Search

Due to their fast computational performance, neural networks (NNs) are nowadays commonly used in the context of remote sensing. The issue of performance is especially important in the context of big data and operational processing. Classical retrieval algorithms often use a radiative transfer model (RTM) as forward model with which an optimization algorithm can then solve the inverse problem of inferring the quantities of interest from the measured spectra. However, these RTMs are usually computationally very expensive and therefore replacing them by a NN is desirable to increase performance. But the application of NNs is not straightforward and there are at least two main approaches:1. NNs used as forward model, where a NN approximates the radiative transfer model and can thus replace it in the inversion algorithm2. NNs for solving the inverse problem, where a NN is trained to infer the atmospheric parameters from the measurement directlyThe first approach is more straightforward to apply. However, the inversion algorithm still faces many challenges, as the spectral fitting problem is generally ill-posed. Therefore, local minima are possible and the results often depend on the selection of the a-priori values for the retrieval parameters.For the second case, some of these issues can be avoided: no a-priori values are necessary, and as the training of the NN is performed globally, i.e. for many training samples at once, this approach is potentially less affected by local minima. However, due to the black-box nature of a NN, no indication about the quality of the results is available. In order to address this issue, novel methods like Bayesian neural networks (BNNs) or invertible neural networks (INNs) have been presented in recent years. This allows the characterization of the retrieved values by an estimate of uncertainty describing a range of values that are probable to produce the observed measurement. We apply and evaluate these new BNN and INN methods for the retrieval of cloud properties from TROPOMI in order to demonstrate their potential as operational algorithms for current (Sentinel-5P) and future (Sentinel-4 and Sentinel-5) Copernicus atmospheric composition missions.

The retrieval of atmospheric composition from space-based measurements, by e.g., Sentinel-5p TROPOMI, is strongly affected by radiative interferences with clouds. Dedicated cloud data products, typically retrieved from measurements by the same sounder, are therefore essential. Cloud information is used to filter data and as input to the modelling of atmospheric radiative transfer and the conversion of slant column densities into vertical column densities. The three main TROPOMI cloud retrieval algorithms are: (i) L2_CLOUD OCRA/ROCINN CAL (Optical Cloud Recognition Algorithm/Retrieval of Cloud Information using Neural Networks; Clouds-As-Layers), (ii) L2_CLOUD OCRA/ROCINN CRB (Clouds-as Reflecting Boundaries), and (iii) the S5P support product FRESCO-S (Fast Retrieval Scheme for Clouds from Oxygen absorption bands for Sentinel). The cloud variables provided by these products (radiometric cloud fraction, cloud (top) height, and cloud albedo/cloud optical thickness) are subsequently used in the retrieval of the TROPOMI trace gas products. The quality of cloud products and trace gas products is routinely assessed by the ESA/Copernicus Atmospheric Mission Performance Cluster (ATM-MPC) validation service, with ad hoc support from Sentinel-5p Validation Team (S5PVT) AO projects. Version upgrades have had a significant impact on the characteristics of S5P cloud data. The change of the wavelength window in the FRESCO product since version 1.4 (‘FRESCO-wide’) leads to a clear increase in the height of low clouds with a large impact on the tropospheric NO2 retrieval (van Geffen, 2022), and improving the validation results regarding the tropospheric and total NO2 column. The first upgrades of the ROCINN products (from v1 to v2.1-v2.3) led to an increase in correlation with CLOUDNET cloud height, but to a more negative bias for the low clouds, with ROCINN CRB cloud height even dropping below the CLOUDNET cloud base height on average. However, this effect seems alleviated with the latest upgrade to v2.4. The impact on the HCHO validation results is investigated but is less clear compared to the NO2 case. To resolve the discontinuities due to the processor version jumps, a full mission reprocessing is currently ongoing and largely carried out for the L2_CLOUD and FRESCO-S products. The reprocessed ROCINN data have a lower dispersion and higher correlation with respect to the CLOUDNET cloud heights. The bias of the L2_CLOUD OCRA/ROCINN CAL CTH becomes more negative, but that of L2_CLOUD OCRA/ROCINN CRB CH bias improves. Finally, we also discuss the impact of the FRESCO-S reprocessing on the validation results.

