Form:CSDMS annual meeting: Difference between revisions
From CSDMS
No edit summary |
No edit summary |
||
| Line 61: | Line 61: | ||
| affiliation = Sandia NL | | affiliation = Sandia NL | ||
| title = DAKOTA: An Object-Oriented Framework for Simulation-Based Iterative Analysis | | title = DAKOTA: An Object-Oriented Framework for Simulation-Based Iterative Analysis | ||
| abstract = The DAKOTA project began in 1994 with the primary objective of reusing software interfaces to design optimization tools. Over nearly 20 years of development, it has grown into an open source toolkit supporting a broad range of iterative analyses, typically focused on high-fidelity modeling and simulation on high-performance computers. Today, DAKOTA provides a delivery vehicle for uncertainty quantification research for both the NNSA and the office of science, enabling an emphasis on predictive science for stockpile stewardship, energy, and climate mission areas.<br><br> | | abstract = The DAKOTA project began in 1994 with the primary objective of reusing software interfaces to design optimization tools. Over nearly 20 years of development, it has grown into an open source toolkit supporting a broad range of iterative analyses, typically focused on high-fidelity modeling and simulation on high-performance computers. Today, DAKOTA provides a delivery vehicle for uncertainty quantification research for both the NNSA and the office of science, enabling an emphasis on predictive science for stockpile stewardship, energy, and climate mission areas.<br /><br /> | ||
Starting with an overview of the DAKOTA architecture, this presentation will introduce processes for setting up iterative analyses, interfacing with computational simulations, and managing high-fidelity workflows. Algorithmic capabilities in optimization, calibration, sensitivity analysis, and uncertainty quantification (UQ) will be briefly overviewed, with special emphasis given to UQ. Core UQ capabilities include random sampling methods, local and global reliability methods, stochastic expansion methods, and epistemic interval propagation methods. This UQ foundation enables a variety of higher level analyses including design under uncertainty, mixed aleatory-epistemic UQ, and Bayesian inference. | Starting with an overview of the DAKOTA architecture, this presentation will introduce processes for setting up iterative analyses, interfacing with computational simulations, and managing high-fidelity workflows. Algorithmic capabilities in optimization, calibration, sensitivity analysis, and uncertainty quantification (UQ) will be briefly overviewed, with special emphasis given to UQ. Core UQ capabilities include random sampling methods, local and global reliability methods, stochastic expansion methods, and epistemic interval propagation methods. This UQ foundation enables a variety of higher level analyses including design under uncertainty, mixed aleatory-epistemic UQ, and Bayesian inference. | ||
}} | }} | ||
Revision as of 16:47, 14 February 2013
CSDMS 2.0: Moving Forward
March 23-25th 2013 Boulder Colorado, USA
Registration deadline: February 14th, 2013
Registration
The online conference registration is a three step process:
|
|
|
|
|
|
Note 1: You only are successfully registered by fulfilling the above steps
Note 2: If you are already registered and want to make changes, then Log in, select your registration record in "participants" and start making changes by clicking "Edit registration".
Objectives and general description
The CSDMS Meeting 2013 is designed to launch CSDMS 2.0 and shape its direction through engaging on the technical and community challenges over the next five years.
The meeting includes: 1) State-of-the art keynote presentations in earth-surface dynamics and modeling; 2) Hands-on clinics related to community models, tools and approaches; 3) Transformative software products and approaches designed to be accessible, easy to use, and relevant; 4) New community initiatives to advance earth-surface process modeling across many disciplines; 5) Breakout sessions for Working and Focus Research Groups to update their strategic plans and define their long, medium and short term goals; 6) Poster Sessions; and more.
Program Schedule updated February 5th
Keynote speakers:
John Atkinson
ARCADIS U.S., Inc.
{{{participants}}}A Coupled ADCIRC and SWAN model of Hurricane Surge and Waves. This presentation will briefly introduce the formulation, numerics, and parallel implementation of the coastal circulation model ADCIRC, discuss the strategy of coupling with the SWAN wave model, and provide background on recent enhancements of the bottom-friction formulation. Several recent applications of the coupled modeling system will be presented. Katy Barnhart
University of Colorado
{{{participants}}}Melting Coasts and Toppled Blocks: Modeling Coastal Erosion in Ice-Rich Permafrost Bluffs, Beaufort Sea, Alaska (with Robert S. Anderson, Irina Overeem, Gary Clow, and Frank Urban)
(thanks to Adam LeWinter and Tim Stanton)
Rates of coastal cliff erosion are a function of the geometry and substrate of the coast; storm frequency, duration, magnitude, and wave field; and regional sediment sources. In the Arctic, the duration of sea ice-free conditions limits the time over which coastal erosion can occur, and sea water temperature modulates erosion rates where ice content of coastal bluffs is high. Predicting how coastal erosion rates in this environment will respond to future climate change requires that we first understand modern coastal erosion rates.
