Remo Dietlicher


Last Name

Dietlicher

First Name

Remo

ORCID

0000-0003-0217-7232

Organisational unit

Filters

2018 - 201952020 - 20261
Reset filters

Search Results

Publications1 - 6 of 6
  • Operational numerical weather prediction with ICON on GPUs (version 2024.10)
    Item type: Journal Article
    Lapillonne, Xavier; Hupp, Daniel; Gessler, Fabian; et al. (2026)
    Geoscientific Model Development
    Numerical weather prediction and climate models require continuous adaptation to take advantage of advances in high-performance computing hardware. This paper presents the port of the ICON model to GPUs using OpenACC compiler directives for numerical weather prediction applications. In the context of an end-to-end operational forecast application, we adopted a full-port strategy: the entire workflow, from physical parameterizations to data assimilation, was analyzed and ported to GPUs as needed. Performance tuning and mixed-precision optimization yield a 5.5x speed-up compared to the CPU baseline in a socket-to-socket comparison. The ported ICON model meets strict requirements for time-to-solution and meteorological quality, in order for MeteoSwiss to be the first national weather service to run ICON operationally on GPUs with its ICON-CH1-EPS and ICON-CH2-EPS ensemble forecasting systems. We discuss key performance strategies, operational challenges, and the broader implications of transitioning community models to GPU-based platforms.
  • A Modeling Study on the Sensitivities of Atmospheric Charge Separation According to the Relative Diffusional Growth Rate Theory to Nonspherical Hydrometeors and Cloud Microphysics
    Item type: Journal Article
    Glassmeier, Franziska; Arnold, L.; Dietlicher, Remo; et al. (2018)
    Journal of Geophysical Research: Atmospheres
    Collisional charge transfer between graupel and ice crystals in the presence of cloud droplets is considered the dominant mechanism for charge separation in thunderclouds. According to the relative diffusional growth rate (RDGR) theory, the hydrometeor with the faster diffusional radius growth is charged positively in such collisions. We explore sensitivities of the RDGR theory to nonspherical hydrometeors and six parameters (pressure, temperature, liquid water content, sizes of ice crystals, graupel, and cloud droplets). Idealized simulations of a thundercloud with two‐moment cloud microphysics provide a realistic sampling of the parameter space. Nonsphericity and anisotropic diffusional growth strongly control the extent of positive graupel charging. We suggest a tuning parameter to account for anisotropic effects not represented in bulk microphysics schemes. In a susceptibility analysis that uses automated differentiation, we identify ice crystal size as most important RDGR parameter, followed by graupel size. Simulated average ice crystal size varies with temperature due to ice multiplication and heterogeneous freezing of droplets. Cloud microphysics and ice crystal size thus indirectly determine the structure of charge reversal lines in the traditional temperature‐water‐content representation. Accounting for the variability of ice crystal size and potentially habit with temperature may help to explain laboratory results and seems crucial for RDGR parameterizations in numerical models. We find that the contribution of local water vapor from evaporating rime droplets to diffusional graupel growth is only important for high effective water content. In this regime, droplet size and pressure are the dominant RDGR parameters. Otherwise, the effect of local graupel growth is masked by small ice crystal sizes that result from ice multiplication.
  • Ice clouds: from ice crystals to their response in a warming climate
    Item type: Doctoral Thesis
    Dietlicher, Remo (2018)
    Atmospheric ice crystals form from a variety of different sources and at different temperatures. Between 0 °C and -38 °C, liquid water and ice crystals can coexist. Cloud ice initiated by freezing of cloud droplets at these temperatures needs to be catalyzed by an ice nucleating particle (INP). Further growth then often involves collisions of ice crystals and cloud droplets or depositional growth of the ice crystals at the expense of the cloud droplets due to the lower water vapor pressure over the ice crystal than over the liquid droplet surface. At temperatures colder than -38 °C, ice can not only originate from freezing of cloud droplets, but also by freezing of deliquesced aerosols and direct deposition of water vapor onto an INP. The complexity of the ice formation processes is reflected in the spread of simulated cloud ice contents in the current generation of global climate models (GCM). This work describes the implementation and first results of a new cloud microphysics scheme in the ECHAM6-HAM2 GCM aimed to reduce the number of weakly constrained parameters involved in the representation of cloud ice formation and evolution. It does no longer rely on heuristic conversion rates between in-cloud ice crystals and precipitating snow but uses only one single, prognostic ice category which better represents