Explore the vast range of titles available.

Biomass burning (BB) aerosols are the largest source of absorbing aerosols on Earth. Coupled with marine stratocumulus clouds (MSC), their radiative effects are enhanced and can cause cloud property changes (first indirect effect) or cloud burn-off and warm up the atmospheric column (semi-direct effect). Nevertheless, the derivation of their quantity and optical properties in the presence of MSC clouds is confounded by the uncertainties in the retrieval of the underlying cloud properties. Therefore, a robust methodology is needed for the coupled retrievals of absorbing aerosol above clouds. Here, we present a new retrieval approach implemented for a Spectro radiometric multi-angle polarimetric airborne platform, the research scanning polarimeter (RSP), during the ORACLES campaign over the Southeast Atlantic Ocean. Our approach transforms the 1D measurements over multiple angles and wavelengths into a 3D image-like input, which is then processed using various deep learning (DL) schemes to yield aerosol single scattering albedos (SSAs), aerosol optical depths (AODs), aerosol effective radii, and aerosol complex refractive indices, together with cloud optical depths (CODs), cloud effective radii and variances. We present a comparison between the different DL approaches, as well as their comparison to existing algorithms. We discover that the Vision Transformer (ViT) scheme, traditionally used by natural language models, is superior to the ResNet convolutional Neural-Network (CNN) approach. We show good validation statistics on synthetic and real airborne data and discuss paths forward for making this approach flexible and readily applicable over multiple platforms.

Marine stratocumulus (MSC) clouds are important to the climate as they cover vast areas of the ocean’s surface, greatly affecting radiation balance of the Earth. Satellite imagery shows that MSC clouds exhibit different morphologies of closed or open mesoscale cellular convection (MCC) but many limitations still exist in studying MCC dynamics. Here, we present a convolutional neural network algorithm to classify pixel-level closed and open MCC cloud types, trained by either visible or infrared channels from a geostationary SEVIRI satellite to allow, for the first time, their diurnal detection, with a 30 min. temporal resolution. Our probability of detection was 91% and 92% for closed and open MCC, respectively, which is in line with day-only detection schemes. We focused on the South-East Atlantic Ocean during months of biomass burning season, between 2016 and 2018. Our resulting MCC type area coverage, cloud effective radii, and cloud optical depth probability distributions over the research domain compare well with monthly and daily averages from MODIS. We further applied our algorithm on GOES-16 imagery over the South-East Pacific (SEP), another semi-permanent MCC domain, and were able to show good prediction skills, thereby representing the SEP diurnal cycle and the feasibility of our method to be applied globally on different satellite platforms.

Aerosol heating due to shortwave absorption has implications for local atmospheric stability and regional dynamics. The derivation of heating rate profiles from space-based observations is challenging because it requires the vertical profile of relevant properties such as the aerosol extinction coefficient and single-scattering albedo (SSA). In the southeastern Atlantic, this challenge is amplified by the presence of stratocumulus clouds below the biomass burning plume advected from Africa, since the cloud properties affect the magnitude of the aerosol heating aloft, which may in turn lead to changes in the cloud properties and life cycle. The combination of spaceborne lidar data with passive imagers shows promise for future derivations of heating rate profiles and curtains, but new algorithms require careful testing with data from aircraft experiments where measurements of radiation, aerosol, and cloud parameters are better colocated and readily available. In this study, we derive heating rate profiles and vertical cross sections (curtains) from aircraft measurements during the NASA ObseRvations of Aerosols above CLouds and their intEractionS (ORACLES) project in the southeastern Atlantic. Spectrally resolved irradiance measurements and the derived column absorption allow for the separation of total heating rates into aerosol and gas (primarily water vapor) absorption. The nine cases we analyzed capture some of the co-variability of heating rate profiles and their primary drivers, leading to the development of a new concept: the heating rate efficiency (HRE; the heating rate per unit aerosol extinction). HRE, which accounts for the overall aerosol loading as well as vertical distribution of the aerosol layer, varies little with altitude as opposed to the standard heating rate. The large case-to-case variability for ORACLES is significantly reduced after converting from heating rate to HRE, allowing us to quantify its dependence on SSA, cloud albedo, and solar zenith angle.

