Archive AGMODELS-L: file log9503, part 1/1, size 225583 bytes:
------------------------------ Cut here ------------------------------
I agree Roger. And I would go on at some length about those who apply
models willy-nilly without a great deal of thought about why they need to
use a model at all or at even greater length about examples where the
modelling seems to have become an end in itself rather than a research
(or decision support) tool but it is far too early in the morning to
raise my blood pressure that much.
Val Snow
________________________________________________________________________
Val Snow CSIRO Division of Water Resources
vos@adl.dwr.csiro.au Private Mail Bag No. 2
61 08 303 8743 (Phone) Glen Osmond, Adelaide
61 08 303 8750 (FAX) South Australia, 5064
On Tue, 28 Feb 1995, Roger Smith wrote:
> I have read the discussions of "spreadsheet based models" with some
> interest. Am I wrong or is this another case of an emphasis on usability,
> presentation, etc, without regard to soundness of soil physics and
> chemistry? The sales pitch in a recent message used all the right words,
> such as on screen, color coding, "powerful" output, but said absolutely
> nothing about how the model figures out pesticide movement in a complex and
> heterogeneous soil profile. Let us turn from magic and salesmanship
> towards process science, and be more honest investigators and less
> salesmen/women.
> (I'll bet this comment gets someone worked up...)
> Roger E. Smith
> roger@lily.aerc.colostate.edu
>
I have a great deal of sympathy for the comments of Roger Smith below.
However there is a place for ALL kinds of efforts in this business of
making our agricultural practices more sound. But as Roger says,
sound physics and chemistry (add plant physiology and micro-
meteorology as well) are an essential basis for good work.
The most complex research models can provide a basis for vastly
simpler application models, when complex interactions can be
approximated for specific sets of conditions. Spreadsheet models
might be a good way to do that approximation if the underlying
processes are understood.
Tom Hodges
Tom Hodges, Cropping Systems Modeler
USDA-ARS email: thodges@beta.tricity.wsu.edu
Rt. 2, Box 2953-A voice: 509-786-9207
Prosser, WA 99350 USA Fax: 509-786-9370
== ## Rent this space ## ==
If this represents anything, it is only my opinion.
On Tue, 28 Feb 1995, Roger Smith wrote:
> I have read the discussions of "spreadsheet based models" with some
> interest. Am I wrong or is this another case of an emphasis on usability,
> presentation, etc, without regard to soundness of soil physics and
> chemistry? The sales pitch in a recent message used all the right words,
> such as on screen, color coding, "powerful" output, but said absolutely
> nothing about how the model figures out pesticide movement in a complex and
> heterogeneous soil profile. Let us turn from magic and salesmanship
> towards process science, and be more honest investigators and less
> salesmen/women.
> (I'll bet this comment gets someone worked up...)
> Roger E. Smith
> roger@lily.aerc.colostate.edu
>
>Hello,
>my name is Helmut Schaefer. I am a chemical engineer, working at Bayer, Germany
>...(snip)
>Currently I am concerned with the problem of kinetic adsorptions effects. There
>are modells on the market which allow the simulation of these effects (i.e
>VARLEACH). But I am not sure about how these models can be parametrized, this
>is
>what is the best experimental setup to gain these model parameters. Has anybody
>expierence in this field?
>...
I have been measuring the kinetics of sorption of pesticides in soils and the
results have become part of the research model RZWQM. The model allows for
three options (a) instantaneoud equilibrium (b) partial instntaneous approach
to equilibrium (c) kinetic approach throughout using the 2-sequential- i
equilibrium model. I suggest (b) seems to be a good compromise and has
also been successful in column leahing studies. RZWQM also allows for a
slower "binding" step that mimics the often-observed increasingly less-
desorbable behavior of pesticides in soils after weeks to months of residence
time.
This is a hot topic in the pesticide modeling area these days. Would you,
Dr. Schaefer, describe how VARLEACH works in this regard? Also, Dr. Smith,
how about OPUS? The fact is, kinetic studies with real soils and pesticides
indicate that even in well-stirred batch studies the sorption process is
complex. How much of this complexity needs to be incorporated into our
models?
Don Wauchope
Hello !
My name is Brian Cooper and I am a PhD student at Purdue University
in West Lafayette Indiana (Go Boilermakers !!!!). I am currently working
on groundwater vulnerabilty models such as DRASTIC and SEEPAGE. These models
take into account various soil attributes as well as certain hydrgeologic
features which include vadose zone and aquifer media, net aquifer recharge,
hydraulic conductivity, and water table depth. I hope to obtain enough data
to map the entire midwest region. I will then varify these maps with
USGS and DNR well monitoring sites. I will also be using pesticide transport
models such as PATRIOT, GLEAMS, and CMLS. If everything goes well I will be
presenting this information at the ASAE conference in Chicago and the AWRA
conference in Houston.
Later
Cooper
In response to Abert Weiss's question about soil heat flux and albedo, I have
just (this morning!) finished a Visual Basic program for calculating
evapotranspiration flux from automatic weather station (AWS) data on hourly
(or shorter) timesteps. This can be used with an AWS with the same config-
uration as quoted (i.e. solar radiation but not net radiation). I have
estimated net radiation and soil heat flux in the following way:-
1. Longwave radiation is estimated from the longwave radiation flux
for clear skies (from Monteith and Unsworth, 1990) reduced according to the
ratio of actual solar radiation / potential for the time of day, year and
location. This is the method used by Campbell Scientific in their on-line
calculation of ET for their AWS and is described in a Technical Note produced
by them.
2. Net radiation is calculated from solar radiation, the albedo of the surface
and the longwave radiation (above). Albedo is assumed to be constant for the
given land cover. (You wanted something more elegant than this didn't you?).
3. Soil heat flux is taken to be 0.1 * net radiation. This is a major
simplification.
The above appears to work OK during the daytime (when ET is taking
place). It is probably not very good for night-time net radiation
or soil heat flux, but then as the stomata are closed and the canopy
resistance is high, the simplification does not have a major effect on ET
and therefore it suits my purposes!
My program for ET will work with almost all AWS that produce an ASCII data
file, and uses the Penman-Monteith method to calculate ET. Default
resistances and albedos are given for standard land covers, but these can
be changed for other cover types. Further details available on request!
---------------------------------------------------------------------
Tim Hess
Department of Water Management
Silsoe College
Cranfield University
Silsoe Tel (0)1525 863292
Bedford, MK45 4DT Fax (0)1525 863000
UK e-mail t.hess@silsoe.cranfield.ac.uk
----------------------------------------------------------------------
I noted with interest (and suppressed glee) the posted-complaints that
yesterday's brief description of TA-DA! was long on "ease-of-use" and short on
physics. Having written various models with a wide-spectrum of physics-based
complexity, I must say that getting the physics right is only a very small part
of writing a USEFUL model. The physics behind TA-DA! are admittedly simple,
but so is F=m*a, which has held up reasonably well over the past few centuries.
The primary technical advantage of producing a model in a spreadsheet
environment is the tremendous ease of accessing, cutting, and pasting vast
amounts of relevant DATA (please disregard the anagram) into the model. But
there are numerous other features making the model more useful, including the
ability to generate graphs and print-outs in virtually any user-defined format,
the portability between platforms, and the immediate interface to the user's
operating system environment.
Comments from others?
comments in Ground Water Modeling Newsletter made by Adrian Brown, IMHO
apply to modeling ag leaching.
... d) Average conductivities, concentrations, and other parameters
are of little use in the evaluation of solute transport: transport
is dominated by non-average conditions.
If this is true is there any utility to current ag models like
PRZM Gleams?
don fontaine
Is Roger Smith saying, "Let's keep our models & plot results in our
labs, because if we make it easy for a non-scientist to use, this
work might show some benefit." ?? I see the point about making sure
the science is OK, but in extension educational efforts, we are
going beyond the printed publication which is not site-specific &
is usually out of date by the time it gets distributed - and programs
that our clients can use and that we can explain to them in workshops
are very important. Bob Peart, Agr. & Biol. Eng., U. Florida.
>I have read the discussions of "spreadsheet based models" with some
>interest. Am I wrong or is this another case of an emphasis on usability,
>presentation, etc, without regard to soundness of soil physics and
>chemistry? The sales pitch in a recent message used all the right words,
>such as on screen, color coding, "powerful" output, but said absolutely
>nothing about how the model figures out pesticide movement in a complex and
>heterogeneous soil profile. Let us turn from magic and salesmanship
>towards process science, and be more honest investigators and less
>salesmen/women.
>(I'll bet this comment gets someone worked up...)
>Roger E. Smith
>roger@lily.aerc.colostate.edu
>
>From all the postings, I would say you were right about getting poeple
worked up, Roger. The debate about simple vs. complex models has been
ongoing for over 20 years now in crop simulation, and is yet unresolved and
probably never will be. The critical point to make here is that models are
built usually with a purpose in mind. If one is trying to explain complex
relationships involving usually non-linear biological, chemical, and
environmental interactions, then the model better have some non-trivial
treatment of those processes. However, if one is trying only to give
insight and solution to a small subset that solves a problem for the end
user, then a simple model may do the job.
>From our 10 years of experience in delivering the GOSSYM/COMAX System to
cotton growers and consultants across the cotton belt for decision support
in cotton crop management, the user interface and ease of use of the
software system is critical to acceptance and continued application. I
agree totally that there must be substance behind the interface.
Another problem that needs to be addressed in this discussion is that
research scientists in ARS do not get credit in the RPES system for
developing user friendly GUI's. At the same time, our end users demand
them, and they are not trivial exercises and take substantial time and
resources to develop, maintain, and extend. We choose to do the development
and maintenance work and to pay the price at evaluation time.
Dr. James M. McKinion Ph. 601-324-4375
USDA-ARS FAX 601-324-4371
Crop Simulation Research Unit Email mckinion@csrumsu.ars.ag.gov
P. O. BOX 5367
Mississippi State, MS 39762
USA
I have built models within spreadsheets and other programs and they have
their uses, especiallly when developing the model. The built in graphics
and dat manipulation facilities are very useful. However, I soon come up
against the limitations of the application program and have to start
programming the model from scratch.
In addition, I find that whether you are using an application program
like a spreadsheet or a programming language, the tools you use can tend
to direct the way the model developes. This is less so when programming
from scratch.
Graham Lewis
On Wed, 1 Mar 1995 08:12:25 -0600,
Bob Peart <peartr@water.agen.ufl.edu> wrote:
> Is Roger Smith saying, "Let's keep our models & plot results in our
>labs, because if we make it easy for a non-scientist to use, this
>work might show some benefit." ?? I see the point about making sure
>the science is OK, but in extension educational efforts, we are
>going beyond the printed publication which is not site-specific &
>is usually out of date by the time it gets distributed - and programs
>that our clients can use and that we can explain to them in workshops
>are very important. Bob Peart, Agr. & Biol. Eng., U. Florida.
This reply is not untypical but more cynical than usually heard when the
subject of model content is raised. Of course I said nothing of the sort.
Model soundness and model utility are neither inversely related, nor or
they mutually exclusive. But what is done when one realizes that the way
nature works is too complex to make into an "easily used" model? Too often
the soundness of process is abandoned, with a variety of rationalizations,
in favor of ease of use, availablity of "parameters", etc. The "user" is
all in favor of ease of use. It is the scientist who is responsible for
what is behind the colored screen interfaces, but who often fools him or
herself in favor of making the model easy to use.
Roger E. Smith
roger@lily.aerc.colostate.edu
On Wed, 1 Mar 1995 08:01:09 -0600,
DONALD D FONTAINE <dfontaine@dow.com> wrote:
>comments in Ground Water Modeling Newsletter made by Adrian Brown, IMHO
>apply to modeling ag leaching.
>
>... d) Average conductivities, concentrations, and other parameters
>are of little use in the evaluation of solute transport: transport
>is dominated by non-average conditions.
>
>If this is true is there any utility to current ag models like
>PRZM Gleams?
>don fontaine
Don: An excellent point. However, not all cases important to pollutant
movement are dominated by variance. The existence of heterogeneity is
often used to trash or at least discount process-based models as being good
for the laboratory only, and implicitly or explicitly to justify the use of
gross simplifications. But I have never seen a demonstration that
simplifications such as "field capacity" based models do a better job of
simulating the effects of heterogeneities.
Roger E. Smith
roger@lily.aerc.colostate.edu
On Wed, 1 Mar 1995 07:22:11 -0600,
DIGUST@ccmail.monsanto.com <DIGUST@ccmail.monsanto.com> wrote:
>I noted with interest (and suppressed glee) the posted-complaints that
>yesterday's brief description of TA-DA! was long on "ease-of-use" and short on
>physics. Having written various models with a wide-spectrum of physics-based
>complexity, I must say that getting the physics right is only a very small part
>of writing a USEFUL model. The physics behind TA-DA! are admittedly simple,
>but so is F=m*a, which has held up reasonably well over the past few centuries.
>
>The primary technical advantage of producing a model in a spreadsheet
>environment is the tremendous ease of accessing, cutting, and pasting vast
>amounts of relevant DATA (please disregard the anagram) into the model. But
>there are numerous other features making the model more useful, including the
>ability to generate graphs and print-outs in virtually any user-defined format,
>the portability between platforms, and the immediate interface to the user's
>operating system environment.
>
>Comments from others?
Indeed, "geting the physics right" has nothing whatever to do with
"usefulness" as such: they are independent objectives, and you comments
merely prove that the latter objective (a perfectly valid one, as long
as it is kept in balance )is dominant in your efforts, as I obsurved.
Roger E. Smith
roger@lily.aerc.colostate.edu
I agree compeltely. Of course it is important to have the science right,
but I have seen to many models in which the user interface was positively
archaic. This has a direct impact on how much *science* can be done with
the model since wrenching a crummy interface takes *lots* of time. It also
has implications for the scientific soundness of the model because it
affects how quickly one can understand and get at the underlying assumptions
of the model. A spreadsheet model has the admirable advantage of having
its working front and center where the user can see and review them. In
addition the interface speads up the volume of science that can be done.
This is important because in many instances the speed with which the
model can be usefully used is the rate limiting factor on getting meaningful
results.
Jonathan Haskett
On Wed, 1 Mar 1995 DIGUST@ccmail.monsanto.com wrote:
> I noted with interest (and suppressed glee) the posted-complaints that
> yesterday's brief description of TA-DA! was long on "ease-of-use" and short on
> physics. Having written various models with a wide-spectrum of physics-based
> complexity, I must say that getting the physics right is only a very small part
> of writing a USEFUL model. The physics behind TA-DA! are admittedly simple,
> but so is F=m*a, which has held up reasonably well over the past few centuries.
>
> The primary technical advantage of producing a model in a spreadsheet
> environment is the tremendous ease of accessing, cutting, and pasting vast
> amounts of relevant DATA (please disregard the anagram) into the model. But
> there are numerous other features making the model more useful, including the
> ability to generate graphs and print-outs in virtually any user-defined format,
> the portability between platforms, and the immediate interface to the user's
> operating system environment.