Arctic coastlines are responding rapidly to climate change. Remotely sensed observations of coastline position indicate that the mean annual erosion rate along a 60-km reach of Alaska’s Beaufort Sea coast, characterized by high ice content and small grain size, doubled from 7 m yr-1 for the period 1955-1979 to 14 m yr-1 for 2002-2007. Over the last 30 years the duration of the open water season expanded from ∼45 days to ∼95 days, increasing exposure of permafrost bluffs to seawater by a factor of 2.5. Time-lapse photography indicates that coastal erosion in this environment is a halting process: most significant erosion occurs during storm events in which local water level is elevated by surge, during which instantaneous submarine erosion rates can reach 1-2 m/day. In contrast, at times of low water, or when sea ice is present, erosion rates are negligible.
We employ a 1D coastal cross-section numerical model of the erosion of ice-rich permafrost bluffs to explore the sensitivity of the system to environmental drivers. Our model captures the geometry and style of coastal erosion observed near Drew Point, Alaska, including insertion of a melt-notch, topple of ice-wedge-bounded blocks, and subsequent degradation of these blocks. Using consistent rules, we test our model against the temporal pattern of coastal erosion over two periods: the recent past (~30 years), and a short (~2 week) period in summer 2010. Environmental conditions used to drive model runs for the summer of 2010 include ground-based measurements of meteorological conditions (air temperature, wind speed, wind direction) and coastal waters (water level, wave field, water temperature), supplemented by high temporal frequency (4 frames/hour) time-lapse photography of the coast. Reconstruction of the 30-year coastal erosion history is accomplished by assembling published observations and records of meteorology and sea ice conditions, including both ground and satellite-based records, to construct histories of coastline position and environmental conditions. We model wind-driven water level set-up, the local wave field, and water temperature, and find a good match against the short-term erosion record. We then evaluate which environmental drivers are most significant in controlling the rates of coastal erosion, and which melt-erosion rule best captures the coastal history, with a series of sensitivity analyses. The understanding gained from these analyses provides a foundation for evaluating how continuing climate change may influence future coastal erosion rates in the Arctic.Chris Duffy
Penn State University
{{{participants}}}PIHM model Michael S. Eldred
Sandia NL
{{{participants}}}DAKOTA: An Object-Oriented Framework for Simulation-Based Iterative Analysis The DAKOTA project began in 1994 with the primary objective of reusing software interfaces to design optimization tools. Over nearly 20 years of development, it has grown into an open source toolkit supporting a broad range of iterative analyses, typically focused on high-fidelity modeling and simulation on high-performance computers. Today, DAKOTA provides a delivery vehicle for uncertainty quantification research for both the NNSA and the office of science, enabling an emphasis on predictive science for stockpile stewardship, energy, and climate mission areas.
Starting with an overview of the DAKOTA architecture, this presentation will introduce processes for setting up iterative analyses, interfacing with computational simulations, and managing high-fidelity workflows. Algorithmic capabilities in optimization, calibration, sensitivity analysis, and uncertainty quantification (UQ) will be briefly overviewed, with special emphasis given to UQ. Core UQ capabilities include random sampling methods, local and global reliability methods, stochastic expansion methods, and epistemic interval propagation methods. This UQ foundation enables a variety of higher level analyses including design under uncertainty, mixed aleatory-epistemic UQ, and Bayesian inference.Courtney Harris
VIMS
{{{participants}}}ROMS & biogeochemistry coupling Kathy Hibbard
Pacific Northwest NL
{{{participants}}}Integrated Assessment Modeling Louis Moresi
Monash University
{{{participants}}}UNDERWORLD model Jaap Nienhuis
WHOI/MIT
{{{participants}}}Growth and Abandonment: Quantifying First-order Controls on Wave Influenced Deltas. What determines the style of river delta growth? How do deltas change after fluvial sediment supply is cut off? River delta evolution is characterized by the progradation and transgression of individual (deltaic) lobes: the delta cycle. We investigate the behaviour of wave-influenced deltas with a simple shoreline model, and quantitatively relate several first-order controls. Mark Schmeeckle
Arizona State University
{{{participants}}}Turbulence- and Particle-Resolving Numerical Modeling of Sediment Transport. Turbulence, bedload, and suspended sediment transport are directly simulated by a coupled large eddy simulation of the fluid and a distinct element method for every sediment grain. This modeling system directly calculates the motion of all grains by resolved turbulence structures. The model directly calculates modification of the flow and turbulence by the grains, such as the effects of grain momentum extraction and density stratification. Simulations such as these can be used in the future to parameterize sediment transport in large-scale morphodynamic simulations. Mauro Werder
Simon Fraser University
{{{participants}}}Modeling channelized and distributed subglacial drainage in 2D This model of the subglacial drainage system simulates the pressurised flow of water at the ice-bed interface of glaciers and ice sheets. It includes both distributed and channelized water flow. Notably the model determines the geometry of the channel network as part of the solution. The resulting channel network is similar to subaerial stream networks with channels carving out hydraulic potential "valleys". However, there are some pronounced differences to subaerial drainage, for example that the time for a network to form (and decay) is on the order of weeks to months; or that, channels originating at point sources can lie on ridges of the hydraulic potential. The model employs a novel finite element approach to solve the parabolic equations for the hydraulic potential simultaneously on the 1D channel network and 2D distributed system.
