Atmospheric Dynamics Modeling Group

Atmospheric Dynamics Modeling Group / Projects


The research projects in our group are focused on the overarching themes of atmospheric modeling and dynamics, and we recently started exploring machine learning concepts for the atmospheric sciences. Our group is highly interdisciplinary with diverse interests in atmospheric science, applied mathematics, computational science / scientific computing and data science.
In particular, we develop new numerical methods (like Adaptive Mesh Refinement (AMR) techniques) and variableresolution approaches for the dynamical cores of atmospheric General Circulation Models (GCMs), analyze and compare existing dynamical cores (especially their behavior at the subgridscale), develop simpler atmospheric models and test cases for dynamical cores that get utilized in communitywide model intercomparison projects, analyze and improve the respresentation of tropical cyclones and Mesoscale Convective Systems (MCSs) in global climate models, investigate stratospheric dynamics with a focus on the QuasiBiennial Osciillation (QBO) and Sudden Stratospheric Warmings (SSWs), and provide insight into physicsdynamics coupling issues. In the past, we also designed cyberinfrastructure tools for the climate sciences. All of our research projects utilize parallel and highperformance computing architectures.

Machine learning approaches and new data science algorithms are an emerging frontier for the atmospheric sciences. We recently started exploring whether newly developed physicsaware machine learning algorithms trained with observations can replace subgridscale physical parameterizations in forecast models, such as the timeconsuming solar radiation code, or the shallow or deep convection cloud schemes. A second, less aggressive approach is to utilize machine learning approaches for the estimation of uncertain parameters in physical parameterizations.
Figure: Machine learning and data science concepts are an emerging frontier for the atmospheric sciences (image credit: Carleton University)

Boulder, CO, June, 6  June, 17 2016
Organized by Paul Ullrich (University of California, Davis), Christiane Jablonowski (University of Michigan), Kevin Reed (Stony Brook University), Colin Zarzycki (NCAR), James Kent (University of South Wales, U.K.), Peter H. Lauritzen (NCAR) and Ramachandran D. Nair (NCAR)
We have organized a summer school and model intercomparison workshop with special focus on the newest nonhydrostatic global models. We invited students, postdocs and the international dynamical core modeling community to join us at the National Center for Atmospheric Research (NCAR, Boulder, CO) for 2 weeks from June/617/2016 for an exciting studentfocused and researchdriven event that has led to an unprecedented dynamical core intercomparison project. The new aspects were that DCMIP2016 focused very strongly on idealized dynamical core test cases with simple moisture feedbacks to assess the physicsdynamics coupling and interactions. As DCMIP2012, DCMIP2016 has been endorsed by the WMO Working Group on Numerical Experimentation (WGNE). The summer school and model intercomparison workshop has been supported by cyberinfrastucture tools like shared workspaces via the Earth System CoG
environment.
DCMIP2016 explores new test techniques for nonhydrostatic models that are also applicable at hydrostatic scales. Examples are a new moist baroclinic wave test case without topography (applicable to both shallowatmosphere and deepatmosphere dynamical core equation sets) and a tropical cyclone test case of intermediate complexity that not only assesses the dynamical core but also includes a 'simple physics' package with a Kesslertype warmrain scheme. These test cases let us investigate nonlinear interactions and the physicsdynamics coupling. The third test is a supercell system on a reducedsize Earth that gives insight into nonhydrostatic motions at low computational cost.
Figure: Cubedsphere and hexagonal computational grids on the sphere.

Boulder, CO, July, 30  August, 10 2012
In the summer of 2012 we (Christiane Jablonowski, Peter Lauritzen, Paul Ullrich, Mark Taylor, Ram Nair) organized a summer school and model intercomparison workshop
with special focus on nonhydrostatic global models which are under development right now. We invited students, postdoctoral researchers and the international dynamical core modeling community to join us at the National Center for Atmospheric Resaearch (NCAR, Boulder, CO) for 2 weeks from July/30August/10/2012 for an exciting studentfocused and researchdriven event that will led to an unprecedented dynamical core intercomparison project, even broader in scale than our 2008 workshop. The event was endorsed by the WMO Working Group on Numerical Experimentation (WGNE). The summer school and model intercomparison workshop was also supported by newly developed cyberinfrastucture tools like shared workspaces that we have developed in collaboration with NOAA and NCAR under an NSF grant.
We explored new test techniques for nonhydrostatic models that are also applicable at hydrostatic scales. Examples are our newly developed tropical cyclone test case of intermediate complexity that not only assesses the dynamical core but also includes a 'simple physics' package. This lets us investigate nonlinear interactions and the physicsdynamics coupling. Other test cases included 'small Earth' experiments that gave insight into nonhydrostatic motions at low computational cost.
Figure: Simulation of a breaking wave on a cubedsphere grid.

