSS-3 VIIRS. Radiometric testing had found nonlinear behavior in the DNB especially at low radiances. The non-linearity was especially problematic the DNB of NOAA-20 VIIRS which resulted is revisions to the ground test program and operating the instrument in a modified “option 21” configuration on-orbit to mitigate the impacts of the non-linearity. Still, non-linearity remains both in NOAA-20 under option 21 and subsequent VIIRS builds, and this nonlinearity is gain, mode, detectors, and sample dependent. In this analysis we look at results from the most recent three VIIRS builds (NOAA-20, JPSS-2, and JPSS-3) where more extensive prelaunch DNB calibration is available to determine the extent on the non-linearity that remains and its effect on the on-orbit calibration. Especially important is the cross-calibration that transfers the low gain stage calibration coefficients calculated from the solar diffuser to the other gain stages. This process leverages low signal samples where non-linear effects are most significant. Tables have been generated to select the optimal linear samples and improve this process.

The Institute of Optical Sensor Systems (OS) at the Robotics and Mechatronics Center of the German Aerospace Center (DLR) has more than 35 years of experience with high-resolution imaging and imaging technology. This paper shows the institute’s scientific results of a hardware driven method to validate the image quality and keep it constant over the whole mission life time. This technology is applicable for highest resolution systems as well as for systems which are foreseen to measure reproducible data series over years. The technology is applicable for panchromatic and multispectral instrument designs. The paper will first define image quality, which is described by modulation transfer function, signal to noise ratio, spatial and spectral resolution, linearity and other key parameters. They are shown the differences in the quality assessment compared to the classic image-based methods. Within an in-orbit initialization phase of the instrument, the full electrical channel is validated by generating a defined amount of electrons instead of electrons coming from the photodiode. The approach is based on charge injection with a reproducible number of electrons which is driving the complete vertical pixel chain. The basic idea is to generate a periodic signal in orbit which can be analyzed with respect to radiation influences. The control timing of the FPGA sensor controller can be operated from ground via commanding. This procedure includes also the initialization mode by controlling the phase adjustment of the CDS sampling. The possible image degradation as a result of typical radiation effects over the mission life time is described in the paper as well as how such effects can be avoided in future by implantation of the proposed method. This new approach enables e.g. linearity test, analysis and alignment and shows the relevance of such a validation technology for high-resolution optical space instruments.

In the past, many spaceborne, Earth viewing radiometers have had a fairly narrow optical field-of-view (FoV) and are scanned in the cross-track direction to provide the required coverage. These (point) scanners swing delicate optics and lack a full instantaneous cross-track coverage. To get around these undesirable issues, the cross-track direction can be optically a wide FoV telescope utilizing freeform optics [Phenis et al., Proc. of SPIE, 12078(12078J), 2021] along with a corresponding focal plane array (FPA). However, there are some unintended and undesirable consequences of wide FoV telescopes particularly when there is a significant lack of symmetry due, in part, to off-axis viewing. One of these consequences is a polarization bias imparted on the light due to the optics as it progresses along the optical chain. Here we develop the polarimetric radiometric uncertainty and a polarimetric error equivalent radiance which can be used in the optical design and radiometric error budget for imaging radiometers.

The planetary boundary layer (PBL) is a key interface of energy exchange between the surface and atmosphere, however, current spaceborne sensors are not optimized to measure in this region. There is significantly more PBL temperature and humidity information content in the microwave spectrum that current satellite instruments resolve. The photonic spectro-radiometer developed under NASA ESTO ACT-20 program capable of fully resolving the microwave spectrum to return all PBL information in the microwave spectrum. A novel photonic integrated circuit is designed having integrated a modulator for up-conversion of signals into optics domain, an arrayed waveguide grating and star couplers with filters.

As the next generation of Earth science programs demand more spectral bands, larger fields of view, faster speeds and reduced size, the optical designer will need to adapt to these new requirements. With the advent of manufacturable freeform optical surfaces, compact high-performance optical systems utilizing these surfaces are becoming practical. Freeform optics provide additional degrees of freedom for the optical designer which allow for more compact optical systems of equal performance, potentially operating at faster speeds or over wider fields of view. While numerous design studies on freeform systems have been published, little has been presented in the open literature on as built freeform systems. In this paper we describe the successful outcome of a hardware development program where we designed, built, aligned, and tested a compact WFOV three-mirror telescope with freeform surfaces. It is important that in addition to good optical performance, excellent stray light control is required in Earth remote sensing systems to minimum calibration errors across spectral bands. While compact size is often emphasized in the design of freeform systems, this needs to be balanced against the requirement for good stray light control. As such, the telescope presented in this paper balances the desire for small size with good stray light control. We present the results of the computer-aided alignment of the telescope along with measured stray light performance.

The Improved Radiometric calibration of land Imaging Systems or IRIS is a compact full-spectrum calibration system that reduces the size, weight and power of conventional on-board radiometric sources into a single flat panel format by combining both carbon nanotube and LED technology within a Jones source design. Introduced in this presentation is a methodology that maintains in-flight traceability through a fusion of the on-board IRIS LED reference with Labsphere’s FLARE vicarious calibration system. The process known as IRIS-V provides SI traceability of the on-board VSWIR calibration system through the mission's lifetime without impacting operational land or coastal image collection.

The Atmospheric Infrared Sounder (AIRS) on the EOS Aqua Spacecraft was launched on May 4, 2002, with a 95% confidence design lifetime of 5 years. Since 2002 AIRS has performed exceptionally well. We now have 20 years of AIRS data. We show that AIRS data are an order of magnitude more accurate and more stable than the 100 mK absolute and 10 mK/yr required to measure climate trends. However, the real limitation on the usable absolute accuracy and stability are time varying residual cloud contamination at all wavelength, and surface contamination effects, particularly in the SW channels. The change in contamination and large interannual variabilities due to ENSO type effects complicate the interpretation of trends for climate. The 20 years of AIRS data contain at least two changes of potential climate significance: The change of the DCC count, but it is opposite in sign to the expected trend, and decrease in the count of SCT clear cases. Both indicate unexpected changes in the cloud structure.

The Atmospheric Infrared Sounder (AIRS) on the EOS Aqua Spacecraft acquires hyperspectral infrared radiances in 2378 channels ranging in wavelength from 3.7-15.4 um with spectral resolution of better than 1200, and spatial resolution of 13.5 km with global daily coverage. The AIRS was designed to measure temperature and water vapor profiles for improvement in weather forecast and improved parameterization of climate processes. Coupling of the polarization emitted from the AIRS scan mirror and the polarization of the spectrometer introduces a scan dependent modulation of the instrument radiometric response. Measurements of the polarization pre-flight were used to determine the current calibration coefficients in Version 5. A Deep Space Maneuver (DSM) of the Aqua Spacecraft, which was performed on September 23, 2021, provides a unique measurement of the radiometric modulation that we can use to derive a new set of polarization parameters for the system. The test involves pitching the Aqua spacecraft in the form of a ‘back flip’ that enables the AIRS instrument (and other instruments on the spacecraft) to view cold space in the Earth view for all scan angles for a good portion of the maneuver. The observed modulation of the data while viewing space is a direct measurement of the mirror polarized emission. The polarization coefficients derived from the maneuver are expected to improve left/right asymmetries and reduce radiometric errors in cold scenes. This paper summarizes the methodologies used, and compares the new polarization parameters to those derived pre-flight and in-orbit using space views.

