Search

Cloud optical properties such as optical thickness along with surface albedo are important inputs for deriving the shortwave radiative effects of clouds from space-borne remote sensing. Owing to insufficient knowledge about the snow or ice surface in the Arctic, cloud detection and the retrieval products derived from passive remote sensing, such as from the Moderate Resolution Imaging Spectroradiometer (MODIS), are difficult to obtain with adequate accuracy – especially for low-level thin clouds, which are ubiquitous in the Arctic. This study aims at evaluating the spectral and broadband irradiance calculated from MODIS-derived cloud properties in the Arctic using aircraft measurements collected during the Arctic Radiation-IceBridge Sea and Ice Experiment (ARISE), specifically using the upwelling and downwelling shortwave spectral and broadband irradiance measured by the Solar Spectral Flux Radiometer (SSFR) and the BroadBand Radiometer system (BBR). This entails the derivation of surface albedo from SSFR/BBR and camera imagery for heterogeneous surfaces in the marginal ice zone (MIZ), as well as subsequent measurement-model inter-comparisons in the presence of thin clouds. In addition to MODIS cloud retrievals and surface albedo from SSFR, we used temperature and humidity data from in-situ data and reanalysis (MERRA-2). The spectral surface albedo derived from the airborne radiometers is consistent with prior ground-based measurements, and adequately represents the surface variability for the study region and time period. Somewhat surprisingly, the primary error in MODIS-derived irradiance fields for this study stems from undetected clouds, rather than from the retrieved cloud properties. In our case studies, about 22 % of clouds remained undetected (cloud optical thickness less than 0.5). The radiative effect of clouds above the detection threshold was −40 W m−2 above cloud, and −39 W m−2 below the cloud layer, and the optical thickness from the MODIS "1621" cloud product was consistent with the reflected and transmitted irradiance observations. This study suggests that passive imagery cloud detection could be improved through a multi-pixel approach, that would make it more dependable in the Arctic. Regardless of the cloud retrieval method, there is a need for an operational imagery-based surface albedo product for the polar regions that adequately captures its temporal, spatial, and spectral variability to estimate cloud radiative effect from space-borne remote sensing. [ABSTRACT FROM AUTHOR]

In this study we developed a neural network (NN) that can be used to relate a large dataset of multi-angular and multi-spectral polarimetric remote sensing observations to retrievals of cloud microphysical properties. This effort builds upon our previous work, which explored the sensitivity of neural network input, architecture, and other design requirements for this type of remote sensing problem. In particular this work introduces a framework for appropriately weighting total and polarized reflectances, which have vastly different magnitudes and measurement uncertainties. The NN is trained using an artificial training set and applied to Research Scanning Polarimeter (RSP) data obtained during the ORACLES field campaign (Observations of Aerosols above Clouds and their Interactions). The polarimetric RSP observations are unique in that they observe the same cloud from a very large number of angles within a variety of spectral bands resulting in a large dataset that can be explored rapidly with a NN approach. The usefulness applying a NN to a dataset such as this one stems from the possibility of rapidly obtaining a retrieval that could be subsequently applied as a first-guess for slower but more rigorous physical-based retrieval algorithms. This approach could be particularly advantageous for more complicated atmospheric retrievals - such as when an aerosol layer lies above clouds like in ORACLES. For the ORACLES 2016 dataset comparisons of the NN and standard parametric polarimetric (PP) cloud retrieval give reasonable results for droplet effective radius (re : R=0.756, RMSE=1.74μm) and cloud optical thickness (τ : R=0.950, RMSE=1.82). This level of statistical agreement is shown to be similar to comparisons between the two most well-established cloud retrievals, namely the the polarimetric cloud retrieval and the bispectral total reflectance cloud retrieval. The NN retrievals from the ORACLES 2017 dataset result in retrievals of re (R=0.54, RMSE=4.77μm) and τ (R=0.785, RMSE=5.61) that behave much more poorly. In particular we found that our NN retrieval approach does not perform well for thin (τ<3), inhomogeneous, or broken clouds. We also found that correction for above-cloud atmospheric absorption improved the NN retrievals moderately - but retrievals without this correction still behaved similarly to existing cloud retrievals with a slight systematic offset. [ABSTRACT FROM AUTHOR]

