This dataset contains over 25,000 human-classified images of snowflakes sorted into the following categories:

AG (aggregate)

CC (columnar crystal)

GR (graupel)

PC (planar crystal)

SP (small particle)

Filenames are in the following format:

YYYY.MM.DD_HH.MM.SS_flake_X_cam_Y_candidate_Z.png

YYYY.MM.DD_HH.MM.SS: datetime raw image was collected

X: trigger event

Y: MASC system camera index that produced raw image

Z: detection index in raw image (multiple flakes are detected in most raw images)

We request that all users of this dataset reference this Zenodo entry and the accompanying JTECH paper:

Key, C. et al. (2021) Advanced Deep Learning-Based Supervised Classification of Multi-Angle Snowflake Camera Images, JTECH

This microwave scattering database comprises the scattering properties of over 7 000 000 aggregates with varying monomer shapes. It contains the Mueller and Amplitude scattering matrix entries at 0 and 180° scattering angles relevant for all radar applications at 5.6, 9.6, 35.6 and 94 GHz and radar elevation angles ranging from -90 to 90°. The scattering properties were calculated using the discrete dipole approximation (implemented in the ADDA code: https://github.com/adda-team/adda). This database can be interfaced to the radar forward operator McRadar (https://github.com/lterzi/McRadar/tree/stochastic_aggregates) or used as a look-up table for any other radar forward operator. More information can be found in the publication "On the geometry of aggregate snowflakes" by Axel Seifert, Fabian Jakub, Christoph Siewert, Leonie von Terzi and Stefan Kneifel.

Organisms have increased in complexity through a series of major evolutionary transitions, in which formerly autonomous entities become parts of a novel higher-level entity. One intriguing feature of the higher-level entity after some major transitions is a division of reproductive labor among its lower-level units in which reproduction is the sole responsibility of a subset of units. Although it can have clear benefits once established, it is unknown how such reproductive division of labor originates. We consider a recent evolution experiment on the yeast Saccharomyces cerevisiae as a unique platform to address the issue of reproductive differentiation during an evolutionary transition in individuality. In the experiment, independent yeast lineages evolved a multicellular “snowflake-like” cluster formed in response to gravity selection. Shortly after the evolution of clusters, the yeast evolved higher rates of cell death. While cell death enables clusters to split apart and form new groups, it also reduces their performance in the face of gravity selection. To understand the selective value of increased cell death, we create a mathematical model of the cellular arrangement within snowflake yeast clusters. The model reveals that the mechanism of cell death and the geometry of the snowflake interact in complex, evolutionarily important ways. We find that the organization of snowflake yeast imposes powerful limitations on the available space for new cell growth. By dying more frequently, cells in clusters avoid encountering space limitations, and, paradoxically, reach higher numbers. In addition, selection for particular group sizes can explain the increased rate of apoptosis both in terms of total cell number and total numbers of collectives. Thus, by considering the geometry of a primitive multicellular organism we can gain insight into the initial emergence of reproductive division of labor during an evolutionary transition in individuality.

This paper presents the development of a physics-based multiple-input-multiple-output algorithm for real-time feedback control of snowflake divertor (SFD) configurations on the National Spherical Torus eXperiment Upgrade (NSTX-U). A model of the SFD configuration response to applied voltages on the divertor control coils is first derived and then used, in conjunction with multivariable control synthesis techniques, to design an optimal state feedback controller for the configuration. To demonstrate the capabilities of the controller, a nonlinear simulator for axisymmetric shape control was developed for NSTX-U which simultaneously evolves the currents in poloidal field coils based upon a set of feedback-computed voltage commands, calculates the induced currents in passive conducting structures, and updates the plasma equilibrium by solving the free-boundary Grad-Shafranov problem. Closed-loop simulations demonstrate that the algorithm enables controlled operations in a variety of SFD configurations and provides capabilities for accurate tracking of time-dependent target trajectories for the divertor geometry. In particular, simulation results suggest that a time-varying controller which can properly account for the evolving SFD dynamical response is not only desirable but necessary for achieving acceptable control performance. The algorithm presented in this paper has been implemented in the NSTX-U Plasma Control System in preparation for future control and divertormore » physics experiments.« less

The wealth and income Lorenz Curves are a fractal phenomena and is best demonstrated using the (Koch Snowflake) fractal.Lorenz distribution is a universal phenomena, inherent in all systems fractal and at all scales. Income and wealth distribution are just two manifestations of this structure and thus is not directly determined by Economic output or growth - they are 'natural'.Publication pending.