This PDF file contains the front matter associated with SPIE Proceedings Volume 11859, including the Title Page, Copyright information, and Table of Contents.

For many years, in-person SPIE Conferences stimulated active and fruitful discussions regarding the remote sensing, its impressive applications and future directions. This year, digital Conference takes place. Conference Chairs and SPIE Organizing Committee welcome all participates and their valuable contributions. These contributions will be highlighted by invited and contributed presentations during two live-stream sessions arranged on Monday and Tuesday. We encourage all speakers to give condensed talks and reserve time for short discussions. It is expected that total duration of talks and subsequent discussions will not exceed 15 minutes for invited presentations and 10 minutes for contributed presentations. Participants can virtually raise their hands to ask questions or submit them in Q and A box. Several factors, such as different time zones and technical issues associated with unacceptable connections, can represent challenges for the arranged live-stream sessions. Well, let us do our best to manage these sessions smoothly and include networking and live elements to our online meeting.

An analysis of the air quality over the Po valley has been performed by using both satellite and in situ observations of NO₂ for the COVID-19 years, 2019-2021. To match satellite observations to those in situ, we have used a geostatistical re-gridding technique. The tools allow us to scale the satellite NO₂ retrievals to a finer spatial resolution, which helps us to perform a better spatial colocation with in situ observations. The satellite data consist of Level 2 (L2) NO₂ retrievals from TROPOMI (the TROPOspheric Monitoring Instrument), whereas in situ observtaions are taken at eleven diverse stations, which are spread over the Po valley. The Po Valley, in the winter 2019/20, has been the first region in Europe to be severely hit by the COVID-19 pandemic. The Italian government introduced severe restriction measures from March to May 2020 (lockdown). We compared TROPOMI NO₂ concentration during winters 2018-19 (no-COVID-19) and the following 2 winters. The observations of TROPOMI, in agreement with the in-situ measurements, saw a significant decrease in the NO₂ concentration in March 2020 after the introduction of the lockdown. But they also found a general decrease in lower tropospheric NO₂ in winter 2019/2020, the warmest winter ever observed that has limited the use of power for residential and commercial heating. NO₂ concentrations raise almost to the pre-COVID-19 values in the 2020/21 winter.

Degrees of freedom or d.o.f. of satellite-based retrievals characterize their independence from the constraints assumed in the inversion process. In the context of Optimal Estimation (OE), the condition is expressed in terms of the background state, which, in a Bayesian meaning, is our best prior knowledge about the parameters we want to estimate. In effect, even if the background is static, it could add artifacts to the retrievals, which modify the seasonal cycle or the spatial patterns of 2-D fields. The issue has been addressed with an analytical treatment based on the OE theory. We derive formulas, which allow us to assess the modulation introduced by d.o.f. variability. The methodology will be exemplified with the help of observations from the Infrared Atmospheric Sounder Interferometer (IASI) onboard the European MetOp satellites. Both time series and 2-D fields of observations will be considered. The analysis is extended to tropical and Mid-Lat regions to exemplify the effect of seasonal variability of d.o.f. The analysis will focus mainly on OCS (carbonyl sulfide) variability in the atmosphere, a new clue to how much carbon plants take up, hence of primary interest to the carbon cycle and the climate. However, our methodology can be applied to any gas or retrieved parameter. For the OCS, we have found that d.o.f. variability is of no concern in the tropics. Still, it becomes crucial at Mid-latitudes where the seasonal cycle can add spurious variability to temporal and spatial patterns.

Experiments for the retrieval of the high-detailed spatial NO2 distribution in the troposphere using measurements of the GSA instrument onboard satellites of Resurs-P series were performed in 2016-2017. The authors developed an algorithm to obtain the tropospheric NO2 2D distribution with the horizontal spatial resolution reaching 2,4 km for the first time at the world level and provided on a grid with a step of 120 m. The high spatial resolution of the NO2 space measurements allowed the identification of local sources of NO2 pollution and their plumes from space observations. To validate the fine structures detected in the NO2 fields of GSA/Resurs-P, we perform comparisons with chemical transport models. The paper presents preliminary results of a comparison with a new model which is based on a numerical-asymptotic approach. The comparison was performed for NO2 observations on September 29, 2016 over Hebei province, the North China Plain. We propose, in particular, a new efficient approach using this model to obtain estimates of emissions from local anthropogenic sources based on GSA/Resurs-P observational data. To validate the coarse structures in the GSA/Resurs-P NO2 field, in this paper, we perform comparisons of our data based on spectral imagery of Tokyo region, Japan, taken in March-April 2017 with observations of OMI/Aura and TROPOMI/Sentinel-5P. The comparison confirmed the reliability of the GSA NO2 fields in general.