Boulder, CO, June 113, 2008
Organized by Peter H. Lauritzen (NCAR), Christiane Jablonowski (University of Michigan), Mark Taylor (Sandia National Laboratories) and Ramachandran D. Nair (NCAR)
The 2week summer colloquium titled "Numerical Techniques for Global Atmospheric Models" surveyed the latest developments in numerical methods for the dynamical cores of Atmospheric General Circulation Models. The objectives of the colloquium were (1) to teach a large group of about 40 graduate students in atmospheric science and mathematics how today's and future dynamical cores are or need to be built, (2) to invite over 10 dynamical core modeling groups to NCAR for an unprecedented studentrun dynamical core intercomparison project, (3) to establish new dynamical core test cases in the community and (4) to invite keynote speakers to NCAR that give lectures on modern numerical techniques and innovative computational meshes.
We have published a book
that is based on the lectures from the 2008 Summer Colloquium:
Lauritzen, P. H., C. Jablonowski, M. A. Taylor and R. D. Nair (Eds.) (2011), Numerical Techniques for Global Atmospheric Models,
Lecture Notes in Computational Science and Engineering, Springer, Vol. 80, 572 pp.
Figure: Selected results of the dynamical core intercomparison project conducted during the 2008 NCAR ASP summer school.
The figure shows a cross section of an advected slotted ellipse after 12 days that was transported up and down and around the sphere via an analytically prescribed 3D wind field. The initial condition and reference solution can be used to assess the characteristics of the advection schemes in the participating models denoted by the acronyms. The horizontal grid spacings are approximately 110 km with 60 levels in the vertical direction (vertical grid spacing is 250 m).

Adaptive Mesh Refinement (AMR) techniques provide an attractive
framework for atmospheric flows since they allow an improved
resolution in limited regions without requiring a fine grid
resolution throughout the entire model domain. The model regions
at high resolution are kept at a minimum and can be individually
tailored towards the research problem associated with atmospheric
model simulations.
The climate system is characterized by complex nonlinear interactions over a broad range
of temporal and spatial scales. Our research objective is to determine
how these multiscales interact and how to use enabling computational tools to
mathematically represent scale interactions in climate models. The research
focuses in particular on scale interactions in the socalled dynamical core of Atmospheric
General Circulation Models (GCM). The dynamical core refers to the fluid dynamics
component of a GCM and encompasses the numerical methods used to solve the
equations of motion on the resolved scales. The research explores how Adaptive
Mesh Refinement (AMR) and other variable resolution grid techniques allow high resolution meshes in regions of
interest like the eye of a tropical cyclone or over mountainous terrain. It thereby suggests
pathways how to bridge the scale discrepancies between local, regional and global
phenomena, a key frontier in climate modeling.
The variableresolution approach is focused on cubedsphere computational meshes that have the
potential to become a standard in future GCMs. Cubedsphere grids offer an almost
uniform grid point coverage on the sphere. They deliver high performance and almost
perfect scaling characteristics on massively parallel computer architectures. The grid is
ideally suited for local grid refinements that are based on DoE's AMR software
framework Chombo from the Lawrence Berkeley National Laboratory (LBNL). Chombo
is under development by DoE's Applied Partial Differential Equations Center (APDEC) under the leadership of our collaborator Dr. Phillip Colella.
Both hydrostatic and nonhydrostatic dynamical core designs are developed, implemented and
assessed in our team using highorder conservative and oscillationfree finitevolume numerical schemes. In addition, we
investigate the numerical schemes in a 2d shallowwater framework that serves as an ideal testbed for 3d model developments.
At a later stage our simulations will also involve an indepth investigation of the validity of physical parameterizations at different
spatial scales. The latter will be done in close collaboration with the National Center for
Atmospheric Research (NCAR).
Figure: Examples of the adaptive mesh refinement and variableresolution
techniques applied to (left, first) a blockstructured FiniteVolume (FV) shallow
water model on a latitudelongitude grid, (middle, second) the blockstructured Chombo shallow water model on the cubedsphere grid, and (right, third) the conformal cubedsphere grid in NCAR's Spectral Element (SE) model. In the model FV all blocks are selfsimilar and contain 9x6
additional grid points. The left figure shows how the refined regions
track the relative vorticity fields of a barotropic instability in the model FV (see StCyr et al., 2008), the middle figure depicts the vorticity field of two merging vortices after four days with the Chombo model (see Ferguson et al., 2016) and the right (third) figure
illustrates a static (nonmoving) variableresolution mesh in the model CAMSE (see e.g. Zarzycki and Jablonowski, 2014, 2015).