>
> Comments from others?
>
>
> Is Roger Smith saying, "Let's keep our models & plot results in our
>labs, because if we make it easy for a non-scientist to use, this
>work might show some benefit." ?? I see the point about making sure
>the science is OK, but in extension educational efforts, we are
>going beyond the printed publication which is not site-specific &
>is usually out of date by the time it gets distributed - and programs
>that our clients can use and that we can explain to them in workshops
>are very important. Bob Peart, Agr. & Biol. Eng., U. Florida.
I would heartily endorse this viewpoint. During the past few years I've
found it necessary to move from model development more and more into the
development of techniques for making this information available to
managers and other non-scientists. We started with spreadsheets and are
moving in the direction of decision support systems based on expert
system techniques.
A strong advantage of spreadsheets in dealing with managers is that they
use a computer language that most managers know.
I've written a couple of papers on this, references follow (some of
these are available on our Web server).
Silvert, William. 1983. Ecological modelling with
electronic spreadsheets. Can. Math. Soc. Appl. Math.
Notes 8:1-8.
Silvert, William. 1984. Teaching ecological modeling with
electronic spreadsheets. Collegiate Microcomputer
2:129-133.
Longhurst, Alan, and William Silvert. 1985. A management
model for the Great Bustard in Iberia. Bustard Studies
2:57-72.
Silvert, WIlliam. 1989. Modelling for managers.
Ecological Modelling 47:53-64.
Silvert, William. 1992. Many risks, many models:
addressing the variety of problems that pollution can
cause. ICES 1992 Mini-Symposium on Ecosystem Modelling
as a Tool to Predict Pollution Associated Risks for the
Marine Environment. Proc. ICES Statutory Meeting, C.M.
1992/Mini:4. 4 p.
Silvert, William. 1992. The Desktop Scientist: Models that
a Manager can Use. Presented at 8th ISEM Int. Conf. on
Ecological Modelling.
Silvert, William. 1994. A decision support system for
regulating finfish aquaculture. Ecological Modelling
75/76:609-615.
Silvert, William. 1994. Decision Support Systems for
Coastal Zone Management. To be presented at Coastal
Zone Canada '94 on 20 September 1994.
Silvert, William. 1994. Putting management models on the
manager's desktop. J. Biol. Systems 2(4):519-527.
Silvert, William. 1994? Decision support systems for
aquaculture licensing. J. Applied Ecology (in press).
Presented at Proc. Workshop on Fish Farm Effluents and
their Control in EC Countries, Hamburg, Germany, Nov.
1992
--
Bill Silvert, Habitat Ecology Div., Bedford Inst. Oceanography
P. O. Box 1006, Dartmouth, Nova Scotia, CANADA B2Y 4A2
Personal InterNet Address: silvert@biome.bio.ns.ca
HED runs a WWW server at URL=http://biome.bio.dfo.ca
>The primary technical advantage of producing a model in a spreadsheet
>environment is the tremendous ease of accessing, cutting, and pasting vast
>amounts of relevant DATA (please disregard the anagram) into the model. But
>there are numerous other features making the model more useful, including the
>ability to generate graphs and print-outs in virtually any user-defined
format,
>the portability between platforms, and the immediate interface to the user's
>operating system environment.
I would agree from a developmental standpoint. In fact, most of the modeling
I've worked on (crop disease models) has actually begun with a spreadsheet.
This type of platform does make it easy to try multiple 'what if' scenerios,
and is an unparalled method of being able to 'visualize' the data before
undertaking a large statistical analysis. Unfortunately, as a DSS or any
kind of an end user management tool, there is simply too much information
displayed. From what I've seen in production agriculture, the majority of
farm managers still see computers as oversized typewriters. With this in
mind, a 'black box' approach with a customized interface and summary
information displays is superiror to several screenfulls of 'raw' data. I
realize that a lot of functionality can be written in to a spreadsheet
application using macros and named ranges, but with today's visual
programming tools and high level languages, implementation as a stand-alone
application is not much more effort than 'programming' the spreadsheet.
On another line of thought, what impact does spreadsheet modeling have on
standardization of data input formats, i.e. weather data, soil data etc. I
know IBSNAT has undertaken an effort to standardize these types of inputs.
IMHO spreadsheets make it easy to design formats and move large blocks of
data, but to import data from a non-standard source, and reformat it is
somewhat time consuming, and definitely error prone. I'd be interested to
hear from others about data standardization.
c 'ya
Woody
/**************************************************************************/
Jeff Woodard Glades Crop Care voice: 407-746-3740
949 Turner Quay fax:
407-746-3775
Jupiter, FL 33458
e-mail: woodard@igc.apc.org
-My opinions are my own-
/**************************************************************************/
At the risk of being somewhat commercialistic, I've got 17 rain guages and
20 temperature probes for which I have no use for. These are the Campbell
Scientific TE-525 switch closure tipping bucket rain guages, and 107-B
thermister temperature probes (including 6 plate gill radiation shields),
some of which have never been out of the box. We obtained this equipment
from an insurance company bankruptcy, and don't have enough weather stations
to use them on.
I'd be interested in selling them (at a lower price than available new), or
trading for some relative humidity or solar radiation probes. I can be
contacted by e-mail at woodard@igc.apc.org.
Thanks
Woody
/**************************************************************************/
Jeff Woodard Glades Crop Care voice: 407-746-3740
949 Turner Quay fax:
407-746-3775
Jupiter, FL 33458
e-mail: woodard@igc.apc.org
-My opinions are my own-
/**************************************************************************/
The spreadsheet models I use to determine input costs are pretty simple.
I copy them from Iowa State University paper-based spreadsheets.
What amazes me is how hard it is to get any modeling at all, good bad or
indifferent, to use in aiding in decision making for the guy who drives
the tractor.
As variable rate technology becomes more avaliable and affordable, it
would seem to me that modelling would help make some kind of sense out of
the many inputs involved in trying to figure out why a certain piece of
ground grew a certain yeild of corn. It sure isn't just fertility, we
have at least figured that out. But admitting the truth of that begs
the more difficult question of what combination of weeds, compaction,
slope, soil, variety, moisture, weather, growing degree days, fertilizer,
cultivation, herbicide, insecticides DID produce the yield.
I'm glad to get all the open models I can.
Since I don't use models to make the decisions, just help me, I figure it
is my responsibility if I use one wrong. My problem is, I can't find
them at all to run on my little 386DX33 with 8MB Ram.
Jim - Farmer - Iowa City, IA,
jnmeade@blue.weeg.uiowa.edu
signoff queiroz@ecn.purdue.edu
Dear ag modellers,
I am working at the International Potato Center CIP's station in Quito,
Ecuador, since 1990 as an associate agronomist, more recently also
doing a PhD at Wageningen on constraints to potato production in
Ecuadorian highlands. Yield-gaps are hoped to serve as a measure of
potential impact of relieving each particular contraint. After some
years of conventional on-farm yield gap trials, I started using
simulation to study the various levels of potential yields that are
relevant on the farmers' fields up here. I am considering the
following levels:
(1) (technologically possible,) defined by intercepted irradiation
(temp sum -{cultivar}-> ground cover -> fraction intercepted)
(2) limited by drought
(3) reduced by late blight
(4) actually harvested by the farmers (with obviously a range of
other constraints not simulated by the model)
We use a modification of LINTUL (Spitters, 1987). We produced a new
version that is based on the physical expansion of the plant, later the
row, finally the closed canopy, supposing a density per volume of leaf
area within the canopy. This was necessary because the distances
between plants and rows vary among the crops. We also made a three-
dimensional hourly sub-model to deal with the varying slope,
orientation, obstruction by the surrounding mountains, etc. site
factors.
Drought affects both light use efficiency and the development of ground
cover, whereas late blight only affects the latter. Going into details
would make this story too long. All the models for the levels 1, 2 and
3 were parameterized (analysis ongoing) using on-station field trials
at two altitudes 3050 and 2700 masl. They are being validated on-farm
in Carchi, a major potato production area, where year-round planting is
possible and we get ample variation in environmental conditions, both
in space and time. We expect/require that the model predict a
considerable part of the variation in the actual yields, and that the
absolute simulated values be slightly above level 4. In the previous
on-farm trials we got five-fold differences in yields, not explained by
differences in production technology. Another "validation" would be if
the model could address part of that variation to topography, weather,
soil and cultivar.
I notice I'm speaking about "we" since the second paragraph. We
include also four Dutch master's thesis students from Wageningen Ag.
Uni, all supervised by Louise Fresco (Prof Agronomy/Tropical Crops), I
myself also co-supervised by Anton Haverkort of AB-DLO, the
Netherlands. Dewi Hartkamp works on (1), Stefanie de Kool and Remco
Roeland on (2), Ingrid Schimmel on (3). Also an Ecuadorian student
belongs to the team; Luis Benavides from Central University of Quito
worked on the "old LINTUL". For email addresses for all of them see
Cc: list.
I prefer modelling in SAS, though I have also tried basic and the FSE
(Wageningen). I find SAS code very informative to non-programmers, and
it's easy to integrate the modelling effort with statistical analysis,
graphics etc.
Any contacts, ideas, questions, etc are welcome.
Best regards,
What is your asking price?
As with Mr. Gustafson, I have also found this spreadsheet controversy
amusing. I've seen this bias, as expressed in some of the recent e-mail, in
other fields as well. It generally takes one or both of two forms.
First, if the tool appears to be simple, then the results produced by the
tool must not be useful. The fallacy in this case is misunderstanding the
capability of the tool. The fact is that there is nothing that can be modeled
in a programming language that can't be modeled in a spreadsheet package such
as Excel 5.0 (in which I happen to do most of my work). When you combine this
capability with a spreadsheet's tremendous flexibility the question should
really be: why is anyone still using anything else?
The second bias is the presumption, as in Mr. Roger Smith's letter (2/28),
that use of a spreadsheet automatically means the abandonment of sound
underlying scientific theory. Why would anyone expect this to be the case?
The same physics and chemistry that goes into Mr. Smith's model can just as
easily (in fact, more easily) go into a spreadsheet model.
Responses, yea or nay, are welcomed.
Eric Fraint
YPTC@aol.com
Eric,
Your analysis seems about right on, and I say again that the result is
that we farmers who need one get neither a simple one nor a complicated
one that we can afford or readily use.
One thing for sure - any model used by the government to regulate us
should be readily available in the public domain, whether we have a
machine capable of running it or not. And if it's too complicated for
some dumb farmer to use, maybe it's too complicated for some dumb farmer
to apply. I refer specifical to RUSLE and similar soil erosion models
that we are expected to comply with by using tillage means, rather than
that we could use for "what-if" analysis on our own to consider the
affect of various practices.
Jim - Farmer - Iowa City, IA,
jnmeade@blue.weeg.uiowa.edu
Someone posted their opinion (as opposed to fact) that putting out
software with a decent User Interface was easy these days because of
the programming tools available.
Really? ;^)
Is this on Mars or Earth? :^)
If that person's opinion were indeed fact, MicroSoft would ship
bug free products on-time! After all, Bill has about $400 Million
after-taxes cash flow, the top 0.1% of the programming talent in
the World and advanced access to its own programming APIs.
Irrespective of progamming tools used, an application hitting 35,000 lines
of source code is not likely to be "easy"-anything. :^) New tools,
even Visual Basic, have *steep* learning curves. They are not
bug free, do not have *all* features desired/needed in a single product,
and make increasing expenditure demands on the end-user machines
(or else it will not run----or will run so slowly as to become shelf-ware).
:^)
--
Steve Modena ab4el@Cybernetics.NET
--------End of Unsent Message
------- Forwarded Message Follows -------
From: "Arkadiusz Salski" <OEKO-PROJEKT/AREK>
Subject: fuzzy-ecology - mailing list
CALL FOR PARTICIPATION
in the fuzzy-ecology mailing list
A mailing list for everybody interested in "fuzzy" topics
in ecology has been established. It is important for our
"fuzzy community" in ecology to keep in touch and have an
opportunity for information exchange and for discussion of
issues regarding fuzzy sets and fuzzy logic applications in
the field of ecology. I would like to invite you to participate
in this new mailing list created for this reason.
If you would like to subscribe to fuzzy-ecology, please send
a message to
maiser@pz-oekosys.uni-kiel.d400.de
with
SUBSCRIBE fuzzy-ecology
in the body of an otherwise empty mail.
I look forward to hearing from you. With kind regards
Arkadiusz Salski
--------------------------
Dr. Arkadiusz Salski
Ecosystem Research Center, University of Kiel
Schauenburgerstr. 112, D-24118 Kiel, Germany
arek@pz-oekosys.uni-kiel.d400.de
>... d) Average conductivities, concentrations, and other parameters
>are of little use in the evaluation of solute transport: transport
>is dominated by non-average conditions.
>
>If this is true is there any utility to current ag models like
>PRZM Gleams?
>don fontaine
Given the season-to-season variability in weather, I have found that
the following approach works best when using the above-mentioned
models to evaluate changes in leaching potential from changing
ag. practices: (1) since model output is given as average annual
values, I use a big weather data set (50-year simulation) and use
the output as a first-tier evaluation. Look at changes between
one scenario versus another; (2) a second-tier evaluation can then
be done to assess probability of a critical event occuring, such as
leaching of an herbicide below the root zone at concentrations considered
unacceptable. This is where some researchers bring in Monte Carlo
or other probabilistic tools. One could also just do model runs
under worst-case scenarios (very high rainfall immediately after
herbicied application, for example). Again, look at relative change
rather than the absolute values of the output. All in all, the models
are quite useful, *if* one approaches their use in this manner.
--
Ruth Shaffer-------------------------- |\ _,,,--,,_ ,) -
"You cannot look at a sleeping cat, /,'.-''' -, ;-;;'
and be tense."-Jane Pauley |,4- ) )-,_ ) /\
(ac698@leo.nmc.edu)-------------------'---''(_/--' (_/-'-----
My 2 cents:
When we talk about developing good models that are both useful and easy
to use, we are talking about *three* different processes: modeling,
decision support, and software engineering. Let's keep them separate, at
least in our discussion, if not in our computer code.
1. Models don't solve problems; they mimic some real phenomenon. Models should
be able to be objectively tested (verified, validated, evaluated...) to see
whether they achieve what they set out to achieve. Whether a model is good
or not is independent of its implementation language.
2. Building a good decision support system involves:
o working with clients (decision makers) to determine their needs;
o identifying decisions that clients need help with;
o determining what information, analyses, and knowledge are needed to make
those decisions well;
o providing either the supporting information or potential decisions
or both to the decision maker.
Models often play important roles in decision support systems because they
provide analysis and information. In DSS, the developer and the clients
need to make choices about what kinds of models are appropriate to the
problem.
