Advanced open-source tools for detecting and analyzing peaks in cloud radar Doppler spectra

<p>Cloud radar Doppler spectra contain vertically highly resolved valuable information about the hydrometeors present in the cloud. A mixture of different hydrometeor types can lead to several peaks in the Doppler spectrum due to their different fall speeds, giving a hint about the size/ density/ number of the respective particles. Tools to separate and interpret peaks in cloud radar Doppler spectra have been developed in the past, but their application is often limited to certain radar settings, or the code not freely available to other users.</p> <p>We here present the effort of joining two methods, which have been developed and published (Radenz et al., 2019; Kalesse et al., 2019) with the aim to make them insensitive to instrument type and settings, and available on GitHub, and applicable to all cloud radars which are part of the ACTRIS CloudNet network.</p> <p>A supervised machine learning peak detection algorithm (PEAKO, Kalesse et al., 2019) is used to derive the optimal parameters to detect peaks in cloud radar Doppler spectra for each set of instrument settings. In the next step, these parameters are used by peakTree (Radenz et al., 2019), which is a tool for converting multi-peaked (cloud) radar Doppler spectra into a binary tree structure. PeakTree yields the (polarimetric) radar moments of each detected peak and can thus be used to classify the hydrometeor types. This allows us to analyze Doppler spectra of different cloud radars with respect to, e.g. the occurrence of supercooled liquid water or ice needles/columns with high linear depolarisation ratio (LDR).</p>

Variability in cirrus cloud properties using a Polly^XT Raman lidar over high and tropical latitudes

Measurements of geometrical and optical properties of cirrus clouds, performed with a multi-wavelength PollyXT Raman lidar during the period 2008 to 2016, are analysed. The measurements were performed with the same instrument, during sequential periods, in three places at different latitudes, Gwal Pahari (28.43∘ N, 77.15∘ E; 243 m a.s.l.) in India, Elandsfontein (26.25∘ S, 29.43∘ E; 1745 m a.s.l.) in South Africa and Kuopio (62.74∘ N, 27.54∘ E; 190 m a.s.l.) in Finland. The lidar dataset was processed by an automatic cirrus cloud masking algorithm, developed in the frame of this work. In the following, we present a statistical analysis of the lidar-retrieved geometrical characteristics (cloud boundaries, geometrical thickness) and optical properties of cirrus clouds (cloud optical depth, lidar ratio, ice crystal depolarisation ratio) measured over the three areas that correspond to subtropical and subarctic regions as well as their seasonal variability. The effect of multiple scattering from ice particles to the derived optical products is also considered and corrected in this study. Our results show that cirrus layers, which have a noticeable monthly variability, were observed between 6.5 and 13 km, with temperatures ranging from −72 to −27 ∘C. The observed differences on cirrus clouds' geometrical and optical properties over the three regions are discussed in terms of latitudinal and temperature dependence. The latitudinal dependence of the geometrical properties is consistent with satellite observations, following the pattern observed with CloudSat, with decreasing values towards the poles. The geometrical boundaries have their highest values in the subtropical regions, and overall, our results seem to demonstrate that subarctic cirrus clouds are colder, lower and optically thinner than subtropical cirrus clouds. The dependence of cirrus cloud geometrical thickness and optical properties on mid-cirrus temperatures shows a quite similar tendency for the three sites but less variability for the subarctic dataset. Cirrus clouds are geometrically and optically thicker at temperatures between −45 and −35 ∘C, and a second peak is observed at lower temperatures ∼-70 ∘C for the subarctic site. Lidar ratio values also exhibit a pattern, showing higher values moving toward the poles, with higher mean values observed over the subarctic site. The dependency of the mid-cirrus temperatures on the lidar ratio values and the particle depolarisation values is further examined. Our study shows that the highest values of the cirrus lidar ratio correspond to higher values of cirrus depolarisation and warmer cirrus. The kind of information presented here can be rather useful in the cirrus parameterisations required as input to radiative transfer models and can be a complementary tool for satellite products that cannot provide cloud vertical structure. In addition, ground-based statistics of the cirrus properties could be useful in the validation and improvement of the corresponding derived products from satellite retrievals.

Coating material-dependent differences in modelled lidar-measurable quantities for heavily coated soot particles

<p>The depolarisation ratio of heavily coated soot particles was previously found to be sensitive to the chemical composition of the coating material, which is reflected by the refractive index. Employing the Discrete Dipole Approximation code ADDA optical calculations were performed with a set of heavily coating soot aggregates with two different coating materials at 355 nm, 532 nm, and 1064 nm. As coating materials sulphate and a toluene-based material were assumed. The soot aggregates were modelled based on results reported from in-situ field measurements and using a coating model, which allows for a tunable transition between film coating and spherical shell coating. The aggregates’ size was varied by increasing the number of soot monomers inside each aggregate from 26 to 1508 in linearly equidistant steps.</p><p>Size-averaged lidar-measureable quantities for the coated aggregates, such as the linear depolarisation ratio, the extinction-to-backscatter ratio (lidar ratio), and the Ångström exponents of the extinction coefficient, the backscatter coefficient, and the extinction-to-backscatter ratio were calculated, and the error of the simulations was estimated. With the exception of the linear depolarisation ratio at 1064 nm these observables do not overlap within the estimated error bounds. As the coating materials result in clearly distinguishable lidar observables, information on the chemical composition of coated soot aerosol can potentially be inferred from lidar measurements.</p><p> </p><p> </p>