MODIS on board the Terra and Aqua spacecrafts have been in operation for 22 and 20 years, respectively. Throughout each mission, Moon observations have been a key component to the on-orbit calibration. For MODIS Collection 7, we have updated the lunar calibration algorithm to provide improvements to the data quality. This includes removing pixel oversampling errors associated with including partial lunar images in addition to full disk images, data masking for the mitigation of crosstalk contamination, and corrections to digital counting errors (sticky bins). These corrections reduce variations in the data while maintaining the long-term trends from Collection 6.

MODIS is an imaging spectroradiometer onboard NASA’s Terra and Aqua satellites with visible to short-wave infrared capability enabled through 20 reflective solar bands (RSB) in the wavelength range of 400 to 2200 nm. RSB gain changes are monitored by using observations of a fully sunlit solar diffuser (SD) for the on-orbit radiometric calibration. SD signal changes, due to degradation of the SD surface, are monitored by using the SD Stability Monitor (SDSM), which has 9 detectors in the wavelength range of 400 to 936 nm. These calibration approaches depend on the geometry of the spacecraft’s orbit. Both Terra and Aqua have recently started to exit their respective constellations and are (or soon will be) in the process of drifting from their nominal orbits. These orbital changes will cause changes in the solar diffuser (SD) and solar diffuser stability monitor (SDSM) viewing geometry and calibration conditions. This, in turn, will drive a variation in the calibration parameters used to calculate reference adjustments for the MODIS reflective solar bands (RSB). We examine the expected effect of orbital drift on the reflectance and transmittance functions, the evaluation of on-going SDSM detector response, the variation of SD surface incidence angle ranges, and the estimation of fully-illuminated observing condition for SD signal monitoring.

Over the MODIS mission, normalized trends of Earth-view digital numbers (EV-dn) over desert calibration sites (PICS) have shown biases in several Terra MODIS reflective solar bands (RSBs) related to the changes in polarization sensitivity. The MODIS Characterization Support Team (MCST) corrects for these effects at the Level 1B stage in Collection 7 with monthly coefficient updates from NASA Ocean Biology Processing Group (OBPG). However, upcoming orbit changes for both MODIS instruments may require more frequent updates and new algorithms. In this work, we present a vicarious calibration algorithm with the potential to characterize polarization sensitivity using a single MODIS granule, across a range of angles-of-incidence (AOI). Marine stratocumulus cloud targets off the coast of South America, measured in near-real-time by Aqua-MODIS and the polarimeter POLDER-3, are used. These clouds strongly polarize light in optical wavelengths, are spatially uniform over wide areas, and are present year-round. After geo-registering both data to ~50 km superpixels, we find the polarized reflectance fit that best matches the cloud microphysics of the POLDER-3 target at 0.865 μm. We then interpolate the fit to the Aqua-MODIS target geometry. We derive polarization sensitivity coefficients for Aqua-MODIS Band 2 (0.858 μm) at a range of AOI using the POLDER-3 retrieval results for six different matchups in 2005. The results suggest that cloud development in the time between Aqua-MODIS and POLDER-3 measurements (~3 min) and simultaneous nadir overpass (SNO) distance are the main error contributions, combined with relatively low polarization sensitivity for Aqua Band 2. Even so, the derived sensitivity coefficients agree with pre-launch values within uncertainty. Therefore, simultaneous, co-incident radiometer and polarimeter data are optimal, such as from OCI and HARP-2/SPEXone on the upcoming NASA Plankton Aerosol Cloud and ocean Ecosystem (PACE) mission.

The Moderate Resolution Imaging Spectroradiometer (MODIS) instruments on-board the Aqua and Terra spacecraft have provided valuable Earth data to the science community for the last 20 and 22 years, respectively. The Spectro-Radiometric Calibration Assembly (SRCA) is an on-board calibrator (OBC) that can characterize the radiometric, spatial, and spectral properties of the MODIS reflective solar bands (RSBs). In radiometric mode, the SRCA is able to monitor the gain trends of the RSBs on a detector level. Nominal radiometric mode measurements are collected during a 10-minute period during spacecraft night using a combination of SRCA halogen lamps. These measurements are intended for deriving on-orbit gains, however due to several 10-watt lamp failures on-orbit, the frequency of SRCA calibrations has been reduced and no longer used in the official L1B LUT algorithm. Once per calendar year the SRCA is operated in radiometric mode over several consecutive orbits using a backup lamp, monitoring the gain changes of the RSBs under unique conditions. This paper will provide insight into the 1W short-term stability of the MODIS RSBs using these calibrations over both the Aqua and Terra missions, along with the long-term trends of these multi-orbit gain observations.

The detectors in the reflective solar bands (RSB) of Terra and Aqua MODIS use a linear relationship to relate the instrument response to the observed top-of-atmosphere radiance. Recently, we reported that Aqua MODIS bands 1 (645 nm) and 2 (858 nm) have deviations from gain linearity that change on-orbit, leading to errors in the NASA Level 1B radiance products for low radiance scenes. In this paper, we expand on these findings to assess the linearity of detector responses in both Aqua and Terra MODIS for all RSB using data from the on-board solar diffuser (SD) and spectro-radiometric calibration assembly (SRCA). We use comparisons of SD observations taken at two different radiance levels: with and without an attenuation screen. The SRCA, operated in radiometric mode, takes observations of RSB detector response at multiple radiance levels using a set of lamps and a neutral density filter. While the lamps do not provide a stable enough source for an accurate radiometric calibration, we can use the relative response with and without the neutral density filter to track changes in gain linearity. Unlike the SD, the SRCA results can be used for nearly all RSB of both instruments. We show that most RSB for both Terra and Aqua MODIS continue to have very linear responses throughout the missions. Some notable exceptions are bands 1 and 2 for both Aqua and Terra, which are used primarily for land imaging applications, and band 26 for Terra, which is used for cirrus cloud identification. The results from the SD and SRCA are in reasonably good agreement with each other.

