ETC2019 17th European Turbulence Conference Submission

16:15 Geophysical and Astrophysical Turbulence 1

16:15 15 mins #119	Effects of Droplet Sedimentation and Wind Shear on Cloud-Top Entrainment Bernhard Schulz, Juan Pedro Mellado Abstract: Stratocumulus clouds are very efficient in cooling the Earth's atmosphere and therefore they are sometimes loosely referred to as 'climate refrigerators' [1}. However, understanding and quantifying the effect of stratocumulus on the Earth's climate remains challenging and one important reason therefor is the limited understanding of cloud-top entrainment. Although it is well understood that droplet sedimentation weakens the entrainment velocity and wind shear enhances the entrainment, there is no consensus on the relevance of each process {2,3}. This poses the question whether wind-shear and droplet-sedimentation effects on cloud-top entrainment can compensate each other. Here, we address this question and study the joint effect of wind shear and droplet sedimentation on cloud-top entrainment in stratocumulus by means of direct numerical simulations, resolving the relevant meter- and submeter-scales. We find that the entrainment reduction by droplet sedimentation is sufficiently strong to completely compensate the entrainment enhancement by wind shear, and thus droplet-sedimentation and wind-shear effects can be equally important for cloud-top entrainment. For instance, for the subtropical conditions considered here, droplet sedimentation alone weakens the entrainment velocity by up to 40\%, while wind shear alone enhances the entrainment velocity by up to 40\% (see Figure 1}). We further find that the sedimentation weakening of the entrainment velocity is nearly shear independent, while the shear enhancement of the entrainment velocity can moderately depend on sedimentation. To quantify these effects an integral analysis of the buoyancy evolution equation is performed, which allows us to decompose the entrainment velocity into separate contributions from mixing, radiative, and evaporative cooling. In particular, this integral analysis shows, that the mechanism by which wind-shear and droplet-sedimentation effects compensate each other strongly depends on the choice of the reference height z_i, which is used to calculate the entrainment velocity w_e = d z_i/d t. In sum, these results imply that the droplet size distribution can substantially affect cloud lifetimes not only because of its effect on rain formation but also because of its effect on cloud-top entrainment. The importance of droplet-sedimentation for cloud-top entrainment emphasizes the need for precise measurements of the droplet size distribution and shows that we need to improve our understanding of how droplet sedimentation interacts with turbulence near cloud boundaries. References: [1] Christopher S Bretherton, Taneil Uttal, Christopher W Fairall, Sandra E Yuter, Robert A Weller, Darrel Baumgardner, Kimberly Comstock, Robert Wood, and Graciela B Raga. The EPIC 2001 stratocumulus study. Bulletin of the American Meteorological Society, 85(7):967–978, 2004. [2] Alberto de Lozar and Juan Pedro Mellado. Reduction of the entrainment velocity by cloud droplet sedimentation in stratocumulus. Journal of the Atmospheric Sciences, 74(3):751–765, 2017. [3] Bernhard Schulz and Juan Pedro Mellado. Wind Shear Effects on Radiatively and Evaporatively Driven Stratocumulus Tops. Journal of the Atmospheric Sciences, 75(9):3245–3263, 2018.
16:30 15 mins #329	SCALE INVARIANT DIFFUSION PARAMETERIZATION IN A MECHANISTIC GENERAL CIRCULATION MODEL Serhat Serhat Can, Urs Schafer-Rolffs, Erich Becker Abstract: Observations by Nastrom & Gage in the upper troposphere showed the existence of continuous energy and enstrophy cascades of Kinetic Energy (KE) across the scales with regard to horizontal wavenumber, and as a result the well known spectral laws of -3 for synoptic and -5/3 for mesoscales were found. Along with the observations, the concept of Lorenz Energy Cycle explains the continuous generation of KE from the conversion of Available Potential Energy (APE) and provides a theoretical framework on how APE is converted into KE then KE is filling the Unavailable Potential Energy (UPE, i.e. TPE - APE) reservoir via irreversible dissipation and finally the generation of APE from UPE. We use the high horizontal and vertical resolution Kühlungsborn Mechanistic general Circulation Model (KMCM) to obtain a realistic KE spectrum without employing any numerical filters or explicit hyperdiffusion. Instead , we parameterize horizontal and vertical momentum diffusion with a newly-developed anisotropic version of the so-called Dynamic Smagorinsky Model (DSM). This scheme takes into consideration the hydrodynamic conservation laws and is also fully consistent with scale invariance. It should be noted that a macro turbulent inertial regime requires the scale invariance for the formulation of such a scheme. In the present study, we completed the DSM with the horizontal diffusion of sensible heat, of which a similar cascade is also observed, see Figure 1. Our focus is on the spectra of the simulated KE and APE and their corresponding budgets. By looking at the contributing terms to these budgets, we aim to provide a picture of the irreversible branches of Lorenz Cycle in the free atmosphere.
