Send mail to:             for the digests or bakeoff
         for comments or help
Anonymous ftp repository: (
Current editor:  Craig Douglas

WWW Sites: or

Today's editor:  Craig Douglas (

Volume 12, Number 2 (approximately February 29, 2002)

Today's topics:

     Errata and Reminder
     Petr Vanek
     Intel Fortran Compiler
     IMACS Workshop on Adaptive Methods for Partial Differential Equations
     Workshop on Multiscale Computational Mechanics
     New entries in bibliography
     Announcing a New Journal in Multiscale Modeling
     Famous Multigrid Related News Recently
        Steve McCormick, Father of a New US National Park
        Austria's Strobl is Gold


Date: Thu, 28 Feb 2002 10:20:01 -0400
From: Craig Douglas 
Subject: Errata and Reminder

        Copper Mountain Conference on Iterative Methods
3/01/02 Hotel reservations

The hotel is basically full.  Make your reservation today if you are going or
ask the hotel for suggestions in a neighboring town with shuttle bus service
to Copper.

The correct URL at MGNet for the meeting is not what was in the last newsletter,


Date: Wed, 13 Feb 2002 18:07:11 +0330 (IRT)
From: Mehdi Rahmani (MS) 
Subject: Petr Vanek

I'm a graduate student in Mechanical Eng.  My thesis is on solution of
Navier-Stokes equation with Algebraic Multigrid methods.  I want to use Petr
Vanek's AMG_code in site.  I have sent some email to these
addresses:  , but
these weren't real addresses.

Pleas help me if you have more information about Mr.Petr Vanek.

Mehdi Rahmani

    Editor's Note: Anyone know Petr's contact information?  I get email asking
    -------------  where he is fairly often.  Jan Mandel says he does not know.


Date: Fri, 15 Feb 2002 17:25:35 -0500 (EST)
From: Dongwoo Sheen 
Subject: Intel Fortran Compiler

Please get the free, but nice intel fortran 90 compiler at the address

There are two versions, one for Red Hat 6.2 and the other for Red Hat 7.1,
probably you need the latter.

You will need to give your name and email address to get them after pressing
Download botton.  Then it mails you passcode to use in installing the "ifc".

    Editor's Note: The free C/C++ compiler is available from
                   Note that these are for noncommercial use only.  Bugs (if
                   any) should be reported directly to Intel.


Date:   Thu, 14 Feb 2002 13:03:30 -0500
From: Ken Jackson 
Subject: IMACS Workshop on Adaptive Methods for PDEs

                  First Announcement and Call for Papers

                             6-9 AUGUST 2002
                       The Fields Institute, Toronto

As part of the Fields Institute's year on "Numerical and Computational
Challenges in Science and Engineering", we are organizing a Workshop on
Adaptive Methods for Partial Differential Equations.  This workshop,
which is co-sponsored by IMACS and the Fields Institute for Research in
Mathematical Sciences, will be held 6-9 August 2002 at the Fields
Institute in Toronto.

Adaptive methods for partial differential equations (PDEs) are the most
effective computational approach for a large class of PDEs that arise
in many important applications in science and engineering. This area
has grown steadily during the past two decades. This workshop will
bring together leading researchers from around the world to address
both theoretical and computational aspects of adaptive methods for PDEs
and to foster stronger collaboration between mathematicians, engineers
and scientists.

* A posterior error estimation
* Adaptive H-p refinement
* Adaptivity with complex geometry
* Implementation of adaptive codes
* Moving mesh techniques and applications
* Adaptive spectral methods
* Nonlinear analysis
* Adaptive modeling 
* Applications of adaptive methods

* Paul Fisher, Argonne National Laboratories, U.S.A
* Joseph E. Flaherty, Rensselaer Polytechnic Institute, U.S.A.
* Benqi Guo, University of Manitoba, Canada (Co-Chairman)
* Kenneth R. Jackson, University of Toronto, Canada (Co-Chairman)
* Robert D. Russell, Simon Fraser University, Canada 

* Mark Ainsworth, Strathclyde University, Scotland
* Ivo Babuska, University of Texas at Austin, U.S.A.
* Martin Berzins, University of Leeds, U.K.
* Anne Bourlioux, University of Montreal, Canada
* Leszek Demkowicz, University of of Texas at Austin, U.S.A.
* Oleg Vassilyev, University of Missouri, U.S.A. 
* Jinchao Xu, Penn State University, U.S.A. 