Often there are several ways to get some piece of the puzzle: one or more
models at different levels of resolution, precision, and complexity,
heuristics, published defaults, etc.
3. Even a good DSS with appropriate and accurate models can fail to be used
because of its design: poor interface, poor flow, failure to take into
account how/when/where the user works, demanding more time, training, or
equipment than the user reasonably expects.
Many features that make software more readily adopted by users (in general)
are built into spreadsheets, so they are one prominant option for delivery
of systems to users.
Ultimately, we hope to solve problems. All three of these activities (with
a few more like good advertising, promotion, support, etc.) are important
toward that end.
--Nick
-----------------------------------------------------------------
Nicholas D. Stone ISIS Lab
Department of Entomology Phone: (703) 231-6885
202 Price Hall FAX: (703) 231-9131
Virginia Tech Internet: nick@vt.edu
Blacksburg, VA 24060 BITNET: nstone@vtvm1
On the WWW:
http://www.ento.vt.edu/Personnel/Professors/NDSInfo.html
-----------------------------------------------------------------
_ __
\ / __/
o |_||__//_
\_/ |\ / \ "Time flies like an arrow;
(_| | >--< \ ) fruit flies like a banana."
/ \_|/____\_/ --Groucho Marx
o | || \\_
_/ \__ \
Hi -
This is Rick Olson, a Biologist at the Crop Simulation Research Unit,
USDA-ARS, Mississippi State. My research involves simulation modeling and
decision-support systems. Enough intro.
I've also been amused by the spreadsheet controversy. It reminds me of the
old (present?) days and the language controversies in bio modeling. You
know, whether C, then C++, then etc. was any good versus good-old-FORTRAN.
When I moved to OOP languages like C++ or Smalltalk, I listened to all kinds
of folk who told me "Well, I can do that in FORTRAN," to which I would reply
"Yes, but why would you WANT to?"
The point is, choose an implementation medium that is right for the job.
Doing OOP in FORTRAN is silly. Likewise, you can, indeed, do anything you
want in Excel, but why would you want to? I use Excel for testing of
equations in model development, etc., due to its "instant gratification"
aspect of changing equation parameters and seeing them change instantly on
the screen. For building complex, graphical interfaces to equally complex,
large simulation models, it just doesn't cut it (IMHO, of course!). There
are plenty of new GUI-building tools, such as Visual Basic, that allow you
to easily construct an interface and then mate with complex simulation
models in any language (as DLLs or DDEs or OLE objects), along with graphing
DLLs, spreadsheet DLLs, etc., etc.
Rick Olson
First, I regret my rather cynical response to Roger Smith's comments
about the spreadsheet - we should be better natured than that -
but I certainly agree with Eric Fraint's comments about the power of
the spreadsheet in simulation work. Bob Peart, U. Florida.
--=====================_794164773==_
Content-Type: text/plain; charset="us-ascii"
TO: Attendees and potential attendees of the the 25th Crop Simulation Workshop
Please find attached a copy of the final program of the Crop Simulation
WOrkshop to be held at the Engineering Research Center, Mississippi Research
and Technology Park, Starkville, MS (Mississippi State) during March
13,14,and 15, 1995. Transportation will be provided from local motels to
and from the ERC. If you have preregistered and have not informed us which
motel/hotel you are staying at and need transportation please advise by
e-mail, phone or fax. If you are going to register on-site, ditto the above.
JAMES
--=====================_794164773==_
Content-Type: text/plain; charset="us-ascii"
Content-Disposition: attachment; filename="BSSG.PRO"
Monday, March 13th
12:00-1:00 p.m. Registration
1:00 p.m. Welcome
J. M. McKinion
Session Moderator - James M. McKinion
1:10 p.m. R. Bruce Curry
25 Years - A Look Back
1:30 p.m. Jack Carlson
Migrating the USDA Field Service Center to a Client-Server
Architecture: Implication for Use of Crop Simulation Models
1:50 p.m. B. M. Jacobson
Combining Crop, Weed, Herbicide, Water Quality, and Economic
Models to Assess Overall Impacts of Management Practices
2:10 p.m. J. R. Barrett
Humanization of Decision Support for Managing U. S. Grain (Corn
and Soybean) Production"
2:30 - 3:00 p.m. Coffee Break
Session Moderator - R. Bruce Curry
3:00 p.m. G. F. Sassenrath-Cole
The Role of Leaf and Canopy Architecture in Determining Canopy
Photon Flux Distribution
3:20 p.m. R. Braddock
Towards a Simple Framework for Modelling Root Growth
Dynamics Under Variable Environmental Conditions
3:40 p.m. Avishay Ben-Porath
Graphical Tracking of Cotton Growth
3:40 p.m. E. Jallas
Graphical Input, Representation, and Biometric Analysis of the
Topology of the Cotton Plant
4:00 p.m. D. C. Akins
C++ Classes for a Generic Population Model
4:20 p.m. G. E. Meyer
Virtual Reality Analysis of Plant Canopies for Evaluating Sensor
Designs for Variable-Rate Application Technology
5:00 p.m. ADJOURN
Transportation to Hotels Available
6:30-7:30 p.m. Reception
(Bar-B-Que Dinner)
Session Moderator - K. J. Boote
7:30-10:00 p.m. Roundtable Discussion
Note: Transportation to Hotels Provided
Tuesday, March 14th
Session Moderator - Harry F. Hodges
8:30 a.m. K. J. Boote
Sensitivity Analyses of CROPGRO Model Response to Climatic
Variables
8:50 a.m. Basil Acock
Testing Crop Simulators as Software: GLYCIM Case Study
9:10 a.m. V. R. Reddy
Estimating Soybean Cultivar-Dependent Parameters for
GLYCIM from Field Data on Crop Growth and Development
9:30 a.m. J. Muhidong
Modelling Cotton Fiber Quality: Importance, History, and
Methods
10:00 a.m. Coffee Break
Session Moderator - K. Raja Reddy
10:30 a.m. R. Gordon
A Cultivar Specific Soil Water Sensitive Model of Potato Leaf
Area Index
10:50 a.m. D. Timlin
2DSPUD, A Modular Two-Dimensional Potato Simulation
Model
11:10 a.m. Hsin-I (Wally) Wu
Use of Fractals in Plant Functional Studies
11:30 a.m. A. Marani
Simulating Nitrogen in the CALGOS Cotton Model
12:00 p.m. Lunch
1:00 - 2:30 p.m. Poster Presentations
& Software Demos
2:30-3:00 p.m. Coffee Break
3:00-4:00 p.m. Poster Presentations
& Software Demos
4:00 p.m. ADJOURN
(Transportation Back to Hotel)
5:30 p.m.-11:30 p.m.
Dinner at Silver Star Casino
(Transportation Provided)
Wednesday, March 15, 1995
Moderator - Jeffrey L. Willers
8:30 a. m. D. S. Fisher
Estimating Variable Rates of Comminution and Passage of Forage
Particles in Models of Ruminant Digestion
8:50 a.m. C. S. Liu
The Effect of Fluometuron Concentration and Temperature on
Transpiration Rates and Some Morphological Characteristics of
Four Weed Species
9:10 a.m. J. A. Landivar
Crop Monitoring System Using a Weather Station Network, Plant
Mapping and Crop Simulation
9:30 a.m. X. N. Wang
The Characteristics of Weather Patterns and Their Regional
Distribution
10:00-10:30 a.m. Coffee Break
Moderator - Michael R. Williams
10:30 a.m. K. R. Reddy
10:50 a. m. S. M. Bridges
Problem-Solving Strategies in a Simulation-Based Expert System
11:10 a. m. R. L. Olson
An Emergent Computational Approach to Ecosystem Modelling
11:30 a.m. Hsin-I (Wally ) Wu
An Integrated Resource Simulation Model for a Semi-Arid
Agroecosystem in the Loess Plateau of Northwestern China
11:50 a.m. ADJOURN
WORKSHOP
ON CROP S IMULATION
Organized by:
Biological Systems Simulation Group
Co-Sponsor ed by:
USDA Agric ulture Research Service
Mississipp i State University
American S ociety of Agricultural Engineers
March 13-1 5, 1995
Mississipp i State, Mississippi
--=====================_794164773==_
Content-Type: text/plain; charset="us-ascii"
Dr. James M. McKinion Ph. 601-324-4375
USDA-ARS FAX 601-324-4371
Crop Simulation Research Unit Email mckinion@csrumsu.ars.ag.gov
P. O. BOX 5367
Mississippi State, MS 39762
USA
--=====================_794164773==_--
>Someone posted their opinion (as opposed to fact) that putting out
>software with a decent User Interface was easy these days because of
>the programming tools available.
Well, Stephen, as the "someone" who posted the opinion, I guess I'll respond
[at least I clearly labeled my post as opinion, while yours was stated as
fact ...]
I feel that software tools used should be appropriate for the problem at hand.
While the learning curve on tools like V. Basic may be steep, they are NOT steep
relative to that for Visual C++, Microsoft's current abortion (even with the
foundation classes). And yes, if you need the complete flexibility offered by
Visual C++, you should use it and bite the bullet as to learning curve. (while
this is still opinion, it is informed opinion, because I've used both
Visuals, and others).
Again IMHO, if you expect to produce state-of-the-art software, you use
state-of-the-art tools. If one can't be bothered to learn them, then fork
over the bucks to hire a programmer who can.
As always, in my humble opinion.
Rick
Steve Modena Says:
>Someone posted their opinion (as opposed to fact) that putting out
>software with a decent User Interface was easy these days because of
>the programming tools available.
>
>Really? ;^)
>
>Is this on Mars or Earth? :^)
>
>If that person's opinion were indeed fact, MicroSoft would ship
>bug free products on-time! After all, Bill has about $400 Million
>after-taxes cash flow, the top 0.1% of the programming talent in
>the World and advanced access to its own programming APIs.
< :^)'s deleted >
I did not mean to imply that programming is easy by my prior post. The point
that I was trying to make was that interface development - either in a
spreadsheet or in a
program - takes some time. Obviously, the nicer the interface, the longer
the time.
Yes, the learning curve is steep for some of the new visual programming
environments (I've been working with Borland's Visual C++), but the user
interface development is greatly simplified. Less than three years ago, if
you wanted to write a 'graphic' user interface (and you didn't work for a
big company that had access to expensive compilers/in-house code) You had to
draw out each individual line, polygon, color, attribute etc etc etc, and
after about a month, you could start making the buttons do something besides
display other buttons. With the package I'm using now, you can construct the
entire basic interface framework -windows application with text editor,
support for multiple documents, file save and retreive, search and replace,
menu system, speed bar, help, fly-by hints, scrollbars, scalable windows -
(admittedly not much application functionality) with two mouse clicks, and
about two minutes of compile time.
Comparing the development of a 'management tool' or model with the
development of operating systems, programming environments, databases, and
desktop publishers by a monolithic software corporation I think is just a
wee bit out of line on the complexity scale. It's kind of like comparing a
fungal spore with an elephant - any likenesses to the actual parties is
purely coincidental -.
/**************************************************************************/
Jeff Woodard Glades Crop Care voice: 407-746-3740
949 Turner Quay fax:
407-746-3775
Jupiter, FL 33458
e-mail: woodard@igc.apc.org
-My opinions are my own-
/**************************************************************************/
Are there references to work demonstrating that a model has worked
shockingly well? Most field studies in my past history were field
dissipation and leaching for registration and were disconnected from
model validation. Is this still the case?
Don Fontaine
1803 Bldg Health and Environmental Sciences Dept
Dow Chemical USA
Midland, MI 48674
(517)636-2179
dfontaine@dow.com
I agree with the comments regarding the use of spreadsheets and
simplified physics for making models useful. Detailed mechanistic models
certainly have their place and can be made user-friendly for day to day
ag production decisions (e.g. Gosym/COMAX). However, these models tend to
scare off the average person.
I believe that many of the complex models could be used to develop
simpler models for use under specific conditions and circumstances. I
also believe that the simpler models can be more practical for decision
making purposes. sometimes I think we researchers get overzealous (self
included) in our use of insignificant digits (e.g. predicting water use
to the 0.01 in when we can apply water with an accuracy of 0.1 or 0.2 in)
I guess I'm saying let the model fit its intended audience.
************************************************************************
* Chris Butts * cbutts@asrr.arsusda.gov
*
* USDA, ARS *****************************
* National Peanut Research Laboratory * Phone: 912-995-7431 *
* 1011 Forrester Dr., SE *****************************
* Dawson, Georgia 31742 * FAX: 912-995-7416 *
************************************************************************
On Fri, 3 Mar 1995 06:37:13 -0600 DONALD D FONTAINE said:
>Are there references to work demonstrating that a model has worked
>shockingly well? ...(snip)
I can tell you I was shocked at how well GLEAMS predicted erosion and
pesticide runoff concentrations and loads from my microplot simulated
runoff experiments (J. Env. Qual. 19:119-125 (1989). It is true the
curve numbers were adjusted to give runoff volumes and the erosivity was
adjusted to fit bare-soil data. What I thought was shocking was that the
resulting predictions of erosion and chemical runoff were generally within
a factor of two for cyanazine and sulfometuron-methyl applied to both bare
AND grassy plots...
Don Wauchope
Hi!
I am just asking, on the behalf of my colleaques, if
someone knows any special mailing lists, dealing with
a) horticulture or
b) nitrogen-15 technique
With regards:
Tapio Salo
Institute of Crop and Soil Science, ARCF
31600 Jokioinen
Finland
tapio.salo@mtt.fi
We have a discussion group for vegetable production. Send a note to
VANVRANKEN@AESOP.RUTGERS.EDU. He can tell you how to subscribe.
>Hi!
>
>I am just asking, on the behalf of my colleaques, if
>someone knows any special mailing lists, dealing with
>
>a) horticulture or
>b) nitrogen-15 technique
>
>
>With regards:
>
>Tapio Salo
>Institute of Crop and Soil Science, ARCF
>31600 Jokioinen
>Finland
>tapio.salo@mtt.fi
>
*************************************
*Milt McGiffen *
*Dept. Botany and Plant Sciences *
*University of California *
*Riverside, CA 92521-0124 *
*Phone:909-787-2430 *
*FAX: 909-787-4437 *
*Email: MILT@UCRAC1.UCR.EDU *
*************************************
In message <21310095344@oeko-projekt.pz-oekosys.uni-kiel.d400.de> writes:
>
> ------- Forwarded Message Follows -------
>
> From: "Arkadiusz Salski" <OEKO-PROJEKT/AREK>
> Subject: fuzzy-ecology - mailing list
>
> CALL FOR PARTICIPATION
> in the fuzzy-ecology mailing list
>
> A mailing list for everybody interested in "fuzzy" topics
> in ecology has been established. It is important for our
> "fuzzy community" in ecology to keep in touch and have an
> opportunity for information exchange and for discussion of
> issues regarding fuzzy sets and fuzzy logic applications in
> the field of ecology. I would like to invite you to participate
> in this new mailing list created for this reason.