Recently, there has been a significant increase in the anthropogenic impact on the environment, including on the atmosphere. Therefore, it is very important to understand the mechanisms of transport of pollutants and to have reliable estimates of the impact of various factors on the transport of atmospheric impurities. Ground-based measuring stations allow local continuous observations, characterized by high accuracy. The main disadvantage of ground-based measurements is the low density of measuring stations, which does not allow reproducing the concentration fields of pollutants. Remote methods include, in particular, satellite observations, the main advantage of which is the ability to cover a large area, but, as a rule, they have rather low spatial resolution. In contrast, this work utilizes new satellite technology providing data with high space resolution. However, for a more detailed description, it is necessary to supplement the measurement data with mathematical modeling of various degrees of complexity. This work is devoted to the construction of a model of NO_x transport from local ground sources with high spatial resolution which take into account chemical transformations. To achieve a high spatial resolution, the model uses a numerical solution of a system of three-dimensional reaction-diffusion-advection equations that takes into account the kinetic equations describing chemical reactions. Information on wind speed, temperature and pressure fields are obtained using the HYSPLIT model. The turbulent transport is described using a first-order closure model, where the turbulent diffusion coefficient parameterization is based on data on the friction velocity and the boundary layer height. Validation of the model was carried out by comparing the results of calculations with high-detailed spatial NO2 distributions obtained using measurements of the GSA instrument onboard the Resurs-P satellite.

The eruption begun after an intensive earthquake activity in the volcanic system of Reykjanes that opened up several kilometer long underground dike crossing the plate boundary of Reykjanes at an angle of around 22°. Quakes of varying intensity caused material damage in the township of Grindavik, the major population center in the neighborhood. The eruption came on March 20th, with put any clear warning, a magma eruption without much ash formation but when reaching the surface, the magma released gasses in a magnitude similar to other volcanic eruption of this type, e.g., Holuhraun (magma from Bardarbunga) 2014 and Surtsey 1963 - 1967. A characteristic SO2 emission in this eruption was measured 6kg/sec of SO2 from each m3 of magma or 2 o/ o o. This is similar to what was observed in the Holuhraun airborne observation campaign and corresponds very well to the estimates for Surtsey. The composition of the volcanic gas is similar too, the main constituent is water, often 90 - 55% of the total gas flow. The magma is 1200 - 1300 °C hot and comes from a very deep source about 20 km down. The possibility exists that the eruption goes on for a long time, widens the conduit and increases in output. An airborne measurement campaign was conducted in a light airplane, TF-VTR, by Dr. Gylfi Árnason, Reykjavik University (RU) with a mobile remote sensing DOAS instrumentation specially adapted for use in this airplane from the Duesseldorf University of Applied Sciences (HSD), Germany. This observation technology has been used with good results during volcanic events in Europe, Japan and America. Four sorties were carried out, measuring the column load of SO2 by flying under the plume in several traverses, each giving about 20 measurements of the SO2 column load. The results are compared to other measurement results from IMO (Icelandic Meteorological Institute) and UI (University of Iceland) and results from previous campaigns 2014 and 1963 - 67 and found similar. In the beginning the eruption output was steady at 5 m3/sec but was increasing in output magnitude and pulsating, making gas flux estimations more difficult. A steady plume in a steady wind follows the dispersion model developed by the authors, but the pulsating plume creates large puffs with high gas concentrations and increased hazards for nearby populations centers. Gas accumulation in a large clouds during calm weather, observed during the 2014 Holuhraun event, does also happen here and increases the risk of serious pollution events. This seriously hampers the possibility of using modeling results only to estimate gas pollution risks, and stresses the need for monitoring the gas flow by airborne measurements of the propagation of the plumes, puffs and accumulated clouds that may threaten the neighborhood.

