GOTM Newsletter 01: (12 Oct 1999)
Dear colleagues,
welcome to the mailing list of GOTM, the General Ocean Turbulence Model.
(http://www.ifm.uni-hamburg.de/~wwwto/ResearchTopics/GOTM/gotm_main.html)
As you now, the very ambitious name of GOTM does not refer to its
present contents, but does give us a guideline for the future. GOTM
seems
to become more popular, since there have been more than 600 logins
to the GOTM web site since July 1999. You can check GOTM's
statistics by clicking on the small graph at the base of the
GOTM main page.
We are now 44 members on the GOTM mailing list coming from all over
the world. It is the hope of the GOTM makers that this number will grow,
and that this helps to further improve and dissiminate GOTM. You find
the GOTM mailing list attached to this mail as an html file.
We are considering to publish the mailing list with full addresses
on the GOTM web page. The intention behind this is to make contact
between turbulence specialists easier. However, if anybody has
objections against publication of full addresses, we will restrict this
to
names and email addresses.
In case you want to send something around to the mailing list, you have
two options:
1. make a group reply to this mail (I heard that this sometimes creates
trouble,
2. email (to gotm@bolding-burchard.com) us the contents of the mail,
you want to send and we will forward it to the list.
If you have any suggestions for GOTM, please report them to us.
Remember, that you may contribute to GOTM and thus become one of the
GOTM makers. If you know other people interested in turbulence modeling,
please tell them about GOTM. However, it should be kept in mind that
GOTM is an informal, non-commercial and non-funded project.
Below, you find some more specific discussion about GOTM and related
topics:
1. What is next in GOTM ?
2. Shall we change GOTM to FORTRAN 90 ? Your opinion is needed.
3. Our webmaster informs ...
4. Problems with the KPP model implementation into GOTM. Who can help ?
5. Recent publications supported by GOTM.
6. Special GOTM edition: Mussels and oysters fed by GOTM.
7. CARTUM, a concerted action on ocean turbulence.
8. Graphical user interface (GUI) for GOTM. Who can make it ?
We wish you all a golden and not too turbulent October,
Yours sincerely,
The GOTM makers
(Hans Burchard
Karsten Bolding
Manuel Ruiz Villarreal
Pierre-Philippe Mathieu)
******************************************************************
1. The release of GOTM2.0 is still under construction. We decided to
wait with the implementation of various further features until it is
clear weather GOTM will be changed into FORTRAN 90 or not (for this
question,
see the discussion under item 2.
The intended improvements for GOTM are listed in the 'roadmap' which
one finds on the GOTM web pages. The essentials of this are discussed
here.
Further suggestions are welcome.
Physical features:
- convective adjustment as simple mixing parameterization for instable
stratification. Very easy to implement, and not really worth for
being part of a turbulence model package. However, this is still
widely used for general ocean circulation models and will inside
GOTM basically be used as a benchmark for higher turbulence models.
- K-profile parameterization (KPP) for vertical mixing coefficients
according
to Large et al. [1994]. The KPP model, which is based on empirical
considerations has become very popular in the ocean modeling
community. We would therefore, although it is not a 'real' turbulence
model, like to have the KPP model inside GOTM, mainly as a
benchmark for higher turbulence models. However, we have not yet been
able to find the bug in our implementation, which does not enable us
to reproduce the results of Large et al. [1994] (see item 4 below).
- prescription of horizontal gradients for temperature and salinity
(and with that density). This will help to include secondary flows
into GOTM driven by internal pressure gradients. Often such
gradients are measured, or they can be estimated, or they can be
constructed for idealized scenarios. These gradients will be read
in from files or may be specified as analytical functions.
- linearization of equation of state: This will be a further option
besides the UNESCO equation. Especially for idealised scenarios,
such an option is desirable.
Boundary conditions:
- surface buoyancy flux for buoyancy equation. Such simple experiments
as pure convection scenarios by using the buoyancy equation will
then be included in GOTM.