We develop, implement and test highorder finitevolume methods for the dynamical cores of atmospheric general circulation models. Our computational grid of choice is the cubedsphere grid that is also depicted above.
This project pays special attention to 3rdorder and 4thorder accurate discretizations that perform well at all spatial scales and all aspect ratios between the horizontal and vertical resolution. The methods are promising candidate for future nonhydrostatic dynamical cores with flexible computational grids. Such grids are for example the blockstructured AMR grids that overlay the cubedsphere geometry. We have developed the nonhydrostatic dynamical core 'MCore' in both a Cartesian channel configuration (Ullrich and Jablonowski, 2012, Mon. Wea. Rev. ) and on a spherical cubedsphere grid (Ullrich and Jablonowski, 2012, J. Comput. Phys.).
Figure: Example simulations with our highorder finitevolume model in a Cartesian channel configuration.
The figure depicts 2D and 3D idealized test cases that assess the model performance at small scales (2D thermal bubble), mesoscales (2D mountain waves) and large planetary scales (3D breaking baroclinic wave in a channel). The results are presented in the paper Ullrich and Jablonowski (2012) in the journal Monthly Weather Review.

This project investigates and improves the representation of tropical cyclones
in the National Center for Atmospheric Research (NCAR) Community Earth System Model
CESM1. CESM is jointly supported by the Department of Energy (DOE) and the National
Science Foundation and contains the Community Atmosphere Model CAM that offers multiple dynamical cores and physics options.
In particular, CAM version 5 contains the finitevolume (FV) dynamical core, the HighOrder Method Modeling Environment (HOMME),
and the spectral transform Eulerian (EUL) and semiLagriangian (SLD) dynamical cores.
The objectives of the research are to (1) simulate the evolution of an
idealized, initially weak vortex into a tropical cyclone in an aquaplanet configuration of CAM, (2)
investigate the sensitivity of tropical cyclone development and structure to varying resolutions and
convective parameterizations within CAM, (3) explore the impact of different numerical schemes
on the evolution of tropical cyclones through utilization of numerous dynamical cores available in
CAM, and (4) use the knowledge gained from these simulations to project the possible effects of climate
change on longterm (decadal) tropical cyclone statistics. Such process studies will reveal the impact and relative importance
of the physical parameterizations and numerical schemes on the simulations of tropical cyclones
in CAM, thereby providing an improved scientific basis for the projections of tropical cyclone
activity under changing climate conditions.
We have developed an idealized tropical cyclone test case based on analytic initial conditions
that spins up intense (up to category5) tropical cyclones over the course of 510 simulation days.
We have also developed a simplified physics package called 'simple physics' for aquaplanet studies
that only contains the most basic driving mechanisms for cyclones such as surface fluxes, boundary layer
diffusion and largescale condendation. This physics package lets us define a test case of intermediate complexity.
Figure: Idealized cyclone simulations with NCAR's CAM 3.1 model in aquaplanet configuration
(using the Finite Volume (FV) dynamical core). The top row displays the magnitude of the
wind near the surface at the resolution 0.25 x 0.25 degrees with an initial maximum wind of 17.8
m/s. The initial radius of maximum wind is 200 km. Results for the initial vortex (day
0), and the vortex after 5 and 10 days are shown. The bottom row displays the wind
magnitude for a vertical cross section through the center latitude of the vortex as a
function of the radius from the center of the vortex [see also Reed and Jablonowski (Mon. Wea. Rev., 2011), Reed and Jablonowski (JAMES 2011), Reed and Jablonowski (Geophys. Lett., 2011) and Reed and Jablonowski (JAMES 2012)].

In this project we assess the characteristics of the QBO and SSWs in
idealized dynamical core simulations with NCAR's Community Atmosphere Model (CAM).
The QBO is a phenomenon that takes place in the
equatorial stratosphere where the zonal wind oscillates between
the westward and eastward phase with a period of about 28 months. SSWs are occasional collapes
of the westerly polar jet in the polar stratosphere. Both phenomena are wave driven and excellent examples of wavemean flow interactions. The
QBO is believed to be driven by gravity waves and equatorially trapped vertically
propagating waves, in particular Kelvin and Mixed RossbyGravity
waves, which act as momentum sources. In nature, these waves are e.g.
triggered by tropical convection. SSWs are driven by upward propagating long Rossby waves (with wavenumber 1 or 2) which originate in the tropospheric midlatitudes.
In order to analyze the QBO behavior and the phenomena that
contribute to its forcing we e.g. use the Transformed
Eularian Mean (TEM) analysis as well as wavenumberfrequency
assessments (WheelerKiladis).
Wavenumberfrequency diagrams reveal if certain wave types
are present in the data, such as Kelvin or mixed RossbyGravity waves.
Kelvin waves are expected to have periods of about 15 days, with
zonal wavenumber of about 12, and usually observed when the mean
zonal flow is easterly. Mixed RossbyGravity waves are expected to have
a 45day period, with a zonal wavenumber of around 4. They are usually
observed when the mean zonal flow is westerly. The figures below illustrate
such analyses for the tropical region, derived from both CAM model data with the semiLangrian dynamical core, as well as reanalysis
data (ERA40). We also apply our analysis technique to Sudden Stratospheric Warming (SSW) events to shed light on the interplay between waves and the mean flow, and investigate the impact of gravity wave drag parameterizations on the dynamicsphysics interaction. These analyses reveal what the impact of the numerical dynamical core design is on the modeled stratospheric circulation.
Figure: Example of an idealized dynamical core simulation with NCAR's semiLagrangian spectral transform model (driven by the socalled HeldSuarez forcing). The figure shows the zonalmean monthlymean zonal wind at the equator. The oscillation resembles the QuasiBiennial Oscillation (QBO) in the stratosphere (see e.g. Yao and Jablonowski, 2013, 2015, 2016).