Clouds have a strong impact on satellite measurements of tropospheric trace gases in the ultraviolet, visible, and near-infrared spectral ranges from space. Therefore, trace gas retrievals rely on information on cloud fraction, cloud albedo, and cloud height from cloud products. In this study, the cloud parameters from different cloud retrieval algorithms for the Sentinel-5 Precursor (S5P) TROPOspheric Monitoring Instrument (TROPOMI) are compared: the Optical Cloud Recognition Algorithm (OCRA) a priori cloud fraction, the Retrieval Of Cloud Information using Neural Networks (ROCINN) CAL (Clouds-As-Layers) cloud fraction and cloud top and base height, the ROCINN CRB (Clouds-as-Reflecting-Boundaries) cloud fraction and cloud height, the Fast Retrieval Scheme for Clouds from the Oxygen A-band (FRESCO) cloud fraction, the interpolated FRESCO cloud height from the TROPOMI NO2 product, the cloud fraction from the NO2 fitting window, the O2–O2 cloud fraction and cloud height, the Mainz Iterative Cloud Retrieval Utilities (MICRU) cloud fraction, and the Visible Infrared Imaging Radiometer Suite (VIIRS) cloud fraction. Two different versions of the TROPOMI cloud products OCRA/ROCINN, FRESCO, and the TROPOMI NO2 product are included in the comparisons (processor version 1.x and 2.x). Overall, the cloud parameters retrieved by the different algorithms show qualitative consistency in version 1.x and good agreement in version 2.x with the exception of the VIIRS cloud fraction, which cannot be directly compared to the other data. Differences between the cloud retrievals are found especially for small cloud heights with a cloud fraction threshold of 0.2, i.e. clouds that are particularly relevant for tropospheric trace gas retrievals. The cloud fractions of the different version 2 cloud products primarily differ over snow- and ice-covered pixels and scenes with sun glint, for which only MICRU includes an explicit treatment. All cloud parameters show some systematic problems related to the across-track dependence, where larger values are found at the edges of the satellite view. The consistency between the cloud parameters from different algorithms depends strongly on how the data are filtered for the comparison, for example, what quality value is used or whether snow- and ice-covered pixels are excluded from the analysis. In summary, clear differences were found between the results of various algorithms, but these differences are reduced in the most recent versions of the cloud data.

Accurate knowledge of cloud properties is essential to the measurement of atmospheric composition from space. In this work we assess the quality of the cloud data from three Copernicus Sentinel-5 Precursor (S5P) TROPOMI cloud products: (i) S5P OCRA/ROCINN_CAL (Optical Cloud Recognition Algorithm/Retrieval of Cloud Information using Neural Networks;Clouds-As-Layers), (ii) S5P OCRA/ROCINN_CRB (Clouds-as-Reflecting Boundaries), and (iii) S5P FRESCO-S (Fast Retrieval Scheme for Clouds from Oxygen absorption bands – Sentinel). Target properties of this work are cloud-top height and cloud optical thickness (OCRA/ROCINN_CAL), cloud height (OCRA/ROCINN_CRB and FRESCO-S), and radiometric cloud fraction (all three algorithms). The analysis combines (i) the examination of cloud maps for artificial geographical patterns, (ii) the comparison to other satellite cloud data (MODIS, NPP-VIIRS, and OMI O2–O2), and (iii) ground-based validation with respect to correlative observations (30 April 2018 to 27 February 2020) from the Cloudnet network of ceilometers, lidars, and radars. Zonal mean latitudinal variation of S5P cloud properties is similar to that of other satellite data. S5P OCRA/ROCINN_CAL agrees well with NPP VIIRS cloud-top height and cloud optical thickness and with Cloudnet cloud-top height, especially for the low (mostly liquid) clouds. For the high clouds, S5P OCRA/ROCINN_CAL cloud-top height is below the cloud-top height of VIIRS and of Cloudnet, while its cloud optical thickness is higher than that of VIIRS. S5P OCRA/ROCINN_CRB and S5P FRESCO cloud height are well below the Cloudnet cloud mean height for the low clouds but match on average better with the Cloudnet cloud mean height for the higher clouds. As opposed to S5P OCRA/ROCINN_CRB and S5P FRESCO, S5P OCRA/ROCINN_CAL is well able to match the lowest CTH mode of the Cloudnet observations. Peculiar geographical patterns are identified in the cloud products and will be mitigated in future releases of the cloud data products.

