HoloGondel: in situ cloud observations on a cable car in the Swiss Alps using a holographic imager

Beck, Alexander; Henneberger, Jan; Schöpfer, Sarah; Fugal, Jacob P.; Lohmann, Ulrike

doi:10.5194/amt-10-459-2017

Cited by 36 publications

(63 citation statements)

References 50 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A random forest model (Breiman, 1996(Breiman, , 2001 constitutes an ensemble of decision trees that are combined through averaging over multiple individual trees. Decision tree approaches have previously been used for cloud particle classification specifically (e.g., Garimella et al, 2016;Bernauer et al, 2016) and are frequently used for any classification prob-lems, such as data from single particle mass spectrometry (e.g., Christopoulos et al, 2018). Ruske et al (2018) recently identified best performance of random forest models when comparing different ML approaches to classify different particle types.…”

Section: Particle Shape Classification: Supervised Machine Learningmentioning

confidence: 99%

“…A major limitation of these devices is the low frame rate on the order of a few tens of particles per second, which limits the number of single particles sampled. This becomes problematic when hydrometeors of different phase are inhomogeneously spatially distributed within clouds (Korolev and Isaac, 2006;Beck et al, 2017). Besides, high frame rates are beneficial for laboratory studies, where particle number concentration easily exceeds a few tens of particles per cubic centimeter, and low frame rates lead to coincidence errors even at moderate number concentrations (Cotton et al, 2010).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A high-speed particle phase discriminator (PPD-HS) for the classification of airborne particles, as tested in a continuous flow diffusion chamber

et al. 2019

View full text Add to dashboard Cite

A new instrument, the High-speed Particle Phase Discriminator (PPD-HS), developed at the University of Hertfordshire, for sizing individual cloud hydrometeors and determining their phase is described herein. PPD-HS performs an in situ analysis of the spatial intensity distribution of near-forward scattered light for individual hydrometeors yielding shape properties. Discrimination of spherical and aspherical particles is based on an analysis of the symmetry of the recorded scattering patterns. Scattering patterns are collected onto two linear detector arrays, reducing the complete 2-D scattering pattern to scattered light intensities captured onto two linear, one-dimensional strips of light sensitive pixels. Using this reduced scattering information, we calculate symmetry indicators that are used for particle shape and ultimately phase analysis. This reduction of information allows for detection rates of a few hundred particles per second.Here, we present a comprehensive analysis of instrument performance using both spherical and aspherical particles generated in a well-controlled laboratory setting using a vibrating orifice aerosol generator (VOAG) and covering a size range of approximately 3-32 µm. We use supervised machine learning to train a random forest model on the VOAG data sets that can be used to classify any particles detected by PPD-HS. Classification results show that the PPD-HS can successfully discriminate between spherical and aspherical particles, with misclassification below 5 % for diameters > 3 µm. This phase discrimination method is subsequently applied to classify simulated cloud particles produced in a continuous flow diffusion chamber setup. We report observations of small, near-spherical ice crystals at early stages of the ice nucleation experiments, where shape analysis fails to correctly determine the particle phase. Nevertheless, in the case of simultaneous presence of cloud droplets and ice crystals, the introduced particle shape indicators allow for a clear distinction between these two classes, independent of optical particle size. From our laboratory experiments we conclude that PPD-HS constitutes a powerful new instrument to size and discriminate the phase of cloud hydrometeors. The working principle of PPD-HS forms a basis for future instruments to study microphysical properties of atmospheric mixed-phase clouds that represent a major source of uncertainty in aerosol-indirect effect for future climate projections.

show abstract

Section: Particle Shape Classification: Supervised Machine Learningmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

A high-speed particle phase discriminator (PPD-HS) for the classification of airborne particles, as tested in a continuous flow diffusion chamber

et al. 2019

View full text Add to dashboard Cite

show abstract

“…Siebert et al 2006), cable cars (e.g. Beck et al 2017), tethered balloon systems (TBS) (e.g. Siebert et al 2003, Maletto et al 2003, Lawson et al 2011, Sikand et al 2013, Canut et al 2016 or launched balloon platforms (e.g.…”

Section: Introductionmentioning

confidence: 99%

Using a holographic imager on a tethered balloon system for microphysical observations of boundary layer clouds