In this paper, we describe an effort to build a new deep edge detection method designed to detect weather-related phenomena such as clouds and planetary boundary layer heights present in backscatter profile imagery. This method builds on the existing deep model called Holistically-Defined Edge Detection (HED), which was shown to perform better than other information theory and convolutional network techniques for edge detection. Though HED outperforms techniques such as Canny Edge detection, HED’s performance is based on it being trained on natural images with very little noise. Weather-related backscatter profiles, such as those generated from LIDAR-based ceilometers, often contain noise. In addition, there is often less of a difference in the pixel density between edges and non-edges, and due to atmospheric dynamics, continuous edges are not always detected in the images. Under these conditions when using HED, subtle but useful edges are lost from side outputs during the fusing process while the network is being trained. Canny Edge detection also does not perform well under these conditions, as it determines edges based on the differences in pixel density. We describe a new edge detection deep network developed specifically for overcoming these issues by applying physics-aware attention mechanisms to the side outputs of the HED learning process. We show how this method is able to learn the subtle edges as opposed to HED or Canny, when used to identify planetary boundary layer heights which involves distinguishing the mixing layer, residual layer, and nocturnal layer in addition to the cloud heights for ceilometerbased backscatter. Though the intent of this network is to learn planetary boundary layer heights and cloud heights, this method could be applied to other weather-related phenomena and applied to backscatter imagery generated from other sources such as satellites.

Clouds and the Earth’s Radiant Energy System (CERES) instruments are scanning radiometers on board the Terra, Aqua, NPP and NOAA20 satellites that use thermistor bolometers as detectors, with a sampling rate of 0.01s. During calibration testing a slow mode of detectors was found which had a magnitude of about three percent of the signal, and a characteristic time of 0.3 sec. To reduce the effect of this mode, a numerical filter was introduced. However, an analysis of the data from the CERES instruments aboard the Terra and Aqua satellites showed a difference of radiances in the forward scan compared to radiances from the backward scan. This difference was due to the fact that parameters of the digital filtered derived based on calibration data were not characterized correctly. A focus of this work is to show results of setting slow mode filter parameters based on in-flight data. A gradient-based minimization strategy is employed to effectively find the filter parameters for each instrument to meet the performance goal, namely the divergence of 0.15% for viewing zenith angles <65 degrees. The newly designed numerical filters are used in processing the newest editions of CERES Level-0 data products, and then in the Erbe-like and SSFs.

Remote sensing from space has become indispensable for climate change monitoring. In this context, we are defining a space mission with the intention to monitor the Earth’s radiation budget and to measure all of its components. In this paper, we focus on how we intend to measure the Outgoing Longwave Radiation (OLR), which is the thermal infrared flux that is reemitted to space by Earth. For that purpose, we will use a wide field-of-view (FOV) thermal camera covering the spectral range from 8 to 14 μm, and featuring a FOV of 140°. This allows observing the Earth from limb to limb from a nominal altitude of 700 km, while accounting for altitude and pointing errors. In addition, we target to achieve high resolution (better than 5 km at nadir) to enable scene identification, while fitting the optics and the detector within 1 CubseSat Unit (1U ~ 1 dm³). Wide FOV thermal camera systems are not widely available, and no commercially available camera can fulfill our target specifications, indicating the need for the development of a custom imaging system design. This paper describes our thermal camera design, evaluates its performance, and discusses the stray-light and tolerancing analysis. Our design combines 3 Germanium lenses with a commercial-off-the-shelf uncooled microbolometer array. To limit the cost and to ease the fabrication, we traded off between the number of elements, the number of aspherical surfaces and the required performance. At all wavelengths and all fields, the optical design performs close to the diffraction-limit. Consequently, we are confident that the wide FOV thermal camera achieves adequate optical performance, featuring a nadir resolution of 4.455 km. As a result, we can safely state that our novel developed thermal camera design goes well beyond the state-of-the-art in the field of thermal cameras, while being specifically dedicated to monitoring Earth’s OLR, and thus improving climate change monitoring.

