Hands-On Research in Complex Systems Schools provide an example of how graduate students and young faculty working in resource-constrained environments can apply key mindsets and methods of tabletop experiments to problems at the frontiers of science. Each day during the Schools’ two-week program, participants work in small groups with experienced tabletop scientists in interactive laboratories on topics drawn from diverse disciplines in science and technology. Using modern low-cost tools, participants run experiments and perform associated data analysis together with mathematical and computational modeling. Participants also engage in other scientific professional activities; in particular, they learn best practices for communicating their results visually, orally, and in writing. In this way, the Hands-On Schools foster the development of scientific leaders in low- and middle-income countries.

Abstract A bolus is a vortex formed by an internal wave propagating upslope with the shoaling wave. The amount of sediment and biota transported by boluses travelling up the world's continental slopes and onto the continental shelves has not been established but could be vital for maintaining nutrient‐rich coastal ecosystems. Previous laboratory experiments and simulations of shoaling boluses have considered a two‐layer density stratification or a linear density profile. We present experiments on bolus formation and propagation in a stratification with a density profile ρ(z) represented by a tanh‐profile, which describes many ocean pycnoclines. Our laboratory experiments examine pycnoclines with thicknesses varying from nearly zero (a two‐layer system) to large thicknesses where the density varies almost linearly with depth. The bolus volume, displacement upslope, and available potential energy are measured as a function of pycnocline thickness. Maximum upslope displacement is found to be twice that observed for a two‐layer stratification.

Weiwei Jin, Corey S. O’Hern, 2, 3 Charles Radin, Mark D. Shattuck, and Harry L. Swinney Department of Mechanical Engineering and Materials Science, Yale University, New Haven, Connecticut 06520, USA Department of Physics, Yale University, New Haven, Connecticut 06520, USA Department of Applied Physics, Yale University, New Haven, Connecticut 06520, USA Department of Mathematics, University of Texas at Austin, Austin, Texas 78712, USA Benjamin Levich Institute and Physics Department, The City College of New York, New York, New York 10031, USA Center for Nonlinear Dynamics and Department of Physics, University of Texas at Austin, Austin, Texas 78712, USA

Many experiments over the past half century have shown that, for a range of protocols, granular materials compact under pressure and repeated small disturbances. A recent experiment on cyclically sheared spherical grains showed significant compaction via homogeneous crystallization (Rietz et al., 2018). Here we present numerical simulations of frictionless, purely repulsive spheres undergoing cyclic simple shear via Newtonian dynamics with linear viscous drag at fixed vertical load. We show that for sufficiently small strain amplitudes, cyclic shear gives rise to homogeneous crystallization at a volume fraction ϕ=0.646±0.001. This result indicates that neither friction nor gravity is essential for homogeneous crystallization in driven granular media. Understanding how crystal formation is initiated within a homogeneous disordered state gives key insights into the old open problem of glass formation in fluids.

An understanding of homogeneous nucleation of crystalline structure from a disordered medium such as a liquid remains an important unsolved problem in condensed matter physics. Guided by the results from a number of experiments on granular and colloidal systems in the past two decades, including in particular observations of homogeneous nucleation in colloidal and granular systems, we suggest an alternative to the statistical mechanics approach to static granular matter initiated by Edwards and Oakeshott in 1989.