Space-born atmospheric composition measurements, like those from Sentinel-5p TROPOMI, are strongly affected by the presence of clouds. Dedicated cloud data products, typically retrieved with the same sensor, are therefore an important tool for the provider of atmospheric trace gas retrievals. Cloud products are used for filtering and modification of the modelled radiative transfer.In this work, we assess the quality of the cloud data derived from Copernicus Sentinel-5 Precursor TROPOMI radiance measurements. Three cloud products are considered: (i) L2_CLOUD OCRA/ROCINN CAL (Optical Cloud Recognition Algorithm/Retrieval of Cloud Information using Neural Networks; Clouds-As-Layers), (ii) L2_CLOUD OCRA/ROCINN CRB (same; Clouds-as Reflecting Boundaries), and (iii) the S5p support product FRESCO-S (Fast Retrieval Scheme for Clouds from Oxygen absorption bands for Sentinel). These cloud products are used in the retrieval of several S5p trace gas products (e.g., ozone columns and profile, total and tropospheric nitrogen dioxide, sulfur dioxide, formaldehyde). The quality assessment of these cloud products is carried out within the framework of ESA’s Sentinel-5p Mission Performance Centre (MPC) with support from AO validation projects focusing on the respective atmospheric gases.Cloud height data from the three S5p cloud products is compared to radar/lidar based cloud profile information from the ground-based networks CLOUDNET and ARM. The cloud height from S5p CLOUD CRB and S5p FRESCO are on average 0.6 km below the cloud mid-height of CLOUDNET measurements, and the cloud top height from S5p CLOUD CAL is on average 1 km below CLOUDNET’s cloud top height. However, the comparison is different for low and high clouds, with S5p CLOUD CAL cloud top height being only 0.3 km below CLOUDNET’s for low clouds. The radiometric cloud fraction and cloud (top) height are compared to those of other satellite cloud products like Aura OMI O2-O2. While the latitudinal variation is often similar, offsets are encountered.Recently, major S5p cloud product upgrades were released for S5p OCRA/ROCINN (July 2020) and for S5p FRESCO (December 2020), leading to a decrease of the ROCINN CRB cloud height and an increase of the FRESCO cloud height on average. Moreover, a major change in the ROCINN surface albedo treatment leads to a clear improvement of the comparison with CLOUDNET at the complicated sea/land/ice/snow site Ny-Alesund.

Sentinel-5 Precursor and Sentinel-4 are atmospheric Copernicus missions focused on trace gas, greenhouse gas, aerosol and cloud retrieval and operate in the UV/VIS/NIR/(SWIR) spectral region. A key ingredient for the retrieval of the aforementioned trace gases and greenhouse gases is a precise knowledge of the presence of clouds. On top of that, clouds are by themselves interesting to measure and monitor because of their contribution to the radiation budget, and hence, impact on climatological applications. In this contribution, we present the algorithms for retrieving the operational cloud products from TROPOMI onboard Sentinel-5 Precursor and the UVN spectrometer onboard Sentinel-4. These are called OCRA (Optical Cloud Recognition Algorithm) and ROCINN (Retrieval of Cloud Information using Neural Networks) and both have their heritage with GOME/ERS-2 and GOME-2 MetOp-A/B/C, where they have already been successfully implemented in an operational environment.OCRA employs a broad band color space approach to determine a radiometric cloud fraction and ROCINN retrieves cloud top height, cloud optical thickness and cloud albedo from NIR measurements in and around the oxygen A-band, taking as a priori input the cloud fraction computed by OCRA.The cloud parameters retrieved by ROCINN are provided for two different cloud models. One which treats clouds more realistically as layers of scattering water droplets (clouds-as-layers, CAL) and one which treats clouds as simple Lambertian reflectors (clouds-as-reflecting boundaries, CRB).The current status of the algorithms is presented along with recent developments and improvements.

The increasing amount of Earth observation data provided by the Copernicus Satellite Sensors, the already operational Sentinel-5 Precursor (S5P) and the upcoming Sentinel-4 (S4), that has to be processed within strict near real time (NRT) requirements demands the use of new approaches to cope with this challenge.In order to solve the inverse problems that arise in atmospheric remote sensing, usually complex radiative transfer models (RTMs) are used. These are very accurate, however also computationally very expensive and therefore often not feasible in combination with the time requirements of operational products. With the recent significant breakthroughs in machine learning, easier application through better software and more powerful hardware, the methods of this field have become very interesting as a way to improve the classical remote sensing algorithms.In this presentation we show a general approach in order to replace the RTM of an inversion algorithm with an artificial neural network (ANN) with sufficient accuracy while at the same time increasing the performance by several orders of magnitude. The several steps, sampling and scaling of the training data, the selection of the ANN architecture and the training itself, is explained in detail. This is then demonstrated at the example of the ROCINN (Retrieval of cloud information using neural networks) algorithm for the operational cloud product of S5P. It is then shown how this approach can also be easily applied to the upcoming S4 mission and how the current algorithm for S5P can be improved by replacing or adding new physical models (e.g. for ice-clouds) in the form of ANNs.The procedure has been continuously developed and evaluated over time and the most important results, in terms of sampling, architecture selection, activation functions and training parameters, are presented. Finally, the huge performance benefits of using an ANN instead of the original RTM also allow for improvements in the inversion algorithm. Several ideas regarding this, e.g. global optimization techniques, are also shown.