16:45 15 mins #305	Fractal reconstruction of sub-grid scales for particle dispersion in large eddy simulation Emmanuel Akinlabi, Marta Wacławczyk, Szymon Malinowski, Juan-Pedro Mellado Abstract: Particle dispersion in turbulent flows have attracted the attention of researchers in recent times due to its wide range of application such as pollutant dispersion, spray dynamics, cloud dynamics etc. Direct numerical simulation (DNS) is the most detailed research tool, used to accurately predict particle trajectories in turbulent flows. For high-Reynolds number flow, this tool requires unrealistic computational resources and as a compromise, Large eddy simulation (LES) is used. LES provides a better representation of large-scale features of the flow while sub-grid (unresolved) scales are modelled. Sub-grid scale model errors can lead to the progressive divergence of particle trajectories when compared with those obtained in experiment or DNS [2]. As a result, particle statistics are either over- or under-estimated [1, 4]. The present work addresses the reconstruction of sub-grid scales in large eddy simulation (LES) of turbulent dispersed flows. We focus on fractal sub-grid model, which is based on the fractality assumption of turbulent velocity field. The fractal model reconstructs sub-grid velocity field from known filtered values on LES grid, by means of fractal interpolation, proposed by Scotti and Meneveau [6]. The characteristics of the reconstructed signal depend on the (free) stretching parameter d, which is related to the fractal dimension of the signal. In [6], the stretching parameter was assumed to be constant in space and time and are obtained from experimental data of homogeneous and isotropic turbulence. However, turbulence at moderate or high-Reynolds’ number possesses intermittency at small scales, which lead to strong variability in its local stretching parameter. To account for the stretching parameter variability, we calculate the probability distribution function of the local stretching parameter from DNS data of stratocumulus top boundary layer (STBL) [5] using an algorithm proposed by Mazel and Hayes [3]. We observe self-similarity in the PDFs of d when the velocity fields are filtered to wave-numbers within the inertial range (see figure 1). By randomly selecting d from its self-similar PDF, we perform a 1D a priori test and compare statistics of the constructed velocity increments with statistics of DNS velocity increments (see figure 2). This idea was applied to Physics of stratocumulus top (POST) airborne data and 3-D LES velocity fields. We observed that the constructed sub-grid scale velocity fields with the random values of d are able to reproduce most of the sub-grid scales and give smaller error in mass conservation when compared to the use of constant values of d.
17:00 15 mins #116	Reactive species in turbulence Wenwei Wu, Lipo Wang, Enrico Calzavarini, Francois Schmitt, Michael Gauding Abstract: In the marine system, many micro-biological species, such as phytho- or zooplanktons, are under the influence of both the turbulent dynamics of their living fluid environment and the reactive interaction between the species or species and nutrition distribution. Such kind of constant density reactive turbulence problem is special in the sense that it is different from either the nonreactive passive scalar in incompressible flows~\cite{Corrsin, Warhaft} or the reactive combustion flows with large change of density~\cite{Nilan}. The specificities of this case are poorly understood yet. \\In this study, a second order reaction is introduced in the form of $R_1+2R_2 \Leftrightarrow P$ with three reactants ($R_1$, $R_2$ and $P$). Direct numerical simulations of isotropic turbulence were performed with the Taylor based Reynolds number about $200$. The Damkholer number $Da$, which is defined as the ratio between the time-scale of the flow (here the dissipative time-scale) and the the characteristic reaction time-scale, is the key parameter to characterize the effects of reaction. At small $Da$, the scalar statistics is close to those of the nonreactive passive cases because of the weak influence from reaction. At very large $Da$, i.e. the limit of a fast reaction, the large scale structure in the scalar field $R_2$ displays sharp front structures as shown by $\nabla R_2$ in Fig.~\ref{totalfig} (a). Interestingly, the nonlinearity has important effects on the $R_2$ scalar. Fig.~\ref{totalfig} (b) demonstrates that with the increase of $Da$ the probability density functions (PDF) of the nonlinear scalar $R_2$ approach to be symmetrical, while for $R_1$ and $P$ the PDFs are not strongly influenced by $Da$. Interestingly, in spite of the noticeable difference from the scalar PDF, other properties, such as the spectra, structure functions and their scaling laws, are only weakly affected.
