I would also like to make it known that a few of us are looking at using your work within integrated modelling frameworks such as: http://www.uvm.edu/giee/SME3, fed by datasets such as http://www.icess.ucsb.edu/modis/modis-lst.html, local meteorological feeds, and http://trmm.gsfc.nasa.gov/. We aspire to run detailed, heterogeneous models such as DSSAT, CROPWAT, PALMS, EPIC, etc., on a landscape level in conjunction with and other linked process models such erosion, household economic, etc on a per-cell basis, out on compute farms like Teragrid. One can imagine how a model of different types of forest growth, technology introduction, and human behavior might drive a model such as: http://www.worldbank.org/nipr/gis/gisother.htm.

Therefore I would ask this group, as it sets forth, to systematically consider:

a) decoupling models from GUIs

b) platform independence (i.e. a farm of Linux servers versus purely Windows desktops)

c) mechanisms to perform variable introspection and setting ("peek" and "poke") under single time-step control

d) checkpointing and persistence, etc.

e) general discussions about shared taxonomies.  For example, I just returned from the OpenGIS technical committee meetings at UN headquaters in NYC, where a representative from FAO was discussing their "Geonetwork" and ISO 19115 classifications, a representative from CAMESA was discussing their taxonomies for South American biodiversity, and folks from the International Working Group on Taxonomic Databases (http://www.tdwg.org/) discussing their work. If all three are fundamentally talking about "corn," they would still be wildly different computer-compatible records.

Anyway, just some items to keep in mind as you evolve this incredibly important body of work.

best regards -

Chris Nicholas

-----Original Message-----
From:   Matthias Langensiepen [mailto:[log in to unmask]]
Sent:   Mon 1/19/2004 8:16 AM
To:     [log in to unmask]
Cc:
Subject:             New discussion forum on plant modelling
Dear members of the DSSAT-L list,

a new interdisciplinary discussion forum on plant
modelling has been launched. Its main purpose is
to enhance communication among researchers working
in different areas of plant science such as eco-physiology,
molecular biology, micrometeorology, soil science, ecology
and agriculture. It is not targeted at a particular
level of biological organization and attempts to
improve our understanding of whole-plant
functioning.

To subscribe send an email to [log in to unmask]
and write in the message body:
SUBSCRIBE PLANTMODELS-L YourFirstName YourLastName

Gerrit asked me to explain the differences between
this new listserver and the DSSAT and AGMODELS listservers
and why it would be interesting for you to subsribe:

DSSAT's and AGMODELS's prime aims are to enhance our
understanding of yield formation processes and
to predict consequences of crop system manipulation.
The complexity of cropping systems models has
considerably increased during the past two decades.
Yet, practical considerations and statistical
restrictions caused the mainstream of crop models
to remain on the system level.

PLANTMODELS-L opens the horizon to deeper levels of
biological organization and delves into complexity.
This will inevitably increase the number and range
of modelling errors - a phenomena which is known
as the complexity paradox. In crop modelling it is
a common practice to keep the scope of models as
simple as possible to reduce such errors. However,
this also means to abandon complexity which cannot be
the goal of quantitative plant science. PLANTMODELS-L
attempts to fill this gap. It regards models as useful
heuristic tools to reduce search in seeking problem
solutions for research issues in ecological plant
physiology which are typically marked by daunting
complexities.

PLANTMODELS-L tries to bring together modellers
from different disciplines and to produce
synergistic effects. Same research issues are often
treated from different viewpoints. Solutions may
be process or system oriented. An interdisciplinary
discussion forum covering different scales and topics
will certainly produce synergistic effects.

Members of the DSSAT and AGMODELS lists will benefit
from joining this new list by receiving complementary
information about modelling plant physiological processes.
Besides, discussions are not only related to a few
economical important plants but to any plant specie
that is of current interest.

I wish you pleasure in participating in the discussions
in case you decide to join PLANTMODELS-L.

Regards,
Matthias Langensiepen

--
Matthias Langensiepen, PhD
Modelling Plant Systems
Faculty of Agriculture and Horticulture
Humboldt-University of Berlin
Berlin, Germany