This study presents and evaluates an updated algorithm for quantification of absorbing aerosols above clouds (AACs) from passive satellite measurements. The focus is biomass burning in the south-eastern Atlantic Ocean during the 2016 and 2017 ObserRvations of Aerosols above CLouds and their interactionS (ORACLES) field campaign deployments. The algorithm retrieves the above-cloud aerosol optical depth (AOD) and underlying liquid cloud optical depth, and is intended to be applied to measurements from sensors including the Sea-viewing Wide Field-of-view Sensor (SeaWiFS), Moderate Resolution Imaging Spectroradiometers (MODIS), and Visible Infrared Imaging Radiometer Suite (VIIRS). Together, these sensors provide around twenty years of observations to date. Airborne NASA Ames Spectrometers for Sky-Scanning, Sun-Tracking Atmospheric Research (4STAR) and NASA Langley High Spectral Resolution Lidar 2 (HSRL2) data collected during ORACLES provide important validation for spectral AOD for MODIS and VIIRS; as the SeaWiFS mission ended in 2010, it cannot be evaluated directly. These 4STAR and HSRL2 comparisons are complimentary and reveal performance generally in line with theoretical expectations. At present the two MODIS-based data records seem the most reliable, although there are differences between the deployments. Data collected in the region from other sources are also used to evaluate some assumptions made in the AAC retrieval. Spatiotemporal patterns in the data sets are very similar, and the time series themselves are very strongly correlated with each other (correlation coefficients from 0.95-0.99). Offsets between the time series are thought to be linked to differences in absolute calibration between the sensors, which can also explain some of the differences in validation results. The time series are also strongly correlated (correlations 0.78-0.94) with quantities such as ultraviolet aerosol index, total column AOD from standard MODIS aerosol products, and active fire detections. This suggests that these quantities may also act as proxies for the above-cloud aerosol load in this region, when AAC retrievals are unavailable. [ABSTRACT FROM AUTHOR]

Nine Pandora Spectrometer Instruments (PSI) were installed at 8 sites in South Korea as part of the KORUS-AQ (Korea U.S.-Air Quality) field study integrating information from ground, aircraft, and satellite measurements for air-quality studies. The PSI made direct-sun measurements of total vertical column NO2, C(NO2), with high precision (0.05 DU, where 1 DU = 2.69 x 1016 molecules/cm²) and accuracy (0.1 DU) that were retrieved using spectral fitting techniques. Retrieval of Formaldehyde (HCHO) total column amounts were also obtained at five sites using the recently improved PSI. The retrievals have with high precision, but possibly lower accuracy than for NO2 because of uncertainty about the optimum spectral window for all ground-based and satellite instruments. PSI direct-sun retrieved values of C(NO2) and C(HCHO) are always significantly larger than OMI retrieved C(NO2) and C(HCHO) for the OMI overpass times (13.5 ± 0.5 hours). In urban areas, PSI C(NO2) averages are at least a factor of two larger than OMI averages. Similar differences are seen for C(HCHO) in Seoul and nearby surrounding areas. Late afternoon values of C(HCHO) measured by PSI are even larger, implying that OMI early afternoon measurements underestimate the effect of poor air quality on human health. The primary cause of the OMI underestimate is the large OMI field of view FOV that includes regions containing low values of pollutants. In relatively clean areas, PSI and OMI are more closely in agreement. C(HCHO) amounts were obtained for five sites, Yonsei University in Seoul, Olympic Park, Taehwa Mtn., Amnyeondo, and Yeoju. Of these the largest amounts of C(HCHO) were observed at Olympic Park and Taehwa Mountain, surrounded by significant amounts of vegetation. Comparisons of PSI C(HCHO) results were made with the Compact Atmospheric Multispecies Spectrometer CAMS during overflights on the DC-8 aircraft for Taehwa Mtn and Olympic Park. In all cases, PSI measured substantially more C(HCHO) than obtained from integrating the CAMS altitude profiles. PSI C(HCHO) at Yonsei University in Seoul frequently reached 0.6 DU and occasionally exceeded 1.5 DU. The semi-rural site, Mt. Taehwa, frequently reached 0.9 DU and occasionally exceeded 1.5 DU. Even at the cleanest site, Amnyeondo, HCHO occasionally exceeded 1 DU. [ABSTRACT FROM AUTHOR]