The first European Sentinel satellite for monitoring the composition of the Earth’s atmosphere, the Sentinel 5 Precursor (S5p), carries the TROPOspheric Monitoring Instrument (TROPOMI) to map trace species of the global atmosphere at high spatial resolution. Retrievals of tropospheric trace gas columns from satellite measurements are strongly influenced by clouds. Thus, cloud retrieval algorithms were developed and implemented in the trace gas processing chain to consider this impact.In this study, different cloud products available for NO2 retrievals based on the TROPOMI level 1b data version 1 and an updated TROPOMI level 1b test data set of version 2 (Diagnostic Data Set 2B, DDS2B) are analyzed. The data sets include a) the TROPOMI level 2 OCRA/ROCINN (Optical Cloud Recognition Algorithm/Retrieval of Cloud Information using Neural Networks) cloud products CRB (cloud as reflecting boundaries) and CAL (clouds as layers), b) the FRESCO (Fast Retrieval Scheme for Clouds from Oxygen absorption bands) cloud product, c) the cloud fraction from the NO2 fitting window, d) the VIIRS (Visible Infrared Imaging Radiometer Suite) cloud product, and e) the MICRU (Mainz Iterative Cloud Retrieval Utilities) cloud fraction. The cloud products are compared with regard to cloud fraction, cloud height, cloud albedo/optical thickness, flagging and quality indicators in all 4 seasons. In particular, the differences of the cloud products under difficult situations such as snow or ice cover and sun glint are investigated.We present results of a statistical analysis on a limited data set comparing cloud products from the current and the upcoming lv2 data versions and their approaches. The aim of this study is to better understand TROPOMI cloud products and their quantitative impacts on trace gas retrievals.

Launched in October 2017, the TROPOspheric Monitoring Instrument (TROPOMI) aboard Sentinel-5 Precursor provides the potential to monitor air quality over point sources across the globe with a spatial resolution as high as 5.5 km × 3.5 km (7 km × 3.5 km before 6 August 2019). The nitrogen dioxide (NO2) retrieval algorithm for the TROPOMI instrument consists of three steps: the spectral fitting of the slant column, the separation of stratospheric and tropospheric contributions, and the conversion of the slant column to a vertical column using an air mass factor (AMF) calculation. In this work, an improved tropospheric NO2 retrieval algorithm from TROPOMI measurements over Europe is presented. The stratospheric estimation is implemented using the STRatospheric Estimation Algorithm from Mainz (STREAM), which was developed as a verification algorithm for TROPOMI and does not require chemistry transport model data as input. A directionally dependent STREAM (DSTREAM) is developed to correct for the dependency of the stratospheric NO2 on the viewing geometry by up to 2 × 1014 molec/cm2. Applied to synthetic TROPOMI data, the uncertainty in the stratospheric column is 3.5 × 1014 molec/cm2 for polluted conditions. Applied to actual measurements, the smooth variation of stratospheric NO2 at low latitudes is conserved, and stronger stratospheric variation at higher latitudes are captured. For AMF calculation, the climatological surface albedo data is replaced by geometry-dependent effective Lambertian equivalent reflectivity (GE_LER) obtained directly from TROPOMI measurements with a high spatial resolution. Mesoscale-resolution a priori NO2 profiles are obtained from the regional POLYPHEMUS/DLR chemistry transport model with the TNO-MACC emission inventory. Based on the latest TROPOMI operational cloud parameters, a more realistic cloud treatment is provided by a clouds-as-layers (CAL) model, which treats the clouds as uniform layers of water droplets, instead of the clouds-as-reflecting-boundaries (CRB) model, in which clouds are simplified as Lambertian reflectors. For the error analysis, the tropospheric AMF uncertainty, which is the largest source of NO2 uncertainty for polluted scenarios, ranges between 20 % and 50 %, leading to a total uncertainty in the tropospheric NO2 column in the 30–60 % range. From a validation performed with ground-based multi-axis differential optical absorption spectroscopy (MAX-DOAS) measurements, the improved tropospheric NO2 data shows good correlations for nine European urban/suburban stations with an average correlation coefficient of 0.78. The implementation of the algorithm improvements leads to a decrease of the relative difference from −55.3 % to −34.7 % on average.