The special significance of green spaces in cities has been fully demonstrated. Green spaces are key elements to improve the quality of urban life. They contribute to human well-being by providing ecosystem services such as climate regulation, capture of pollutants and flood control. They also promote contact between residents and community integration, and offer a favourable place for health, relaxation and contemplation of nature. Greener environments have lower crime rates. They tend to have a positive effect on people and induce mental vitality. Green spaces can help to reduce urban heat islands and generate true cold islands in urbanized environments. Finally, urban parks provide economic value for cities, including an increase in the value of properties in their proximity. The literature on urban climate has highlighted the singular importance of urban greenery for mitigating urban heat islands (UHI) and extreme temperatures. Urban vegetation plays a fundamental role in adapting to climate change in cities. Green areas register lower temperatures than the rest of urban spaces and have a cooling effect that spreads to their surroundings to create a real “cool island” effect. The World Health Organization recommends that green spaces (of at least 0.5 hectares) should be accessible within a 300 m linear distance of residences. However, the concepts of “park” and “green area” are vague and imprecise, and there is no clear consensus about what should be considered urban greenery. Highly artificialized spaces, with high proportions of sealed soil and no vegetation, are often considered “parks” in urban planning. Likewise, forest spaces, with few artificial elements, are usually considered urban “parks”. The use of satellite images has helped to study urban vegetation. Indicators such as the normalized difference vegetation index (NDVI) and many others allow us to understand the extent and quality of greenery, and to assess its impact on day and night temperatures. In this context, the aim of this study was to examine the extent of vegetation in Barcelona Metropolitan Area (636 km², 3,303,927 inhabitants) from various satellite sensors and indicators of greenness, to determine the thresholds from which it is possible to speak with rigor of urban green.

In many applications, ground-leaving reflectance data is needed to fully understand the desired phenomenology in the obtained observations. Measured sensor data is altered by spectrally dependent effects due to the atmosphere and are commonly removed using an atmospheric compensation algorithm. Most existing atmospheric compensation algorithms are specifically designed to operate on single image products. We present a simple atmospheric compensation algorithm that takes advantage of multiple temporal images over a single site for an improved surface reflectance result. The algorithm identifies psuedo-invariant feature pixels common across an entire time series and adjusts the compensation for an improved result over single image compensation. Initial results show our temporal approach outperforms Sen2Cor, improves reflectance retrieval accuracy of Sentinel-2 products by 3.5% percent, and yields an overall accuracy of 5.6% percent relative to RadCalNet ground truth.

The Earth-observing Visible Infrared Imaging Radiometer Suite (VIIRS) has 14 reflective solar bands (RSBs), covering wavelengths from 412 to 2250 nm. The first VIIRS instrument is aboard the Suomi National Polar-orbiting Partnership (SNPP) satellite launched on October 28, 2011. The RSBs use a sunlit onboard solar diffuser (SD) to calibrate, with the dark count obtained through viewing the Space View (SV) port. Stray light contaminations in the SV and/or the SD View, if present, could negatively impact on the radiometric calibration accuracy. Here, we assess the stray light contamination in the SNPP VIIRS Space View, focusing on band M1 (412 nm).

The first Visible Infrared Imaging Radiometer Suite (VIIRS) instrument onboard the Suomi National Polar-orbiting Partnership (S-NPP) satellite has been collecting Earth observations since 2011. Despite challenges created by ongoing telescope throughput degradation and uneven solar diffuser (SD) degradation, the on-orbit radiometric calibration of the reflective solar bands (RSBs) has been successfully carried out utilizing observations of the sun-lit SD panel. Recently, an increase of striping in S-NPP VIIRS Sensor Data Record (SDR) RSB imagery has been observed, especially for the short-wavelength bands M1 to M4. This observation is also consistent with an increase, during multiple years of on-orbit operation, in the divergence of the on-orbit SD-based calibration factors (F-factors) among the different detectors for those bands. To reduce the observed striping, the NOAA STAR VIIRS SDR Cal/Val team utilized measurements over Deep Convective Clouds (DCC) to develop correction factors that can be applied during the operational automated processing of the S-NPP VIIRS solar calibration measurements in IDPS (RSBAutoCal). In this study, we perform analysis to quantify striping in S-NPP VIIRS M1-M4 imagery using a detector-level cumulative histogram approach over multiple selected cases of homogenous land and ocean targets. We also quantitatively evaluate the impact of DCC-based striping correction by performing striping comparisons using operational (without striping correction) and reprocessed (with striping correction) data. Furthermore, we examine the dependence of the striping correction performance on the detector gain stage involved. The results suggest that the applied corrections can effectively reduce striping for the majority of cases, especially in M1 and M2, where a striping reduction on the order of ~2% can be achieved.

VIIRS day-night band (DNB) covers a wavelength range from 500 nm to 900 nm, has three gain stages enabling a dynamic range of 7 orders of magnitude, and is calibrated by a solar diffuser. In this paper, the calibration uncertainty of the DNB is analyzed for both SNPP and NOAA-20 VIIRS instruments. It is shown that the uncertainties of the DNB for all gain stages, detectors, half angle mirror sides, and aggregation modes are much smaller than the uncertainty specifications of the band, which is 5%, 10%, and 100% for low, middle, and high gain stage, respectively.

The S-NPP and NOAA-20 VIIRS instruments have successfully operated since their launches on October 28, 2011 and November 18, 2017, respectively. A panchromatic channel onboard VIIRS is referred to as the day-night band (DNB), was designed with a large dynamic range and high sensitivity, such that its detectors can make observations during both daytime and nighttime. The DNB uses an onboard solar diffuser (SD) panel for low gain stage calibration, and the SD observations are also carefully selected to compute gain ratios between low-to-mid and mid-to-high gain stages. In this paper, we present the S-NPP and NOAA-20 VIIRS DNB calibration performed by the NASA VIIRS Characterization Support Team (VCST) to generate the calibration coefficient look up tables (LUTs) for the latest NASA Level 1B Collection 2 products. This activity supports the NASA Earth science community by delivering consistent VIIRS sensor data products via the Land Science Investigator-led Processing Systems. The DNB stray light contamination and its different behavior have been highlighted between two instruments. Its estimate and correction methods as well as performance are illustrated.

The Visible Infrared Imaging Radiometer Suite (VIIRS) was successfully launched on November 18, 2017, onboard the National Oceanic and Atmospheric Administration-20 (NOAA-20) satellite. With its activation on November 28, 2017, VIIRS started producing Solar Diffuser (SD) observations and performing on-orbit radiometric calibration for accurate Sensor Data Records (SDRs). After the initial adjustment of the calibration coefficients called F-PREDICTED LUTs, the F-PREDICTED LUT has been set to a constant level for each detector in the Reflective Solar Bands (RSB) after April of 2018 because no noticeable degradation has been observed. Meanwhile, the NOAA VIIRS SDR team has been checking the validity of the F-PREDICTED LUT by comparing with the lunar calibration coefficients (F-factors), Deep Convective Cloud (DCC) trends. Recently, small but meaningful upward trends (indicative of responsivity degradation) were observed in some RSB F-factors and the NOAA-20 VIIRS F-PREDICTED LUT was updated to replace the current constant F-factors for M1-M6 and I1. The changes were based on the lunar calibration trends that were validated by the daily and monthly time series of DCC observations. The changes will smoothly mitigate the observed degradation in two years to provide the best radiometric calibration for the current VIIRS SDR products. By monitoring the lunar and DCC trends frequently, a follow-on update is expected when there is a significant long-term trend change between current F-PREDICTE LUT and lunar or DCC trends.