The NASA Clouds and the Earth’s Radiant Energy System (CERES) project has provided the climate community more than a 20-year record of top of the atmosphere (TOA) reflected shortwave (SW) and emitted longwave (LW) fluxes. The fluxes are used to monitor the Earth’s energy balance as well as for climate model validation and cloud feedback studies. One of the largest uncertainties in climate models is the response of clouds in feedback studies. Reducing this uncertainty requires a more stringent validation of model-generated fluxes by cloud-type. Rather than relying on radiative transfer model generated fluxes from observed cloud-type retrievals, the CERES FluxByCloudTyp (FBCT) product relies on MODIS empirical narrowband to broadband relationships to convert the cloudy portions of the CERES footprints into fluxes. The overcast footprint and sub-footprint cloud layer fluxes are then stratified by 7 pressure layers and 6 optical depth bins and temporally averaged into daily and monthly regional cloud-type fluxes. The CERES project analyzed a total of 19 MODIS channels for use in the next generation FBCT product narrowband to broadband conversion. This was accomplished by comparing the narrowband to broadband RMS errors from all possible 5-channel combinations. The SW and LW optimized channels were selected by also considering analogous geostationary, MODIS and VIIRS imager channels as well as channel combinations that provided the lowest RMS errors across all surface types. The optimized channel combinations will be tested in the FBCT algorithm to investigate any narrowband to broadband dependencies as a function of cloud fraction, effective pressure, optical depth, PW, solar and view angle, surface type.

The Umov effect is the inverse correlation between the maximum of the linear polarization of the light scattered on an object and the geometrically albedo of this object. The importance of studying this effect should be considered in the context of one of the complex and important tasks of remote sensing: determining the concentration of particles in optically thin clouds. Since the intensity of the scattered light depends on two unknown quantities: the concentration of particles in the cloud and the phase function of the particles of the cloud. Then to retrieve the concentration from the measured signal, it is necessary to know the phase function in advance. In real observations, the phase function is, usually, not known. The Umov effect will make it possible to estimate some necessary unknown characteristics of particles in a cloud, which determine the phase function. This paper is devoted to the study of this effect for particles with sizes much larger than the wavelength of the incident light. The report presents a solution to the problem of light scattering by randomly oriented particles of irregular shape for particles with sizes of 100 and 200 microns, for a wavelength of 0.532 microns. The solution was obtained within the both frameworks: the physical optics method and the geometric optics approximation. It was found that if the imaginary part of the refractive index less than 0.001, the Umov effect is observed with good accuracy for particles of size from 100 to 200 microns. However, the Umov effect is violated when the imaginary part of the refractive index is greater than 0.001.

Cloud microphysics in terms of their liquid/ice water content and particle size are the principal factors addressed to study and understand the behavior behind the climate change phenomenon. Based on remotely sensed measurements, in the last decades, some evidence exists that an increase in temperature leads to an increase in cloud liquid water content (CLWC). The temperature dependence of ice water content (CIWC) is also evident from measurements of midlatitude cirrus clouds. Hence, innovative methods, such as those based on the use of Artificial Intelligence (AI) allowing a more relevant investigation of how clouds influence the hydrological cycle and radiative components of the Earth's climate system, are required. This work investigates the capability of a statistical regression scheme of CLWC and CIWC, implemented through the use of a multilayer feed-forward neural network (NN). The whole methodology is applied to a set of simulated IASI-NG L1C and MWS acquisitions, covering the global scale. The NN regression analysis shows good agreement with the test data. The retrieved cloud liquid water and ice profiles have an accuracy of 20 to 60% depending on the given layer. Finally, the layer with the maximum concentration is accurately identified.

This paper presents the results of calculating and analyzing the light scattering matrix of aggregates of atmospheric hexagonal ice particles located in cirrus clouds. Two types of basic particle shapes for aggregates are considered: a hexagonal column and a hexagonal plate. For both forms, two types of particle arrangement in aggregates were chosen: compact and non-compact. As a result, 4 sets of aggregates were built: compact hexagonal columns, non-compact hexagonal columns, compact hexagonal plates, and non-compact hexagonal plates. Each set consists of 9 aggregates differing in the number of particles in them, and the particles in each individual aggregate have the same shape and size, but different spatial orientation. The light scattering matrices for all aggregates were calculated for the case of arbitrary orientation in the geometric optics approximation. Dependences of the first element of the matrix on the number of particles in aggregate, with different types of particle arrangement, and for two types of shapes are given.

