Search

The Edition 2 (Ed2) cloud property retrieval algorithm system was upgraded and applied to the MODerate-resolution Imaging Spectroradiometer (MODIS) data for the Clouds and the Earth’s Radiant Energy System (CERES) Edition 4 (Ed4) products. New calibrations for solar channels and the use of the 1.24- $\mu \text{m}$ channel for cloud optical depth (COD) over snow improve the daytime consistency between Terra and Aqua MODIS retrievals. Use of additional spectral channels and revised logic enhanced the cloud-top phase retrieval accuracy. A new ice crystal reflectance model and a CO2-channel algorithm retrieved higher ice clouds, while a new regional lapse rate technique produced more accurate water cloud heights than in Ed2. Ice cloud base heights are more accurate due to a new cloud thickness parameterization. Overall, CODs increased, especially over the polar (PO) regions. The mean particle sizes increased slightly for water clouds, but more so for ice clouds in the PO areas. New experimental parameters introduced in Ed4 are limited in utility, but will be revised for the next CERES edition. As part of the Ed4 retrieval evaluation, the average properties are compared with those from other algorithms and the differences between individual reference data and matched Ed4 retrievals are explored. Part II of this article provides a comprehensive, objective evaluation of selected parameters. More accurate interpretation of the CERES radiation measurements has resulted from the use of the Ed4 cloud properties.

The Clouds and Earth’s Radiant Energy System (CERES) has been monitoring clouds and radiation since 2000 using algorithms developed before 2002 for CERES Edition 2 (Ed2) products. To improve cloud amount accuracy, CERES Edition 4 (Ed4) applies revised algorithms and input data to Terra and Aqua MODerate-resolution Imaging Spectroradiometer (MODIS) radiances. The Ed4 cloud mask uses 5–7 additional channels, new models for clear-sky ocean and snow/ice-surface radiances, and revised Terra MODIS calibrations. Mean Ed4 daytime and nighttime cloud amounts exceed their Ed2 counterparts by 0.035 and 0.068. Excellent consistency between average Aqua and Terra cloud fraction is found over nonpolar regions. Differences over polar regions are likely due to unresolved calibration discrepancies. Relative to Ed2, Ed4 cloud amounts agree better with those from the Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observations (CALIPSO). CALIPSO comparisons indicate that Ed4 cloud amounts are more than or as accurate as other available cloud mask systems. The Ed4 mask correctly identifies cloudy or clear areas 90%–96% of the time during daytime over nonpolar areas depending on the CALIPSO–MODIS averaging criteria. At night, the range is 88%–95%. Accuracy decreases over land. The polar day and night accuracy ranges are 90%–91% and 80%–81%, respectively. The mean Ed4 cloud fractions slightly exceed the average for seven other imager cloud masks. Remaining biases and uncertainties are mainly attributed to errors in Ed4 predicted clear-sky radiances. The resulting cloud fractions should help CERES produce a more accurate radiation budget and serve as part of a cloud property climate data record.

The foundation for future space mission Climate Absolute Radiance and Refractivity Observatory (CLARREO) is the ability to produce climate change benchmark records and provide on-orbit calibration standard through the highly accurate and Systeme Internationale -traceable observations. The accuracy of CLARREO measurements is set to 0.3% $(k=2)$ for spectrally resolved reflectance. The instrument sensitivity to polarization and polarization of reflected light at the top of atmosphere are the sources for systematic uncertainty. In this paper, we estimate radiometric errors due to polarization effects for CLARREO benchmark and reference intercalibration observations. Data from the Polarization and Anisotropy of Reflectance for Atmospheric Sciences coupled with Observations from Lidar (PARASOL) instrument, a spaceborne polarimeter, have been used in combination with the orbital modeling of Earth's sampling. For the CLARREO benchmark data, we used simulated annual nadir sampling for the polar orbit with 90° inclination, and for the intercalibration with cross-track sensors on the JPSS, such as CERES and VIIRS, we simulated on-orbit matched data sampling. Selected PARASOL data over one full solar year provided polarization parameters in visible (VIS) spectral range. For estimating polarization in near infrared (NIR) spectral range, we used a radiative transfer model. Our results show that to limit error contribution due to polarization to half of the allowed total, the sensitivity to polarization of CLARREO reflected solar instrument should not exceed 0.5% $(k=2)$ in spectral range from VIS to NIR.

The Climate Absolute Radiance and Refractivity Observatory (CLARREO) mission was recommended by the National Research Council in 2007 to conduct highly accurate and International System of Unit-traceable decadal change observations and provide an on-orbit intercalibration standard with high accuracy for relevant Earth observing sensors. The goal of reference intercalibration is to enable rigorous observations of critical climate change variables, including reflected broadband radiation, cloud properties, and changes in surface albedo, including snow and ice albedo feedback, to be made consistently among different sensors. This requires the CLARREO Reflected Solar Spectrometer (RSS) to provide highly accurate spectral reflectance measurements to establish an on-orbit reference with a radiometric accuracy requirement better than 0.3% $(\mathrm{k} =2) $ for existing sensors. In this paper, MODTRAN-simulated top-of-atmosphere spectral data and spectral measurements from the SCIAMACHY instrument on Envisat are used to determine sensitivity of intercalibration uncertainty on key design parameters of the CLARREO spectrometer: spectral range, sampling and resolution. Their impact on intercalibration uncertainty for MODIS and VIIRS imagers is estimated for various surface types (ocean, vegetation, desert, snow, deep convective clouds, clouds and all-sky) . Results indicate that for the visible to near-infrared spectral region (465–856 nm) , the RSS instrument under current design concept produces uncertainties of 0.16% for the spectral range and 0.3% for the sampling and resolution. However, for the water vapor absorption bands in the short wavelength infrared region (1242–1629 nm) , the same requirement is not met for sampling and resolution due to their high sensitivity to the influence of atmospheric water vapor.