>
> If you would like to subscribe to fuzzy-ecology, please send
> a message to
> maiser@pz-oekosys.uni-kiel.d400.de
> with
> SUBSCRIBE fuzzy-ecology
> in the body of an otherwise empty mail.
> I look forward to hearing from you. With kind regards
> Arkadiusz Salski
> --------------------------
> Dr. Arkadiusz Salski
> Ecosystem Research Center, University of Kiel
> Schauenburgerstr. 112, D-24118 Kiel, Germany
Hallo ,
I am not a crop modeller, so I hope the question is not entirely
meaningless.
I am trying to look at resource availability for birds living
in agricultural areas. A critical window in a given year seems
to be the period between sowing date and emergence of cereals.
Would it be possible to predict, at least at a large scale
the sowing date for, say spring and winter
wheat and barley in a given year?
(or I suppose I should say the date before which it is unlikely
there is any sowing)
Which data would be needed?
Any feed-back would be greately appreciated.
Alessandro Gimona
geo293@abdn.ac.uk
Allessandro Gimona wrote asking for information about
sowing dates of cereals. Much of the data she seeks is
available at least for Europe. For any any region there is
generally an earliest and latest normal date of sowing
which will vary with the crop and with the cropping
system. Within that period it is possible to develop rules
for predicting the date of sowing in a particular field.
The rules given below should be applicable to north-west
Europe. Note that a crop cannot be sown until after the
previous crop is harvested and the ground prepared! The
first autumn sowing on a farm will take place as soon after
the earliest possible date as is possible, provided that
the soil is dry enough (2 consecutive days without rain?).
If it is still early in the period then a farmer may be
prepared to wait for conditions to improve but as time goes
on it is more important to sow in time than to get a good
seedbed. Really early sowings of winter wheat (early
September) seem to have gone out of fashion in the UK
because of the increased costs of crop protection. Note
that sowing takes place over a number of days on a farm
because it takes time to sow a field (about 2 ha/hr).
Farmers will have established a priority for sowing fields
based on crop and possibly soil type. In spring time,
before April the farmer can usually afford to wait for
better weather. As a rule of thumb, sowing will take place
after two days without rain provided the mean soil
temperature exceeds 5 degrees C. Crop emergence generally
occurs about 120 day degrees (base temperature =0C) after
sowing.
I have more data that might be helpful.
Graham Russell
Dr G. Russell
University of Edinburgh
Institute of Ecology & Resource Management (Agriculture Building)
West Mains Road
Edinburgh EH9 3JG
SCOTLAND
International phone +44 31 535 4063 Fax +44 31 667 2601
UK phone 031 535 4063 Fax 031 667 2601
Dr G. Russell
University of Edinburgh
Institute of Ecology & Resource Management (Agriculture Building)
West Mains Road
Edinburgh EH9 3JG
SCOTLAND
International phone +44 31 535 4063 Fax +44 31 667 2601
UK phone 031 535 4063 Fax 031 667 2601
I agree with Graham Russell's post. Those temperature guidelines
also apply in North America. Some years spring sowing may be delayed
locally by extremely dry soil in non-irrigated areas.
Most US states publish yearly ag statistics booklets with planting
and harvesting info for major crops. The National Ag Statistics
Service (NASS) and state Ag statistics services co-publish them.
I don't have an address for NASS, but it could probably be found by
telnetting to fedworld.gov or possibly someone has an email address
for Alessandro?
Tom
Tom Hodges, Cropping Systems Modeler
USDA-ARS email: thodges@beta.tricity.wsu.edu
Rt. 2, Box 2953-A voice: 509-786-9207
Prosser, WA 99350 USA Fax: 509-786-9370
== ## Rent this space ## ==
If this represents anything, it is only my opinion.
On Tue, 7 Mar 1995 geo293@aberdeen.ac.uk wrote:
> Would it be possible to predict, at least at a large scale
> the sowing date for, say spring and winter
> wheat and barley in a given year?
>
> Alessandro Gimona
> geo293@abdn.ac.uk
>
NASS crop statistics files, and some state files, can be
found on WorldWideWeb I think at Cornell University, but if
you have access via MOSAIC, Netscape, or some other browser,
you could simply use Web Crawler with the search terms
"crop yield" and poke around through the "http" addresses
that are returned. I know that I've accessed this; there's
tons of this stuff out there on the Web, but it does take
patience and persistence to find it. If we weren't in the
midst of a building move at the moment, I'd have the exact
Http-addresses. Sorry.
Ron Kickert
Battelle PNL rn_kickert@pnl.gov
Richland, WA
I'm a subscriber and a marketing communications exec. who's trying to figure
out why people use electronic media. If you don't mind, can you tell me why
you're here, and whether or not the medium has met your expectations.
Thanks
Chris Meyer
Rumrill-Hoyt
60 Corporate Woods
Rochester, NY 14623
(716) 272-6220
Fax (716) 272-6300
1. I can use it anytime, anywhere - today I'm at a conference where
computers with connection to e-mail is available.
2. I can send more detailed and accurate messsages than by phone.
3. Much less formal than having a letter sent, letterhead, envelope,
etc. Much quicker. speelling & format not as improt
important. For cexxx example, I probably would not answer your question
if you sent me a nornoal form letter.
Bob Peart, Agr. Eng. Dept., U. Florida.
The following reply merits comment. I really feel that messages sent to
a mailing list should be neatly formatted, properly spelled, and easy to
read. This kind of sloppiness really inhibits the use of maiing lists.
I know that whenever I get some letters like this on lists that I run
there is an increase in the number of people who unsubscribe.
Bill Silvert
>3. Much less formal than having a letter sent, letterhead, envelope,
>etc. Much quicker. speelling & format not as improt
>important. For cexxx example, I probably would not answer your question
>if you sent me a nornoal form letter.
--
Bill Silvert, Habitat Ecology Div., Bedford Inst. Oceanography
P. O. Box 1006, Dartmouth, Nova Scotia, CANADA B2Y 4A2
HED runs a WWW server at URL=http://biome.bio.dfo.ca
>We have a discussion group for vegetable production. Send a note
>to
>VANVRANKEN@AESOP.RUTGERS.EDU. He can tell you how to subscribe.
>*************************************
>*Milt McGiffen *
>*Dept. Botany and Plant Sciences *
>*University of California *
>*Riverside, CA 92521-0124 *
>*Phone:909-787-2430 *
>*FAX: 909-787-4437 *
>*Email: MILT@UCRAC1.UCR.EDU *
>*************************************
I am student from Slovakia studding Horticulture on SUNY
Cobleskill. Perhaps you could suggest my Internet resources
concerning FRUIT PRODUCTION.
Thanks for a help,
Martin
Hi
I'm the manager of our Agriculture unit at The Open Polytechnic of New
Zealand Private Bag 31914 Lower Hutt, New Zealand. Internet address:
leedav@mhs.topnz.ac.nz
I've only just joined and I'm new to internet so can't quantify any
benefits yet but I'm sure there is a place for the new technology.
Agriculture is part of our School of Natural Resources that includes
Horticulture & Enviroment and we are interested in Sustainable
Agriculture as a new area for course development.
Cheers
Can someone help me with the commands to unsubscribe?
vpierce@umce.umext.maine.edu
In message Wed, 8 Mar 1995 14:25:47 -0600,
bill@biome.bio.dfo.ca (Bill Silvert) wrote:
...
> read. This kind of sloppiness really inhibits the use of maiing lists.
^^^^^^
Yeah, and maybe letter writers should stay out of the Screech too, eh?!
David in Ithaca
*****@*****@*****@*****@*****@*****@*****@*****@*****@*****@*****@****
>From the desk of: David C. Hall, Ithaca, NY
dch5@cornell.edu
David C. Hall, DVM, MSc
Dept. Agricultural Economics
411 Warren Hall, Cornell University
Ithaca, NY, U.S.A. 14853 Home phone/fax: (607)256-3248
*****#*****#*****#*****#*****#*****#*****#*****#*****#*****#*****#****
In response to Bill Silvert, I'm in general agreement, but please explain
what a "maiing list" is.
Pete Jamieson
Crop & Food Research
Lincoln
New Zealand
jamiesonp@crop.cri.nz
thank you
thank you
thank you
thank you
Attention Mechanistic Modelers.
A group academic purchse of Matlab is brewing at Clemson.
Please contact David Moline ASAP if interested.
>> dmoline@eng.clemson.edu
The MATLAB group purchase has done very well, but I still need some
information from several of you. It is important that you get a response to
me ASAP so we can get this stuff in the mail. For those of you who are new
to purchase group there is some info at the end of this e-mail. Those of you
who decide you're not interested should let me know ASAP as well, please.
==>All buyers must give me info about (1) what you want, (2) what
type of machine (PC, Mac, PowerMac), (3) a mailing or house address (so the
Mathworks knows you are a real person), and (4) how you want to pay (check,
credit card, or purchase order).<==
If you have sent the information, thanks. Purchase orders will be
arranged shortly for signatures, etc. If you can't remember what you sent,
check the summary below. Be sure to verify that I have you down for the
right software. I will be calling some of you, but please make an effort to
get the information to me. If you wish, you may give the information to a
secretary (especially if purchase orders are involved), and send me his/her
name--I'll take it from there.
Those of you who want to drop off payments or credit card numbers may
also leave it with a secretary or make arrangements with me. Checks can be made
out to the Mathworks or to me, David Moline (in which cases I'll give you a
receipt, for records sake, and write one check to the Mathworks).
Here's where prices and numbers of copies stand (the second number
on some of them includes unconfirmed buyers). Use these in figuring your
costs. Include shipping and handling is about $1 per buyer.
copies price
Matlab 22-28 $187.50
Simulink 3 $495
Symbolic Math Toolbox 11-13 $100
Control Toolbox 3/5 $195-$150
Signal Processing Toolbox 3 $195
any other toolbox 1 $195
==>You'll notice that Simulink, Signal Processing Toolbox and
Control Toolbox are close to 5-buyer prices ($150 for the toolboxes, $365
for Simulink). Anybody who wants to jump in on one of those at the last
minute here if free to do so--maybe we can get them to 5 buyers.<==
Anyway, here is a summary of what I have from everybody; it is
alphabetical by e-mail address. I've left out/shortened some of the
personal details of course, but check to be sure I have things correct;
correct me if I'm wrong on anything. A question mark means either I didn't
get the info (account number in the case of some purchase orders), I didn't
get a confirmation, or I lost track of it (as I know I did in one case).
The general format is as follows:
All right guys, I wrote "maiing" when I meant "mailing". Enough already.
Shall I send a couple of dozen "thank you" messages to everyone who
pointed out my spelling error?
Bill
>In response to Bill Silvert, I'm in general agreement, but please explain
>what a "maiing list" is.
And just how do we spell the word "mailing" today?
On Wed, 8 Mar 1995, Bill Silvert wrote:
> The following reply merits comment. I really feel that messages sent to
> a mailing list should be neatly formatted, properly spelled, and easy to
> read. This kind of sloppiness really inhibits the use of maiing lists.
> I know that whenever I get some letters like this on lists that I run
> there is an increase in the number of people who unsubscribe.
>
> Bill Silvert
>
> >3. Much less formal than having a letter sent, letterhead, envelope,
> >etc. Much quicker. speelling & format not as improt
> >important. For cexxx example, I probably would not answer your question
> >if you sent me a nornoal form letter.
>
> --
> Bill Silvert, Habitat Ecology Div., Bedford Inst. Oceanography
> P. O. Box 1006, Dartmouth, Nova Scotia, CANADA B2Y 4A2
> HED runs a WWW server at URL=http://biome.bio.dfo.ca
>
Jim - Farmer - Iowa City, IA,
jnmeade@blue.weeg.uiowa.edu
I agree with Bill on two counts. First, users should
take the time to learn how to use mail properly. This
includes not replying to everyone on the list when the
message is clearly meant as a response to one individual.
Second, many of us get far too much mail to have to filter
out sarcastic responses that serve no purpose beyond venting
frustration. List servers such as agmodels are designed for
discussion and dissemination of pertinent information. Let's
make an effort to cut down on the irrelevant mailings.
Was there a Biometeorology listserve group mentioned earlier in the group?
I recall one listed either here on in sci.geo.meteo or sci.geo.hydro...
============================================================================
Bill Capehart (wjc@essc.psu.edu) "For a list of all the ways technology
Penn State Meteorology has failed to improve the quality of
University Park, PA 16802 life, please press three" Alice Kahn
===================== /* John Galt, Call Your Office */ ====================
Could anyone tell me where to ftp or extract the weekly historical
Palmer Drought Index or Crop Moisure Index, say in 1988?
Thanks.
Zaitao Pan
Meteorology
Iowa State University
I am a new subscriber to the system and wish information on agricultural
models pertaining to research and development decisions. Can anyone out
there indicate possible sources. I have reviewed the literature and found a
few empirical studies that have estimated rates of return to research
investment for a few commodities. The Australians and the U.S. appeared
to have done most of this work. I am interested in models that can be
adapated quite readily to a PC computer .
Regards Richard Carew (Rcarew@EM.AGR.CA)
The following is information of USDA - NASS Statistics in Cornell U.,
but the one I know is a gopher site.
gopher://usda.mannlib.cornell.edu
Tung
*************************************************************************
Ching-pin Tung
Graduate Student Phone: +1-607-257-1241
Civil & Environmental Engineering Fax : +1-607-257-1241
Cornell University Email: ct16@cornell.edu
Ithaca, NY 14850
USA
*************************************************************************
Has anyone calibrated EPIC (Erosion-Productivity Impact Calculator) for potato cropping?
We want to use EPIC to evaluate nitrate leaching under intensive potato cropping in Quebec. Not much litterature was found on potato growth parameters used in EPIC. It seems that EPIC was mostly validated for grain crops.
Is this model suitable for simulating potato crops? Has there been any effort to integrate other growth models in EPIC in the same way that SIMPOTATO was integrated in CERES for potato simulations?
Also, EPIC seems to overestimate organic matter mineralization during winter. Is this possible? While the soil is frozen (Soil temperature is below 0), nitrogen is constantly released from the organic matter pool. Maybe the soil temperature factor set at
EPIC is certainly interesting for all the components it integrates although model ajustments for Northern conditions may be necessary. I'll be glad to hear from any potato or soil modeler out there!
Marc-Olivier Gasser
Departement des sols
Pavillon Comtois
Universite Laval, Quibec.
G1K 7P4, Canada
mogasser@grr.ulaval.ca
Tel.: 418 656 3627
Try SIMPOTATO at ftp.tricity.wsu.edu
On Thu, 9 Mar 1995, Richard Carew wrote:
> I am a new subscriber to the system and wish information on agricultural
> models pertaining to research and development decisions. Can anyone out
> there indicate possible sources. I have reviewed the literature and found a
> few empirical studies that have estimated rates of return to research
> investment for a few commodities. The Australians and the U.S. appeared
> to have done most of this work. I am interested in models that can be
> adapated quite readily to a PC computer .