Jerry Paul Gollub was a remarkably creative physicist who conducted foundational experiments in nonlinear dynamics, including the first observation and characterization of the transition from order to chaos in fluid systems. His work contributed greatly to the understanding of complex dynamical behavior. His lucid papers and lectures made him widely sought as a lecturer at universities and international conferences. Jerry Paul Gollub LISA J. GODFREY/HAVERFORD COLLEGEPPT|High-resolutionJerry was born in Saint Louis, Missouri, on 9 September 1944 and died in Haverford, Pennsylvania, on 8 June 2019. He graduated with an AB degree from Oberlin College in 1966 and with a PhD from Harvard University in 1971; his thesis adviser was Michael Tinkham. His primary appointment during the entirety of his professional career, from 1971 to 2012, was at Haverford College. In 1997 he was named the John and Barbara Bush Professor of Natural Sciences. His steadfast commitment to undergraduate education and to Haverford led him to decline offers of professorships at several major research universities.In his Haverford laboratory, Jerry mentored more than 100 undergraduates and worked with postdocs and with graduate students from the University of Pennsylvania, where he was an adjunct professor. Because of the world-class research he conducted, in 1986 he became the first recipient of the American Physical Society (APS) Prize for a Faculty Member for Research in an Undergraduate Institution. He also held visiting appointments at the University of Paris VII in 1985, École Normale Supérieure in 1991, and the Weizmann Institute of Science in 1997–98.My fondest memories of Jerry are from the period 1974–75, when we collaborated on experiments designed to test Lev Landau’s 1944 prediction that the transition to turbulence would occur through an infinite sequence of instabilities, each adding a new frequency to the motion, as the Reynolds number was increased. In our intense collaboration, Jerry and I developed a deep friendship and a research style and purpose that we sustained throughout our careers.Our experiments, conducted at the City College of New York, yielded time series of the fluid velocity measured at a point between concentric cylinders. As the inner cylinder rotation rate (proportional to the Reynolds number) was increased, power spectra of the velocity time series revealed a transition from time-independent flow to a state characterized by a single frequency, in accord with the Landau picture. At a higher Reynolds number, a second frequency component, incommensurate with the first, appeared in the power spectrum, just as Landau had anticipated. With further increase in the Reynolds number, however, the spectra contained increasing broadband noise but no additional discrete frequency components. The noisy behavior differed from the expected Landau scenario, but the observation was consistent with models and analyses of chaos developed in the 1960s and 1970s.In the 1980s Jerry and his group at Haverford developed a technique for visualizing spatial patterns in convecting fluids. With it, they investigated how competition between different spatiotemporal modes led to chaos. In the late 1980s, Jerry used particle-tracking methods to characterize chaotic mixing in time-periodic convective flows, the formation of fractal and dendritic structures in solidification processes, and pattern formation and chaos in surface waves. He also conducted a series of imaginative experiments on the dynamics of granular materials. In 2008–9 Jerry was a Leverhulme Visiting Professor at Cambridge University, where he collaborated with Raymond Goldstein, who had developed green algae as model organisms for biological fluid dynamics. Their experiments revealed diffusive yet non-Gaussian tracer statistics in suspensions of swimming microorganisms.From 2000 to 2002, Jerry served as cochair of the National Research Council committee that produced the study Learning and Understanding: Improving Advanced Study of Mathematics and Science in U.S. High Schools. In 2005–8 he served on the National Academy of Sciences governing council.Jerry was awarded the 2003 APS Fluid Dynamics Prize “for his elucidation of chaos, instabilities, mixing and pattern formation.” He also served for three decades in many APS elected and appointed positions, including on the council and the executive board.Jerry is greatly missed by his many friends, students, postdocs, and other collaborators throughout the world. His innovative experiments on the complex dynamics of systems driven away from thermodynamic equilibrium will have lasting influence. He will also be remembered as a tireless advocate for high school and undergraduate science education, especially in physics, and for his contributions to the general good of the broad scientific community.© 2019 American Institute of Physics.

An understanding of homogeneous nucleation of crystalline structure from a disordered medium such as a liquid remains an important unsolved problem in condensedmatter physics. Guided by the results from a number of experiments on granular and colloidal systems in the past two decades, including in particular observations of homogeneous nucleation in colloidal and granular systems, we suggest an alternative to the statistical mechanics approach to static granular matter initiated by Edwards and Oakeshott in 1989.

Internal gravity wave energy contributes significantly to the energy budget of the oceans, affecting mixing and the thermohaline circulation. Hence it is important to determine the internal wave energy flux $\boldsymbol{J}=p\,\boldsymbol{v}$, where $p$ is the pressure perturbation field and $\boldsymbol{v}$ is the velocity perturbation field. However, the pressure perturbation field is not directly accessible in laboratory or field observations. Previously, a Green’s function based method was developed to calculate the instantaneous energy flux field from a measured density perturbation field $\unicode[STIX]{x1D70C}(x,z,t)$, given a constant buoyancy frequency $N$. Here we present methods for computing the instantaneous energy flux $\boldsymbol{J}(x,z,t)$ for an internal wave field with vertically varying background $N(z)$, as in the oceans where $N(z)$ typically decreases by two orders of magnitude from the pycnocline to the deep ocean. Analytic methods are presented for computing $\boldsymbol{J}(x,z,t)$ from a density perturbation field for $N(z)$ varying linearly with $z$ and for $N^{2}(z)$ varying as $\tanh (z)$. To generalize this approach to arbitrary $N(z)$, we present a computational method for obtaining $\boldsymbol{J}(x,z,t)$. The results for $\boldsymbol{J}(x,z,t)$ for the different cases agree well with results from direct numerical simulations of the Navier–Stokes equations. Our computational method can be applied to any density perturbation data using the MATLAB graphical user interface ‘EnergyFlux’.