Mixed-phase clouds involve complex processes. The characterization of these processes requires novel measurement techniques. One approach is to utilize polarimetric Doppler cloud radars, since their measured variables are sensitive to specific fall velocities and shapes of hydrometeors. Polarimetric retrievals for particle shape and orientation detection already exist, but provide so far only information about the main particle population. In state-of-the art polarimetric approaches, elevation (range-height indicator, RHI) scans of observed differential reflectivity ZDR and correlation coefficient RHV of the main particle population are compared with modeled angular dependencies of differential reflectivity and correlation coefficient. The retrieval output is a pair of the polarizability ratio and the degree of orientation corresponding to the best agreement between measured and modeled angular dependences of the polarimetric variables. It is aim of the herein study to extend an existing main-peak retrieval. Basis of the approach is, that the Doppler spectrum is split into five parts. The retrieval is then applied separately on each part. In the frame of this paper the application of the shape and orientation approach to RHI scans (30° to 90° elevation) of spectrally resolved polarimetric Ka-band cloud radar observations of ZDR and RHV for all 5 parts of Doppler spectra are demonstrated and results of observed spectrally resolved shape distributions will be presented. Case studies will be shown which demonstrate well that ZDR and RHV behave differently for the different parts of the Doppler spectra. From our study we conclude that the extended approach enables one to track how ice particle shapes change during precipitation from cloud top to melting layer. By applying best-guess density estimates, the retrieved polarizability ratios will be converted to geometrical axis ratios. Interpretations of the pathway of such a hydrometeor evolution will be presented by considering in addition the presence of supercooled liquid droplets, which allows to distinguish between riming and aggregation processes.

Influence of aerosol particles in the cloud formation and microphysical properties, commonly known as aerosolcloud interaction (ACI), has emerged as a research field of special interest since the low-level knowledge on this topic causes the largest uncertainties in climate projections. Active remote sensing already provided promising results combining backscatter lidars and cloud radars to characterize atmospheric aerosol and cloud droplets. The retrieval of aerosol microphysical properties is of high interest for ACI studies. However, complex inversion techniques are usually required. Mamouri and Ansmann (2016) proposed to derive the so-called extinctionto- number-concentration factors to retrieve aerosol number concentration from aerosol extinction in a more straightforward way. These factors can be easily obtained from the relationship between the aerosol number concentration and the aerosol optical depth from AERONET measurements. However, experimental data are not always available. This study demonstrates that it is also possible to obtain the extinction-to-number-concentration factors using pre-calculated optical properties. To do so, MOPSMAP, open-source software based on a data set of pre-calculated single-particle optical properties, is used. Using a study case (9 July 2021), it is shown that number concentration of aerosol-particle with radius larger than 250 μm (n₂₅₀) derived with MOPSMAP is within the uncertainty range of the one derived with AERONET.

This work focuses on combining polarized Micro-Pulse Lidar (P-MPL) and AERONET measurements in the GRASP (Generalized Retrieval of Atmosphere and Surface Properties) code for the retrieval of optical and microphysical properties of aged smoke particles. In particular, a few smoke cases corresponding to an aged Canadian smoke plume observed at El Arenosillo/Huelva (Spain) during 7-8 September 2017 were selected. Both the GRASP-derived columnar and height-resolved optical and microphysical properties are compared with AERONET retrievals and vertical lidar-retrieved profiles, respectively. Linear regression analysis, mean fractional bias and relative differences between both datasets are the statistical proxies used for assessing the degree of agreement by comparing the vertical profiles. The inter-comparison analysis of the columnar-integrated properties (e.g. total volume concentration, effective radius, particle volume size distribution, single scattering albedo and complex refractive index) indicates that GRASP retrievals are consistent with those provided by AERONET for the smoke event examined in this study. By analysing the height - resolved properties (e.g. the total particle backscatter coefficient and the total volume concentration) , the degree of the agreement also shows high confidence between GRASP retrievals and P-MPL-derived variables, which were retrieved by using an alternative and well-validated polarization-based methodology.

Long-term records of aerosol optical depth (AOD) with high quality, suitable temporal continuity and spatial coverage are of immense interest to climate-related research activities. Both satellite- and ground-based measurements of AOD are typically provided by instruments with different designs, and distinct data acquisition and processing schemes. Thus, the corresponding AOD records likely have different accuracy, spatial coverage, and temporal resolution. Several studies have been focused on the synergy of multi-sensor satellite AOD products. Here we combine multi-year (1997-2018) AOD records available from four collocated ground-based instruments deployed at the mid-continental Southern Great Plains (SGP) Central Facility supported by the U.S. Department of Energy Atmospheric Radiation Measurement (ARM) Program. We demonstrate how to minimize drawbacks (patchy spots) and to maintain benefits (high quality) of these records. Our demonstration finds a combined AOD obtained at two wavelengths (500 and 870 nm), with high temporal resolution (1-min), and provides the user with an estimate of the AOD uncertainty. Finally, we highlight expected applications of the merged dataset and its future extensions.