Ramelli¹,

Beck²,

Henneberger³

et al. 2019

Preprint

Self Cite

View full text Add to dashboard Cite

Conventional techniques to measure boundary layer clouds such as research aircrafts are unable to sample in orographic or densely-populated areas. In this paper, we present a newly developed measurement platform on a tethered balloon system (HoloBalloon) to measure in situ vertical profiles of microphysical and meteorological cloud properties up to 1 kilometer above ground. The main component of the HoloBalloon platform is a holographic imager, which uses digital in-line holography to image cloud particles in a velocity independent sample volume, making it particularly well suited for measure-5 ments on a balloon. The unique combination of holography and balloon-borne measurements allows observations with high spatial resolution, covering cloud structures from the kilometer down to the millimeter scale.We present observations of a supercooled low stratus cloud (high fog event) during a Bise situation over the Swiss Plateau in February 2018. In situ microphysical profiles up to 700 m altitude above the ground and at temperatures down to -8°C and wind speeds up to 15 m s −1 were performed. We were able to capture unique microphysical features from the kilometer down to the 10 meter scale. For example, we observed cloud regions with decreased cloud droplet number concentration (<0.5 CDN C) and cloud droplet size at scales of 30-50 meters. These cloud inhomogeneities could arise from adiabatic compression and heating and subsequent droplet evaporation in descending air parcels. Moreover, we observed conditions favorable for the formation of boundary layer waves and Kelvin-Helmholtz instability at the cloud top. This potentially influenced the cloud structure on a scale of 10-30 kilometers, which is reflected in the variability of the CDNC. 15

show abstract

“…Cloud particle imaging probes (e.g. CPI (Lawson et al, 2001), CIP (Baumgardner et al, 2001) and PHIPS-HALO (Abdelmonem et al, 2011) directly capture a 2D particle image of single cloud particles, whereas digital in-line holography (e.g., HOLODEC, , HOLIMO 2, Henneberger et al (2013), HOLIMO 3G, Beck et al (2017) and…”

mentioning

confidence: 99%

“…Particles between the lens and the camera scatter the light which interferes with the plane reference wave. The interference pattern is captured by the camera as a hologram from which images in different distances to the camera can be reconstructed.Figure adaptedfromBeck et al (2017).…”

mentioning

confidence: 99%

A Convolutional Neural Network for Classifying Cloud Particles Recorded by Imaging Probes

Touloupas¹,

Lauber²,

Henneberger³

et al. 2019

Preprint

Self Cite

View full text Add to dashboard Cite

During typical field campaigns, millions of cloud particle images are captured with imaging probes. Our interest lies in classifying these particles in order to compute the statistics needed for understanding clouds. Given the large volume of collected data, this raises the need for an automated classification approach. Traditional classification methods that require extracting features manually (e.g. decision trees and support vector machines) show reasonable performance when trained and tested on data 5 coming from a unique dataset. However, they often have difficulties to generalize to test sets coming from other datasets where the distribution of the features might be significantly different. In practice, we found that each new dataset requires labeling a huge amount of data by hand using those methods. Convolutional neural networks have the potential to overcome this problem due to their ability to learn complex non-linear models directly from the images instead of pre-engineered features, as well as by relying on powerful regularization techniques. We show empirically that a convolutional neural network trained on cloud 10 particles from holographic imagers generalizes well to unseen datasets. Moreover, fine-tuning the same network with a small number (256) of training images improves the classification accuracy. Thus, the automated classification with a convolutional neural network not only reduces the hand-labeling effort for new datasets but is also no longer the main error source for the classification of small particles.

show abstract

HoloGondel: in situ cloud observations on a cable car in the Swiss Alps using a holographic imager

Cited by 36 publications

References 50 publications

A high-speed particle phase discriminator (PPD-HS) for the classification of airborne particles, as tested in a continuous flow diffusion chamber

A high-speed particle phase discriminator (PPD-HS) for the classification of airborne particles, as tested in a continuous flow diffusion chamber

Using a holographic imager on a tethered balloon system for microphysical observations of boundary layer clouds

A Convolutional Neural Network for Classifying Cloud Particles Recorded by Imaging Probes

Contact Info

Product

Resources

About