We present an experiment on crystallization of packings of macroscopic granular spheres. This system is often considered to be a model for thermally driven atomic or colloidal systems. Cyclically shearing a packing of frictional spheres, we observe a first order phase transition from a disordered to an ordered state. The ordered state consists of crystallites of mixed fcc and hcp symmetry that coexist with the amorphous bulk. The transition, initiated by homogeneous nucleation, overcomes a barrier at 64.5% volume fraction. Nucleation consists predominantly of the dissolving of small nuclei and the growth of nuclei that have reached a critical size of about ten spheres.

We examine numerically the conversion of barotropic tidal energy into internal waves by flow over an isolated seamount and over systems of periodically and randomly distributed 1100 m tall seamounts with Gaussian profiles. The simulations use the Massachusetts Institute of Technology general circulation model (MITgcm) to calculate for an infinitely deep ocean the dependence of the energy conversion on seamount slope, seamount separation, tidal direction, and the size and aspect ratio of the simulation domain. For neighboring seamounts with a slope greater than the internal wave beam slope, wave interference reduces the conversion relative to that calculated for an isolated seamount, and relative to that predicted by linear theory for a seamount of slope less than the beam slope. The conversion by an individual seamount in a system of random seamounts separated by an average distance of 18 km is found to be suppressed by 16% relative to the conversion by an isolated seamount. This study provides insight into tidal conversion by ocean seamounts modeled as Gaussian mountains with slopes both smaller and larger than the beam slope. We conclude that the total energy conversion by all seamounts (peak height >= 1000 m) and knolls (peak height 500-1000 m), taking into account interference affects, is of the order of 1% of the total barotropic to baroclinic energy conversion in the oceans, which is about twice as large as previous estimates.

The propagation of sound in a density-stratified fluid is examined in an experiment with a tank of salty water whose density increases continuously from the fluid surface to the tank bottom. Measurements of the height dependence of the fluid density are used to calculate the height dependence of the fluid salinity and sound speed. The height-dependent sound speed is then used to calculate the refraction of sound rays. Sound propagation in the fluid is measured in three dimensions and compared with the ray analysis. This study provides a basis for laboratory modeling of underwater sound propagation in the fluctuating stratified oceans.

The movie shows the nucleation of spheres in the shear cell. For visualization the spheres that are in a crystalline state are shown in a rotating side view. Color indicates the crystal symmetry. At the top right the current state is specified by the global packing fraction, shear cycle, and the number of crystalline spheres. The end of the plateau in the packing fraction is marked by the appearance of the first growing nucleus, which is encircled in the rotating animation and depicted at a constant viewing perspective on the lower right. The wire frame indicates the inner part of the cell, while the green and brown frames are parallel to the shear walls. The bottom of the interrogation volume is cut non-orthogonally because of optical accessibility.

Determination of energy transport is crucial for understanding the energy budget and fluid circulation in density varying fluids such as the ocean and the atmosphere. However, it is rarely possible to determine the energy flux field J = pu, which requires simultaneous measurements of the pressure and velocity perturbation fields p and u, respectively. We present a method for obtaining the instantaneous J(x, z, t) from density perturbations alone: A Green's function-based calculation yields p; u is obtained by integrating the continuity equation and the incompressibility condition. We validate our method with results from Navier-Stokes simulations: The Green's function method is applied to the density perturbation field from the simulations and the result for J is found to agree typically to within 1% with J computed directly using p and u from the Navier-Stokes simulation. We also apply the Green's function method to density perturbation data from laboratory schlieren measurements of internal waves in a stratified fluid and the result for J agrees to within 6% with results from Navier-Stokes simulations. Our method for determining the instantaneous velocity, pressure, and energy flux fields applies to any system described by a linear approximation of the density perturbation field, e.g., to small-amplitude lee waves and propagating vertical modes. The method can be applied using our MATLAB graphical user interface EnergyFlux.