The program will consist of invited lectures (45 minutes each), and
contributed talks (30 minutes each).  If you wish to contribute a talk,
please send a one-page abstract, written in English, related to the
topics of the conference, by 30 April 2002. The abstract should
include:  names(s) and affiliation(s) of author(s), as well as the
address, e-mail address, phone and fax numbers of the contact person.
Authors are kindly requested to submit their abstracts via e-mail in
plain LaTeX (or plain AMS Tex) to:

                 Prof. Benqi Guo
                 Mathematics Department, University of Manitoba
                 Fax: 204 - 474 7611

For more details about the workshop, please see our webpage


Date: Mon, 18 Feb 2002 15:53:30 +0100
From: David Dureisseix 
Subject: Workshop on Multiscale Computational Mechanics


"Multiscale Computational Mechanics for material and Structures"

18-19-20 September 2002, Cachan, France


- Multiscale methods: theory and computation
- Experimental tests and Identification for multiscale modelling
- Multiscale modelling of damage and fracture
- Verification and validation of multiscale models
- Engineering applications
- Hierarchical multiscale models
- Adaptive parallel computational strategies for multiscale problems
- Multiscale multiphysics problems
- Coupled continuum - atomistic models.


The Workshop will be devoted to recent advances in Multiscale Computational
Mechanics for Material and Structures (MCM) and their impacts on the
next-generation material - structure - fabrication design which will require
an integrated approach where the distinction between the material and the
structure is completely removed.  Multiscale modelling and associated
computational strategies is extremely promising to elaborate efficient and
robust engineering tools for predicting - for example - damage evolution up to
and including final fracture for composite structures.  Various scientific
areas are involved; Multiscale Computational Mechanics for Materials and
Structures should be necessarily built on the synergy of such areas which have
advanced until now nearly independently as Material Science and Computational
Mechanics.  The Workshop is intended to be a meeting ground for the various
contributors, including material scientists, mecanicians involved in testing
and computation, mathematicians involved in computation, and design engineers.
The Workshop should provide answers to questions such as:
- What is currently being used or could be used in the near
  future to solve engineering problems?
- What are the benefits and drawbacks of MCM?
- What are the key scientific issues of MCM?


Laboratoire de Mécanique et Technologie
ENS de Cachan/CNRS/Paris 6 University
61 avenue du Président Wilson
94235 CACHAN CEDEX France
Phone : (33) 1 47 40 22 41/22 53
Fax : (33) 1 47 40 27 85
E-mail :

Civil Engineering, Mechanical Engineering,
Aerospace Engineering Center
Rensselaer Polytechnique Institute
NY 12180 TROY - USA
Phone : 1 518 276 6191
Fax : 1 518 276 4833
E-mail :






The speakers of the Workshop are well-known specialists.  The papers will be
published in a book form.  Participants are invited by the scientific and
organizing committee.  The Workshop will be organized to promote a wealth of
stimulating discussions.  They are invited to participate actively with few
transparencies to the discussions sessions.


Web address:
Email address:
               (with subject: Workshop MCM3)


Date: Tue, 19 Feb 2002 16:39:37 +0100
From: David Dureisseix 
Subject: New entries in bibliography

Here are some new entries related to domain decomposition and/or multilevel
scheme, from our team.