> Regards Richard Carew (Rcarew@EM.AGR.CA)
>
>
Bill did you realize you had a misspelling on your last letter?
Jim.
>The following reply merits comment. I really feel that messages sent to
>a mailing list should be neatly formatted, properly spelled, and easy to
>read. This kind of sloppiness really inhibits the use of maiing lists.
>I know that whenever I get some letters like this on lists that I run
>there is an increase in the number of people who unsubscribe.
>
>Bill Silvert
>
>>3. Much less formal than having a letter sent, letterhead, envelope,
>>etc. Much quicker. speelling & format not as improt=7F=7F=7F=7F=7F=7F
>>important. For cexxx example, I probably would not answer your question
>>if you sent me a nornoal form letter.
>
>--
>Bill Silvert, Habitat Ecology Div., Bedford Inst. Oceanography
>P. O. Box 1006, Dartmouth, Nova Scotia, CANADA B2Y 4A2
>HED runs a WWW server at URL=3Dhttp://biome.bio.dfo.ca
You can try the following two numbers (voice). For the National Climate
Data Center (704) 259-0682 and for the Climate Analysis Center (301) 763-8071.
Albert Weiss
ALBERT WEISS, DEPT. OF AGRICULTURAL METEOROLOGY
UNIVERSITY OF NEBRASKA-LINCOLN, LINCOLN,NE 68583-0728
VOICE (402)472-6761, FAX (402)472-6614
E-MAIL AGME011@UNLVM.UNL.EDU
Robert,
Thank you for your answer, just back from holidays (carnaval - mardi gras,
quite important in this latitude).
Best wishes
Bernardo
I recently received an announcement over the AGMODELS-L mailing list
for the upcoming Crop Simulation Workshop (held this week at
Mississippi State). Although I am unable to attend this meeting,
one of the talks interests me greatly. It is:
C++ Classes for a Generic Population Model, by D. C. Akins
Can anyone on this mailing list give me coordinates for this researcher?
I'd also be interested in references to publications or people doing
similar work.
Scott Ray
DowElanco
sray@dowelanco.com
I am also unable to attend the Crop Simulation Workshop (held this week at
>Mississippi State). Can anyone offer the group a list of speakers, their
talks, and where they are from?
*************************************
*Milt McGiffen *
*Dept. Botany and Plant Sciences *
*University of California *
*Riverside, CA 92521-0124 *
*Phone:909-787-2430 *
*FAX: 909-787-4437 *
*Email: MILT@UCRAC1.UCR.EDU *
*************************************
Scott Ray writes:
>I recently received an announcement over the AGMODELS-L mailing list
>for the upcoming Crop Simulation Workshop (held this week at
>Mississippi State). Although I am unable to attend this meeting,
>one of the talks interests me greatly. It is:
>
> C++ Classes for a Generic Population Model, by D. C. Akins
>
>Can anyone on this mailing list give me coordinates for this researcher?
>I'd also be interested in references to publications or people doing
>similar work.
I might mention a paper I wrote a couple of years ago which address this
issue, although the examples in the paper use the object-oriented
extensions of Turbo Pascal instead of C++
The paper describes how to develop a population model in TP and how to
use inheritance in defining population classes.
Silvert, W. 1993.
Object-oriented ecosystem modelling.
Ecological Modelling 68:91-118.
--
Bill Silvert, Habitat Ecology Div., Bedford Inst. Oceanography
P. O. Box 1006, Dartmouth, Nova Scotia, CANADA B2Y 4A2
Personal InterNet Address: silvert@biome.bio.ns.ca
HED runs a WWW server at URL=http://biome.bio.dfo.ca
>
>I recently received an announcement over the AGMODELS-L mailing list
>for the upcoming Crop Simulation Workshop (held this week at
>Mississippi State). Although I am unable to attend this meeting,
>one of the talks interests me greatly. It is:
>
> C++ Classes for a Generic Population Model, by D. C. Akins
>
>Can anyone on this mailing list give me coordinates for this researcher?
>I'd also be interested in references to publications or people doing
>similar work.
>
>Scott Ray
>DowElanco
>sray@dowelanco.com
>
>
>
Scott.
Dennis Akins is an Agricultural Engineer working in the Insect Modeling
Project of the Crop Simulation Research Unit, USDA-ARS at Mississippi State,
MS. Dr. Rick Olson of the CSRU is also working in this area. Their
telephone numbers are 601-324-4365(Dennis) and 601-324-4367(Rick).
James
Dr. James M. McKinion Ph. 601-324-4375
USDA-ARS FAX 601-324-4371
Crop Simulation Research Unit Email mckinion@csrumsu.ars.ag.gov
P. O. BOX 5367
Mississippi State, MS 39762
USA
help
I don't know the E-mail address but you can call D. C. Akins at
Mississippi State: 601-324-4365.
>I agree compeltely. Of course it is important to have the science right,
>but I have seen to many models in which the user interface was positively
>archaic. This has a direct impact on how much *science* can be done with
>the model since wrenching a crummy interface takes *lots* of time
Introduction
I've been following the discussions about the need for friendly user
interfaces with some interest and decided this might be a good time to
introduce myself.
My name is Caroline Parker and I work for a 'department' within
Loughborough University (UK - Midlands) called the HUSAT Research
Institute. Our area of interest is the human factors of advanced
technology, in all domains. My background is psychology with some AI and
computing thrown in for good measure. Prior to joining HUSAT (some 6 years
ago) I worked as a 'knowledge engineer'(!) with what was then the UK's
National Vegetable Research Station.
OK, what has this to do with Agmodels you might ask. Well to cut a long
story short, one of the two agricultural projects I'm currently involved in
is looking at the reasons for the lack of uptake of decision support tools
(mostly model based) by the grower/farmer population. One of the biggies
identified so far is the impenetrability of some user interfaces. Another
is the failure to answer the questions the user really wanted answering.
Neither of these factors is related to the quality of the underlying
science - the user never gets that far. That doesn't mean the robustness
of the model in 'the field' is not important, continued use of the program
depends on it. Growers and farmers do however seem to be willing to use
tools which aren't perfect as long as they have some idea of the confidence
they can put in them. Of all the users I have met they are the most
accustomed to working with uncertainty (probably from long association with
the vagaries of the weather!).
I would be very grateful if anyone out there who has experience of
incorporating their models into programs for use by farmers would talk to
me and tell me all about it (the good and the bad). In return I'm happy to
share any insights I've gained or suggest methods that might prove useful.
Please email me directly on c.g.parker@lut.ac.uk & not reply through
agmodels as I'm sure that most people will quickly get bored with the
discussion!
Caroline
-------
Caroline Parker
HUSAT Research Institute
Loughborough University
Tel: +509 611088
Fax: +509 234651
Dear All:
I am trying to incorporate modules in an undergrad crop-environment
interaction course that show plant-environment linkages through physical
and
physiological processes. For example, can concepts involving air
temperature, canopy energy load, soil water availability and
evapotranspiration be shown better with a STELLA model than in a traditional
lecture setting with overheads?. I would like to hear from STELLA II users
about their applications in crop physiology/ecology and plant-environment
interactions.
Thanks in advance for your reply.
***********************************
dpk@psu.edu
Daniel P. Knievel
116 Agricultural Sciences & Industries
University Park, PA 16802
USA
Phone:
Office: (1-814) 865-1547
FAX: (1-814) 863-7043
***********************************
On Tue, 14 Mar 1995 04:33:55 -0600 Caroline Parker said:
...(SNIP)
Actually I suspect most of us would be interested in the discussion...
Don Wauchope
Well personally I'd very interested in the discussion as I'm involved
in the development of a decision support system for viticulture.
While this DSS relies essentially on downloaded weather data (either
remotely or manually) thereby introducing some calculated risk in the
prediction of pests and disease, the thing which concerns me most is that
almost all of the knowledge engineers, computer programmers etc.,
discount the possibility that growers will not get past the user
interface to test the knowledge contained within the system. My role
is to incorporate grower input into development, and, amongst other
things, ensure that the DSS meets the needs of the end perceived
beneficiaries. While my findings thus far suggest that, indeed in
Australia at least, and for the time being, adoption of a DSS in
viticulture will not be undertaken by the growers, rather by
consultants who will then pass on the information to the growers on
a fee for service basis.
I would be happy to share my findings with anyone who is conducting
similar research.
Scott
..
S.Glyde
Ron Potter Centre - Winery,
Charles Sturt University
WAGGA WAGGA 2650
NSW AUSTRALIA
Tel: 069 332 728
Fax: 069 332 107
We have had similar experience as previous posts with farmers,
consultants. We are in the processof developing several models, expert
systems for peanut production and harvest. The one closest to release,
EXNUT, is an expert system for irrigated peanut production decisions. A
few of the farmers who were in on the ground floor of the testing and
validation are using the system. A few consultants use the expert
system as well as some county extension agents. I think much of our
problem comes from many growers not having thier own computers or that
they don't use them in their daily, or even weekly, farm management
decisions and record keeping.
Most extension agents like the expert system and its user interface.
However, the agent does not have time to run the expert system for all
his farmers' fields.
We have put considerable effort in programming graphiical user
interfaces, with minimal typing required. These changes have been
implemented because of feedback from users, primarily growers.
My personal opinion is that farmers, just to maintain average yield and
quality, will have to start using computers in their bookkeeping,
pesticide records, and management decisions.
************************************************************************
* Chris Butts * cbutts@asrr.arsusda.gov
*
* USDA, ARS *****************************
* National Peanut Research Laboratory * Phone: 912-995-7431 *
* 1011 Forrester Dr., SE *****************************
* Dawson, Georgia 31742 * FAX: 912-995-7416 *
************************************************************************
The preceding statements are my opinion and do not necessarily reflect
those of my employer.
>system as well as some county extension agents. I think much of our
>problem comes from many growers not having thier own computers or that
>they don't use them in their daily, or even weekly, farm management
>decisions and record keeping.
I too, have developed a peanut model (for prediction of web blotch favorable
conditions BLOTCHCAST), and had similar experiences in grower acceptance.
Out of a group of about 50-75 growers, only 2 had computers, and both of
them used them primarily for CD-ROM encyclopedias for their children. As far
as being used for farm decisions, one of the growers used a spreadsheet to
calculate the cheapest products for fertilizer mixes, and the other had a
pretty slick connection with a Tandy (Yes, Tandy beleive it or not) weather
station. I wouldn't call either case a situation where the phrase 'using the
computer as a management tool' applied. Our original plan was to develop a
software package that individual growers could use on their computers, but
that quickly had to be scrapped so that we could place something that would
actually be used.
>Most extension agents like the expert system and its user interface.
>However, the agent does not have time to run the expert system for all
>his farmers' fields.
With BLOTCHCAST, we designed a software interface (sacrificing some site
specific model detail) for use by the extension agent, where the region was
broken into smaller areas. In the center of each of these smaller areas, we
placed a weather station, which was downloadable via radio-telemetry (and
phone modem) from the extension agent's computer. He would then run the
preditcions for each smaller region, and record this information on a
telephone answering machine that growers could call and listen for the
regional prediction closest to their fields. While not perfect, the system
seemed to work quite well. The peanut grower association actually funded the
research, purchase of weather stations (with help from the SCS), and a
computer for the extension agent. The agent was more than happy to upgrade
his 8088 to a 486 class machine.
>We have put considerable effort in programming graphiical user
>interfaces, with minimal typing required. These changes have been
>implemented because of feedback from users, primarily growers.
I would agree, GUI's seem to receive much more use, and are learned much
more easily than text based typing-required interfaces..
/**************************************************************************/
Jeff Woodard Glades Crop Care voice: 407-746-3740
949 Turner Quay fax:
407-746-3775
Jupiter, FL 33458
e-mail: woodard@igc.apc.org
-My opinions are my own-
/**************************************************************************/
Several points have been raised in response to my original email which I
would like to bring back to the group.
1. Use of models by consultants rather than growers
S. Glyde says
>my findings thus far suggest that, indeed in
>Australia at least, and for the time being, adoption of a DSS in
>viticulture will not be undertaken by the growers, rather by
>consultants who will then pass on the information to the growers on
>a fee for service basis.
I can confirm this finding to a certain extent. In both the vegetable and
arable domains I have found that the type of decision requiring the support
of biological/meteorological models rests most frequently in the hands of
the consultants. However this does not mean that some farmers, our fellow
agmodellers for example, are not interested in using them . Larger farms,
which I am constantly being told will inherit the earth, employ farm
managers, highly trained people who are definitely potential users.
The reasons for this lack of uptake by most small/medium sized farmers may
be as follows. Farming is time intensive and stressful. Unless the
software model is going to offer a substantial increase in profits/cost
savings etc. then the time required to learn how to use it, enter the data,
analyse the results will not be found. The type of decisions which are
supported by models - pest prediction, irrigation schedules, spray
management etc. etc. are in areas in which many farmers have limited
scientific knowledge. This coupled with the large financial implications of
success or failure lead many to seek outside support from specialists (ie
consultants). Most will not be willing to trade that accessible and
accountable (if costly) human being for the non-prescriptive support of a
program - however friendly. (I should say here that I have found no
support at all for prescriptive software)
2) Usability vs. usefulness
One of my colleagues has a wonderful overhead which shows a truly
beautifully designed interface - every button in the right place -
intuitive to use - light on colour etc. but utterly useless because no-one
needs the functions it so tastefully provides. How many 'technology
transfer' projects have identified the questions which the potential users
really need answering. Many, in my experience, have a smart idea which
they are trying to find a market for. This is often because of pressure
from funding bodies and brings me to my next point.......
3) Who's job is it?
It sounds as if all research institutes in this area are being squeezed
financially, and told to get something practical out to the industry to
justify their existence. I know this is true in the UK. None of the
funding bodies seem to be aware of the extra costs involved in doing this
properly and will all be very surprised and dismayed when the 'products'
fail. No commercial company would dream of developing a product without
adequate market research and pre-launch testing so why are you being asked
to do it? While I believe that an understanding of the needs of the
industry is essential for anyone writing software for it I also feel that
it is unreasonable to expect someone with specialised training in biology
to suddenly perform as computer programmer/user interface designer/human
factors specialist/market researcher and deliver a scientifically sound,
appropriate and usable package for the same price previously paid for
delivering learned papers. Ooops a bit of a rant there.
4) Solutions?
Apart from getting lots more funding?!!
Here are a few fairly low cost suggestions based on the user centred design
approach with apologies for those who may find them simplistic.
- In lieu of a full scale user requirements survey, ask a small number of
potential users to contribute to the development of the system.