The dynamical core of atmospheric General Circulation Models (GCMs) comprises the fluid dynamics component of every climate and weather forecasting model.
Tests of GCMs and, in
particular, tests of their dynamical cores are important steps
towards future model improvements. They reveal the influence of an
individual model design on climate and weather simulations and
indicate whether the circulation is described representatively by
the numerical approach. However, testing a global 3D atmospheric
model and it dynamical core is not straightforward.
In the absence of nontrivial analytic solutions, the model
evaluations most commonly rely on intuition, experience and model
intercomparisons.
We develop
idealized test cases for dynamical cores and conduct
international Dynamical Core Model Intercomparison Projects (DCMIP) (e.g. at NCAR in June 2008,
DCMIP2012 in August 2012 and DCMIP2016 in June 2016). An example of a
dynamical core test is the evolution of a baroclinic wave that is
also depicted in the figure below. In addition, we work on test
cases with intermediate complexity that include simple moisture
feedbacks, e.g. for tropical cyclonelike simulations.
Figure: Surface pressure field at day 9 of the baroclinic
instability test case simulated with 9 different dynamical cores.
The tests starts with balanced initial conditions that are overlaid
by a Gaussian hill perturbation. The grid imprint of the
cubedsphere and icosahedral grids can be seen in the Southern
Hemisphere (GEOSFVCUBE, GME, ICON, OLAM). Spectral ringing
appears in CAMEUL and HOMME. The baroclinic wave test is
documented in Jablonowski and Williamson (QJ, 2006) and in the
Jablonowski and Williamson NCAR Technical Report TN469+STR
(2006).

This project assesses and quantifies the
role of unresolved subgridscale mixing processes in the dynamical
cores of stateoftheart General Circulation Models (GCMs). The
representation of the subgrid scale in GCMs is complex. Besides
physical processes to be represented, numerical errors manifest
themselves as subgridscale diffusion, and mixing is used to assure
numerical stability and to compensate for numerical dispersion
errors. The quantitative effect of physical mixing is therefore
conflated with mixing processes associated with filtering, implicit
and explicit diffusion and numerical errors. This raises new
questions concerning the accuracy, stability and climate
sensitivity of today's climate modeling approaches.
We focus our research on the numerical schemes used in NCAR's
Community Atmosphere Model, version 5 (CAM 5). This model has
options for four very different dynamical cores. These
are the spectraltransform Eulerian and semiLagrangian dynamical
cores as well as the Finite Volume (FV) and HOMME spectral element dynamics package. All four are
well established to propagate resolved, longscale, structures with
credible accuracy. However, they all treat smallscales
differently, from the point of view of both physical processes and
numerical construction. We perform a set of numerical experiments
with increasing complexity. In particular, we analyze the
subgridscale characteristics of idealized dynamical core
experiments like baroclinic instability studies, perform longterm
HeldSuarez climate runs with idealized forcing functions and
assess aquaplanet simulations with intermediate complexity. In
addition, the subgridscale mixing in tracer transport experiments
is investigated. We develop and apply evaluation techniques and test whether the lessons learned in simplified experiments are
predictors of the
performance in climate models. In 2011 we published a 113page Springer book chapter on the pros and cons of diffusion, filters and fixers in GCMs that sheds
light on the many subgridscale mechannisms in today's weather and climate models.
Figure:Surface pressure (hPa) and 850 hPa meridional velocity (m/s) at day 9 of the growing
baroclinic wave test case of Jablonowski and Williamson (2006) from the CAM FV dynamical
core at a 1x1 degree resolution (about 110 km) with 26 vertical levels. Simulations with (a,b): default secondorder divergence damping, (c,d) fourthorder
divergence damping, (e,f) no divergence damping. The unusual contour interval for the v
wind emphasizes the very weak oscillations in (d).

Latest update: July 27th, 2018