Accurate knowledge of cloud properties is essential to the measurement of atmospheric composition from space. In this work we assess the quality of the cloud data derived from Copernicus Sentinel-5 Precursor (S5P) TROPOMI radiance measurements: cloud top height and cloud optical thickness (retrieved with the S5P OCRA/ROCINN_CAL algorithm), cloud height (S5P OCRA/ROCINN_CRB and S5P FRESCO) and radiometric cloud fraction (all three algorithms). The analysis combines: (i) the examination of cloud maps for artificial geographical patterns, (ii) the comparison to other satellite cloud data (MODIS, NPP-VIIRS and OMI O2-O2), and (iii) ground-based validation with respect to correlative observations (2018-04-30 to 2020-02-27) from the CLOUDNET network of ceilometers, lidars and radars. Peculiar geographical patterns were identified, and will be mitigated in future releases of the cloud data products. Zonal mean latitudinal variation of S5P cloud properties are similar to that of other satellite data. S5P OCRA/ROCINN_CAL agrees well with NPP VIIRS cloud top height and cloud optical thickness, and with CLOUDNET cloud top height, especially for the low (mostly liquid) clouds. For the high clouds, S5P OCRA/ROCINN_CAL cloud top height is below the cloud top height of VIIRS and of CLOUDNET, while its cloud optical thickness is higher than that of VIIRS. S5P OCRA/ROCINN_CRB and S5P FRESCO cloud height are well below the CLOUDNET cloud mean height for the low clouds, but match on an average better with the CLOUDNET cloud mean height for the higher clouds. As opposed to S5P OCRA/ROCINN_CRB and S5P FRESCO, S5P OCRA/ROCINN_CAL is well able to match the lowest CTH mode of the CLOUDNET observations.

The measurement of atmospheric composition from space requires also a precise knowledge regarding the appearance of clouds within the observed scene, and if present, a quantification of cloud properties such as cloud fraction, cloud height and cloud optical thickness/cloud albedo. The Copernicus mission Sentinel-5 Precursor, being operational since 2 years, covers the UV/VIS/NIR/SWIR spectral region. By covering the spectral region of the Oxygen A-band in the NIR, it provides an excellent prerequisite to retrieve the cloud parameters mentioned above. The same holds true for the anticipated Sentinel-4 mission (foreseen launch in 2023). In this contribution we present the most recent advances in the algorithms for retrieving the operational cloud products from TROPOMI onboard Sentinel-5 Precursor and from the future Sentinel-4/UVN onboard MTG-S. The applied cloud retrieval algorithms OCRA (Optical Cloud Recognition Algorithm) and ROCINN (Retrieval of Cloud Information using Neural Networks) have their heritage with GOME/ERS-2 and GOME-2 on MetOp-A/B/C, where they have already been successfully implemented in an operational environment. OCRA uses a broad band color space approach in the UV/VIS in combination with a set of cloud-free reflectance background composite maps to determine a radiometric cloud fraction while the ROCINN algorithm retrieves the cloud top height, cloud optical thickness and cloud albedo from NIR measurements in and around the oxygen A-band, taking as a priori input the cloud fraction computed by OCRA. ROCINN includes two different cloud models. One which treats clouds more realistically as layers of scattering water droplets (clouds-as-layers, CAL) and another one which treats clouds as simple Lambertian reflectors (clouds-as-reflecting boundaries, CRB). Substantial improvements to the algorithms have been implemented recently, some of which will be presented here, e.g. improved background maps, inclusion of cloud phase, retrieval of surface properties using machine learning. Further validation efforts via satellite-to-satellite comparisons with VIIRS on Suomi-NPP have been carried out and consolidate the product quality.

Satellite measurements of tropospheric or total column trace gas species, including those from Sentinel-5p TROPOMI, are affected by the presence of clouds. Therefore, cloud data products retrieved with the same sensor play an essential role, as they allow the data provider to take an estimated cloud impact on the trace gas retrieval into account. Examples are the modification of the radiative transfer and associated quantities such as the air mass factor, and the partial masking of the measurement scene. Evidently, the accuracy of these corrections relies on the accuracy of the retrieved cloud properties, like radiometric cloud fraction (CF), cloud top height (CTH) or cloud height (CH), and cloud optical thickness (COT) or cloud albedo (CA).We consider here three S5p TROPOMI-based cloud products: (i) L2_CLOUD OCRA/ROCINN CAL (Optical Cloud Recognition Algorithm/Retrieval of Cloud Information using Neural Networks; Clouds-As-Layers), (ii) L2_CLOUD OCRA/ROCINN CRB (Clouds-as Reflecting Boundaries), and (iii) the S5p support product FRESCO-S (Fast Retrieval Scheme for Clouds from Oxygen absorption bands). These are input to the S5p operational processors for several trace gas products, including ozone columns and profile, total and tropospheric NO2, formaldehyde, sulfur dioxide. The quality assessment of these cloud products is carried out within the framework of ESA’s Sentinel-5p Mission Performance Centre (MPC) with support from AO validation projects focusing on the respective trace gases.In this work, cloud height (from S5p CLOUD CRB and S5p FRESCO algorithms) and cloud top height (from S5p CLOUD CAL) S5p data is validated with radar/lidar-based cloud profile information from the ground-based networks CLOUDNET and ARM at 17 sites. For some sites the comparison is difficult due to e.g., orography or snow/ice cover. S5P and CLOUDNET report similar cloud height variations at several sites, and the correlation between the S5p cloud products and CLOUDNET can be high (Pearson R up to 0.9). However, there is a site-dependent negative bias of the S5p cloud (top) height with respect to the CLOUDNET data: up to -2.5 km for S5p CLOUD CAL cloud top height and -1.5 km for S5p CLOUD CRB and S5p FRESCO cloud height. The dependence on other parameters measured by S5p and CLOUDNET (e.g., radiometric cloud fraction, cloud phase,…) is investigated.