Inducement, if required, to be a cheap/free version of the software. You
might want 2/3 for a start-up 'workshop' to define the questions they need
answering. A couple more to check the software at an early stage and
perhaps the first lot to use the 'pre-release' version over a longer period
of time. Even small numbers of users can help to identify major problems
which might prevent software from being used (1).
- Use early consultation to identify potential 'translation problems' with
the inputs and outputs to the model. Are you using the same language as
the user or focusing on the same parts of the problem? Are your units of
measurement the same as those used in the industry (e.g. dry weight is not
used by farmers); are the models recommendations based on non-standard
experimental equipment; what is the industry language for the yield of that
crop? etc. etc.
- Use rapid prototyping/slide shows to present your ideas to users before
investing too much in hard code.
- Write 'liaison with users' into future proposals!
Any more?
(1) Virzi, R. A. (1992). Refining the Test Phase of Usability Evaluation:
How Many Subjects Is Enough. Human Factors, 34(4), 457-468.
-------
Caroline Parker
HUSAT Research Institute
Loughborough University
Tel: +509 611088
Fax: +509 234651
Dennis,
I have NSERL Reports 2 and 9. Is this all there is to WEPP
documentation, or is there more?
Specifically, I am interested in the rill width calculations,
distribution of shear between soil and residue, and adjustments to
infiltration parameters based on tillage or residue cover.
NSERL Report 9 does not speak to these and so I am wondering if
NSERL Report 2 said is still a valid source of info on these matters
for WEPP.
Thanks,
Greg McIsaac
Senior Research Specialist
Agricultural Engineering
University of Illinois at Urbana-Champaign
On March 16, Greg McIsaac (GFM@age2.age.uiuc.edu) wrote:
> Dennis,
>
> I have NSERL Reports 2 and 9. Is this all there is to WEPP
> documentation, or is there more?
>
> Specifically, I am interested in the rill width calculations,
> distribution of shear between soil and residue, and adjustments to
> infiltration parameters based on tillage or residue cover.
>
> NSERL Report 9 does not speak to these and so I am wondering if
> NSERL Report 2 said is still a valid source of info on these matters
> for WEPP.
>
> Thanks,
>
> Greg McIsaac
> Senior Research Specialist
> Agricultural Engineering
> University of Illinois at Urbana-Champaign
>
Greg:
I am assuming that I was the "Dennis" intended by your original message.
I am not sure why you posted your original message to the agmodels list,
but since you did I will reply there, since others may be interested in
the answer.
There were a few addendums to NSERL Report Number 2 over the years 1990-91,
each of which brought the technical documentation up to the current status of
the WEPP model at the time. A revised draft of the technical documentation
was assembled in late 1992, but was not very widely distributed. This was
mainly a draft for chapter authors to work from in revising their chapters.
The final WEPP model documentation is still undergoing revision. If
you are able to browse on the World-Wide-Web, you can view/transfer
the most recent model technical documentation drafts which we have.
Enter the following URL:
http://pasture.ecn.purdue.edu/~snelson/wepp/wepptut/ahtml/doc.html
This will put you on our WEPP Documentation Page. I would not trust NSERL
Report 2 for most details. I don't believe that the distribution of shear
between soil and residue has changed, though some of the friction
factor estimates have changed. The rill width calculations are now
only a function of flow discharge (not also soil texture). All of
the infiltration parameter adjustment equations have changed.
Unfortunately, we have not yet received the Soil Chapter back which
will contain most of these revised equations.
I hope that this helps. I have had trouble getting final documentation
out of many of the chapter authors, but recently we have gotten most of the
chapters in. I expect that by early May we should have revised drafts
of most chapters available. Final documentation will be completed by
July of this year.
Some other URL's for your and the agmodel group's information:
NSERL Home Page (to obtain WEPP/CPIDS software) is -
http://purgatory.ecn.purdue.edu:20002/NSERL/nserl.html
WEPP CD-ROM Page (work in progress for August 1995 CD-ROM release) is -
http://pasture.ecn.purdue.edu/~snelson/wepp/wepptut/main.html
My personal Home Page is -
http://soils.ecn.purdue.edu/~flanagan
If any of the agmodels group does look over our WWW materials, we would
appreciate any constructive comments on how we could improve them. In
particular, comments to us now on how we might improve the CD-ROM can
be evaluated and incorporated into the HTML documents before we cut
the actual CD-ROM this summer.
Thanks.
Dennis Flanagan
-------------------------------------------------------------------------------
Dennis C. Flanagan
Agricultural Engineer Phone: (317) 494-8673
USDA-Agricultural Research Service FAX: (317) 494-5948
National Soil Erosion Research Laboratory email: flanagan@ecn.purdue.edu
1196 Building SOIL
West Lafayette, IN 47907-1196
-------------------------------------------------------------------------------
>3) Who's job is it?
>It sounds as if all research institutes in this area are being squeezed
>financially, and told to get something practical out to the industry to
>justify their existence. I know this is true in the UK. None of the
Well, I would agree (with my experiences in the US) that the research
institutes are being squeezed financially, especially in the area of general
laboratory equipment, which many funding sources are unwilling to provide
for through grants. Getting something practical out to industry is entirely
a different matter though. While practical products gone to industry and
community involvement are nice kudos to have, tenure and promotion boards
seem to focus mostly on publication lists.
Who's job is it is a really good question though. In the US, it would seem
that basic technology/knowledge should be created at the university, and the
concepts passed to extension where the final -hopefully practical- version
of that technology is transferred to the community at large. With Ag
Extension personell and budgets rapidly falling, time and resources become
pretty scarce. At the rate of technology development today, a basic set of
development tools/resources is an absolute necessity to construct a final
product. Major universities have many more resources (in terms of expertise
and equipment) than do extension offices i.e. Sun workstation vs PC-XT. It
seems to me that the most efficient means of development/technology transfer
would be to provide incentives for this to be done at the universities where
all of the resources are in place already. Otherwise, it's kind of like
Jeopardy, where where the university says: 'Here's the answer', and it's up
to extension to figure out what the question was.
/**************************************************************************/
Jeff Woodard Glades Crop Care voice: 407-746-3740
949 Turner Quay fax:
407-746-3775
Jupiter, FL 33458
e-mail: woodard@igc.apc.org
-My opinions are my own-
/**************************************************************************/
Dear List Reader :
A DSSAT list server has been established through the efforts of
Dr. Gerrit Hoogenboom, University of Georgia, Griffin to serve users and
those interested in DSSAT and the crop models contained therein. For
the record, the following crop models are currently accessible under the
DSSAT shell. They include the CERES family of models for maize, wheat,
rice, barley, sorghum, and millet; the CROPGRO series of models--SOYGRO
(soybeans), PNUTGRO (peanut/groundnut), and BEANGRO (dry beans); and the
root crop models, CROPSIM-cassava and SUBSTOR-potato.
We trust this mode of communication will allow each user to share
concerns, successes, messages, and thoughts related to DSSAT and its
application. Additionally, information on updates and modifications will
be announced here as well as through conventional means.
Instructions to subscribe follows:
To subscribe, send a message to: listserv@uga.cc.uga.edu
in the body of the message state : subscribe DSSAT
Firstname Lastname.
Dr. Hoogenboom will serve as "caretaker" of the listserver.
We look forward to meeting you on the "net".
Gordon Y. Tsuji
ibsnat@uhunix.uhcc.hawaii.edu
Unsubscribe me please
Hello, I'm Bernard Budde. I work on 'emission and water quality modeling'
at Delft University of Technology, down here in the Lowlands. I'm a
'reader' of this list.
Recently Dr. Dennis Flanagan provided this list with
several interesting WWW references, related to the National Soil
Erosion Research Laboratory. For convenience I have listed these references
in a WWW page, which you are invited to use. Its URL is:
http://dutcg16.tudelft.nl/~bernard/misc/agmodels.html
Regards, Bernard
---------------------------------------------------------------------
bernard.budde@ct.tudelft.nl phone: +31 15 782969
http://dutcg16.tudelft.nl/~bernard/iwm/ fax : +31 15 785559
Everyone is born with a bird in his heart (Chapman)
---------------------------------------------------------------------
are you working with Ceres Rice in Europe?
emailme!
Bye.
Position: Coord, Campus Computing Associate
Institution: U of Maryland, College Park
Location: Maryland
***************************************************************************
COORDINATOR
Campus Computing Associate
The University of Maryland at College Park is searching for a
discipline oriented computing support person for the College
of Agriculture. The incumbent will be responsible for
assisting and promoting the use of effective educational
computing technologies in the college, facilitating the use of
networking resources for teaching and research, and will be a
liaison between the college and the campus academic computing
support unit. The incumbent must hold a bachelor's degree;
Master's or Ph.D. degree is desirable. A minimum of 5 years'
experience in using information technology. Experience in a
discipline related to Agriculture or Biological Science is
desirable; should have experience in DOS and/or Windows
applications, preferably in a networked environment.
Experience with multiple platforms/operating systems is a
plus. Experience working in an academic environment is
strongly desirable. Must have excellent interpersonal and
communications skills, be able to work well in a collegial
relationship with faculty and researchers, and be flexible,
yet highly motivated. To be assured of maximum consideration,
applicant should apply by May 8, 1995. Please submit
application (consisting of a letter of application, resume,
and names of three professional references) to:
Angela Walker, AGRI Search
Computer Science Center
University of Maryland
College Park, Maryland 20742-2411
The University of Maryland at College Park actively subscribes
to a policy of equal employment opportunity, and will not
discriminate against any employee or applicant because of
race, age, sex, color, physical or mental handicap, religion,
national origin, or political affiliation. Minorities and
women are encouraged to apply.
Note: The following departments are in the College of Agriculture:
Agriculture and Resource Economics
Agricultural Engineering (incl Biological Engineering)
Agronomy (incl Soil Science)
Animal Sciences (incl Aquaculture and Biometrics)
Horticulture (incl Landscape Architecture)
Poultry Sciences
Veterinary Sciences
The program in Natural Resource Management is also in the college.
The departments of Botany and Entomology are outside the college
but have strong ties to the college.
Dear Ag. Modellers,
I need to estimate runoff from a small catchment (50 ha) for our
experimental farm. I wonder whether there is a computer model that will
estimate runoff during a year on rainfall event basis. I shall greatly
appreciate any help on this.
Thanks.
Basant Maheshwari
________________________________________________________________
Dr. B.L. Maheshwari Email
b.maheshwari@uws.edu.au
School of Agri. & Rural Development Tel. (61+45) 701 235
or 885 652
University of Western Sydney Fax (61+45) 885 538
Richmond, NSW 2753, AUSTRALIA
________________________________________________________________
Dear Members,
I have a question about MATLAB, but first I would like to
introduce myself.
My name is Hans Gijzen. I am working at the Department of
Horticulture of the Wageningen Agricultural University. Until
last year I was at the AB-DLO (formerly CABO-DLO). My field
is the development and validation of greenhouse crop models
(light transmission of the greenhouse cover, light
interception, crop photosynthesis, transpiration, dry matter
production).
At present I am working on a comprehensive model of the
greenhouse-crop system, not only simulating crop growth, but
also the greenhouse climate, based on outside weather
conditions. The work is actually not the development of new
modules for rates of all kind of processes, but
rather the putting together of modules developed at various
places. The work is a part of a project in collaboration with
Israeli colleagues. Together with one of those colleagues, DR
Shabtai Cohen, I am working on a more advanced design of the
model than of the models of the greenhouse-crop system known
to us. It is being implemented in FORTRAN. We are aiming to apply
the philosophy of object orientation as much as possible,
although FORTRAN is obviously not the language best suited
for this purpose.
The question I have is: what are the experiences of people of
using MATLAB (SIMULINK)? I believe it is popular in the
technical sciences, but how about using it as an environment
for modelling physiological processes and crop growth?
Some specific points:
- Can a model be kept well structured and the functionality
within a model be kept clear, also when the model is
increasing in size and complexity?
- Can one easy work on one part of the model, while the
complexity of the rest is hidden?
- Is it easy for a novice user (like a student) to work with
a model build in MATLAB (i.e. it is not only doing runs, but
also doing minor adaptations)?
- Is it easy to do sensitivity analysis, validation, and
parameter calibration with a model?
- I was told MATLAB is not object-oriented. Is this a major
drawback of the package?
- Can the execution speed of a model build in MATLAB be a
problem?
Thanks for any response.
---------------------------------------------
Hans Gijzen
Dept. of Horticulture
Wageningen Agricultural University
Wageningen, the Netherlands
E-mail: GIJZEN@RCL.WAU.NL
----------------------------------------------
Basant,
Several daily timestep (somewhat close to event basis) models exist which
can handle years of simulation, depending on your specific needs. Most of
the US developed models rely on the Curve Number method developed by the
USDA Soil Conservation Service. This is an empiracal approach based on
soil hydrologic group (A, B, C, or D with A generally being sandier and D
having higher clay values) and basic cropping or management practices such
as contoured, row crops. Some of the models grouped by scale are:
(a.)field-scale (one homogenous unit area):ALMANAC, EPIC, NLEAP, PRZM2,
SPUR (b.) hillslope-scale: APEX, GLEAMS, WEPP and (c.) watershed-scale:
HSPF SWAT/SWRRBWQ
I am sending you a paper and software (MOD-C-LECT) by surface mail. This
may be helpful in selecting the proper model(s) depending on your specific
needs.
Ray H. Griggs
Blackland Research Center
Texas Agricultural Experiment Station
808 East Blackland Road
Temple, TX 76502
(817) 770-6631
email griggs@brcsun0.tamu.edu
>Dear Ag. Modellers,
>
>I need to estimate runoff from a small catchment (50 ha) for our
>experimental farm. I wonder whether there is a computer model that will
>estimate runoff during a year on rainfall event basis. I shall greatly
>appreciate any help on this.
>
>Thanks.
>
>Basant Maheshwari
>
>
>________________________________________________________________
>Dr. B.L. Maheshwari Email
>b.maheshwari@uws.edu.au
>School of Agri. & Rural Development Tel. (61+45) 701 235
>or 885 652
>University of Western Sydney Fax (61+45) 885 538
>Richmond, NSW 2753, AUSTRALIA
>________________________________________________________________
>
Sorry, I forgot to specify the subject when sending my previous mail.
This message is the same as that one.
--------------------------------
Dear Members,
I have a question about MATLAB, but first I would like to
introduce myself.
My name is Hans Gijzen. I am working at the Department of
Horticulture of the Wageningen Agricultural University. Until
last year I was at the AB-DLO (formerly CABO-DLO). My field
is the development and validation of greenhouse crop models
(light transmission of the greenhouse cover, light
interception, crop photosynthesis, transpiration, dry matter
production).