The VIIRS instrument on the NOAA-20 satellite continues to have excellent performance in its reflective solar bands (RSB), with detector gains changing by less than 0.5% over the first four and a half years of the mission. The calibration of the RSB for NASA’s Collection 2 Level 1B product relies primarily on data from the on-board solar diffuser (SD). Recently, we updated the calibration algorithm to include corrections to the long-term gain trends based on regularly scheduled lunar observations. Earlier in the NOAA-20 mission, due to insufficient lunar data, the long-term correction applied to the SD trends was based on scaled results from the first VIIRS instrument on the SNPP satellite. In this paper, we show updated NOAA-20 data trends of the solar diffuser degradation and the RSB gains derived from every-orbit SD observations. We also show updated lunar data and describe our algorithm for using the lunar data to correct the long-term gain trends. The gain differences between the previous algorithm, using scaled SNPP lunar data, and new algorithm, directly using NOAA-20 lunar data, are largest for bands M2 (0.4%), M3 (0.4%), M4 (0.5%), and I1 (0.6%), and negligible (<0.2%) for the other RSB. The updated algorithm began to be implemented in forward production of the NASA VIIRS Level 1B product starting in September 2021 and will allow for more accurate tracking of the long-term RSB gain in the future.

Six channels of the Advanced Baseline Imager (ABI) onboard the Geostationary Operational Environmental Satellite (GOES) sense radiance in the visible and near infrared (VNIR) spectrum. Two ABI have been launched and in service, one with GOES-16 at 75.2oW and the other with GOES-17 at 137.2oW. Unlike the GOES infrared channels that have had onboard calibration since the 1970’s, ABI is the first GOES instrument that is equipped with onboard calibration for its RSB. This paper reviews the operational calibration of the ABI VNIR channels, including initial post-launch calibration, correction for elevation angle variation, and correction for azimuth (beta) angle variation. GOES-17 suffers from a compromised cooling subsystem such that the instrument temperature, including that for the VNIR Focal Plane Module, varies more than designed. The impact and mitigation of this thermal stress will also be described. Finally, GOES-18 will be launched in 1 March 2022. It is expected that some preliminary calibration results of GOES-18 will be available for discussion. Declaimer: The scientific results and conclusions, as well as any views or opinions expressed herein, are those of the authors and do not necessarily reflect those of NOAA or the Department of Commerce.

The Advanced Baseline Imager (ABI) instrument is the key payload onboard NOAA’s series of Geostationary Operational Environmental Satellites (GOES-R) that provides high-quality earth imagery to improve the weather forecasts and environmental change studies over the Western Hemisphere. GOES-16 was launched on 19 November 2016 and became the GOES-East satellite at 75.2 West on 18 December 2017. As the first satellite in the series, it overcame a number of anomalies in the first few years of operation but has performed well since the calibration was stabilized in April 2019. On the other hand, GOES-17, launched on 1 March 2018 and became GOES-West at 137.2 West on 12 February 2019, suffered a malfunction of the cooling system, which called for quite a different operation and calibration configurations. This talk is a summary of the on-orbit radiometric calibration performances for the GOES-16/17 ABI Infrared (IR) channels. The GOES-16 IR radiance is well calibrated and very stable at various temporal and spatial scales after the two major IR ground system updates in its early in-orbit time. The overall radiometric calibration accuracy of GOES-17 IR channels during the stable period is comparable with that of GOES-16. One exception is that the bias to the reference is relatively large for GOES-17 Ch16 due to the shift of its spectral response function caused by the elevated operational temperature. This talk will also include the impacts of the major calibration events for the GOES-17 ABI IR data after its operation, including the implementation of the predictive calibration (pCal) algorithm in July 2019 to improve the radiometric calibration accuracy at satellite night time, salvaged imagery with the cooling timeline in the peak thermal stress days around the eclipse seasons, change of FPM set-point temperatures in late 2021 and early 2022, and the large-scale best detector select (BDS) updates in December 2021.

Landsat-9, launched on September 27, 2021, carries the Thermal Infrared Sensor (TIRS). The Landsat-9 TIRS is a close copy of the Landsat-8 TIRS instrument; it is a two spectral-band, pushbroom sensor with three Sensor Chip Assemblies (SCAs) that cover the 15-degree field-of-view. The primary radiometric change between the instruments is the addition of baffling in the Landsat-9 TIRS telescope to mitigate the stray light issue that has impacted the radiometric quality of Landsat-8 TIRS. The on-orbit radiometric performance is monitored using the on-board variable temperature blackbody and views of deep space. Maneuvers to look at and around the moon have provided an assessment of the stray light. The absolute calibration is monitored by vicarious calibration techniques by teams at NASA/Jet Propulsion Lab and the Rochester Institute of Technology. Landsat-9 completed a three-month commissioning phase in January 2022 and has been operational since February 2022. The instrument has demonstrated excellent radiometric performance, as assessed from the on-orbit measurements. The TIRS instrument is radiometrically stable to 0.1% within a power cycle, and has noise levels below 0.1K. The lunar scans and the vicarious calibration data provide evidence that the stray light has been effectively mitigated.

Landsat 9 (L9) was launched on September 27, 2021, from Vandenberg Space Force Base in California. The U. S. Geological Survey (USGS) released Level-1 data, geometrically orthorectified and radiometrically calibrated imagery in digital numbers that can be scaled to Top-of-Atmosphere reflectance, and Level-2 data, geometrically orthorectified and radiometrically calibrated surface reflectance imagery, to the public on February 10, 2022. From September 27, 2021 to early January of 2022, the satellite and its two instruments, the Operational Land Imager (OLI) and the Thermal Infrared Sensor (TIRS), were in their commissioning phase, updating key radiometric and geometric calibration parameters for both the spacecraft and the instruments. The data acquired during the commissioning phase of the spacecraft and instruments were reprocessed with the newly determined post-launch calibration parameters prior to the releasing of the data to the public. After the public release of the data, the calibration parameters of the sensors and the spacecraft continue to be monitored to ensure the data released to the public is of the same high quality as previous Landsat data products. This paper discusses three key geometric performance aspects of the L9 spacecraft and its instruments during its early mission time frame (September 27, 2021 to June 27, 2022) including geodetic accuracy, geometric accuracy, and within band registration accuracy of the L9 products generated.