This paper presents the operational cloud retrieval algorithms for the TROPOspheric Monitoring Instrument (TROPOMI) on board the European Space Agency Sentinel-5 Precursor (S5P) mission scheduled for launch in 2017. Two algorithms working in tandem are used for retrieving cloud properties: OCRA (Optical Cloud Recognition Algorithm) and ROCINN (Retrieval of Cloud Information using Neural Networks). OCRA retrieves the cloud fraction using TROPOMI measurements in the ultraviolet (UV) and visible (VIS) spectral regions, and ROCINN retrieves the cloud top height (pressure) and optical thickness (albedo) using TROPOMI measurements in and around the oxygen A-band in the near infrared (NIR). Cloud parameters from TROPOMI/S5P will be used not only for enhancing the accuracy of trace gas retrievals but also for extending the satellite data record of cloud information derived from oxygen A-band measurements, a record initiated with the Global Ozone Monitoring Experiment (GOME) on board the second European Remote-Sensing Satellite (ERS-2) over 20 years ago. The OCRA and ROCINN algorithms are integrated in the S5P operational processor UPAS (Universal Processor for UV/VIS/NIR Atmospheric Spectrometers), and we present here UPAS cloud results using the Ozone Monitoring Instrument (OMI) and GOME-2 measurements. In addition, we examine anticipated challenges for the TROPOMI/S5P cloud retrieval algorithms, and we discuss the future validation needs for OCRA and ROCINN.

An improved tropospheric nitrogen dioxide (NO2) retrieval algorithm from the Global Ozone Monitoring Experiment-2 (GOME-2) instrument based on air mass factor (AMF) calculations performed with more realistic model parameters is presented. The viewing angle-dependency of surface albedo is taken into account by improving the GOME-2 Lambertian-equivalent reflectivity (LER) climatology with a directionally dependent LER (DLER) dataset over land and an ocean surface albedo parametrization over water. A priori NO2 profiles with higher spatial and temporal resolutions are obtained from the IFS(CB05BASCOE) chemistry transport model based on recent emission inventories. A more realistic cloud treatment is provided by a Cloud-As-Layers (CAL) approach, which treats the clouds as uniform layers of water droplets, instead of the current Clouds-as-Reflecting-Boundaries (CRB) model, which assumes the clouds as Lambertian reflectors. Improvements in the AMF calculation affect the tropospheric NO2 columns on average within ±15 % in winter and ±5 % in summer over largely polluted regions. In addition, the impact of aerosols on our tropospheric NO2 retrieval is investigated by comparing the concurrent retrievals based on ground-based aerosol measurements (explicit aerosol correction) and aerosol-induced cloud parameters (implicit aerosol correction). Compared to the implicit aerosol correction through the CRB cloud parameters, the use of CAL reduces the AMF errors by more than 10 %. Finally, to evaluate the improved GOME-2 tropospheric NO2 columns, a validation is performed using ground-based Multi-AXis Differential Optical Absorption Spectroscopy (MAXDOAS) measurements at the BIRA-IASB Xianghe station. The improved tropospheric NO2 dataset shows good agreement with coincident ground-based measurements with a correlation coefficient of 0.94 and a relative difference of −9.9 % on average.

The results of the ATHLI16 (ATHens Lidar Intercomparison) campaign, held in Athens from 26/09 to 07/10 2016 are presented. The campaign was performed within the Lidar Calibration Centre activities (EU H2020 ACTRIS-2 project) to assess the performance of the EOLE lidar system (NTUA, Athens, Greece), operating within EARLINET, by comparing against the EARLINET reference lidar system MUSA (CNR-IMAA, Potenza, Italy). For both lidars only products retrieved by the EARLINET Single Calculus Chain have been compared.