Determining the direct aerosol radiative effect (DARE) of absorbing aerosols above clouds from satellite observations alone is a challenging task, in part because the radiative signal of the aerosol layer is not easily untangled from that of the clouds below. In this study, we use aircraft measurements from the NASA ObseRvations of CLouds above Aerosols and their intEractionS (ORACLES) project in the southeastern Atlantic to derive it with as few assumptions as possible. This is accomplished by using spectral irradiance measurements (Solar Spectral Flux Radiometer, SSFR) and aerosol optical depth (AOD) retrievals (Spectrometer for Sky-Scanning, Sun-Tracking Atmospheric Research, 4STAR) during vertical profiles (spirals) that minimize the albedo variability of the underlying cloud field – thus isolating aerosol radiative effects from those of the cloud field below. For two representative cases, we retrieve spectral aerosol single scattering albedo (SSA) and the asymmetry parameter (g) from these profile measurements and calculate DARE given the albedo range measured by SSFR on horizontal legs above clouds. For mid-visible wavelengths, we find SSA values from 0.80 to 0.85 and a significant spectral dependence of g. As the cloud albedo increases, the aerosol increasingly warms the column. The transition from a cooling to a warming top-of-aerosol radiative effect occurs at an albedo value (critical albedo) just above 0.2 in the mid-visible wavelength range. In a companion paper, we use the techniques introduced here to generalize our findings to all 2016 and 2017 measurements and parameterize aerosol radiative effects.

In this paper, we use observations from the NASA ORACLES (ObseRvations of CLouds above Aerosols and their intEractionS) aircraft campaign to develop a framework by way of two parameterizations that establishes regionally representative relationships between aerosol-cloud properties and their radiative effects. These relationships rely on new spectral aerosol property retrievals of the single scattering albedo (SSA) and asymmetry parameter (ASY). The retrievals capture the natural variability of the study region as sampled, and both were found to be fairly narrowly constrained (SSA: 0.83 ± 0.03 in the mid-visible, 532 nm; ASY: 0.54 ± 0.06 at 532 nm). The spectral retrievals are well suited for calculating the direct aerosol radiative effect (DARE) since SSA and ASY are tied directly to the irradiance measured in the presence of aerosols – one of the inputs to the spectral DARE. The framework allows for entire campaigns to be generalized into a set of parameterizations. For a range of solar zenith angles, it links the broadband DARE to the mid-visible aerosol optical depth (AOD) and the albedo (α) of the underlying scene (either clouds or clear sky) by way of the first parameterization: P(AOD, α). For ORACLES, the majority of the case-to-case variability of the broadband DARE is attributable to the dependence on the two driving parameters of P(AOD, α). A second, extended, parameterization PX(AOD, α, SSA) explains even more of the case-to-case variability by introducing the mid-visible SSA as a third parameter. These parameterizations establish a direct link from two or three mid-visible (narrowband) parameters to the broadband DARE, implicitly accounting for the underlying spectral dependencies of its drivers. They circumvent some of the assumptions when calculating DARE from satellite products or in a modeling context. For example, the DARE dependence on aerosol microphysical properties is not explicit in P or PX because the asymmetry parameter varies too little from case to case to translate into appreciable DARE variability. While these particular DARE parameterizations only represent the ORACLES data, they raise the prospect of generalizing the framework to other regions.