The Landsat 9 satellite was launched on September 27, 2021 to continue systematic imaging of the Earth’s land surfaces. Together with Landsat 8, it provides coverage of the entire Earth every 8 days. Landsat 9 carries the Operational Land Imager (OLI), which is practically a copy of the Landsat 8 OLI, and the Thermal Infrared Sensor (TIRS). In this paper we demonstrate the excellent radiometric performance of Landsat 9 OLI over its first several months of operations on orbit. On-board calibrator data were used to assess the sensor’s radiometric performance characteristics. All spectral bands are radiometrically stable to within 0.1%. The signal-to-noise performance is stable and is 3 to 8% better than Landsat 8. The bias stability is better than 1 DN (Digital Number). The validation of the absolute calibration performed with surface measurements indicated the OLI was calibrated to within 5% in spectral radiance and 3% in reflectance. Still, a comparison between Landsat 9 and Landsat 8 OLI derived top-of-atmosphere reflectance indicated small disagreements between the instruments in all spectral bands, The absolute radiometric calibration of Landsat 9 OLI was adjusted to be in closer agreement with Landsat 8 OLI before products were released to the public.

The Operational Land Imager2 (OLI2), is a visible to short-wave infrared pushbroom imaging radiometer. The OLI2 generates data in 9 spectral bands by sub-selecting the operational detectors within the focal plane array for each spectral band. This sub selection of detectors, also known as the operational flight detector-select-map, was determined pre-launch based on various radiometric performance characteristics. Due to new updates made in OLI2 it is possible to repeat similar extended radiometric characterization of all detectors on the focal plane while the system is on-orbit. This new characterization uses the OLI2 on-orbit calibration devices (Shutter and Stim lamp) observations while toggling the detector select maps from the operational setting to cycle through all possible detectors. Another special new ability for OLI2 was the addition of the stim-lamp non-linearity characterization collects. Both of these new abilities have been successfully executed on-orbit. In this paper and presentation, we present the results from these new characterization capabilities that extended the dynamic range of the non-linearity characterization as well as the on-orbit radiometric characteristics for the full set of focal plane detectors. Such new characterizations further enhance the information on the state of health and aging of the focal plane, as well as enable transfer of radiometric calibration from the operational detector-select to any other detector select. Overall these characterizations help to assure a robust high quality radiometric performance of OLI2 that should last far into the mission life time and beyond.

Combining images from multiple Earth Observing (EO) satellites increases the temporal resolution of the data; overcoming the limitations imposed by low revisit time and cloud coverage. However, this requires an inter-calibration process, to ensure that there is no radiometric difference in top-of-atmosphere (TOA) observations and to quantify any offset in the respective instruments. In addition, combining vicarious calibration processes to the inter calibration of instruments can provide a useful mechanism to validate and compare data from multiple sensors. The Radiometric Calibration Network (RadCalNet) provides automated surface and top-of atmosphere reflectance data from multiple participating ground sites that can be used for instrument vicarious calibration. We present comparative analysis of the Landsat 8 and Landsat 9 Operational Land ImagerOLI) sensors and validate the data by comparing them to measurements from RadCalNet sites as a quantitative inter-calibration approach. The presented process provides SItraceable inter calibration methodology and quantifies the offset and uncertainty in the OLI and MSI instruments to assess if the data can be reliably cross corelated and used by the scientific community.

The Operational Land Imager 2 (OLI-2) is flying on the Landsat 9 satellite that launched on September 27, 2021. OLI-2 collects data in nine spectral bands in the solar reflective region with a ground sample distance (GSD) of 30 m for all bands except the panchromatic band, which has a 15 m GSD. OLI-2 is the second in a series of instruments flying aboard Landsat platforms. Its predecessor, the Operational Land Imager (OLI), onboard the Landsat 8 satellite, has been operational since 2013. Both satellites fly in a sun-synchronous polar orbit at an altitude of 705 km. The OLI-2 telescope completed alignment in December 2017 at Ball Aerospace. In this paper we present a summary of the as-designed and as-built performance of the telescope that demonstrates compliance to the OLI-2 requirements. In addition, we provide a comparison of the OLI-2 telescope performance to the reported OLI telescope performance. Performance parameters discussed include measurement of effective focal length, modulation transfer function, optical throughput, polarization, stray light, and pointing.

Spectral characterizations were made of the Ocean Color Instrument (OCI) short-wave infrared (SWIR) Detection Subassembly (SDS) responses (940–2260 nm) prior to their integration. Using modulated output light from a Fourier transform spectrometer, the in-band relative spectral responses of the nine different configurations of SDSs were found along with out-of-band (OOB) sensitivity. From these spectral responses, the center wavelengths (λ₀), full widths at half of the maximum, full widths at 1% of the maximum, and OOB rejection ratios were determined. All spectral parameters are within requirements. There are 2–8 repeats of each configuration, and the 1 σ spread among repeats is largest for the 1250 nm and 1615 nm high-gain configurations and is greater than 1 nm. The engineering requirement is for these values to be within ±4 nm and ±10 nm, respectively, of the nominal λ₀. There is also a λ₀ temperature dependence, which is expected. This temperature dependence is nearly a linear function of wavelength with a 9.5 × 10⁻³ nm K⁻¹ relationship on average.

The OCI (Ocean Color Instrument) is the main sensor on the upcoming PACE (Plankton Aerosol Cloud ocean Ecosystem) mission. OCI has two hyperspectral CCD sensors covering 340nm to 885nm and 9 SWIR (Short Wave IR) bands from 940nm to 2260nm. SWIR bands have nominal 1km ground pixel size and CCD bands have native 1/8 km ground pixel size in diagnostic mode that will be aggregated into 1km pixels to improve SNR and meet the data rate constraints. OCI has a rotating telescope that is synchronized to the readout of the CCD and SWIR detectors. Full pre-launch system level testing for the OCI ETU (Engineering Test Unit) was completed in June 2021. With time-delayed scan mode, a sub-pixel level time-delay step is applied to the detector readout. This sub-pixel level time-delay step causes a sub-pixel level shift in the start of the data collection. After collecting time-delay step scans with different step sizes, a scan profile with sub-pixel resolution can be constructed. 1/8 and 1/4 of CCD pixel resolutions were achieved using this mode. In this paper, the OCI time-delayed scan mode will be described as well as how it was used to calculate OCI’s high spatial resolution PSF (Point Spread Function), IFOV (instantaneous Field of View), MTF (Modulation Transfer Function), and BBR (Band to Band Registration).

The Ocean Color Instrument (OCI), the primary payload of the Plankton, Aerosol, Cloud, ocean Ecosystem (PACE) observatory, will collect data to monitor the health of Earth’s oceans and atmosphere. The Short-Wave Infrared (SWIR) Detection Assembly (SDA) was built and characterized by the Utah State University Space Dynamics Laboratory (SDL) and is a subsystem of OCI. The SDA measures seven bands centered at 940, 1038, 1250, 1378, 1615, 2130, and 2260 nm, with standard- and high-gain varieties for the 1250 and 1615 nm bands, resulting in nine total detection configurations in the SWIR. The delivery of high-quality science data is critically dependent upon accurately characterizing the linearity of the SDA. Two metrology techniques were employed to measure the linearity and characterize the frequency-dependent linearity uncertainty of the system. The first technique used superposition linearity measurements to determine the DC linearity, and the second technique involved an oscillating small-signal response at seven frequencies to determine the frequency-dependent linearity. Discrepancies between the DC and frequency-dependent linearities constrain the uncertainty between the two. Examining the difference between these two methods for all SDA channels, we find most channels experience an uncertainty below 0.2% with a worst-case measurement uncertainty of 0.31%. Averaging SDA channels with similar detectors, optical filters, and electronics to simulate the flight-like data products yields a worst-case frequency-dependence linearity uncertainty of 0.12%, demonstrating minimal frequency dependence, implying an excellent linearity knowledge. Detailed performance knowledge, including linearity performance, verifies data quality and builds confidence in the success of the PACE mission.