A comprehensive analysis of the seasonal variability of the optical properties of Saharan dust aerosols over Athens, Greece, is presented for a 17-year time period (2000-2016), as derived from multi-wavelength Raman lidar measurements (57 dust events with more than 80 hours of measurements). The profiles of the derived aerosol optical properties (aerosol backscatter and extinction coefficients, lidar ratio and aerosol Ångström exponent) at 355 nm are presented. For these dust events we found a mean value of the lidar ratio of ~52±13 sr at 355 nm and of ~58±8 sr (not shown) at 532 nm (2-4 km a.s.l. height). For our statistical analysis, presented here, we used monthly-mean values and time periods under cloud-free conditions. The number of dust events was greatest in late spring, summer, and early autumn periods. In this paper we also present a selected case study (04 April 2016) of desert dust long-range transport from the Saharan desert.

The EOLE multi-wavelength aerosol Ramandepolarization lidar, and the AIAS depolarization lidar, in synergy with a sun photometer (CIMEL), were used, in the period 2007-2016, to provide the vertical profiles of the aerosol optical properties over Athens, Greece. More than 30 biomass burning events (fresh and aged smoke particles) were observed, with smoke layers between 1.5 up to 4-5 km height, while their duration ranged from 1-3 days. Lidar ratio (LR) values ranged from 40-105 sr (at 355 nm) and from 40-100 sr (at 532 nm), while the linear particle depolarization ratio (LPDR) at both 355 and 532 nm, remained a) at 355 nm/532 nm) ranged from 0.3 to 2.1. Additionally, a case of a near-range transport of biomass burning aerosols arriving over Athens up to 4 km height, between 27 and 28 June 2016, was studied. For this case, we found LRs of the order of 70±5 sr (355 nm) and 65±15 sr (532 nm) and AEa(355 nm/532 nm) around 1.

Measurements of cloud condensation nuclei (CCN) concentrations (cm−3 ) at five levels of supersaturation between 0.2–1%, together with remote sensing profiling and aerosol size distributions, were performed at an urban background site of Athens during the Hygroscopic Aerosols to Cloud Droplets (HygrA-CD) campaign. The site is affected by local emissions and long-range transport, as portrayed by the aerosol size, hygroscopicity and mixing state. Application of a state-of-the-art droplet parameterization is used to link the observed size distribution measurements, bulk composition, and modeled boundary layer dynamics with potential supersaturation, droplet number, and sensitivity of these parameters for clouds forming above the site. The sensitivity is then used to understand the source of potential droplet number variability. We find that the importance of aerosol particle concentration levels associated with the background increases as vertical velocities increase. The updraft velocity variability was found to contribute 58–90% (68.6% on average) to the variance of the cloud droplet number, followed by the variance in aerosol number (6–32%, average 23.2%). Therefore, although local sources may strongly modulate CCN concentrations, their impact on droplet number is limited by the atmospheric dynamics expressed by the updraft velocity regime.

The international experimental campaign Hygroscopic Aerosols to Cloud Droplets (HygrA-CD), organized in the Greater Athens Area (GAA), Greece from 15 May to 22 June 2014 provided an unprecedented record of data on aerosols, clouds and meteorology. HygrA-CD brought together different active/passive remote sensing and in situ instrumentation/teams for the purpose of fostering our understanding on the physico-chemical properties of aerosols and their impact on the cloud formation in the lower troposphere, with emphasis on the top of the Planetary Boundary Layer (PBL). The use of well-established numerical weather prediction and atmospheric modeling, such as the Flexible particle (FLEXPART) dispersion model and the Weather Research and Forecasting (WRF) model enabled us to simulate air mass back-trajectories and atmospheric variables in the PBL, such as PBL height (PBLH), relative humidity, wind fields etc. This paper presents the major findings and highlights of the HygrA-CD campaign.

Combined multi-wavelength aerosol Raman lidar and sun photometry measurements were performed during the HYGRA-CD campaign over Athens, Greece during May-June 2014. The retrieved aerosol optical properties (3 aerosol backscatter at 355-532-1064 nm and 2 aerosol extinction profiles at 355-532 nm) were used as input to an inversion code to retrieve the aerosol microphysical properties (effective radius reff and number concentration N ) using regularization techniques. Additionally, the volume concentration profile was derived for fine particles using the LIRIC code. In this paper we selected a complex case study of long-range transport of mixed aerosols (biomass burning particles mixed with dust) arriving over Athens between 10-12 June 2014 in the 1.5-4 km height. Between 2-3 km height we measured mean lidar ratios (LR) ranging from 45 to 58 sr (at 355 and 532 nm), while the Angstrom exponent (AE) aerosol extinction-related values (355nm/532nm) ranged between 0.8-1.3. The retrieved values of reff and N ranged from 0.19±0.07 to 0.22±0.07 μm and 460±230 to 2200±2800 cm-3 , respectively. The aerosol linear depolarization ratio (δ) at 532 nm was lower than 5-7% (except for the Saharan dust cases, where δ~10-15%).