A Recipe to Obtain Lidar Polarisation Calibration Parameters G, H and K

The accuracy of the polarisation calibration is of prime importance for aerosol classification using lidars. We present a detailed description how to obtain the calibration parameters introduced in 2016 [1] accounting for various effects of non-ideal optics, lasers and atmospheric conditions. We find that crucial parameters such as the rotation angle of the plane of polarisation of the Laser (RotL) as well as the degree of linear polarisation (DOLP) influence the volume linear depolarisation ratio significantly.

On the spectral depolarisation and lidar ratio of mineral dust provided in the AERONET version 3 inversion product

Knowledge of the particle lidar ratio (Sλ) and the particle linear depolarisation ratio (δλ) for different aerosol types allows for aerosol typing and aerosol-type separation in lidar measurements. Reference values generally originate from dedicated lidar observations but might also be obtained from the inversion of AErosol RObotic NETwork (AERONET) sun/sky radiometer measurements. This study investigates the consistency of spectral Sλ and δλ provided in the recently released AERONET version 3 inversion product for observations of undiluted mineral dust in the vicinity of the following major deserts: Gobi, Sahara, Arabian, Great Basin, and Great Victoria. Pure dust conditions are identified by an Ångström exponent <0.4 and a fine-mode fraction <0.1. The values of spectral Sλ are found to vary for the different source regions but generally show an increase with decreasing wavelength. The feature correlates to AERONET, retrieving an increase in the imaginary part of the refractive index with decreasing wavelength. The smallest values of Sλ=35–45 sr are found for mineral dust from the Great Basin desert, while the highest values of 50–70 sr have been inferred from AERONET observations of Saharan dust. Values of Sλ at 675, 870, and 1020 nm seem to be in reasonable agreement with available lidar observations, while those at 440 nm are up to 10 sr higher than the lidar reference. The spectrum of δλ shows a maximum of 0.26–0.31 at 1020 nm and decreasing values as wavelength decreases. AERONET-derived δλ values at 870 and 1020 nm are in line with the lidar reference, while values of 0.19–0.24 at 440 nm are smaller than the independent lidar observations by a difference of 0.03 to 0.08. This general behaviour is consistent with earlier studies based on AERONET version 2 products.

Saharan dust and biomass burning aerosols during ex-hurricane Ophelia: validation of the new UK lidar and sun-photometer network

On 15–16 October 2017, ex-hurricane Ophelia passed to the West of the British Isles, bringing dust from the Sahara and smoke from Portuguese forest fires that was observable to the naked eye and reported in the national press. We report here detailed observations of this event using the UK operational lidar and sun-photometer network, established for the early detection of aviation hazards. The observations, taken continuously over a period of 30 hours, show a complex picture, dominated by several aerosol layers at different times, and clearly correlated with the passage of different air-masses associated with the intense cyclonic system. A similar evolution was observed at several sites, with a time delay between them explained by their different location with respect to the storm. The event commenced with a hallow dust layer at 1–2 km in altitude, and culminated in a deep and complex structure that lasted 12 hours at each site, correlated with the storm’s warm sector. For most of the time, the aerosol detected as mineral dust, as highlighted by depolarisation measurements, but an intense smoke layer was observed towards the end of the event, lasting around 3 hours at each site. The aerosol optical depth AOD) during the whole event ranged from 0.2 to 2.9, with the larger AOD correlated to the intense smoke plume. Such a large AOD is unprecedented in the United Kingdom according to AERONET records or the last 20 years. The Raman lidars permitted the measurement of the aerosol extinction coefficient at 355 nm, the particle depolarisation ratio (PDR) and the lidar ratio (LR), and made possible the separation of the dust (depolarising) aerosol from other aerosol types. A specific extinction has also been computed to provide an estimate of the atmospheric concentration of both aerosols separately, which peaked at 500 &amp;pm; 100 μg m−3 for the dust and 600 &amp;pm; 100 μg m−3 for the smoke. Back-trajectories computed using the Numerical Atmospheric dispersion Modelling Environment (NAME) were used to identify the sources and strengthen the conclusions drawn from the observations. The UK network represents a significant expansion of the observing capability in Northern Europe, with instruments evenly distributed across Great Britain, from Camborne in Cornwall to Lerwick in the Shetland Islands, and this study represents the first attempt to demonstrate its capability and validate the methods in use. Its ultimate purpose will be the detection and quantification of volcanic plumes, but the present study clearly demonstrates the advanced capabilities of the network.