17:15 15 mins #43	Pair Dispersion in Canopy Flow Turbulence Ron Shnapp, Yardena Bohbot-Raviv, Eyal Fattal, Alex Liberzon Abstract: The interaction between the Earths' surface and the atmosphere occurs, in part, through the turbulent flow dynamics of the air in the atmospheric surface layer, so-called canopy flows. Therefore, it is important to understand turbulent dispersion and mixing in canopy flows, in the context of modeling of air pollution, or fluxes of heat, momentum, and admixtures (Britter & Hanna (2003)). In the Lagrangian framework, fluid properties are prescribed at moving fluid particle positions, making it a natural approach to investigations of turbulent dispersion and transport phenomena (Toschi & Bodenschatz (2009)). However, the direct interaction of high momentum air flow with Earths' rough terrain generates highly inhomogeneous flow statistics, with a characteristic vertically asymmetrical momentum transport that seems to be dominated by large scale coherent structures. To this end, the effect of inhomogeneity and anisotropy on the dispersion of Lagrangian fluid particles is yet unclear. We study Lagrangian dispersion within a wind tunnel canopy flow, through the three-dimensional particle tracking velocimetry (3D-PTV) method. An application of real-time image analysis on hardware (Shnapp et al. (2019), Bohbot-Raviv et al. (2017)) allowed us to capture millions of tracer particles' trajectories (Fig.1), sufficient to estimate statistical moments of Lagrangian translations. The obtained dataset allows to directly estimate 3D dispersion of Lagrangian tracer particles, inside and immediately above the canopy layer. Therefore, we can now extract dispersion parameters, and extend and validate existing models, without relying on prior simplifications or premises. Mixing and spreading of physical quantities is strongly related to the mutual dynamics of pairs of tracer particles; the relative position dynamics are studied through the so-called pair dispersion (Sawford (2001)). We investigate the pair dispersion empirically through the Lagrangian dataset in the canopy flow model. At the short time scales, we observe behavior that is very similar to what is seen in the homogeneous turbulent flows - a Ballistic regime that is governed by the Eulerian distribution of spatial velocity increments through the Eulerian structure function. Interestingly, we find that the turbulent flows' anisotropy has a very small effect on this short time behavior. At longer times, which approach the Lagrangian integral timescale, we observe a change in the scaling laws of pair dispersion that is strongly dependent on the initial separation distance; we examine this effect in the context of a recently proposed Lagrangian intermittency parameter (Shnapp & Liberzon (2018)), the results of which are currently under investigation. Lastly, we see observe that large flow scales have a significant effect on the rate of separation of Lagrangian particles.
17:30 15 mins #240	ANALYSIS OF THE TURBULENT ENERGY SPECTRA OBTAINED DURING THEWADIS-2 SOUNDING ROCKET CAMPAIGN Victor Avsarkisov, Boris Strelnikov Abstract: In the present study we analyze recent sounding rocket mesospheric measurements of fluctuations in density of neutral air obtained during WADIS-2 sounding rocket campaign conducted in March 2015 at the Andøya Space Center ($16^\circ$E, $69^\circ$N). A detailed description of WADIS mission and overview of measured parameters can be found in Strelnikov (2017).
17:45 15 mins #550	Unravelling wave-vortex interactions and geophysical turbulence phenomenology at oceanic mesoscales Jim Thomas Abstract: The oceanic mesoscale, with scales of the order of 100 kms, is constrained by the ef- fects of rapid rotation and strong stratification. Consequently, the flow field at these scales is an intertwined mix of fast evolving internal gravity waves and slow evolving vortices, popularly known as mesoscale eddies. Well established geophysical turbulence phenomenology at these scales completely ignores internal gravity waves, although fast evolving waves are commonly found along with eddies in oceanic observations. In this talk I will present a new set of interactions between fast internal gravity waves and slow vortical fields in the limit of rapid rotation and strong stratification. The findings point in the direction of a new geophysical turbulence phenomenology that involves significant energy exchange between waves and the vortical field.
18:00 15 mins #464	Toward Understanding the Multi-Scale Coupling in Global Oceanic Flows Hussein Hussein Aluie, Mahmoud Sadek, Chayut Teeraratkul, Matthew Hecht, Geoffrey Vallis Abstract: Large-scale currents and eddies pervade the ocean and play a prime role in the general circulation and climate. The coupling between scales ranging from $O(10^4)$ km down to $O(1)$ mm presents a major difficulty in understanding, modeling, and predicting oceanic circulation and mixing, where our constraints on the energy budget suffer from large uncertainties. Identifying the energy sources and sinks at various scales and geographic locations can reduce such uncertainty and yield insight into new parameterizations of nonlinear physical processes. To this end, we develop the coarse-graining framework to analyze the multi-scale dynamics on the sphere. This is made possible by generalizing the definition of convolution to ensure that our filtering operators and spatial derivatives on the sphere commute, thereby allowing us to derive the PDEs governing any sets of scales. The approach is very general, allows for probing the dynamics simultaneously in scale and in space, and is not restricted by usual assumptions of homogeneity or isotropy. We will demonstrate the application of this framework to satellite altimetry data and to strongly eddying high-resolution simulations using General Circulation Models. This work was supported by NSF Grant OCE-1259794 and NASA grant 80NSSC18K0772.