Measurements of cloud condensation nuclei (CCN) concentrations between 0.2-1.0% supersaturation and aerosol size distribution were performed at an urban background site of Athens during HygrA-CD. The site is affected by local and long-range transported emissions as portrayed by the external mixing of the particles, as the larger ones appear to be more hygroscopic and more CCN-active than smaller ones. Activation fractions at all supersaturations exhibit a diurnal variability with minimum values around noon, which are considerably lower than unity. This reinforces the conclusion that the aerosol is mostly externally mixed between “fresher”, less hygroscopic components with more aged, CCN active constituents.

Multi-wavelength aerosol Raman lidar measurements with elastic depolarization at 532 nm were combined with sun photometry during the HYGRA-CD campaign over Athens, Greece, on May-June 2014. We retrieved the aerosol optical [3 aerosol backscatter profiles (baer ) at 355-532-1064 nm, 2 aerosol extinction (aaer ) profiles at 355-532 nm and the aerosol linear depolarization ratio (δ ) at 532 nm] and microphysical properties [effective radius (reff ), complex refractive index (m ), single scattering albedo (ω )]. We present a case study of a long distance transport (~3.500-4.000 km) of biomass burning particles mixed with dust from the Russian Federation-Kazakhstan regions arriving over Athens on 21-23 May 2014 (1.7-3.5 km height). On 23 May, between 2-2.75 km we measured mean lidar ratios (LR) of 35 sr (355 nm) and 42 sr (532 nm), while the mean Angstrom exponent (AE) aerosol backscatter-related values (355nm/532nm and 532nm/1064nm) were 2.05 and 1.22, respectively; the mean value of δ at 532 nm was measured to be 9%. For that day the retrieved mean aerosol microphysical properties at 2-2.75 km height were: reff =0.26 μm (fine mode), reff =2.15 μm (coarse mode), m =1.36+0.00024i, ω =0.999 (355 nm, fine mode), ω =0.992(355 nm, coarse mode), ω =0.997 (532 nm, fine mode), and ω =0.980 (532 nm, coarse mode).

The international experimental campaign Hygroscopic Aerosols to Cloud Droplets (HygrA-CD), organized in the Greater Athens Area (GAA), Greece from 15 May to 22 June 2014, aimed to study the physico-chemical properties of aerosols and their impact on the formation of clouds in the convective Planetary Boundary Layer (PBL). We found that under continental (W-NW-N) and Etesian (NE) synoptic wind flow and with a deep moist PBL (~ 2–2.5 km height), mixed hygroscopic (anthropogenic, biomass burning and marine) particles arrive over the GAA, and contribute to the formation of convective non-precipitating PBL clouds (with droplets of ~ 16–20 μm mean diameter) with vertical extent up to 500 m. Under these conditions, high updraft velocities (1–2 m s− 1) and cloud condensation nuclei (CCN) concentrations (~ 2000 cm− 3 at 1% supersaturation), generated clouds with an estimated cloud droplet number of ~ 600 cm− 3. Under Saharan wind flow conditions (S-SW) a shallow PBL (< 1–1.2 km height) develops, leading to much higher CCN concentrations (~ 3500–5000 cm− 3 at 1% supersaturation) near the ground; updraft velocities, however, were significantly lower, with an estimated maximum cloud droplet number of ~ 200 cm− 3 and without observed significant PBL cloud formation. The largest contribution to cloud droplet number variance is attributed to the updraft velocity variability, followed by variances in aerosol number concentration.

In this paper we studied the temporal evolution of the Planetary Boundary Layer height (PBLH) over the basin of Athens, Greece during a 5-year period (2011-2016) using data from the EOLE Raman lidar system. The lidar data (range-corrected lidar signals-RCS) were selected around 12:00 UTC and 00:00 UTC for a total of 332 cases: 165 days and 167 nights. Extended Kalman filtering techniques were used for the determination of the PBLH. Moreover, several well established techniques for the PBLH estimation based on lidar data were also tested for a total of 35 cases. Comparisons with the PBLH values derived from radiosonde data were also performed. The mean PBLH over Athens was found to be of the order of 1617±324 m at 12:00 UTC and of 892±130 m at 00:00 UTC, for the period examined. The mean PBLH growth rate was found to be about 170±64 m h-1 and 90±17 m h-1, during daytime and nighttime, respectively.