  author =      "P. Ladeveze and D. Dureisseix",
  title =       "A micro / macro approach for parallel computing
                 of heterogeneous structures",
  journal =     "International Journal for Computational Civil and
                 Structural Engineering",
  volume =      "1",
  year =        "2000",
  pages =       "18--28",
  author =      "D. Dureisseix and C. Farhat",
  title =       "A numerically scalable domain decomposition method
                 for the solution of frictionless contact problems",
  journal =     "Int. J. Numer. Meth. Engng.",
  volume =      "50",
  year =        "2001",
  pages =       "2643--2666",
  author =      "P. Ladeveze and O. Loiseau and D. Dureisseix",
  title =       "A micro-macro and parallel computational strategy
                 for highly heterogeneous structures",
  journal =     "Int. J. Numer. Meth. Engng.",
  volume =      "52",
  year =        "2001",
  pages =       "121--138",

Best regards,

LMT CACHAN / ENS CACHAN  fax : (33) 1 47 40 27 85
61 Av du Pdt Wilson      tel : (33) 1 47 40 53 86
F-94235 CACHAN CEDEX     e-mail :
FRANCE         (depuis la France, remplacer (33) 1 par 01)


Date: Wed, 20 Feb 2002 15:25:01 -0500
Subject: Announcing a New Journal in Multiscale Modeling

Now Accepting Submissions!

Multiscale Modeling and Simulation
A SIAM Interdisciplinary Journal

Thomas Y. Hou, California Institute of Technology, Editor-in-Chief

Announcing the first journal to provide a comprehensive forum for
multiscale research.

Centered around multiscale phenomena, Multiscale Modeling and
Simulation (MMS) is an interdisciplinary journal focusing on the
fundamental modeling and computational principles underlying various
multiscale methods. Featuring a distinguished, international editorial
board, MMS publishes new ideas and methodologies that can be used in
various application fields. Particularly emphasized is the interplay
between analysis and modeling, modeling and simulation, and
mathematics and various applications.

By its nature, multiscale modeling is highly interdisciplinary, with
developments occurring independently across fields.  A broad range of
scientific and engineering problems involve multiple scales.
Traditional monoscale approaches have proven to be inadequate, even
with the largest supercomputers, because of the range of scales and
the prohibitively large number of variables involved. Thus, there is a
growing need to develop systematic modeling and simulation approaches
for multiscale problems. MMS will provide a single broad,
authoritative source for results in this area.

MMS bridges the growing gap in communication between mathematics,
chemistry, physics, engineering, computer science, environmental
science, and more.

As part of the family of high-quality, distinguished SIAM journals,
MMS will benefit from global distribution, accelerated electronic
publication, and low subscription prices.

Submit now to the only journal to welcome papers from all disciplines
that use multiscale methods

MMS accepts submissions via anonymous ftp. Complete electronic
submission instructions, instructions for authors, and a detailed
editorial policy are available at Please review the
instructions before submitting your paper.

Contact with questions or for more information.


Date: Sun, 10 Feb 2002 19:15:01 -0400
From: Craig Douglas 
Subject: Famous Multigrid Related News Recently

***** This ends the regular part of the issue.
***** Stop here if you only want science.

I apologize to any of you who only want to hear about serious multigrid.  You
might consider this spam, which I apologize in advance for.

This week has been wonderful for a well known multigridder and a nice place
for a multigrid conference series in the news even though neither is actually
in the respective article.

The Steve McCormick in the Colorado news story is the Other Steve McCormick in
Boulder (no relation).  The authentic one is not quoted.  (I was disappointed
since the Nature Conservancy was formed years ago to save the river 75m from
my house and I had mistakenly thought the two were the same).  In any case,
the person behind the new national park is a Steve McCormick of Boulder.  I
heartily recommend visiting there.

Ulrich Langer has been running workshops in Austria's Strobl for some years.
The town did not ski to victory, but the town is a wonderful place for a
conference, a hike in the mountains, or a family vacation.  It, along with a
few other multigrid conference series (Copper Mountain, European Multigrid,
GAMM Parallel Multigrid, GAMM-Leipzig/Kiel, Lawrence Livermore National Labs,
...) all rate Gold with me, but I was happy to see that the AP thought so,
too, and singled Strobl out for publicity.  Now, let's all root for the skier

Fixed font is necessary for the second news article; sorry.


Sand dunes may be next national park

Great Sand Dunes would be 56th U.S. national park

WASHINGTON (AP) --Congress' efforts to convert the tallest sand dunes in
North America into a Colorado national park and wildlife refuge have
taken a big step with a private conservation group buying a huge ranch
next to the area.