Ship-borne aerosol profiling with lidar over the Atlantic Ocean: from pure marine conditions to complex dust–smoke mixtures

The multi-wavelength Raman lidar PollyXT has been regularly operated aboard the research vessel Polarstern on expeditions across the Atlantic Ocean from north to south and vice versa. The lidar measurements of the RV Polarstern cruises PS95 from Bremerhaven, Germany, to Cape Town, Republic of South Africa (November 2015), and PS98 from Punta Arenas, Chile, to Bremerhaven, Germany (April/May 2016), are presented and analysed in detail. The latest set-up of PollyXT allows improved coverage of the marine boundary layer (MBL) due to an additional near-range receiver. Three case studies provide an overview of the aerosol detected over the Atlantic Ocean. In the first case, marine conditions were observed near South Africa on the autumn cruise PS95. Values of optical properties (depolarisation ratios close to zero, lidar ratios of 23 sr at 355 and 532 nm) within the MBL indicate pure marine aerosol. A layer of dried marine aerosol, indicated by an increase of the particle depolarisation ratio to about 10 % at 355 nm (9 % at 532 nm) and thus confirming the non-sphericity of these particles, could be detected on top of the MBL. On the same cruise, an almost pure Saharan dust plume was observed near the Canary Islands, presented in the second case. The third case deals with several layers of Saharan dust partly mixed with biomass-burning smoke measured on PS98 near the Cabo Verde islands. While the MBL was partly mixed with dust in the pure Saharan dust case, an almost marine MBL was observed in the third case. A statistical analysis showed latitudinal differences in the optical properties within the MBL, caused by the down-mixing of dust in the tropics and anthropogenic influences in the northern latitudes, whereas the optical properties of the MBL in the Southern Hemisphere correlate with typical marine values. The particle depolarisation ratio of dried marine layers ranged between 4 and 9 % at 532 nm. Night measurements from PS95 and PS98 were used to illustrate the potential of aerosol classification using lidar ratio, particle depolarisation ratio at 355 and 532 nm, and Ångström exponent. Lidar ratio and particle depolarisation ratio have been found to be the main indicator for particle type, whereas the Ångström exponent is rather variable.

On the spectral depolarisation and lidar ratio of mineral dust provided in the AERONET version 3 inversion product

Knowledge of the particle lidar ratio (Sλ) and the particle linear depolarisation ratio (δλ) for different aerosol types allows for aerosol typing and aerosol-type separation in lidar measurements. Reference values generally originate from dedicated lidar observations but might also be obtained from the inversion of AERONET sun/sky radiometer measurements. This study investigates the consistency of spectral Sλ and δλ provided in the recently released AERONET version 3 inversion product for observations of undiluted mineral dust in the vicinity of major deserts: Gobi, Sahara, Arabian, Great Basin and Great Victoria deserts. Pure dust conditions are identified by an Ångstöm exponent

Ship-borne aerosol profiling with lidar over the Atlantic Ocean: From pure marine conditions to complex dust-smoke mixtures

The multiwavelength Raman lidar PollyXT have been regularly operated aboard the research vessel Polarstern on expeditions across the Atlantic Ocean from North to South and vice versa. The lidar measurements of the Polarstern cruises PS95 from Bremerhaven to Cape Town (November 2015) and PS98 from Punta Arenas to Bremerhaven (April/May 2016) are presented and analysed in detail. The latest setup of PollyXT allows improved coverage of the marine boundary layer (MBL) due to an additional near-range receiver. Three case studies provide an overview of the detected aerosol over the Atlantic Ocean. In the first case, marine conditions were observed near South Africa on the autumn cruise PS95. Values of optical properties (depolarisation ratios close to zero, lidar ratios of 23 sr at 355 nm and 532 nm) within the MBL indicate pure marine aerosol. A layer of dried marine aerosol, indicated by an increase of the particle depolarisation ratio to about 10 % at both wavelengths and thus confirming the non-sphericity of these particles, could be detected on the top the MBL. On the same cruise, an almost pure Saharan dust plume was observed near the Canary Islands, presented in the second case. The third case deals with several layers of Saharan dust partly mixed with biomass-burning smoke measured on PS98 near the Cape Verde Islands. While the MBL was partly mixed with dust in the pure Saharan dust case, an almost marine MBL was observed in the third case. A statistical analysis showed latitudinal differences in the optical properties within the MBL, caused by the down-mixing of dust in the tropics and anthropogenic influences in the northern latitudes whereas the optical properties of the MBL in the southern hemisphere correlate with typical marine values. The particle depolarisation ratio of dried marine layers ranged between 4–9 %. Night measurements from PS95 and PS98 were used to illustrate the potential of aerosol classification using lidar ratio, particle depolarisation ratio and Ångström exponent. Lidar ratio and particle depolarisation ratio have been found to be the main indicator for the particle type, whereas the Ångström exponent is rather variable.