In most parts of the electromagnetic spectrum, solar radiation cannot penetrate clouds. Therefore, cloud detection and masking are essential in image preprocessing for observing the Earth and analyzing its properties. Because clouds vary in size, shape, and structure, an accurate algorithm is required for removing them from the area of interest. This task is usually more challenging over bright surfaces such as exposed sunny deserts or snow than over water bodies or vegetated surfaces. The overarching goal of the current study is to explore and compare the performance of three Convolutional Neural Network architectures (U-Net, SegNet, and DeepLab) for detecting clouds in the VENμS satellite images. To fulfil this goal, three VENμS tiles in Israel were selected. The tiles represent different land-use and cover categories, including vegetated, urban, agricultural, and arid areas, as well as water bodies, with a special focus on bright desert surfaces. Additionally, the study examines the effect of various channel inputs, exploring possibilities of broader usage of these architectures for different data sources. It was found that among the tested architectures, U-Net performs the best in most settings. Its results on a simple RGB-based dataset indicate its potential value for any satellite system screening, at least in the visible spectrum. It is concluded that all of the tested architectures outperform the current VENμS cloud-masking algorithm by lowering the false positive detection ratio by tens of percents, and should be considered an alternative by any user dealing with cloud-corrupted scenes.

The total effect of aerosols, both directly and on cloud properties, remains the biggest source of uncertainty in anthropogenic radiative forcing on the climate. Correct characterization of intensive aerosol optical properties, particularly in conditions where absorbing aerosol is present, is a crucial factor in quantifying these effects. The southeast Atlantic Ocean (SEA), with seasonal biomass burning smoke plumes overlying and mixing with a persistent stratocumulus cloud deck, offers an excellent natural laboratory to make the observations necessary to understand the complexities of aerosol–cloud–radiation interactions. The first field deployment of the NASA ORACLES (ObseRvations of Aerosols above CLouds and their intEractionS) campaign was conducted in September of 2016 out of Walvis Bay, Namibia. Data collected during ORACLES-2016 are used to derive aerosol properties from an unprecedented number of simultaneous measurement techniques over this region. Here, we present results from six of the eight independent instruments or instrument combinations, all applied to measure or retrieve aerosol absorption and single-scattering albedo. Most but not all of the biomass burning aerosol was located in the free troposphere, in relative humidities typically ranging up to 60 %. We present the single-scattering albedo (SSA), absorbing and total aerosol optical depth (AAOD and AOD), and absorption, scattering, and extinction Ångström exponents (AAE, SAE, and EAE, respectively) for specific case studies looking at near-coincident and near-colocated measurements from multiple instruments, and SSAs for the broader campaign average over the month-long deployment. For the case studies, we find that SSA agrees within the measurement uncertainties between multiple instruments, though, over all cases, there is no strong correlation between values reported by one instrument and another. We also find that agreement between the instruments is more robust at higher aerosol loading (AOD400>0.4). The campaign-wide average and range shows differences in the values measured by each instrument. We find the ORACLES-2016 campaign-average SSA at 500 nm (SSA500) to be between 0.85 and 0.88, depending on the instrument considered (4STAR, AirMSPI, or in situ measurements), with the interquartile ranges for all instruments between 0.83 and 0.89. This is consistent with previous September values reported over the region (between 0.84 and 0.90 for SSA at 550nm). The results suggest that the differences observed in the campaign-average values may be dominated by instrument-specific spatial sampling differences and the natural physical variability in aerosol conditions over the SEA, rather than fundamental methodological differences.