Labsphere has created automated vicarious calibration sites using convex mirror technology in the new FLARE (Field Line-of-sight Automated Radiance Exposure) Network. FLARE has been operational for over two years, with network expansion and performance validation against industry standards and common methods for calibration and validation (cal/val) of 350-2500nm optical Earth Observation Systems (EOS). The FLARE point sources provide absolute and traceable data, creating a new tool in harmonization of satellites with ground sampling distances (GSD) of 0.3m to 60m. This paper provides an overview of the FLARE system and presents findings and improvements in operational hardware and software performance. Once commissioned, all FLARE nodes have been repeatedly targeting Landsat 8, Landsat 9 (starting 2022), and Sentinel 2A/B. This has produced a multi-year archive of radiometric and spatial calibration imagery. Landsat and Sentinel are the premier reference programs for Earth Observation performance and utilize both on-board calibration equipment and on-ground reference sites such as RadCalNet and PICS. This work compares the results of the FLARE technique to current official radiometric coefficients and spatial performance metrics for these satellites. Discussion will center on new insights gleaned from the archive analysis and FLARE’s contribution to the community’s capability for data fusion, instrument harmonization, and the potential to support the concept of Analysis Ready Data (ARD) for easier data use and information extraction. Finally, the future progression of FLARE sites, capabilities, and activities will be outlined.

The Empirical Line Method (ELM) is a widely applied technique of achieving absolute atmospheric correction assuming a linear relationship between the raw Digital Numbers (DNs) or at-sensor radiance and surface reflectance measurements collected in-situ. The ELM measures reference targets of known reflectance in an image. Labsphere has created an automated vicarious calibration system using the SPecular Array Radiometric Calibration (SPARC) mirror technology in the new Field Line-of-sight Automated Radiance Exposure (FLARE) network. In the FLARE system the known reflectance targets are convex mirrors - because of that it is titled Mirror based Empirical Line Method (MELM). In this context, the objective of this work is to present the initial results of the MELM using one the FLARE network system. The FLARE system evaluated in this work is the Alpha Node located at Arlington, SD. Initially, the data collected in 2020 and 2021 with the Alpha FLARE concomitant with the OLI sensor overpass on-board the Landsat-8 satellite were used in the assessment. In summary, the surface reflectance image product available to download for OLI sensor were compared directly with the surface reflectance image resulting from the MELM method. The preliminary results showed the mean absolute error data between the surface reflectance from the OLI Level-2 product image and the surface reflectance from the MELM was lower than 0.01 for the Blue, Green, Red and SWIR-1 bands; lower than 0.03 for the for the NIR and SWIR-2 spectral bands; and around 0.05 for Coastal Aerosol band (all in reflectance units). These results suggest the MELM technique using FLARE has great potential for reflectance surface evaluation of orbital sensors.

In March 2020, the Remote Sensing Group (RSG) of the Wyant College of Optical Sciences at the University of Arizona deployed a ground-viewing radiometer (GVR) equipped with linear motion to support its Radiometric Calibration Test Site (RadCaTS). Prior to the development and deployment of a GVR with linear motion, all GVRs were stationary radiometers. The GVRs consist of 8 spectral channels covering a wavelength range of 400 nm to 1550 nm. Each GVR, including the one with linear motion, are automated systems designed for long-term standalone operation. This paper presents a two-year post-deployment summary and analysis of the GVR fitted with autonomous daily linear motion, GVR 23. Incorporating linear motion to a GVR increases the spatial sample size of the GVR. A larger spatial sample provides RSG with an improved representation of the surface under measurement. The current linear motion system operates autonomously between 16:00 UTC and 22:00 UTC. This work describes the current system design, the data acquired from the radiometer, issues that have risen, and future improvements.

The Remote Sensing Group (RSG) of the Wyant College of Optical Sciences at the University of Arizona currently has a single hyperspectral instrument, a Spectrometer Arduino Mega (SpAM), deployed to its Radiometric Calibration Test Site (RadCaTS). Results have shown SpAM to be a robust and accurate instrument, however its high degree of customization makes it difficult to productively reproduce. The Hamamatsu C12880MA is a commercial micro-spectrometer that may provide a solution to RSG’s need for an easily deployable and reproducible hyperspectral instrument. The C12880MA is an ultra-compact grating-based spectrometer that operates in the visible and near-infrared (VNIR). This work presents the initial development of the C12880MA, which involves prototyping and characterizing the device for automated field deployment. The micro-spectrometer is prototyped using a device-specific evaluation circuit, measurement software, and a custom 3D printed electrostatic discharge (ESD) safe housing. It is characterized in RSG’s laboratory and auxiliary facilities. The eventual goal for this device is to become an autonomous standalone system that can be easily deployed and integrated into the RadCaTS suite of instruments. The results from this work will determine the efficacy of the instrument as well as its potential for future deployment. The daily hyperspectral measurements from this device, if deployed, will supplement the current data, and reduce the uncertainty of RadCaTS results.

The Radiometric Calibration Test Site (RadCaTS) was developed by the Remote Sensing Group of the Wyant College of Optical Sciences at the University of Arizona in the early 2000s. During the prototyping phase, RadCaTS was used to supplement the in situ data that were routinely collected by on-site personnel using the traditional reflectance-based approach. A data processing methodology was developed, tested, and compared to the reflectance-based results during this stage. The experience gained in this process resulted in the development of radiometrically-stable, all-weather, groundviewing radiometers (GVRs), the first of which were deployed in 2012. Additional upgrades over the past ten years have included a satellite uplink station, upgraded Cimel CE-318T solar-lunar photometer, and a GVR with linear motion. This work provides an overview of RadCaTS, describes the lessons learned during the past ten years of operation, and also a summary of the radiometric calibration and validation results for such sensors as Landsat 7 ETM+, Landsat 8 and 9 OLI, Terra and Aqua MODIS, SNPP and NOAA-20 VIIRS, Sentinel-2A and -2B MSI, and GOES-16 and -17 ABI.