- surface fluxes (precipitation - evaporation, P-E) for salinity
equation.
Implementing this BC is very easy, prescribing P-E however
somewhat complicated.
- relaxation of surface temperature and salinity to measurements as
simple
parameterization of surface fluxes. If information on P-E is missing,
such boundary conditions are often a solution.
Biological features:
- simple seagrass dynamics according to Verduin and Backhaus [1999].
A prototype for this already exists and works. Although a bit too
special for a turbulence model, this shows the possibilities of
water column model and does on the other side not hamper the
performance of the model if not used. (This is also to make the
biologists on the mailing list happy !)
- simple phytoplancton model. This model implementation also exists as
prototype. However, the set-up of scenarios with initial and
varification data is not yet finished. (For further discussion,
see item 6 below.)
Informatic features:
- reorganization of namelist input (Karsten, please add something)
- imposing a new system for time calculation
- restructuring code for easier inclusion of new features
2. FORTRAN 90 for GOTM ? We need your input.
Another matter to discuss is whether we should change the programming
language of GOTM from F77 to F90/F95. The main reason is that it will
-
technically - be much easier to include new features without
disturbing
the overall program structure using the F90 Module concept. Since
today
all comercial Fortran compilers can handle both F77 and F90 this is
not
a major problem. Also in the Free Software world there are free F90
compilers available for non-commercial use.
We would like to get some comments on the change. Please send us an
email.
3. Our webmaster (Pierre-Philippe Mathieu) informs,
The GOTM web pages are under heavy construction in the moment.
A beta-version already exists, but is (due to technical problems)
not yet linked to the Hamburg (Germany) server. This beta version
may now be viewed at
http://home13.inet.tele.dk/kbk/
The beta version already offers:
- frame web page.
- a list of web links
- information about the GOTM team.
The following features are intended to be included soon:
- an automatic mailing list.
- the publications in electronic format.
Any comments/suggestions/advice are warmly welcome at
"pp.mathieu@altavista.net".
4. It has been tried by Hans Burchard to implement the KPP model
into GOTM. This is easy in principal, because the paper
by Large et al. [1994] gives all the information needed.
However, a simple test case (free convection) is not at all
reproduced by the implementation. This implementation with
the test case and visualisation (gnuplot script) can be
downloaded from the GOTM main page (you will find it at the
bottom of the page). It would be very beneficial if anybody
(great work for master or PhD students) could find the bug.
Acknowledgments (as always with GOTM) are guaranteed.
Comments should be sent to Hans Burchard
(burchard@ifm.uni-hamburg.de).
5. Some papers have recently been written using GOTM.
Reprints may be requested by mailing to gotm@bolding-burchard.com.
[1] Burchard, H., O. Petersen and T.P. Rippeth 1998.
Comparing the performance of the Mellor-Yamada and
k-epsilon two-equation turbulence models.
J. Geophys. Res. 103, 10543-10554.
[2] Burchard, H., O. Petersen 1999. Models of turbulence in the
marine environment - A comparative study of two-equation
turbulence models. J. Marine Systems 21, 29-53.
[3] Burchard, H. 1999. Recalculation of surface slopes as forcing
for numerical water column models of tidal flow.
Appl. Math. Mod. 23, 737-755.
I should be reminded that the GOTM report may be downloaded as
postscript file from the GOTM web page or requested directly from
the GOTM makers (gotm@bolding-burchard.com)
6. Special GOTM edition: Mussels and oysters fed by GOTM.
---------------------------
* Biological pelagic model:
---------------------------
A biological model with 4 compartments (Phyto/Zooplankton, Nutrients,
Detritus) has been implemented in GOTM in a modular way.
Its main originality is that biological processes are handled via
a formalism of "operator" functions which provides the following
advantages:
(1) It offers a set of biological parameterizations that enables
an easy choice/inclusion of various ecological processes,
(2) It enhances lisibility and limits the risk of bugs,
(3) It is a practical approach for inverse modelling.