Clinics:
Peter Burgess & Chris Jenkins
Royal Holloway, UK & Univ. of Co.
{{{participants}}}Carbonate clinic Gary Clow
USGS
{{{participants}}}Introduction to the Weather Research & Forecasting (WRF) System, a High-Resolution Atmospheric Model Scott Peckham
University of Colorado
{{{participants}}}Introduction to the Basic Model Interface and CSDMS Standard Names In order to simplify conversion of an existing model to a reusable, plug-and-play model component, CSDMS has developed a simple interface called the Basic Model Interface or BMI that model developers are asked to implement. In this context, an interface is a named set of functions with prescribed function names, argument types and return types. By design, the BMI functions are straightforward to implement in any of the languages supported by CSDMS, which include C, C++, Fortran (all years), Java and Python. Also by design, the BMI functions are noninvasive. A BMI-compliant model does not make any calls to CSDMS components or tools and is not modified to use CSDMS data structures. BMI therefore introduces no dependencies into a model and the model can still be used in a "stand-alone" manner. Any model that provides the BMI functions can be easily converted to a CSDMS plug-and-play component that has a CSDMS Component Model Interface or CMI.
Once a BMI-enabled model has been wrapped by CSDMS staff to become a CSDMS component, it automatically gains many new capabilities. This includes the ability to be coupled to other models even if their (1) programming language, (2) variable names, (3) variable units, (4) time-stepping scheme or (5) computational grid is different. It also gains (1) the ability to write output variables to standardized NetCDF files, (2) a "tabbed-dialog" graphical user interface (GUI), (3) a standardized HTML help page and (4) the ability to run within the CSDMS Modeling Tool (CMT).
This clinic will explain the key concepts of BMI, with step-by-step examples. It will also include an overview of the new CSDMS Standard Names, which provide a standard way to map input and output variable names between component models as part of BMI implementation. Participants are encouraged to read the associated CSDMS wiki pages in advance and bring model code with specific questions. See
1) BMI Page: BMI_Description
2) Standard Names Page: CSDMS_Standard_NamesIrina Overeem
University of Colorado
{{{participants}}}CMT clinic Thomas Hauser & Monte Lunacek
University of Colorado
{{{participants}}}Python for Matlab users clinic This workshop is a hands-on introduction to using Python for computational science. Python is a powerful open source interpreted language that has been adopted widely in many application areas. The goal of this workshop is to teach participants how to use Python as an open source alternative for MATLAB in their computational workflows. While we will demonstrate how to implement MATLAB-based scientific computing workflows in Python, attendees are not required to have MATLAB or Python experience. The goal of this tutorial is to show how an open source alternative to MATLAB can be used productively for computational science research. In the first part of this workshop we will introduce basic Python concepts and iPython with a focus on migrating from MATLAB to Python. We will show how the Python modules Numpy and Scipy, for scientific computing, and Matplotlib, for plotting, can make Python as capable as MATLAB for computational science research. In the second part of the tutorial we will discuss on how to interface Python with compiled languages like C or Fortran to improve performance of numerical codes. Additionally we will show how to use distributed parallel computing on a supercomputer from interactive python notebooks.