In this paper, the authors aim to study a few meteorological elements, viz. rainfall, the freezing level height (HFL), bright band height (BBH), and the bright band intensity (BBI) at the subtropics, with special emphasis on the STHP. The study reveals a stunning finding. Unlike the low latitudes, the bright band height (BBH) lies above the freezing level height (HFL) at the subtropics. Besides, the study shows that as the latitude increases, the probability of BBH>HFL increases. The probability of BBH>HFL is the maximum at the STHP, in comparison to the lower latitudes. The investigation shows that the STHP is not the region of the least rainy belt in the 35N-35S region always. However, in most of the months, and most of the years, it shows very low rain. The study shows the daily, monthly, and yearly variations of rainfall, HFL, BBH, and BBI, and their correlations. The authors aim in particular, to find out if the maximum (minimum) rainfall in this region corresponds to the maximum (minimum) BBI and BBH. The HFL, BBH, and BBI for the period 1999-2002, and 2007-2008 have been obtained from the data product 2A23 of the precipitation radar onboard the Tropical Rainfall Measuring Mission (TRMM) satellite. The rainfall for the period 2004-2008 has been obtained from the data product 2A12 of the Tropical Microwave Imager (TMI) onboard the TRMM. Besides, the authors investigated the surface temperature, surface pressure, relative humidity, and precipitable water over a few locations in the subtropics. The study shows the interrelations of rainfall with surface elements. These surface elements have been obtained from the Integrated Global Radiosonde Archive (IGRA). The study also investigates the impact of the El Nino/La Nina on HFL.

At the Siberian lidar station, long-term measurements of the ozone vertical distribution are continued at the sensing wavelength pairs of 299 and 341 nm, 308 and 353 nm. The report presents a quasi-three-year seasonal model of vertical ozone profiles formed in recent years, obtained using the differential absorption lidar complex of the Siberian lidar station and the Aura, MetOp satellites in the upper troposphere - stratosphere. A typical seasonal feature of the vertical ozone distribution in Western Siberia is presented. The ozone profiles were retrieved using the vertical temperature distribution from the meteorological satellite data. We analyzed how the existing sets of absorption cross sections influence the deviation of the ozone profiles, retrieved with their application, from the Krueger model and the quasi-three-year model. Since 2021, measurements have been performed at the lidar station with different spatial resolutions from 10 m to 100 m. An analysis and estimation of how different spatial resolutions influence the error of retrieving ozone profile from lidar and satellite measurements in 2021 was conducted.

In order to investigate atmospheric aerosol properties in the mountainous regions of Japan, aerosol observations by the NASA/AERONET field campaign were conducted around Nagano prefecture in Japan since March 2020. This field campaign is called DRAGON/J-ALPS because the target area includes the mountains known as the Japanese Alps. One of the objectives of DRAGON/J-ALPS is to understand the spatial distribution of aerosol properties in mountainous regions. The aerosol concentration levels in the J-ALPS sites are usually not too high. Possible reasons are low local emissions and aerosol advection blocked by the high mountains. However, there are days when the concentration is higher than usual. One of the reasons for this is the advection of “yellow dust” from the continent. Another factor is the local emissions. The shape of the basin surrounded by mountains and meteorological conditions may also promote the retention of aerosols. This work is expected not only to introduce the results of the J-ALPS, but also to provide insights into aerosol advection and local emission in closed mountain areas.

In this work, characterization of biomass burning aerosols generated by large-scale wild fire events is dealt with the advantages of multi-channel measurements including near-UV and/or significance of simultaneously loading polarized and non-polarized bands of GCOM-C/SGLI. At first, advantage of near–UV bands by SGLI is shown. The simple color composite images with the three primary colors shifted to shorter wavelengths as (R, G, B): (443, 412, 380 nm) than usual clearly demonstrate the smoke behavior caused by wild fires. Next the index AAI, which is defined as the ratio of the satellite observing reflectance R at two bands of 412 and 380 nm, indicates the presence of biomass burning aerosols (BBAs). Then the mutual use of radiance and polarization is effective in radiative transfer simulations for retrieval of severe BBAs. The obtained results seem to suggest the difficult task of simultaneous analysis of aerosols and clouds in a hazy scene.