Biomass burning emissions often contain brown carbon (BrC), which represents a large family of light-absorbing organics that are chemically complex, thus making it difficult to estimate their absorption of incoming solar radiation, resulting in large uncertainties in the estimation of the global direct radiative effect of aerosols. Here we investigate the contribution of BrC to the total light absorption of biomass burning aerosols over the South-East Atlantic Ocean with different optical models, utilizing a suite of airborne measurements from the ORACLES 2018 campaign. An effective refractive index of black carbon (BC), meBC=1.95+ikeBC, that characterizes the absorptivity of all absorbing components at 660 nm wavelength was introduced to facilitate the attribution of absorption at shorter wavelengths, i.e. 470 nm. Most values of the imaginary part of the effective refractive index, keBC, were larger than those commonly used for BC from biomass burning emissions, suggesting contributions from absorbers besides BC at 660 nm. The TEM-EDX single-particle analysis further suggests that these long-wavelength absorbers might include iron oxides, as iron is found to be present only when large values of keBC are derived. Using this effective BC refractive index, we find that the contribution of BrC to the total absorption at 470 nm (RBrC,470) ranges from ∼8 %–22 %, with the organic aerosol mass absorption coefficient (MACOA,470) at this wavelength ranging from 0.30±0.27 to 0.68±0.08 m2 g−1. The core–shell model yielded much higher estimates of MACOA,470 and RBrC,470 than homogeneous mixing models, underscoring the importance of model treatment. Absorption attribution using the Bruggeman mixing Mie model suggests a minor BrC contribution of 4 % at 530 nm, while its removal would triple the BrC contribution to the total absorption at 470 nm obtained using the AAE (absorption Ångström exponent) attribution method. Thus, it is recommended that the application of any optical properties-based attribution method use absorption coefficients at the longest possible wavelength to minimize the influence of BrC and to account for potential contributions from other absorbing materials.

Southern Africa produces a third of global biomass burning emissions, which have a long atmospheric lifetime and influence regional radiation balance and climate. Here, we use airmass trajectories to link different aircraft observations to investigate the evolution of biomass-burning aerosols during their westward transport from Southern Africa over the south-eastern Atlantic, where a semi-permanent stratocumulus cloud deck is located. Our results show secondary organic aerosol formation during the initial 3 days of transport, followed by decreases in organic aerosol via photolysis before reaching equilibrium. Aerosol absorption wavelength dependency decreases with ageing, due to an increase in particle size and photochemical bleaching of brown carbon. Cloud processing, including aqueous-phase reaction and scavenging, contributes to the oxidation of organic aerosols, while it strongly reduces large diameter particles and single-scattering albedo of biomass burning aerosols. Together, these processes resulted in a marine boundary layer with fewer yet more oxidized and absorbing aerosols.

NASA NeMO-Net, The Neural Multimodal Observation and Training Network for global coral reef assessment, is a convolutional neural network (CNN) that generates benthic habitat maps of coral reefs and other shallow marine ecosystems. To segment and classify imagery accurately, CNNs require curated training datasets of considerable volume and accuracy. Here, we present a citizen science approach to create these training datasets through a novel 3D classification game for mobile and desktop devices. Leveraging citizen science, the NeMO-Net video game generates high-resolution 3D benthic habitat labels at the subcentimeter to meter scales. The video game trains users to accurately identify benthic categories and semantically segment 3D scenes captured using NASA airborne fluid lensing, the first remote sensing technology capable of mitigating ocean wave distortions, as well as in situ 3D photogrammetry and 2D satellite remote sensing. An active learning framework is used in the game to allow users to rate and edit other user classifications, dynamically improving segmentation accuracy. Refined and aggregated data labels from the game are used to train NeMO-Net’s supercomputer-based CNN to autonomously map shallow marine systems and augment satellite habitat mapping accuracy in these regions. We share the NeMO-Net game approach to user training and retention, outline the 3D labeling technique developed to accurately label complex coral reef imagery, and present preliminary results from over 70,000 user classifications. To overcome the inherent variability of citizen science, we analyze criteria and metrics for evaluating and filtering user data. Finally, we examine how future citizen science and machine learning approaches might benefit from label training in 3D space using an active learning framework. Within 7 months of launch, NeMO-Net has reached over 300 million people globally and directly engaged communities in coral reef mapping and conservation through ongoing scientific field campaigns, uninhibited by geography, language, or physical ability. As more user data are fed into NeMO-Net’s CNN, it will produce the first shallow-marine habitat mapping products trained on 3D subcm-scale label data and merged with m-scale satellite data that could be applied globally when data sets are available.