At present I am working on a comprehensive model of the
greenhouse-crop system, not only simulating crop growth, but
also the greenhouse climate, based on outside weather
conditions. The work is actually not the development of new
modules for rates of all kind of processes, but
rather the putting together of modules developed at various
places. The work is a part of a project in collaboration with
Israeli colleagues. Together with one of those colleagues, DR
Shabtai Cohen, I am working on a more advanced design of the
model than of the models of the greenhouse-crop system known
to us. It is being implemented in FORTRAN. We are aiming to apply
the philosophy of object orientation as much as possible,
although FORTRAN is obviously not the language best suited
for this purpose.
The question I have is: what are the experiences of people of
using MATLAB (SIMULINK)? I believe it is popular in the
technical sciences, but how about using it as an environment
for modelling physiological processes and crop growth?
Some specific points:
- Can a model be kept well structured and the functionality
within a model be kept clear, also when the model is
increasing in size and complexity?
- Can one easy work on one part of the model, while the
complexity of the rest is hidden?
- Is it easy for a novice user (like a student) to work with
a model build in MATLAB (i.e. it is not only doing runs, but
also doing minor adaptations)?
- Is it easy to do sensitivity analysis, validation, and
parameter calibration with a model?
- I was told MATLAB is not object-oriented. Is this a major
drawback of the package?
- Can the execution speed of a model build in MATLAB be a
problem?
Thanks for any response.
---------------------------------------------
Hans Gijzen
Dept. of Horticulture
Wageningen Agricultural University
Wageningen, the Netherlands
E-mail: GIJZEN@RCL.WAU.NL
----------------------------------------------
We have several people here at OSU using Matlab, mostly for teaching projects.
Our experience has been that the Simulink package is pretty nice for doing
relatively simple simulations but begins to break down for more sophisicated
problems. It is not object-oriented in any significant sense. It is pretty
easy to work with and modify, very nice for students, and supports multiple
hardware/OS environments. Not particularly fast unless your running your models
on big hardware. Relatively inflexible when it comes to I/O, linkages with
other programs.
If you are serious about sophisticated models, object-orientation, and
performance, we have found C++ to be a very good choice. It offers full
object-orientation (ideal for complex modelling work), good performance, great
development environments, and is much more readable, flexible and maintainable
than FORTRAN. After working in an object-based language, you quickly realize
the limitations of FORTRAN for doing modeling. I would think hard about
proceeding with a FORTRAN implemenation.
============================================================================
John Bolte
Bioresource Engineering
Oregon State University
============================================================================
If you know any place in the internet where I can get wheather
information sets for Europe, please emailme.
Bye!
On Tue, 21 Mar 1995 21:12:00 -0600 Basant Maheshwari said:
>Dear Ag. Modellers,
>
>I need to estimate runoff from a small catchment (50 ha) for our
>experimental farm. I wonder whether there is a computer model that will
>estimate runoff during a year on rainfall event basis. I shall greatly
>appreciate any help on this.
>
>Thanks.
>
>Basant Maheshwari
>
>
One widely used model which provides daily runoff values is CREAMS and/or
it's successor GLEAMS. You will need to know the slope(s) of the catchment
in detail as well as soil permeability, daily rainfall, etc. Contact Ralph
Leonard (sewrl@tifton.cpes.peachnet.edu) and he will tell you how to get a
copy. If you have problems contact me back directly.
Don Wauchope
>________________________________________________________________
>Dr. B.L. Maheshwari Email
>b.maheshwari@uws.edu.au
>School of Agri. & Rural Development Tel. (61+45) 701 235
>or 885 652
>University of Western Sydney Fax (61+45) 885 538
>Richmond, NSW 2753, AUSTRALIA
>________________________________________________________________
>
>
>
> If you know any place in the internet where I can get wheather
> information sets for Europe, please emailme.
> Bye!
I would appreciate a copy of any responses to this request.
David Gustafson
Monsanto
digust@ccmail.monsanto.com
Here is a post that appeared on sci.agriculture recently regarding
choice of language for math intensive applications. Apologies for
the long lines.
Tom
Tom Hodges, Cropping Systems Modeler
USDA-ARS email: thodges@beta.tricity.wsu.edu
Rt. 2, Box 2953-A voice: 509-786-9207
Prosser, WA 99350 USA Fax: 509-786-9370
== ## Rent this space ## ==
If this represents anything, it is only my opinion.
Below is a note I just sent to the chairman of a physics department which
requires freshmen to take a programming class offered by the engineering
department. That course has traditionally taught Fortran 77, but there is now
a push on from the engineering school to switch to teaching C++. As usual,
the arguments being mustered revolve around criticisms of Fortran 77 by
people who are too out of touch to know much of anything about Fortran 90.
In either case, I was asked by one of the senior physics faculty to contribute
my two cents worth about what language they should be teaching. The appended
note is the one I sent and it summarizes our corporate experience with C++ and
some of our feelings about Fortran 90. Others may find this interesting since
this question is arising throughout the academic community.
John
----
Dr. John K. Prentice
Quetzal Computational Associates
3701 San Mateo N.E., Suite I, Albuquerque, NM 87110-1249 USA
Phone: 505-883-3706 E-mail: john@quetzalcoatl.com
========= my note to the physics department
Colston mentioned to me that the Engineering school is considering changing
Eng 120 to teach C++ instead of Fortran and that you were soliciting comments
about this with regard to physics students. Even though I am not associated
with the university, I wanted to contribute my thoughts, which come from a
commercial as well as a research perspective. Perhaps they will be of some
value to you.
As you know, Quetzal Computational Associates specializes in computational
science. We currently have projects computational physics, earth sciences,
and agriculture for clients which include DoE and DoD laboratories as
well as commercial clients as diverse as hazardous waste companies and
grain companies. During the last year, we have developed numerical methods
and codes based on them for modeling contaminant flow in porous media,
modeling the structural mechanics of resonant sonic drilling rigs, modeling
bistatic ground penetrator radar propagation in partially saturated soils,
modeling solid dynamics at high strain rates, modeling the phenological
development of corn and soy beans, modeling solar insolation in the
photosynthetically active spectrum based on first principle atmospheric
physics, and developed neural networks for the detection of dust clouds from
satellite imagery.
One of our largest computational physics projects is the development of
advanced methods for modeling solid dynamics based on first principle physics.
This is a multi-year, multi-million dollar project. This code
numerically solves the partial differential equations for continuum solid
dynamics using a hybrid finite volume/finite element technique, coupled
to advanced equations of state and constitutive models for the solids and
fluids in the calculation. This code and others we work with are huge number
crunching codes, a modest 3d simulation will take 100 or more hours of Cray
C-90 time to complete a single calculation. By any standard, they are amongst
the largest computational physics simulations being done anywhere in the
world. In addition, we are on the forefront in the application of parallel
computing to these problems. We have projects to develop parallel versions
of our codes for a diverse collection of computers, including networks of
UNIX workstations, the IBM SP-2, the Cray T3D, and the Intel Paragon.
For all of these projects, we employ Fortran 90 as our main language. We
do some development work in Fortran 77, C, and C++, but we are moving away
from those languages as quickly as possible. There are many reasons for
our choice of Fortran 90, but first let me say a bit about why we are not
enthusiastic about C++. The biggest strength of C++ is probably the
availability of relatively inexpensive and high quality C++ compilers for PCs.
But that is a pretty minor consideration in our business and it is outweighed by
the enormous liabilities we have observed with C++. First, we regard C++ as
the weakest of the object oriented languages. Objective C is a far more
solid and well designed OOPS language, C++ is really some OOPS capability
slapped on top of C. C++ is consequently extremely inefficient, inconsistent,
overly large, and enormously difficult to program in. The experience of our
clients mirrors our own, and in fact many DoE and DoD laboratories are
finding that their headlong rush to C++ has been a hideously expensive
mistake. I know of several C++ scientific coding projects in the DoE that
consumed millions of dollars and tens of man-years, only to be abandoned
because the resulting code was enormously inefficient on both traditional
serial computers and on their large parallel supercomputers. Similar
horror stories abound throughout the programming community at this point.
Bill Gates claimed that his biggest mistake in designing their new NT
operating system was adopting C++ for the graphics coding, the resulting
code took years longer to write than it should have and ran terribly
slow. While OOPS is a solid development in the computer science community,
I think it is fair to say that C++ is destined to be a passing fad, much
like Pascal and Ada before it.
The main reason C++ has attracted the attention it has in the scientific
community is because Fortran 77 was a terribly outdated language. The
many weaknesses of Fortran 77 were solved with Fortran 90 however. Fortran 90
has every feature in C that is important to scientific programming and most of
the features of an object oriented language (it lacks only inheritance and that
is likely going to be added in Fortran 2000). However, unlike C and C++,
Fortran 90 is designed to generate executable codes that are highly
optimized and thus run extremely fast. An example is pointers. Pointers
are integral to C and C++ programming and because the compiler cannot
determine whether a pointer is aliased, it is impossible for it to
determine interprocedural dependencies. The result is a significant
degradation in optimization and extremely slow execution speeds (for
most scientific codes, C and C++ generally produce code which is commonly
an order of magnitude slower than Fortran 90 codes, based on the benchmarks
we and others have done). Fortran 90 pointers are designed to give the
functionality of pointers, but with restrictions that eliminate issues
such as aliasing. From a programming perspective however, an even more
important point is that Fortran 90 has more natural ways of expressing
the functionality that C and C++ require pointers to express. Because of
this, Fortran 90 is a more natural language to program in and the time
required for debugging codes is a fraction of that required by C and C++
(C++ is much worse than C, provided you are really employing an OOPS
paradigm, since you find yourself spending alot of debug time going up
and down inheritance trees). Another important point is that the time
required to learn Fortran 90 is much less than the time to learn either
C or C++.
Fortran 90 has another major advantage over C or C++. Modern scientific
computing, and computing in general, is moving toward the use of
parallel computers. Even PCs and workstations now come with multiple
processors, so parallelism is something that everyone from an accountant
to a physicist is encountering now. A major problem in programming
parallel computers however is the linear memory model that is inherent
to all procedural programming languages, with the singular exception of
Fortran 90. A linear memory model is one that assumes that consecutive
elements of an array are consecutive in memory. This was a reasonable
assumption on traditional computers, but it is completely incorrect on
a parallel computer. Only Fortran 90 has addressed this problem and
providing standardized language support for parallelism. This support includes
array syntax and many intrinsics for doing array operations varying
from reduction operations such as array sums to matrix operations. With
the use of Fortran 90 operator overloading and polymorphism, one can
significantly extend the number of operations that avoid any reliance
on the linear memory model. The fact that Fortran 90 moved away from
a linear memory model is the main reason that it has become the base
for so many data parallel languages such as Vienna Fortran, Fortran D,
CRAFT, and High Performance Fortran. The availability of data parallel
dialects of Fortran 90 is an especially large factor in favor of Fortran 90.
Compilers for High Performance Fortran, for example, are now coming on
the market for virtually every machine out there (including networks
of workstations) and writing parallel codes in this language is
straightforward. Of particular importance is that porting a Fortran 90
code to High Performance Fortran is extremely straightforward and codes
written in High Performance Fortran can be run unaltered on a Fortran 90
compiler (with the exception of one HPF construct, the forall, which is
being put into Fortran 95).
My own opinion is that scientists today need to know more than one language
or one computing paradigm. And I think it is entirely reasonable that students
learn C++ before they graduate, though even more important is that they
learn how to program MATLAB and a computer algebra system such as Maple
or Macsyma. But the issue is what freshmen should learn as their first
language and for that I would recommend Fortran 90 hands down. It is
a better language for scientific programming and is both easier to learn and
use than the alternatives. It is also much more likely to be the language
students will be employing in their jobs upon graduation and it is
the most promising route currently developing for the programming of parallel
computers.
John
Howdy,
I was just going to reply to the original sender, but after
thinking about it I figured it would be of interest to the
general population of Ag modelers.
The NCDC (National Climatic Data Center) has the summary of
day surface environmental data for 1994 archived at thier
anonymous FTP site ftp.ncdc.noaa.gov in a dirrectory
labeled "globalsod". This contains lots of data from all over
the world. (27Meg per month, compressable to 5meg). I
forget the full dirrectory path, but if any one has problems
finding it I can look it up and pass it on to them.
The NCDC also has the historical data sets from 1991
back on CDROM for $120. I am trying to get the years
1992 and 1993 compiled but I recieved a note back from
them saying there is not enough interest in it, but they
would compile the data for me for some $15K a month.
If anyone has access to the data over the US for the
years 1992 and 93, I would greatly appreciate if they
would help me get access to it. I know earthinfo inc.
has it, but charges $2400 for the continental US and
this project can not afford that much of an expence.
Hope this is of some help, and thanks in advance for
the help in finding the 1992 and 93 data sets.
Daniel R. Fuka
Quetzal Computational Associates
dan@quetzalcoatl.com
> From: ricbraga@isa0.isa.utl.pt
> To: Multiple recipients of list <agmodels-l@unl.edu>
> Content-Length: 114
>
> If you know any place in the internet where I can get wheather
> information sets for Europe, please emailme.
> Bye!
>
>
Dear Agmodelers:
My name is Ferhan Ozadali. I am a graduate student toward Ph.D degree at
Iowa State University. Is there anybody out there working on microbial
growth or production modeling? If there is, would you e-mail me? Does
anybody know a very user friendly model? I used modified logistic and
modified Gompertz models before. Any information would be greatly
appreciated.
Thanks,
Ferhan Ozadali
burcu@iastate.edu
--------
I would also be interested in finding WWW sites for
European weather data. The UK is especially difficult.
F.A.O. have a disc of monthly averages for the
Mediterranean zone of Europe as well as Africa, Asia, South
America but that is not online as far as I know.
Graham
Dr G. Russell
University of Edinburgh
Institute of Ecology & Resource Management (Agriculture Building)
West Mains Road
Edinburgh EH9 3JG
SCOTLAND
International phone +44 31 535 4063 Fax +44 31 667 2601
UK phone 031 535 4063 Fax 031 667 2601
>
> If you know any place in the internet where I can get wheather
> information sets for Europe, please emailme.
> Bye!
>
>
There is a mailing list called CLIMLIST <CLIMLIST@psuvm.psu.edu> where
you probably get an answer on your question.
Also, the DKRZ in Hamburg has some information on the WWW. Send an e-mail
with your inquiry to www@dkrz.d400.de
---------------------------------------------------------------------
Peter Becker-Heidmann : : /YY\ : :
Institut fuer Bodenkunde : |YYYY|. :
Allende-Platz 2, D-20146 Hamburg, GERMANY : . \YY/. : :
Phone: +49 40 4123 2003, Fax: +49 40 4123 2024 __:__:_||__:__:__
E-Mail: PBeckerH@Uni-Hamburg.de SOILSOILSOILSOILS
---------------------------------------------------------------------
Can anyone supply me with information about the Lunteren Workshop where
15 wheat models were compared against two common data sets? I would like
to know the model name, contact person, and an e-mail address for this
person. I read about this workshop in the Global Change Newsletter of
June 1994.
