Recent  

We describe in detail the design and calibration of the new Raman channels for the WALI lidar system, going over the important sources of bias and uncertainty on retrieved temperature profiles. For the first time their impact is investigated using horizontal shots in a homogenous atmosphere: the magnitude of the highlighted biases can be much larger than the targeted absolute accuracy of 1 °C. Actual measurement errors are quantified using radiosoundings launched close to the lidar site.

Mineral dust and particles of biological origin are important types of ice-nucleating particles (INP) that can trigger ice formation of supercooled cloud droplets. Heat treatments are used to “detect” the presence of biological INP in samples collected from the environment as the activity of mineral INP is assumed unchanged, yet this has not been fully assessed. We show that the ice-nucleating ability of some minerals can change after heating and discuss how INP heat tests should be interpreted.

The Support Vector Machine can effectively represent the increasing effect of rain in affecting wind speeds. This research provides correction of deviations that are skew to Gaussian-like features caused by rain in Ku-band scatterometer wind. It demonstrates the effectiveness of a Machine Learning method when used based on elaborated analysis of the model establishment and result validation procedures. The corrected winds provide information previously lacking, while vital for nowcasting.

An all-sky camera is used to obtain the relative sky radiance, and this radiance is used as input in an inversion code to obtain aerosol properties. This paper is really interesting since it pushes forward the use and capability of sky cameras for more advance science purposes. Enhanced aerosol properties can be retrieved with accuracy using only a sky camera, but synergy with other instruments providing aerosol optical depth could even increase the powerfull of these low-cost instruments.

We present a formalism that relates the vertical column density (VCD) of the oxygen collision complex O₄ to surface values of temperature and pressure, based on physical laws. In addition, we propose an empirical modification which also accounts for surface relative humidity (RH). This allows for simple and quick but still accurate calculation of the O₄ VCD without the need for constructing full vertical profiles, which is expected to be useful in particular for MAX-DOAS applications.

While the Aerosol Mass Spectrometer provides high time resolution characterization of the overall extent of oxidation, the extensive fragmentation of molecules and specificity of the technique has posed challenges toward deeper understanding of molecular structures in aerosols. This work demonstrates how functional group information can be extracted from a suite of commonly-measured mass fragments using collocated infrared spectroscopy measurements.

Oxidative Potential (OP) of particulate matter (PM) is an important indicator of PM toxicity. However, no automated instrument has ever been developed to provide a rapid high-throughput analysis of the cell-based OP measurements. Here, we developed a first-of-its kind semi-automated instrument for measuring oxidative potential using rat alveolar cells. We also developed a dataset on the intrinsic cellular OP of several compounds commonly known to be present in the ambient PM.

A dataset from the RadSnowExp represents the first-ever triple-frequency radar reflectivities at X-Ka-W-band combined with almost perfectly co-located and coincident airborne in situ microphysics probes. Results confirm the main findings of previous modelling works with radar dual frequency ratios (DFRs) occur at different zones of the DFR plane associated with different ice habits. Moreover, the dataset could be used to produce look-up tables for retrieving microphysical properties.

Broad supersaturation scanning CCN (BS2-CCN) system, in which particles are exposed to a range of supersaturation simultaneously, can measure a broad range of CCN activity distribution with a high-time resolution. We describe how the BS2-CCN system can be effectively calibrated and which factors can affect the calibration curve. For evaluating the BS2-CCN system, inter-comparison experiments between typical DMA-CCN and BS2-CCN measurement were performed.

Soot particles with varying amounts of secondary organic matter coating were generated and used to compare a series of aerosol absorption measuring instruments: filter-based and photoacoustic instruments as well as photothermal interferometers. Significant deviations in the response of the instruments were observed depending on the amount of secondary organic coating. The system can be used for the inter-comparison and characterisation of instruments.

A Random Forest (RF) model framework for FY-4A quantitative precipitation estimation during daytime/nighttime is established by using FY-4A multi-band spectral information, cloud parameters, high-density precipitation observations, and physical quantities from reanalysis data. The RF model of FY-4A QPE has high precipitation retrieval accuracy, especially for precipitation intensities of 0.5–10 mm/h, which has advantages for near-real-time monitoring of precipitation over East Asia in future.

Heating by solar radiation is the dominant error source for daytime temperature measurements by radiosondes. This paper describes a new laboratory setup to characterise this radiation error for pressures and ventilation speeds that are typical for the conditions between the surface and 35 km altitude. This characterisation is the basis for the radiation correction that is applied in the GRUAN data processing for the RS41 radiosonde. The GRUAN data product is compared to that of Vaisala.

This study shows how precision error and bias in column methane retrieval change with different instrument specifications, and the impact of spectrally complex surface albedos on retrievals. We show how surface interferences can be mitigated with an optimal spectral resolution and a higher polynomial degree in a retrieval process. The findings can inform future satellite instrument designs to have robust observations capable of separating real CH₄ plume enhancements from surface interferences.

Machine learning has become an important tool for pattern recognition in many applications. In this study, we used a neural network to improve the data quality of OCO-2 measurements made at northern high latitudes. The neural network was trained and used as a binary classifier to filter out bad OCO-2 measurements in order to increase the accuracy and precision of OCO-2 XCO₂ measurements in the Boreal and Arctic regions.