The Nature Conservancy said it's spending $31.28 million to purchase the
97,000-acre Baca Ranch and two 14,000-foot peaks in southern Colorado,
next to Great Sand Dunes National Monument and Preserve.  The ranch's
plentiful aquifer keeps the 750-foot-high dunes intact and nourishes the
San Luis Valley's farmers and ranchers.

The conservation group's purchase prevents the land from being sold to
someone else.  Once the deed is turned over to the federal government,
the 39,000-acre monument and preserve is to be combined with about
43,000 acres of the ranch, including Kit Carson Peak and Challenger
Point, to form half of what federal officials plan for the new Great
Sand Dunes National Park and Preserve.

Some 54,000 acres of the ranch will be used to make up about
three-fifths of the planned Baca National Wildlife Refuge, where hunting
will be allowed.

'Truly heroic scale'

"This project represents conservation and partnership at a truly heroic
scale," said Steve McCormick, the group's president.

The last time Congress created a national park was in 1999 with the
conversion of another Colorado national monument into Black Canyon of
the Gunnison National Park.  Great Sand Dunes would be the nation's 56th
national park.

Interior Secretary Gale Norton said Wednesday she strongly supports its
creation in her home state, where she was Colorado's attorney general
before joining the Bush administration.

"This is public-private partnership at its best," she said.

Colorado Gov.  Bill Owens said the ranch purchase is the largest land
preservation effort in the state's history and would protect "the
history, culture and way of life of the entire San Luis Valley."

Once the park is created, Colorado's Taylor Peak would be the only
remaining 14,000-footer left in private hands in North America.

Elk, coyotes and bighorn sheep

Spurred by Colorado lawmakers, Congress and the Clinton administration
in 2000 authorized creating the national park and buying the ranch.
They also created Great Sand Dunes National Preserve from 42,000 acres
of the Rio Grande National Forest.

Spread between sagebrush grasslands and the snowy Sangre de Cristo
Mountains, the wind-shaped dunes rise from a valley floor at more than
7,000 feet above sea level.  The area is abundant with elk, coyotes and
bighorn sheep that pass among the aspen, spruce and pine trees.

Congress already has appropriated $10.2 million for the park.  Yale
University has agreed to donate up to $4 million in profits from the
sale of its interest in the ranch.  Sen.  Wayne Allard, Republican of
Colorado, said the only thing remaining for Congress is to appropriate
at least $17 million to repay The Nature Conservancy.

Congress has until 2005 to complete funding for the ranch's purchase.
If the deadline is missed, the deal would likely be renegotiated.

Yale and other parties with a stake in the ranch had promoted a
controversial plan to sell its water to the fast-growing cities along
Colorado's Front Range north and south of Denver.  But Colorado voters
in 1998 overwhelmingly defeated initiatives to aid that plan.

Allard said the idea for creating the national park and wildlife refuge
blossomed from a 1998 town meeting in Alamosa, Colorado, in which
residents said they wanted to both protect the area's natural resources
and create a bigger draw for tourists.

"This is an example of how land-use issues should be handled from the
local level up," he said.  "Now we have an entire ecosystem that's
enclosed and protected here."


Downhill dream

Austrian Strobl captures gold; Kjus takes silver

Posted: Sunday February 10, 2002 1:26 PM
Updated: Sunday February 10, 2002 6:18 PM

                                                         [Fritz Strobl]Fritz
SNOWBASIN, Utah (AP) -- That an Austrian won the men's
downhill was no surprise. That this Austrian won was     Strobl won with a
the shocker.                                             time of 1 minute,
                                                         39.13 seconds. AP
Fritz Strobl, a part-time cop with no catchy nickname like the Herminator,
raced down the dizzyingly steep 1.9-mile Grizzly course Sunday in less time
than it takes to boil water.

His daredevil ride over, Strobl stared at the scoreboard and raised his arms
in triumph when he saw his time and realized he had beaten heavily favored
teammate Stephan Eberharter, who finished third.