the spectrum of ice crystals in clouds. Because precipitating snow is no longer diagnosed, the trajectory of ice crystals must be fully prognostic. Numerical stability of vertical advection is achieved by an adaptive time step in the microphysics routine which leads to an increase in computation time of roughly 25%. The new scheme significantly reduces the conceptual complexity of the model. Tuning parameters for the ice crystal fall speeds and the conversion to snow are no longer needed. With the introduction of a new cloud cover parameterization the high bias of high cloud cover in the base model version ECHAM6.3-HAM2.3 could be reduced. Overall, the new model is in reasonable agreement with observations in key variables while some deficiencies remain. New model diagnostics are introduced to disentangle the relative importance of ice formation pathways to provide a sound cause-and-effect relation between the simulated cloud fields and the process parameterizations. This analysis revealed that immersion and contact freezing in supercooled liquid clouds only dominate ice formation in roughly 5% of the simulated clouds, a small fraction compared to roughly 64% of the clouds governed by freezing in the cirrus temperature regime below -38 °C. Furthermore, we could demonstrate that even in the mixed-phase temperature regime between -38 °C and 0 °C, the dominant source of ice is the sedimentation of ice crystals that originated in the cirrus regime. The new scheme is used to assess changes in the cloud fields in response to a warming climate. The equilibrium response of the global mean surface temperature to an instantaneous doubling of atmospheric carbon dioxide concentrations is found to be 3.8 °C which is within the spread of the current generation of GCMs but substantially larger than the base model version ECHAM6.3-HAM2.3 with a value of 2.5 °C. This difference could be narrowed down to different cloud optical depth feedbacks and needs further investigation. Even though clouds are predominantly glaciated already below temperatures of roughly -5 °C, the cloud phase feedback is suppressed. Since most cloud ice is formed in clouds with a large vertical extent and high optical thickness, phase transitions do not significantly increase the optical depth of the cloud.
  • A high-speed particle phase discriminator (PPD-HS) for the classification of airborne particles, as tested in a continuous flow diffusion chamber
    Item type: Journal Article
    Mahrt, Fabian; Wieder, Jörg; Dietlicher, Remo; et al. (2019)
    Atmospheric Measurement Techniques
    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.
  • A High Speed Particle Phase Discriminator (PPD-HS) for the classification of airborne particles, as tested in a continuous flow diffusion chamber
    Item type: Working Paper
    Mahrt, Fabian; Wieder, Jörg; Dietlicher, Remo; et al. (2019)
    Atmospheric Measurement Techniques Discussions
    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 2D 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 micron. 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 micro meter. 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 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. We conclude that PPD-HS constitutes a powerful new instrument to size and discriminate phase of cloud hydrometeors and thus study microphysical properties of mixed-phase clouds, that represent a major source of uncertainty in aerosol indirect effect for future climate projections.
  • Elucidating ice formation pathways in the aerosol-climate model ECHAM6-HAM2
    Item type: Journal Article
    Dietlicher, Remo; Neubauer, David; Lohmann, Ulrike (2019)
    Atmospheric Chemistry and Physics
    Cloud microphysics schemes in global climate models have long suffered from a lack of reliable satellite observations of cloud ice. At the same time there is a broad consensus that the correct simulation of cloud phase is imperative for a reliable assessment of Earth's climate sensitivity. At the core of this problem is understanding the causes for the inter-model spread of the predicted cloud phase partitioning. This work introduces a new method to build a sound cause-and-effect relation between the microphysical parameterizations employed in our model and the resulting cloud field by analysing ice formation pathways. We find that freezing processes in supercooled liquid clouds only dominate ice formation in roughly 6 % of the simulated clouds, a small fraction compared to roughly 63 % of the clouds governed by freezing in the cirrus temperature regime below −35 ∘C. This pathway analysis further reveals that even in the mixed-phase temperature regime between −35 and 0 ∘C, the dominant source of ice is the sedimentation of ice crystals that originated in the cirrus regime. The simulated fraction of ice cloud to total cloud amount in our model is lower than that reported by the CALIPSO-GOCCP satellite product. This is most likely caused by structural differences of the cloud and aerosol fields in our model rather than the microphysical parametrizations employed.
Publications1 - 6 of 6