Higher spatial and temporal resolution in Earth observation (EO) imagery is beneficial for many applications, including real-time hazard monitoring and agriculture. SI-traceable intercomparison between concurrent Earth imaging satellites can help harmonizing and improving spatial or temporal resolution of their data products. To that end, the Committee on Earth Observation Satellites (CEOS) initiated the Radiometric Calibration Network (RadCalNet) to provide automated surface and top-of-atmosphere TOA reflectance data from multiple participating ground sites to the worldwide user community. Here, we report results of the intercomparison between Landsat-8 Operational Land Imager (OLI) and National Oceanic and Atmospheric Administration-20 (NOAA-20) Visible Infrared Imaging Radiometer Suite (VIIRS) in the wavelength range from 400 nm to 2500 nm using RadCalNet as a common reference. The results of this study indicate that RadCalNet is a valuable tool for harmonization of EO imagery even in the case where the sensors have different temporal and spatial resolutions. Both OLI and VIIRS are calibrated to well-within their absolute radiometric calibration uncertainties and the two sensors agree with each other to better than 2% using RadCalNet sites as a common reference.

The CEOS recommended Libya-4 Pseudo Invariant Calibration Site (PICS), located at 28.55° N and 23.39° E, has been extensively used for post-launch radiometric calibration and stability assessment of high, medium, and low-resolution satellite imagers, including MODIS and VIIRS. The NASA LaRC CERES Imager and Geostationary Calibration Group (IGCG) utilizes Libya-4 to perform an independent assessment of the radiometric stability of the MODIS and VIIRS L1B products, which are used in scene identification to convert CERES broadband radiances into fluxes. The site is also used for absolute radiometric scaling between MODIS, VIIRS, and geostationary imagers to ensure consistent cloud and radiative flux retrievals. The Libya-4 clear-sky TOA observed reflectances from sun-synchronous sensors are modeled as a function of solar angle. The model provides observed relative reflectances within 1% for bands 1-6 and 1.8% for band 7 (2.1-μm). The Terra-MODIS, Aqua-MODIS, and NPP-VIIRS Libya-4 TOA observed relative reflectances are shown to fluctuate in tandem. The residual reflectance variability is associated with cloudy and dust storm events as well as seasonal variations of atmospheric parameters, such as precipitable water (PW) and ozone. By correlating the Aqua-MODIS TOA reflectances with PW and solar angle, the 0.64-μm, 0.86-μm, 1.24-μm, and 2.1-μm relative reflectance variability is reduced by 10%, 25%, 20%, and 50%, respectively. The relative reflectance dependency with ozone was minimal. Bright reflectance outliers were associated with large AOD events, whereas darker outliers were related to cloud events. The improved Libya-4 approach provides Aqua-MODIS C6.1 channel stability assessments that range between 0.5% for the 1.6-μm band and 0.9% for the 0.46-μm band.

The NASA Clouds and the Earth's Radiant Energy System (CERES) energy balanced and filled (EBAF) product provides top of atmosphere SW and LW fluxes for monitoring the Earth’s energy budget and to validate climate models. The current EBAF Ed4.1 products, based on the Terra and Aqua CERES instrument observed radiances, rely on MODIS cloud properties to determine the scene-selected angular distribution model used to convert the CERES radiances into fluxes. The CERES EBAF Ed4.2 product will be based on NOAA-20 CERES observations beginning with the data month of April 2022. A seamless transition of fluxes and clouds can only occur if the analogous MODIS and VIIRS channels are properly inter-calibrated. The spectral response functions (SRF) of these bands differ noticeably and will require scene-dependent spectral band adjustment factors (SBAF) for proper radiometric scaling between them. VIIRS I1 and M5 same-granule reflectance measurements provide the optimal opportunity to validate SBAFs over many surface and cloud conditions. The CERES project maintains SCIAMACHY-, GOME-2-, and Hyperion-based scene-stratified hyper-spectral reflectance measurements that can be convolved with sensor pair SRFs to compute the corresponding SBAFs. The SCIAMACHY 2^nd order and GOME-2 linear fit SBAFs were optimal in providing spatially uniform M5/I1 spectrally corrected reflectance ratios over all-sky tropical ocean scenes, which corrected both clear-sky and bright cloud reflectances simultaneously by varying the SBAF as a function of the reflectance. Dome-C and deep convective cloud (DCC) SBAFs had a small M5/I1 SBAF reflectance correction of ~1.03, whereas Libya-4 had a large M5/I1 SBAF reflectance correction of ~1.085. DCC, Dome-C and Libya-4 are spectrally uniform spatial targets with site M5/I1 reflectance ratio spatial homogeneity within 0.2%. The impact of cloud contamination from both the cloud tops and shadows over Libya-4 reduced the M5/I1 reflectance ratio. Over Dome-C, cloud contamination did not significantly shift the M5/I1 reflectance ratio. The Hyperion spectral reflectances are too coarse to be convolved with the narrow M5 SRF, which resulted in M5/I1 SBAFs that differed significantly from those based on SCIAMACHY and GOME-2.

The Visible Infrared Imaging Radiometer Suite (VIIRS) plays an important role in Earth observations and climate studies. Multiple VIIRS instruments have been built, including one onboard the Suomi-NPP spacecraft and another onboard the NOAA-20 spacecraft. These instruments have been extensively tested pre-launch in ambient environment and in a thermal vacuum chamber. One of the important tests in the ambient environment provides the characterization of the straylight response of the instrument. The straylight rejection requirement states that for the spacecraft in an operational, nadir-facing attitude, the VIIRS sensor response to any straylight striking the sensor on any surface (except the entrance aperture and within the sensor field of view) from any angle shall be less than 1% of the response to the given typical spectral radiances. It is challenging to replicate the straylight from the operational environment in a clean room; therefore, some modeling plus a special laboratory setup is necessary. The straylight test is briefly summarized and the test results from five VIIRS instruments built in the past 15 years are compared. It is shown that the straylight performance remained consistent among the VIIRS instruments and they meet the requirement by a healthy margin at the beginning of life, which indicates expected low levels of straylight on-orbit.

The two VIIRS instruments onboard the Suomi NPP and NOAA-20 spacecraft collect data in 22 spectral bands from 0.4 μm to 12.5 μm. Both instruments have exhibited a known artifact in the behavior of the capacitive transimpedance amplifier (CTIA) that manifests in a double-valued (rollover) response in the Earth view imagery. This behavior was identified and well characterized prelaunch and is predominant for the single-gain band M6. The rollover phenomenon is also observed in the low-gain stages of the dual-gain reflective solar bands and thermal bands M12 and I4. The NASA VIIRS L1B processing currently has a rollover flagging scheme that is based on fixed thresholds in the digital numbers (DN) that approximately correspond to the designed maximum radiance of the band. In this work, we briefly review the prelaunch results, current on-orbit flagging methodology, and propose improvements for future implementation in the NASA Level 1B products. Results indicate significant improvement via more accurate flagging for the rollover pixels in band M6 as well as other RSB. The methodologies developed in this work could also be applicable to future VIIRS instruments to be launched aboard the JPSS-2-4 spacecrafts.