Albert Weiss
ALBERT WEISS, DEPT. OF AGRICULTURAL METEOROLOGY
UNIVERSITY OF NEBRASKA-LINCOLN, LINCOLN,NE 68583-0728
VOICE (402)472-6761, FAX (402)472-6614
E-MAIL AGME011@UNLVM.UNL.EDU
The results of the Lunteren workshop are in press or
possibly even published as
"predicting crop yields under global change" 1995. J.
Goudriaan et al. Woodshole Conference proceedings.
Graham Russell
Dr G. Russell
University of Edinburgh
Institute of Ecology & Resource Management (Agriculture Building)
West Mains Road
Edinburgh EH9 3JG
SCOTLAND
International phone +44 31 535 4063 Fax +44 31 667 2601
UK phone 031 535 4063 Fax 031 667 2601
This probably refers to the GCTE Wheat. A contact is John Ingram, GCTE
Focus 3 Officer, Oxford, UK. His e-mail is:
john.ingram@plant-sciences.oxford.ac.uk
>Can anyone supply me with information about the Lunteren Workshop where
>15 wheat models were compared against two common data sets? I would like
>to know the model name, contact person, and an e-mail address for this
>person. I read about this workshop in the Global Change Newsletter of
>June 1994.
> Albert Weiss
>
Dr. Jeffrey J. Lee jeffl@mail.cor.epa.gov
US EPA 503 754-4578
Environmental Rsearch Laboratory FAX: 503 754-4799
200 SW 35 St.
Corvallis,OR 97333
USA
I read the message regarding the use of FORTRAN 90, C, and C++ and found
it a very interesting discussion. I have been struggling with a similar
decision although not in the teaching realm. We are developing models
and expert systems for ag production management and post harvest processing.
Is FORTRAN 90 available for the PC?
************************************************************************
* Chris Butts * cbutts@asrr.arsusda.gov
*
* USDA, ARS *****************************
* National Peanut Research Laboratory * Phone: 912-995-7431 *
* 1011 Forrester Dr., SE *****************************
* Dawson, Georgia 31742 * FAX: 912-995-7416 *
************************************************************************
As of 2 years ago, only Salford was advertising a full fortran 90
compiler. Watcom and Lahey were advertising many 90 extensions in
there pc compilers. I haven't seen more recent ads, but I expect
there are several 90 compilers out now.
Tom
Tom Hodges, Cropping Systems Modeler
USDA-ARS email: thodges@beta.tricity.wsu.edu
Rt. 2, Box 2953-A voice: 509-786-9207
Prosser, WA 99350 USA Fax: 509-786-9370
== ## Rent this space ## ==
If this represents anything, it is only my opinion.
On Fri, 24 Mar 1995, Chris Butts wrote:
> I read the message regarding the use of FORTRAN 90, C, and C++ and found
> it a very interesting discussion. I have been struggling with a similar
> decision although not in the teaching realm. We are developing models
> and expert systems for ag production management and post harvest processing.
>
> Is FORTRAN 90 available for the PC?
>
Chris,
I know of one F90 compiler for the PC available right now.
This is the Lahey F90 compiler. Lahey is also resposible for
one of the better F77 compilers commercially available for
the PC (or so I am told, most of our research is done on
workstations, so I am not to familiar with PC compilers.)
If you want more information I imagine that Tom Lahey
would be about the best place to start. His phone number
is 702 831-2500 work, or 702 831-8123 fax. I am not sure
of his email address.
If you need any information on F90 compilers for workstations
I can be of more help. Or if there is any interest in High
Performance Fortran (HPF) discussions, I would be willing to throw
in my two cents. Quetzal is initiating a HPF benchmarking
study in the near future.
Hope this is of some help,
Daniel Fuka
Quetzal Computational Associates
dan@quetzalcoatl.com
>
> Is FORTRAN 90 available for the PC?
I saw Lahey FORTRAN 90 for DOS advertised in SciTech's software catalog
(1-800-622-3345). Price: $859.- ($799.- academic).
H. Scherm
Iowa State Univ.
On Fri, 24 Mar 1995 12:04:14 -0600 Chris Butts wrote:
>I read the message regarding the use of FORTRAN 90, C, and C++ and found
>it a very interesting discussion. I have been struggling with a similar
>decision although not in the teaching realm. We are developing models
>and expert systems for ag production management and post harvest processing.
>
>Is FORTRAN 90 available for the PC?
We recently purchased a Fortran 90 compiler for PC's from Salford Software. They say
it's a full ISO/ANSI standard Fortran compiler that produces 32 bit code. I haven't
worked with it very much yet but it looks promising from what I've read. It's also
quite a bit different from FTN77. It can handle pointers and structures like C can and
also has some special array manipulation functions and a 'case' function.
Salford also sells a MS-Windows library for Fortran that's fairly easy to use to create
simple windows and dialog boxes. I don't think it has the flexibility to make very
complex Windows programs though. You can use it with FTN77 but you have to be creative
when you want to access C type structures. I think it's a good alternative if you want
to add some Windows capabilities to an existing Fortran program without messing with C
code.
------------------------------------------------
Dennis Timlin
USDA-ARS Systems Research Lab
Bldg 007, Rm 008
10300 Baltimore Ave
Beltsville, MD 20705
301-504-6255
fax 301-504-5823
DTimlin@ASRR.ARSUSDA.GOV
03/24/95 15:40:12
> Is FORTRAN 90 available for the PC?
Another Fortran90 PC compiler is NDP Fortran version 4.5.0 from
Microway, Research Park, Box 79, Kingston, Massachusetts 02364
It includes the option of Pentium optimization and has a graphics
library.
*******************************************************************
* *
* Alister Metherell email METHEREA@LINCOLN.AC.NZ *
* AgResearch *
* C/o Soil Science Department phone (direct dial in) *
* P.O. Box 84 64-3-325 3888 *
* Lincoln University phone (via operator) *
* New Zealand 64-3-325 2811 ext 7888 *
* fax 64-3-325 2944 *
*******************************************************************
We are working
with a colza model, and we want to adjust the parameters to our field
data. This must be a very common problem in modelling? I would be very
interested to know how other people have approached this problem, and get
references if possible.
It seems to me that there are several aspects to this problem. First of all,
what parameters do you choose to adjust? It is out of the question to
adjust them all. There are too many, and there are no doubt some very
large correlations among the estimators, which will degrade
predictive ability of the model. Secondly, how do you
use the different kinds of field data (LAI, biomass, yield, etc)?
Do you just minimize the sum of squares for all the data? That
doesn't seem reasonable, for lots of reasons. (Heterogeneous
variances, covariances among the data, plus the fact that we
are not really interested in predicting LAI for example, so why
treat it as being as important as yield). Finally, there is
the numerical problem of actually calculating the adjusted
values, once a criterion has been chosen. I think that this ia a relatively
minor problem, once the preceding problems have been solved. We will use the
NAG numerical algorithm library.
Thanks in advance for any suggestions, or comments, or bibliography.
Daniel Wallach
Institut National de la Recherche Agronomique
wallach@toulouse.inra.fr
Available Fortran 90 PC compilers: Lahey, NDP/Microway
Due out soon (3 months according to sales rep.): Watcom
Lahey's complier was reported to be buggy (on the Fortran newsgroup)
but has been out about 6 months now so should be better. Haven't
heard much on NDP/Microway.
Nigel Pickering
Visiting Assistant
Dept. Agric. Engr pick@water.agen.ufl.edu (Internet)
12 Frazier-Rogers Hall pick@nervm (Bitnet)
Museum Road (904) 392-4092 (FAX)
University of Florida (904) 392-7719 (Voice)
Gainesville, FL 32611-0570 (904) 392-6180 (Voice)
>
>We are working
>with a colza model, and we want to adjust the parameters to our field
>data. This must be a very common problem in modelling? I would be very
>interested to know how other people have approached this problem, and get
>references if possible.
\
>Daniel Wallach
>Institut National de la Recherche Agronomique
>wallach@toulouse.inra.fr
>
>
Daniel,
Our Group has developed a new methodology for model parameteriztion using
Genetic Algorithms from the field of Artificial Intelligence. Please see
the paper:
Sequeira, R.A., R. L. Olson, J. L. Willers, and J. M. McKinion. 1994.
Automating the parameterization of mathematical models using genetic
algoritms. Computers and Electronics in Agriculture 11(1994)265-290.
Not only will the methodology help in model parameterization, but it will
also provide statistically significant measures of goodness of fit ond
improvements thereof.
Ron Sequeira can be reached at (601)324- 4345 or via e-mail at
sequeira@marlin.csrumsu.ars.ag.gov.
James
Dr. James M. McKinion Ph. 601-324-4375
USDA-ARS FAX 601-324-4371
Crop Simulation Research Unit Email mckinion@csrumsu.ars.ag.gov
P. O. BOX 5367
Mississippi State, MS 39762
USA
GREETINGS!
Below at the end, I want to present the complete research
programme of Department of Theoretical Production Ecology,
Wageningen Agricultural University.
I will visit the University of Guelph from May 14 to May 22.
Please contact Peter Kevan <pkevan@uoguelph.ca> or Franco di
Giovanni <francodi@uoguelph.ca> for making an appointment if
you'd like to meet me in person.
Looking forward to meeting you at Guelph,
...___... ...___... ...___... ...___.... ...___...
Sincerely Yours, dr Meindert D. de Jong I'm for hire!!!
E-mail: Vlinders@RCL.WAU.NL ^^^^^^^^^^^^^^^
Voice: (+31) 8370 21937 FAX: (+31) 8370 23110
Photos of research & researchers, c.v.:
de Jong et al. 1990. Risk Analysis for Biological Control:
A Dutch Case Study in Biocontrol of Prunus serotina by the
Fungus Chondrostereum purpureum. Plant Disease 12:189-194.
Send request for reprint to: AB-DLO, P.O.Box 14,
6700 AA Wageningen, Netherlands
================== research programme ========================
Department of Theoretical Production Ecology, WAU, Netherlands
The research programme of the Depertment of Theoretical
Production Ecology concentrates on matter and energy flows
from a production ecology perspective. The programme has four
themes: (1) Soil and climate, (2) Crop growth and development,
(3) Pests, diseases and weeds and (4) Agricultural production
systems. The distinction in themes is hierarchical, in that
one theme contributes to at least one subsequent theme.
Overview of Research Projects
Theme 1: Soil and climate
Research in this theme involves soil physics, soil microbiolo-
gy and micro-meteorology. Results are largely used to support
research in other themes. This applies especially to the
micrometeorological work, which has been used in the simulati-
on of crop growth and in calculations of the uptake by crops
of air polluting components.
Soil biology research focuses on partial anaerobiosis in
unsaturated soil and on microbial activity in connection with
the nitrogen economy of the soil and the consequences for crop
growth. In addition, the effect of partial anaerobiosis on
production of gases which contribute to the greenhouse effect
is studied. Projects:
. Soil biota interactions with respect to nutrient
transformation
. Vegetation-atmosphere models
. Factors effecting emission of nitrous oxide from grasslands
in the removal of nitrate from soil by denitrification
. Quantitative assessment of growth and N-utilization in
catch crops with regard to a higher nitrogen efficiency in
plant production systems
Theme 2: Crop growth and development
Crop growth under potential and limiting conditions is central
to the Department's research activities. Research projects
cover a range of scales in time and space. For example, detai-
led crop physiological studies are carried out into the beha-
viour of leaf stomata and into the kinetics of ion fluxes in
plant roots. At the crop level, growth and production is
studied in relation to different levels of availability of
water and nitrogen. Research in this theme resulted in simula-
tion models of monocultures as well as mixed vegetations on
different soil types.
Furthermore, research is done to analyse the increase of
carbon dioxide in the atmosphere and its effect on crops and
vegetations.
The morphogenetic development of crops is studied to provide
quantitative insight in crop form and fuction. In particular,
the development of leaf area and plant height in crops and
weeds is investigated in relation to the increase in dry
matter. Projects:
. Effects of climate changes on the production potentials of
winter wheat, silage and grain maize in the E.C.
. Diurnal and seasonal CO2 exchange and its components in
grasslands
. Developments, updating and application of crop growth models
used in research
. Simulation and systems analysis for rice production
. Quantification of the energy costs of maintenance and
transport processes in crops
. Global carbon cycle
. Harvesting the sun's energy using agro-ecosystems
. Analysis and modelling of nitrogen dynamics in soil, crop
and air of irrigated rice
. Adaptability and yielding ability of potato crops in
(sub)tropical regions
. Genotype-environment interaction and yield gap analyses in
wheat production using simulation modelling
Theme 3: Pests, diseases and weeds
In this theme the interaction between crops and growth reduc-
ting factors is emphasized. On the one hand, this leads to
studies on the effects of growth reducing factors on crop
growth and production based on analysis of injury in crop
physiological and crop ecological terms. Insights gained in
the first two research themes are used extensively in these
studies. On the other hand, the temporal and spatial dynamics
of plant pathogens, insects, mites, and weeds are studied. The
dynamics of populations are explained on the basis of behavi-
our of individuals or integrated control. Also, aspects which
are specifically related to decision support are addressed,
such as sampling techniques and risk associated with uncert-
ainty in predictions. Projects:
. Theoretical and experimental research on the nature and
level of yield depression in crop canopies due to pest and
disease causing organisms
. Improvement of the use of pesticides in wheat
. Population dynamics of pests and disease in crops;
impact on crop growth and production
. Crop-weed interactions
Theme 4: Agricultural production systems
In theme 4 knowledge acquired in the first three themes is
integrated and combined with agronomic and economic considera-
tions to explore the possibilities for agriculture under
different constraints. These studies are carried out at the
levels of farm, region and nation. Interactive multi-goal
linear programming is used as an integrative framework.
Because of the innovative nature of the approach, development
of methodology is an important topic of this theme. Projects:
. Developments of methods for designing, validation and
optimization of integrated cropping systems
. Perspectives for environment-oriented production of
flower bulbs
. Methodology, development for quantification of
LUST's regional and farm household models
#
We are planning a section on user interfaces for agronomic
models for the software scene at the next American Society of
Agronomy meeting in St. Louis. The deadline is approaching
(April 8) for submission of title summaries. Oral, Poster, and
Computer demonstrations are invited on types of interfaces,
principles of design, interface development tools, and use of
interfaces. I encourage anyone who is doing work in this area to
submit a title summary. Of special interest are papers that
discuss how a particular type of interface delevoped, the design
philosophy, etc. Please e-mail me if you need a title summary
form or more information.
------------------------------------------------
Dennis Timlin
USDA-ARS Systems Research Lab
Bldg 007, Rm 008
10300 Baltimore Ave
Beltsville, MD 20705
301-504-6255
fax 301-504-5823
DTimlin@ASRR.ARSUSDA.GOV
03/28/95 14:04:02
------------------------------ Cut here ------------------------------