Users should install the enthought python distribution https://www.enthought.com if they want to follow the hands on part of the clinic.Mary Hill
USGS
{{{participants}}}Toward Transparent, Refutable Hydrologic Models in Kansas or Oz. Numerical models are critical to integrating knowledge and data for environmental systems and understanding future consequences of management decisions, weather variability, climate change, and so on. To attain the transparency and refutability needed to understand predictions and uncertainty and use models wisely, this clinic presents a strategy that emphasizes fundamental questions about model adequacy, sensitivity analysis, and uncertainty evaluation, and consistent use of carefully designed metrics. Emphasizing fundamental questions reveals practical similarities in methods with widely varying theoretical foundations and computational demands. In a field where models take seconds to months for one forward run, a credible strategy must include frugal methods for those in Kansas who can only afford 10s to 100s of highly parallelizable model runs in addition to demanding methods for those in Oz who can afford to do 10,000s to 1,000,000s of model runs. Advanced computing power notwithstanding, people may be in Kansas because they have chosen complex, high-dimensional models, want quick insight into individual models, and/or need systematic comparison of many alternative models. This class will briefly review the fundamental questions, demonstrate relations between existing theoretical approaches, and address challenges and limitations. Students will be able to examine a model constructed using FUSE and compare results from computationally frugal method evaluations conducted in class and demanding methods for which results are provided. Xiaofeng Liu
UT San Antonio
{{{participants}}}Modeling of Earth Surface Dynamics and Related Problems using OpenFOAM®. This clinic aims to introduce the open source computational fluid dynamics (CFD) platform, OpenFOAM®, to the earth surface dynamics research community and to foster collaborations. OpenFOAM® is essentially a computational toolbox which solves general physical models (differential equations) using finite volume method. This short clinic is tailored to be suitable for an audience at various levels (from beginners to experienced code developers). It will provide an overview of OpenFOAM. We will demonstrate its usage in a variety of applications, including hydrodynamics, sedimentation, groundwater flows, buoyant plumes, etc. Participants can also bring the problems in their fields of interest and explore ways to solve them in OpenFOAM®. Knowledge of C++, object-oriented programming, and parallel computing is not required but will be helpful. Eckart Meiburg & students
University of California, SB
{{{participants}}}TURBINS using PETSc This clinic will provide information on how laboratory scale flows and field scale flows can be simulated by direct numerical simulations (DNS) and large-eddy simulations (LES) using parallel, high-performance computing facilities. DNS results, from the software TURBINS, of gravity and turbidity currents propagating over complex sea floor topography will be discussed. The use of the PETSc software package within the DNS simulations will be highlighted. LES results of high Reynolds number gravity and turbidity currents, and reversing buoyancy currents over a flat topography will be discussed. Issues relevant to LES such as grid resolution, grid convergence, subgrid models and wall-layer modeling will also be discussed. Helena Mitasova
North Carolina State Univ.
{{{participants}}}Modeling and analysis of evolving landscapes in GRASS GIS This clinic will introduce participants to GRASS6.4.3 with special focus on terrain modeling, geomorphometry, watershed analysis and modeling of landscape processes such as surface water flow and erosion/deposition. The hands-on section will explore lidar-based terrain models, multiple surface visualization, analysis of coastal lidar time series and visualization of terrain evolution using space-time cube. Overview of new capabilities in the GRASS7 development version will also be provided.
Notice: The participants will be expected to download and install GRASS6.4.3 as well as the practice data sets from the provided web site prior to the clinic.Ad Reniers
University of Miami
{{{participants}}}Xbeach clinic Hari Rajaram
University of Colorado
{{{participants}}}Numerical Methods clinic
Participants
Interested to see who registered already for the meeting?
Preliminary list of participants as of today:
Reimbursement
Within its budget, CSDMS intends to support member applicants to attend the annual meeting. Towards this goal, we encourage members to fully or partially cover their expenses if capable. We additionally thank those in the industry and agency fields for understanding that 1) we cannot compensate federal agency participants since our own funding is from NSF, and 2) we request that our industrial/ corporate participants cover their own costs thereby allowing more academic participants to attend.
To the extent possible, CSDMS intends to reimburse the registration fee, lodging (shared rooms at 100% and single rooms at 50% at Millennium Harvest House Hotel), and a limited amount of travel expenses of qualified registrants - those members who have attended all three days of the meeting and are not industry or federal employees.
Registration by 14 February 2013 is required to secure your space at the meeting and your accommodations at the meeting hotel. There are limited number of rooms at a discounted rate being held by the Millennium Harvest House Hotel, Boulder. (Note: The discounted rate is only available via this website registration.) We ask that you register as soon as possible.
Important for foreign travelers requesting reimbursement: If you need a visa to travel to USA, select a business visa. If you need an invitation letter, please email csdms@colorado.edu soonest. Also indicate whether specific wording is required in the letter. Second, we will need to copy the entry stamp in your passport sometime during the meeting as proof that you were here on business as required by US tax laws for reimbursement (especially when dealing with airfare.)
Student Scholarships
The application period for the student scholarship is now closed.