The routines have been tested and fully documented. The next step
is to validate parameters through comparison with biogeochemical
in-situ data at various oceanographic stations (ex. Papa, Flex).
Although not yet released on the GOTM web site, anyone interested in
using the model or collaborating to its validation is invited to
contact P.P Mathieu at "pp.mathieu@altavista.net".
---------------------------
* Biological benthic model:
---------------------------
The pelagic biological model has been coupled to a benthic model which
represents the main processes in the benthic population (e.g. filtration
& rejection by mussels). The implementation of the numerical model has
fully benefited from the general formalism of GOTM (especially the easy
use of Dirichlet or Neuman boundary conditions). If you are interested
in this model, do not hesistate to contact P.P. Mathieu.
7. CARTUM, a concerted action on geophysical turbulence.
The members of the GOTM user group will be added to the CARTUM
interest group which is informed on an irregular basis on
activities related to CARTUM.
Below, the executive summary for CARTUM is given. For more information,
please
visit the CARTUM home page at
http://www.ifm.uni-hamburg.de/~wwwto/ResearchTopics/CARTUM/carthome.htm.
Executive summary of CARTUM:
Comparative Analysis and Rationalization of Second-Moment
Turbulence Models (CARTUM)
Concerted Action of the Commission of the European Communities,
Directorate General XII for Science, Research and Development
Marine Science and Technology, MAST III, 1994 - 1998
In CARTUM 40 scientists from 24 Institutions and 15 countries
cooperate
Scientific Coordinator: Prof. Dr. Jürgen Sündermann, Co-Coordinator:
Dr. Helmut Baumert
Project Manager: Dr. Ilse Hamann
Executive Summary
Today a wealth of microstructure observations have been made in the
open ocean, on the shelf and in estuaries. Simultaneously
high-resolution
numerical experiments on different realistic flow configurations
(Direct
Numerical Simulation, DNS, of turbulence, Large Eddy Simulation, LES)
are increasing in number at a fast rate. More and more turbulence
closures are being implemented in numerical circulation models with
second-moment closure schemes playing the role of workhorses in most
cases.
Except for a few papers authored by CARTUM Partners, some of
them still in print, there is no attempt to bring microstructure
observations, closure models and direct numerical simulations together.
Hence, one would think that significant progress in the representation
of
small-scale processes in circulation models could be gained from
systematic comparisons of the above methods.
Consequently, CARTUM sets out to meet the challenge to merge
theoretical aspects, new field observations, laboratory measurements,
and computer-generated data sets with computational and numerical
considerations with the ultimate goal of significantly improving
second-moment closure models. CARTUM aims at bringing together
the critical mass of unresolved problems, diverse but pertinent
datasets,
and a host of scientists ranging from world-leading experts in the
field of
marine turbulence to younger researchers with specific experience.
CARTUM will provide
- improved models for the computation of turbulence
- characteristics (e.g. Kolmogorov, Batchelor and Thorpe scales)
in the upper mixing layer,
- general comparisons between the two most prominent
representatives of second-moment closure models,
- either a unified canonical community model or a family of
significantly improved models for different applications and/or
conditions.
Contacts:
baumert@hydromod.de
hamann@ifm.uni-hamburg.de
8. Graphical user interface (GUI) for GOTM. Who can make it ?
It would greatly help to increase the ease of use of GOTM
if we would have a graphical user interface for GOTM,
through which the namelist.inp files are edited, the
program started and afterwards the results displayed. This GUI
should preferably be written in JAVA. This could be e.g. an
interesting task for a diploma thesis for computer
scientists. Of course, this work cannot be paid, but the
programmer will be acknowledged and will become a GOTM author.
If you know somebody interested, please contact Karsten Bolding
(Karsten@Bolding-Burchard.com).