This study is about the development of a deep neural network to make very short-term predictions of torrential rains at the urban scale (meso-γ). The new polarimetric Phased Array Weather Radar (MP-PAWR) operating at Saitama (Japan) since 2018 is used. Thanks to the unique spatio-temporal resolution of the measurements, the precursors of torrential rains are detected aloft more than 20 minutes before the rain occurs. With this information, we aim at the prediction of surface precipitation with a lead time of 20 min, a horizontal resolution better than 500 m within a radius of 25 km around the instrument. Two supervised neural networks are considered to extrapolate radar reflectivity (Z_H) at the altitude of 600 m. The first model (model-1) is based on a technique developed for mesoscale predictions from observations at a single altitude. It uses horizontal (2D) convolutions in gated recurrent time iterations and a multilayer encoder-decoder (EC/DC) architecture. The technique is adapted to consider 3 radar parameters and 11 altitudes up to 10 km, in the same way as RBG channels in video analysis. The second model (model-2) uses similar architecture but with 3D spatial convolutions to properly describe the vertical motions between adjacent layers. The input to the models consist in 20 min long time series of ZH, Doppler velocity and differential reflectivity observations (30 sec sampling). The models are trained using all the rain events observed between August and October 2018, and are assessed using local heavy rains observed over a period of 1-hour on July, 24, 2018. The beginning of the rain is first predicted with a lead time of about 5 min, and its evolution is fairly well reproduced to lead times up to about 10 min. Results quickly degrades for longer lead times. We found that a deeper network with 4 layers EC/DC gives better 20 min predictions than a model with 3 layers, but final results were not yet obtained at the time of writing. Regarding lead-times of 10 min, model-2 gives critical success indexes (CSIs) of 0.60 and 0.40 for pixels with Z_H> 10 dBZ and 37 dBZ, which is comparable or better than results presented in other studies. For lead-times of 20 min, CSIs dropped to 0.28 and 0.10, respectively, and no other studies was found for comparison. Model-1 clearly shows poorer performance, especially for high Z_H. However, this approach demands much less calculations and the training lasts only 2 weeks long, namely half of the time spent for model-2. Therefore, it is worth further studying both approaches and potential improvements are discussed.

All sky-cameras are devices with a very high potential in order to study atmospheric phenomena and were originally designed to obtain the cloud cover. However, methods based in different approaches produce significant differences in the results. State-of-art methods usually offer better performance, thanks to computer vision and machine learning (ML) techniques, than traditional algorithms based on channel ratios using both fixed and adaptive thresholds to classify the pixels of one image as cloud or cloud free. We have developed a cloud cover adaptive threshold algorithm base on Probability Density Function (PDF) of the Blue to Red Ratio (BRR), standing out in: simplicity; ease of implementation; compatibility with any sky-camera in terms of technical requirement and type of image acquisition. The goal of this study is to compare our algorithm with a most fashionable method based on Machine learning, discovering the pros and cons of each one and weather ultimately less can be more. The comparison has been done using a set of 1-year HDR imagery database, representing a wider range of atmospheric scenarios such as clear sky, cloudy, partly cloudy and different types of aerosol conditions and clouds such as cirrus, cumulus, stratum and nimbus. To stablish a quantitative comparison of both methods, a limited set of images has been chosen. The PDF method show better agreement than our ML implementation, with a better performance for all weather conditions, in comparison against our cloud cover database.

A commercial all-sky-camera is employed to derive a whole-sky product of Cloud Optical Depth. The methodology consists in a radiative closure combining measurements of the blue and red channels with libRadtran 1D monochromatic radiance simulations. Besides, a matrix of data quality Flags is obtained for every COD image. The data quality Flags indicate the reliability of the retrieval at each pixel, and gives information about the method to solve the radiance monochromatic ambivalence. In addition, the Flags product also indicates the presence of out-of-range radiances with respect to the RT simulations. Such out-of-range radiances are related to the neglection of horizontal radiation transport in the 1D plane-parallel approach. A set of around 2000 images during 2020 have been analyzed and COD has been obtained for each pixel. The COD shows values ranging from 0 to 130, with around 83% of the cases between 5 and 30. Our COD results have been validated using the zenith COD retrieval from AERONET. Our results present a very good agreement with AERONET cloud mode retrieval and show a correlation factor of 0.94 and a slope of 0.99 respectively.