NO₂ plays a central role in the atmospheric photochemical processes and requires accurate measurements. This research presents NO₂ data obtained via chemiluminescence using a photolytic converter from airborne studies around Cabo Verde and laboratory investigations. We show the limits and error-proneness of a conventional blue light converter in aircraft measurements affected by humidity and NO levels and suggest the use of a new quartz converter for more reliable results.

We develop candidate Bayesian algorithms to evaluate the next-generation NASA Cloud Convection Precipitation (CCP) observing system. The hybrid Bayesian algorithm combines the Bayesian Monte Carlo Integration and optimization methods, and it uses the optimal estimation methodology and ensemble approaches to maximize the posterior probability density function. Quantitative assessments for synergistic radar and radiometer capability in retrieving ice cloud microphysics are provided.

To study what is happening in the atmosphere, it is important to be able to measure the molecules and clusters present in it. In our work, we studied an artifact that happens inside a mass spectrometer, in particular, the fragmentation of clusters. We were able to quantify the fragmentation and retrieve the correct concentration and composition of the clusters using our dual (experimental and theoretical) approach.

Chlorine dioxide (OClO) is used as an indicator for chlorine activation. We present a new DOAS retrieval algorithm for OClO from measurements of TROPOMI on the Sentinel-5P satellite. To achieve a substantially improved accuracy for the weak absober OClO we consider several additional fit parameters accounting for various higher order spectral effects. The retrieved OClO slant column densities are compared with ground based zenith sky measurements.

Understanding and improving the quality of data generated from low-cost sensors is a crucial step in using these sensors. We show how averaging time, choice of reference instrument, and the observation of higher pollutant concentrations can all impact the perceived performance of low-cost sensors in an evaluation. The influence of these factors should be considered when comparing one sensor to another or when determining whether a sensor can produce data that fits a specific need.

Measuring the shape, size and mass of a large number of snowflakes is a challenging task; hard to achieve in an automatic and instrumented manner. We present a method to retrieve these properties of individual snowflakes using as input a triplet of images/pictures automatically collected by a Multi-Angle-Snowflake-Camera (MASC) instrument. Our method, based on machine learning, is trained on artificially generated snowflakes and evaluated on 3d-printed snowflake replicas.

This study investigates a method for 24-h cloud cover calculation using a camera-based imager and supervised machine learning methods. The cloud cover is calculated by learning the statistical characteristics of the ratio, difference, and luminance using RGB channel of the image with a machine learning model. The proposed approach is suitable for nowcasting because it has higher learning and prediction speed than the method in which the many pixels of a 2D image are learned.

This paper presents an extended H₂O/HDO total column data set from short-wave infrared measurements by TROPOMI including cloudy and clear-sky scenes. Coverage is tremendously increased compared to previous TROPOMI HDO data sets. The new data set is validated against recent ground-based FTIR measurements from the TCCON network and against aircraft measurements over the ocean. The use of the new data set is demonstrated with a case study of a cold air outbreak in January 2020.

We proposed a method to derive a convective boundary layer height using insects and Bragg scatter in radar observations and investigated the consistency of these retrievals between different radar frequencies (5, 35 and 94 GHz). This method can be applied to radars at other measurement stations and serve as additional way to estimate the boundary layer height during summer. The entrainment zone was also observed by the 5 GHz radar above the boundary layer in a form of a Bragg scatter layer.

CO is an important atmospheric gas that influences both air quality and the climate. Here, we compare CO measurements from the TROPOMI instrument with those from ACE-FTS and an Arctic ground-based FTS at Eureka, Nunavut to further characterize the accuracy of TROPOMI measurements. CO columns from the instruments agree well, but show larger differences at high latitudes. Despite this, the results fall within the TROPOMI accuracy target, indicating good data quality at high-latitudes.

In this work, we examine different methods for monitoring the calibration of the radar differential reflectivity Z_DR. First, we implemented a well-known procedure to calibrate and monitor the Z_DR offset using vertical profiles of polarimetric variables. Then, we propose a novel, operational method to calibrate Z_DR using quasi-vertical profiles of polarimetric variables. We observed an excellent agreement between both methods, as the errors (MAE and RMSE) are within 0.1 dB.

Large microwave surface emissivities (ε) make more diffculties in widely using satellite microwave data over land. Usually ground-based radiometers are fixed to scan field to obtain the temporal evolution of ε over single land-cover area. To obtain the long-term temporal evolution of ε over different land-cover surfaces simultaneously, we devoloped a ground mobile observation system to enhance in situ ε observations, and presented some preliminary results.

NO₂ maps from airborne imaging remote sensing often appear much smoother than one would expect from high-resolution model simulations of NO₂ over cities, despite the small ground-pixel size of the sensors. Our case study over Zurich, using the newly implemented building module of the MYSTIC radiative transfer solver, shows that 3D effect can explain the smearing and that building shadows cause a noticeable underestimation of and noise in the measured NO₂ columns.

This manuscript presents the main algorithms of the Aeolus level 2 aerosol optical properties product. The processing chain was developed under contract with ESA. We show that the ALADIN instrument, although primarily designed to retrieve atmospheric winds, is also able to provide valuable information about aerosol and clouds optical properties. The algorithms are detailed and validation on simulated and real examples are shown.

Radiosondes keep measuring and transmitting atmospheric data after balloon burst – potentially providing an extra profile at minimal cost. We show that a subset of RS41 descent data is currently usable, and similar quality to the ascent data. Parachutes reduce the temperature and pressure biases seen. In the near future improved processing should mean that a larger subset is usable.