One of the Climate Absolute Radiance and Refractivity Observatory (CLARREO) mission objectives is to provide a high accuracy calibration standard on orbit to enable inter-calibration of existing sensors. In order to perform an accurate inter-calibration of imaging radiometers, such as VIIRS, one must take into account instrument sensitivity to polarization of incoming light. Even if the sensitivity to polarization of an instrument is established or known on orbit, the knowledge of the polarization state of reflected light is required to make relevant radiometric corrections. In the case when coincident polarimetric measurements are not available, we propose to use a combination of empirical and theoretical models to predict the polarization of solar reflected light at the top-of-atmosphere. We used observations from on-orbit polarimeter PARASOL to derive a global set of empirical Polarization Distribution Models (PDM) as a function of scene type and viewing geometry. The PDM accuracy for the mean values is estimated to match the 3% PARASOL uncertainty in its polarization measurements. The instantaneous single sample uncertainty of the prototype PDMs for the linear degree of polarization is contained within 15%. We also present the formalism and numeric estimates for resulting uncertainty for inter-calibration of an imaging radiometer with the CLARREO reference observations, including uncertainty due to instrument sensitivity to polarization. The uncertainty estimates consider a range of scenarios with varying data sampling, uncertainty of polarization, and imaging radiometer sensitivity to polarization. These results are used to recommend CLARREO mission requirements relevant to reference inter-calibration and polarization effects.

For pt.I see ibid., vol.37, no.1, p.543-54 (1999). Advanced Very High Resolution Radiometer (AVHRR) images acquired over the Arctic Ocean often show strong bidirectional reflectance of snow on sea ice. The observed anisotropic reflectance characteristics of these images are consistent with the theoretical analysis presented in Part I of this study. The anisotropic reflectance at the top of atmosphere (TOA) is simulated by radiative transfer modeling and the results show a good model-observation agreement. Based on these modeling results, a method was developed to correct the effect of anisotropic reflectance on AVHRR images in channels 1 and 2. Results show the method is effective and efficient. Comparisons of TOA snow reflectance before and after anisotropic correction show that the systematic and large variation of snow reflectance across the scan lines of an image due to satellite viewing/illumination geometry can either be eliminated or greatly reduced by applying the correction algorithm.

Relatively little work on cloud shadow detection has been published and many of these papers deal with restricted geometries. Here, arbitrary viewing and illumination conditions are considered. A means is provided to extend more restricted treatments of cloud shadow detection and removal to the general case.

The bidirectional reflectance anisotropy in Advanced Very High Resolution Radiometer (AVHRR) Channel 1 and 2 spectral regions for snow- and ice-covered surfaces was investigated using a comprehensive radiative transfer model for the coupled atmosphere, snow, ice, and ocean system. The model has been developed to include surface roughness effects. The importance of the scattering characteristics of snow and the surface roughness of sea ice to the bidirectional reflectance of these surfaces is presented. The modeled anisotropic factor, however, shows less sensitivity to aerosol loading and snow soot contamination than to solar elevation, snow phase function, and ice surface roughness. Results show high anisotropic reflectance factors (ARFs) and different reflectance features for snow and ice surfaces in both the AVHRR Channel 1 and 2 spectral regions. The model-observation comparison indicates that the model is able to predict the general characteristics of the bidirectional reflection of snow and sea ice. The introduction of surface roughness in the model can well explain the extremely strong forward or high anisotropic reflectance of sea ice.

A collocated nadir observation technique has been used as part of the geostationary meteorological satellite (GMS) pathfinder project and is now employed at the Australian Bureau of Meteorology Research Centre to calibrate the visible infrared spin scan radiometer (VISSR) instrument used on GMS-5. It uses satellite-to-satellite cross calibration to bypass many of the problems inherent in the absolute calibration of satellite instruments. Orbital calculations determine where and when the VISSR and advanced very high resolution radiometer (AVHRR) instruments on the GMS and National Oceanic and Atmospheric Administration (NOAA) satellites, respectively, are observing the same Earth-atmosphere scene from the same direction at the same time. Results show that, with careful selection of calibration scenes, the satellites are observing the same radiation field, allowing a fundamental count-to-count calibration of the instrumentation. This calibration can subsequently be related to radiance. It should be possible to cross correlate all geostationary satellites currently employed using a similar method. This paper makes the following three contributions. 1) Using radiative transfer modeling, a sound theoretical basis for the calibration technique is provided. 2) Behavior of the GMS-5 VISSR instrument is tracked for the Pathfinder benchmark period (July 1, 1995-June 30, 1996). 3) Detailed theoretical and experimental comparison of the GMS-4 and GMS-5 VISSR instruments, which takes into account their different spectral response functions, is provided.