Lab Home   |  Phone
 
 

Computer, Computational, & Statistical Sciences, CCS

CONTACTS

New Website Announcement!


New Website Announcement!

The Data Science at Scale Team has published a new website and is no longer updating this website. The new website is at:

http://datascience.lanl.gov/

In the future you will be automatically redirected to the new website, but for now you can stay at the old website. If you would like to proceed to the new website, you can click the above link.


Highlights


Understanding Heat Transport in Ocean Simulations will Improve our Understanding of Climate Change --> movie available


>>>>>Movie Download<<<<<
	
Movie Captions
	Eddies transport heat from the Equator to the Poles; they are important to the world's climate. 
	We focus on one of the climate community's canonical methods for identifying eddies: the 
	Okubo-Weiss parameter. In general, this parameter divides the ocean into regions dominated 
	by vorticity, regions dominated by strain or deformation, and a background where neither 
	effect is dominant. Regions dominated by vorticity are potentially eddies, though vorticity 
	dominance can be caused by other effects, such as sharp turns in the mean flow. Because 
	Okubo-Weiss is difficult to interpret directly, it has been normalized to its standard 
	deviation and colored based on the oceanographic rule of thumb that the significant activity 
	is 0.2 standard deviations from 0. 

	In this movie, we visualize the Okubo-Weiss parameter by removing all voxels in the 
	three-dimensional simulation data above the −0.2 standard deviations threshold. The remaining 
	voxels, which according to Okubo-Weiss comprise the eddies and other high-vorticity features, 
	are then painted as solid cubes colored by normalized Okubo-Weiss value, from red (lower 
	vorticity) to yellow (higher vorticity). These high-vorticity features are placed on a 3D 
	bathymetric map of the ocean floor also extracted from the simulation data. The three-dimensional 
	shapes of the eddies are now made clear: here in a region containing the Gulf Stream, several
	strong eddies reach very deeply into the ocean, while smaller eddies remain near the surface, 
	and the Gulf Stream dominates near the surface.

Understanding Heat Transport in Ocean Simulations will Improve our Understanding of Climate Change


Problem

	-Difficult to understand heat transport 
	in the ocean from raw simulation results 
	-Eddies transport heat from equator to 
	pole, so are important to the climate

Solution

	-Compute and measure features 
	-Global ocean simulation cannow resolve 
	eddies atdiameters of about 100 km
	-Areas with negative Okubo-Weiss are 
	dominated by vorticity and are therefore 
	eddy-like
	-Extract eddies from LANL's Parallel 
	Ocean Project (POP) simulation, using a 
	metricfrom the oceanography community:
the Okubo-Weiss (ow) parameter, (stretching strain)2 + (shear strain)2 - (vorticity)2
		
Impact

	-Having a better understanding of heat transport in the ocean will improve our understanding 
	of the effects of global climate change
	-Identifying eddies allows empirical comparisons between simulations and observations, 
	increasing our trust in our simulations

Scientific Visualization Critical for Quickly Finding Cosmological Simulations Coding Errors

Problem

	-Difficult to find subtle coding mistakes in complex cosmological simulations
	-Existing statistical and numerical analysis checks do not always identify these errors
	
Solution

	-Visualizing the results of an erroneous code can quickly reveal subtle errors
	-In the cosmology simulation below, a coding error caused velocity vectors with extremely 
	large y-components, shown as long red arrows
	-The cosmologist writing the code says, "...the first test I always do for checking new 
	code developments in detail is visualizing the particles and their velocities. You very 	
	quickly get a large amount of information and understand much more intuitively what is going 
	on. It is very helpful."
	
Impact

	-Visualization plays a key role in producing verified simulation codes
	
      
  Erroneous Simulation Output     Visualization Reveals the Errors           Corrected Output

Automated distance visualization reduces time to visual understanding


Problem
	Climate scientists at LANL perform daily simulations at ORNL.  It takes minutes to hours 
	of their time (approximately 23 minutes per one full resolution, single field data set) 
	to copy the data for local analysis.
	
Solution
	Automatically compress data after it is generated with data precision guarantees.  The 
	remote simulation data is tracked, managed, and pre-loaded into the scientist’s visualization 
	tool for fast local access at a desired accuracy.

Impact
	Data is automatically available on the scientist’s desktop for local viewing increasing 
	productivity. Our approach reduces the network transfer time to seconds/minutes with a 
	guaranteed maximum data error for visualization and analysis.

	Mat Maltrud, Ocean/Climate Scientist, LANL: "This new distance visualization technology 
	will increase our productivity by significantly reducing the amount of time spent in analyzing 
	our remote data." 
	   

Operated by Los Alamos National Security, LLC for the U.S. Department of Energy's NNSA

Inside | © Copyright 2009-10 Los Alamos National Security, LLC All rights reserved | Disclaimer/Privacy | Web Contact