"It's sensational. I didn't expect it," said Strobl, who grabbed a cell
phone to call his wife and two young sons in Austria with the news. "I was
just thinking of racing down the course, not of winning."

The lanky Strobl finished in 1 minute, 39.13 seconds to become the sixth
Austrian to win the downhill in the 15 races since Alpine skiing made its
Olympic debut in 1948.

      Strobl Factfile        The race was a bust for Daron Rahlves,
                             America's best medal hope who boasted Saturday
Age: 29                      that he'd knock off the favored Austrians. He
Country: Austria             finished 16th, trailing less-heralded teammate
Started skiing at the age    Marco Sullivan by seven spots.
of four, taught by his
father. A consistent         "It's a tough one to swallow. I'm still kind of
performer with six World     in a daze right now," said Rahlves, the
Cup downhill wins (and one   reigning world champion in super giant slalom.
in super-G), he had never    "It was just a very poor performance. The good
previously featured at a     thing is I've still got the Super G coming up
major championships, his     on Saturday."
best performance a 4th
place, missing the bronze    Lasse Kjus of Norway finished second for his
by 0.01 seconds. His last    fourth Olympic medal, matching his downhill
World Cup win came in        silver medal from 1998. He was a fifth of a
December in Bormio, Italy.   second behind Strobl. Eberharter was
A policeman by trade.        six-hundredths of a second behind Kjus.
Past Olympics: 1998 --
11th in downhill. He         "I have not failed. Many racers don't have a
missed the 1992 Games with   medal. It was not the best run, but it was not
a knee injury.               the worst. It was a bronze run," Eberharter
Other successes: 4th in      said.
the downhill at the 1997
world championships, 6th     Strobl, known as "Friz the Cat," never seemed
in 2001 world                to struggle on a bear of a course that featured
championships downhill.      dazzling drops and extremely tight turns. Even
Reaction: "I don't think     Eberharter said it reminded him of a roller
this will change my life.    coaster: "You have a lot of up and downs, left
I was happy before."         and rights."
-- Reuters
                             From the top of the course, exactly 9,289 feet
                             above sea level on Mount Ogden, skiers gazed at
the skyline of Salt Lake City about 35 miles away - and anti-terrorism
snipers surveyed the action.

Injected with water to make it icier and faster, Grizzly begins with a
74-degree jump- - akin to falling off the side of a cliff - and includes a
spot where racers fly almost 150 feet in the air.

Amazingly, not a single skier failed to finish the course despite speeds of
more than 80 mph. Frenchman Pierre-Emmanuel Dalcin crossed the finish line
with one ski -- after losing the other one on the way down.

Racers were cheered at the bottom of the course by flag-waving,
cowbell-ringing fans, many of whom arrived after the start of the race
because of massive traffic backups leading to the ski area.

Some fans who paid up to $95 per ticket did not arrive until an hour after
the start of the race, long after the top skiers had finished and the medals
were decided.

Eberharter made a few mistakes in the middle
of the course, spraying snow in the air as his
skis bobbled a bit on the icy slope. By
contrast, Strobl hardly seemed to leave a       
trace as his skis bit into the snow.            
Even without the injured Herman Maier, the      
popular Herminator, Austria dominated the       
event, as usual -- with three skiers among the  
top six finishers.                              
Eberharter, skiing ninth, got a fast start,     
but had trouble controlling his skis halfway    
down the course. His time was the best up to    
that point, but his lead did not last long.     
Strobl was up next, and when his time was
posted, Eberharter was stone-faced. Three       
skiers later, Kjus sped past Eberharter into    
second place.                                   
"It was not really my race. I made two or       
three small mistakes, in particular coming      
into the final wall," said Eberharter, who      
remains the favorite for the Super G and is a   
medal contender in the giant slalom.

The laid-back Strobl , who hopes to run his parents' farm when he retires,
said he does not expect the Olympic title to change his life.

"This is just sensational. But life was beautiful before this and it will
continue to be beautiful," he said. "Things won't change much."


End of MGNet Digest