The radiative closure methodologies to obtain Cloud Optical Depth (COD) from Remote Sensing techniques have traditionally relied on one-dimensional (1D) assumptions. These assumptions might be far away from the radiation transport over a realistic three-dimensional (3D) atmosphere, especially in cloudy conditions, as the natural inhomogeneities of clouds are not conveniently represented and treated in 1D models. The differences between the 1D and 3D approaches manifests in the 3D effects: a) the plane-parallel albedo bias and, b) the horizontal transport effect. The plane-parallel albedo bias is usually addressed by means of the Independent Pixel approximation (IPA), that considers each pixel radiatively independent from the others. Nevertheless, the IPA neglects the horizontal transport, entailing bias in the retrievals. In this work, we use the advantages of 3D radiative transfer (RT) to analyze COD and parameterize the 3D biases in terms of the plane-parallel approach. Detailed 3D RT simulations using MYSTIC are performed over two Highly Resolved Large Eddy Simulations cloud fields of known optical thickness. The output radiance is analyzed by a 1D IPA inversion retrieval based on a radiative closure to obtain the COD. The comparison between the retrieved COD fields for diverse illumination conditions and the real COD allow us to study the 3D effects separately and evaluate the retrieval. Our results show radiation enhancement in cloud edges depending on solar, viewing and cloud geometries, that induces a COD underestimation. The 1D approach works well for overcast conditions and underestimates the COD in broken clouds scenarios.

This paper presents the results of calculation and analyzes the light scattering matrix of random oriented ice particles of non-convex shape (hollow column) with cavity angles from 0 to 50 degrees for lidar wavelengths of 0.355, 0.532, and 1.064 microns and refractive indices of 1.3249, 1.3116, and 1.3004. The calculation was carried out within both physical and geometrical optics approximation methods for particle sizes varied from 10 to 100 microns. As a result, it is shown that differential scattering cross-section for non-convex shape (hollow column) demonstrates a power-law dependence on the particle size. However, the linear depolarization ratio has no simple dependence on particle size and is practically independent of wavelength for small particles (L<50μm). The linear depolarization ratio increases from 0.2 up to 0.5–0.8 with an increase of the cavity angle of the crystal. The elements of the light scattering matrix depending on scattering and cavity angle are given.

The aim of this study is to predict the main aerosol properties in the atmosphere, Aerosol Optical Depth (AOD) and Angstrom Exponent (AE), with the aid of machine learning techniques and images from an All-Sky camera. Two different machine learning techniques have been used in this work: a random forest (RF) and an artificial neural network (ANN) with target values furnished by AERONET database. HDR images from the All-Sky camera sited in Burjassot (Spain) have been used. All of them have been taken in a clear-sky condition (without clouds) and with different aerosol depth. Selected images come out with a range from 0 to 0.5 of AOD at 500 nm as reference. The data in the groundbased station are available since the 10th of February of 2020 to the 31th March of 2021 in almost one year of samples. We have developed two ways of building signals combined with the two machine learning methods. Firstly, a signal generated from scattering angles in a single image which is obtained as the average of relative irradiance (RGB) using 100 random points in each scattering angle isoline, obtaining 29 values for each signal. Secondly, the signal has been generated in the same way but from zenith angles isolines of a single image. The main result obtained is that we improve significantly the state of art results of not calibrated images. For example, the red channel improves the percentage of predicted AOD values within the AERONET uncertainties from 62% to 90%-93% using an ANN and the zenith method.

The paper presents a solution to the problem of light scattering by hexagonal atmospheric plates and columns, as well as irregularly shaped particles with sizes from 10 to 100 microns. The solution is presented in the form of a databank of light backscattering matrices. The solution was obtained for typical wavelengths used in laser sensing problems: 0.355, 0.532, 1.064 μm; as well as for the wavelengths of the near infrared range: 1.55, 2 and 2.15 μm. At wavelengths of 0.532 and 1.064 μm, in addition to the refractive index of ice, the refractive index of the dust aerosol was used: 1.48+i•0.002 and 1.6+i•0.002, respectively. The solution was obtained within the framework of the physical optics method developed by the authors. Based on the calculated light backscattering matrices, the values of the color and linear depolarization ratios were obtained. It is shown that the power laws previously identified by the authors are violated in the presence of absorption, in particular, for hexagonal particles with sizes up to 100μm, with an imaginary part of the refractive index greater than i•0.0004, significant deviations from the power law are observed. For irregularly shaped particles at wavelengths for which there is no absorption, smooth power law dependences are seen.