The Visible Infrared Imaging Radiometer Suite (VIIRS) instruments onboard the Suomi National Polar-orbiting Partnership (SNPP) and NOAA-20 (N20) satellites have been operational since their launch on October 28, 2011 and November 18, 2017, respectively. VIIRS has 22 spectral bands with wavelengths ranging from visible (VIS) to long-wave infrared (LWIR). Observations from both the SNPP and N20 VIIRS sensors have been used to develop a wide range of data products that have benefited a number of studies of the Earth’s atmosphere, land, and oceans. Among these 22 VIIRS bands, the day/night band (DNB) is a visible/near-infrared panchromatic band. It has three gain stages: low-gain stage (LGS), medium-gain stage (MGS), and high-gain stage (HGS), which allows us to study the Earth at any time of day or night. With its HGS, the DNB can also observe reflected lunar radiances at night. This research uses numerous daily observations of the reflected lunar radiances at night from Dome-C to investigate the long-term calibration stability of the DNB and the calibration consistency between the two VIIRS sensors. The VIIRS DNB measured lunar radiances are compared to those predicted by the GIRO (GSICS Implementation of the ROLO) model.

The Joint Polar Satellite System 4 (JPSS-4) is the follow-on for the Suomi-National Polar-orbiting Partnership (S-NPP) and Joint Polar Satellite Systems 1-3 (JPSS-1, -2 and -3) missions. A primary sensor on both JPSS and S-NPP spacecrafts is the Visible-Infrared Imaging Radiometer Suite (VIIRS) that provides valuable weather and climate products to the user community. VIIRS covers the Reflective Solar Band (RSB) and Thermal Emissive Band (TEB) spectral regions and contains a Day Night Band (DNB) that uses Lunar illumination at night. VIIRS provides top-of-atmosphere radiance, reflectance, and brightness temperature within the Sensor Data Records (SDRs) that are used in sea surface temperature, cloud characterization, land surface properties and ocean color/chlorophyll Environmental Data Record (EDR) products. The SDR calibration is performed using unpolarized sources such as a Solar Diffuser (SD) for the RSBs or an On-Board Calibrator BlackBody (OBCBB) for the TEBs. Earth scenes with polarizing properties will create radiometric bias errors within the SDRs based on how sensitive VIIRS is to polarized illumination and must be corrected in some EDR algorithms. This paper will discuss the JPSS-4 VIIRS polarization characterization methodology, polarization sensitivity results and compare its performance to its predecessors S-NPP and JPSS-1 through -3 VIIRS.

Eutrophication of water is becoming more and more serious, especially Case 2 waters in inland lakes. Chlorophyll-a (Chla) concentration is an important characterized parameter of inland lakes. Its monitoring can effectively reflect the eutrophication degree and changes of inland lakes. This study takes Chagan Lake as the research area, which is located in Jilin Province, Northeast China. Using Sentinel-2 images combined with the in-situ measurements at the same day to establish the inversion model of Chl-a concentration through single-band and band-ratio methods. In the process, models with these two methods using in-situ hyperspectral data which is corresponding to bands on Sentinel-2 are also developed. The experimental results show that the band-ratio method has a better effect on inhibiting the effects of total suspended sediments (TSS) and colored dissolved organic matter (CDOM) compared with single band method. In-situ model (R² = 0.69) and model for Sentinel-2 (R² = 0.69) have RMSE at 2.54 ug L^-1 and 2.51 ug L^-1, respectively. Thus, it is feasible to retrieve Chl-a in inland lakes from northeast China by remote sensing using the band-ratio method, which is of great significance for monitoring and early warning of water eutrophication.

Snow is an essential element in surface climate studies. Snow measurement using remote sensing data gradually has become mainstream with the continuous development of remote sensing technology. Synthetic aperture radar (SAR) has the capability of all-day, all-weather ground observation, and combines high spatial resolution, interferometry, and polarization imaging. C-band SAR images are more sensitive to snow characteristics and are an effective data source for obtaining the spatial distribution of snow in areas with complex terrain. In recent years, the methods of using machine learning and deep learning to invert snow parameters such as snow depth and snow water equivalent have been widely used. In this paper, the study area is the farmland area of Northeast China. The relationship between snow depth and snow parameters is analyzed by a deep learning algorithm and machine learning algorithm using multi-temporal Sentinel-1 Cband SAR data, measured snow depth data, and so on. The inversion model based on the type of farmland subsurface is established to invert the snow depth in the farmland area of northeast China and output the map. The aim is to generate surface-based snow depth inversion results with less error and higher accuracy. The average MAE of the model is 2.06cm, RMSE is 2.96 cm, R² is 0.72, the maximum absolute error is 10.206 cm, and the minimum absolute error is 0.045cm for the farmland mixed group feature screening dataset.

This paper proposes a method of water body extraction from remote sensing images based on deep learning. Using Landsat images to establish a joint probability model of convolution neural network in which the full connection layer is replaced by the multi-layer perceptron and combined with spectral information. This paper trains the CNN model firstly and then puts the extracted features into the MLP classifier. Since the full connection layer of the CNN can obtain full information, using MLP to replace the full connection layer to train the neural network has a higher identification ratio. In addition, due to the insufficient utilization of spectral information from remote sensing images by CNN, and NDWI can highlight the water body characteristics in remote sensing images, using the joint probability model can effectively improve the water body extraction accuracy. In this paper, two water bodies in Jilin Province of China, Jinyue Lake, and Chagan Lake, are selected as the study areas. The water body extraction method based on a joint probability model of convolutional neural network and spectral information is compared with the common water body extraction methods, and the new algorithm has a kappa coefficient of 0.86 and an overall accuracy of 93.17% for the water body extraction of Jinyue Lake, and it has a kappa coefficient of 0.86 and an overall accuracy of 93% for the water body extraction of Chagan Lake. The new algorithm has improved the kappa coefficient and overall accuracy compared with other water body extraction methods, which proves the high identification ratio.

Using remote sensing images to automatically extract road network information has become an imposing method because manually labeling roads is a time-consuming and laborious task. Traditional road extraction is regarded as a pixel-based segmentation method, which predicts the probability of each pixel on the road. It often neglects the pixel’s neighboring information and the topology property of the road network, so some branches of roads cannot be recognized by this method. To mitigate the problem, this paper proposes a new road extraction method with attention and topology modules, TSELinkNet. It extracts channel features and topology features from the images and then integrates them with spatial features to acquire a comprehensive feature. This method is conducted on two different types of road datasets including urban and rural areas. Also, we compare the predicted results from TSE-LinkNet with other results from existing methods using Precision, Recall, F1 and mean intersection over union (mIoU) evaluation metrics. In both datasets, TSE-LinkNet improves Precision metrics by 0.61% and 2.58% respectively. In the urban road dataset, other metrics arise greatly as well, where Recall, F1 and mIoU increase1.28%, 14.18% and 4.46% sharply. The extracted roads from TSE-LinkNet have better connectivity compared with other methods in qualitative results. Experimental results showed that this method has a satisfying ability to extract roads from complex-topology urban areas.

This PDF file contains the front matter associated with SPIE Proceedings Volume 12232 including the Title Page, Copyright information, Table of Contents, and Conference Committee Page.