• TABLE OF CONTENTS
HIDE
 Title Page
 The problem
 The challenge
 Approach
 Program goal
 Systems-based research
 Data bases. expert systems and...
 Using farmer knowledge in decision...
 Prototype decision support...
 Objective
 Project outputs
 Work plan by objective/schedul...
 Attachment






Title: Strategic plan for the Indonesia TropSoils program
CITATION PAGE IMAGE ZOOMABLE PAGE TEXT
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00055445/00001
 Material Information
Title: Strategic plan for the Indonesia TropSoils program
Physical Description: Book
Language: English
Creator: TropSoils
Publisher: TropSoils
 Subjects
Subject: Farming   ( lcsh )
Agriculture   ( lcsh )
Farm life   ( lcsh )
 Notes
Funding: Electronic resources created as part of a prototype UF Institutional Repository and Faculty Papers project by the University of Florida.
 Record Information
Bibliographic ID: UF00055445
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.

Table of Contents
    Title Page
        Title Page
    The problem
        A 1
    The challenge
        A 1
    Approach
        A 2
    Program goal
        A 3
    Systems-based research
        A 3
    Data bases. expert systems and simulation models
        A 3
        A 4
    Using farmer knowledge in decision making
        A 5
    Prototype decision support system
        A 6
    Objective
        A 7
    Project outputs
        A 8
    Work plan by objective/schedule
        A 8
        A 9
        A 10
        A 11
    Attachment
        B 1
        B 2
        B 3
        B 4
        B 5
        B 6
        B 7
        B 8
        B 9
        B 10
        B 11
        B 12
        B 13
        B 14
        B 15
        B 16
        B 17
        B 18
        B 19
        B 20
Full Text









STRATEGIC PLAN FOR THE INDONESIA

TROPSOILS PROGRAM









STRATEGIC PLAN FOR THE INDONESIA

TROPSOILS PROGRAM



The Problem

A backlog of potentially beneficial soil management practices remains

unused because the people who need them most do not have the time, resources or

skills to test every promising practice in each biophysical and socioeconomic

setting. The ecological range over which farmers operate is so great that even

the largest research network cannot hope to deal with the unique

characteristics of every farming system. Today the cost of testing new crops,

products and practices in existing farming systems far exceeds the cost to

produce the crop, product or practice. And if the new field of biotechnology

delivers on its promise to produce even more innovations, the inability to move

technology from research centers to farmer fields will continue to be the major

bottleneck to agricultural development.



The Challenge

The Agency for International Development working with U.S. and host

country institutions is developing the means to overcome bottlenecks that now

prevent rapid integration of new crops, products and practices into existing

farming systems to make them more efficient and productive. But at the same

time, AID, the U.S. institutions and the host country agencies are faced with

declining resources to carry out their responsibilities. The new challenge for

the group, and particularly for the U.S. institutions, is to provide more

service of better quality with fewer resources. To do so the U.S. institutions

with their large technical base must provide the scientific leadership to

create with the USAID mission and host country agencies the research structure









STRATEGIC PLAN FOR THE INDONESIA

TROPSOILS PROGRAM



The Problem

A backlog of potentially beneficial soil management practices remains

unused because the people who need them most do not have the time, resources or

skills to test every promising practice in each biophysical and socioeconomic

setting. The ecological range over which farmers operate is so great that even

the largest research network cannot hope to deal with the unique

characteristics of every farming system. Today the cost of testing new crops,

products and practices in existing farming systems far exceeds the cost to

produce the crop, product or practice. And if the new field of biotechnology

delivers on its promise to produce even more innovations, the inability to move

technology from research centers to farmer fields will continue to be the major

bottleneck to agricultural development.



The Challenge

The Agency for International Development working with U.S. and host

country institutions is developing the means to overcome bottlenecks that now

prevent rapid integration of new crops, products and practices into existing

farming systems to make them more efficient and productive. But at the same

time, AID, the U.S. institutions and the host country agencies are faced with

declining resources to carry out their responsibilities. The new challenge for

the group, and particularly for the U.S. institutions, is to provide more

service of better quality with fewer resources. To do so the U.S. institutions

with their large technical base must provide the scientific leadership to

create with the USAID mission and host country agencies the research structure









to move quickly and efficiently, the huge backlog of underexploited

agroproduction technology from research centers to farmer fields.



Approach

Recent advances in computer technology and artificial intelligence present

the Indonesia TropSoils program with an opportunity and responsibility to

measureably increase the flow of agroproduction technology from research

centers to farmer fields. These advances include personal computers with near

mainframe capability, data base management software to store large quantities

of information in readily retrievable form, and new developments in the

artificial intelligence field covering such areas as theorem proving, game

playing, machine learning, pattern recognition, natural language processing,

robotics, machine cognition and expert systems.

These advances in information technology enables agricultural researchers

to replace slow and costly trial and error research with systems-based

research. The central concept of systems-based research is that the whole

system -must-be- understood in-order -to-evaluate-changes-iTr-any-- iigle-compenen~--

of the system. This approach brings together existing acknowledge of the

farming system, identifies major components and processes and their

interactions, and seeks to identify the bottlenecks to improve performance.

Until recently this approach could not be applied to complex systems

consisting of large number of interdependent and interacting factors. Advances

in information technology now enable users to organize a minimum set of

relevant data to simulate complex process in agricultural systems so that

costly and time consuming trial and error adjustments can be avoided.








Program Goal L2 c- ( J

The principal goal of the TropSoils program is to uncover principle which

will enable resource-poor farmers to adopt soil management practice that will

increase farm productivity and-family--income and, at-the-same time preserve

land quality for future generations. The research strategy is designed to

ensure that social, cultural, economic and environmental factors that enhance

adoption of soil management innovations are made an integral part of the

research plan. To achieve its goal, the program conducts a significant part of

its research in farmers' fields, using systems-based research.



Systems-Based Research

The basic aim of systems-based research is to enable users to apply

knowledge of specific processes in complex agricultural systems and then to be

able to utilize this knowledge to obtain a comprehensive understanding of

the way the systems operates as a whole. A thorough understanding of systems

operation is the basis for developing expert systems and simulation models that

mimic, and therefore, predict the behavior and.performance of each part as

it interacts with other parts in the system. Understanding and prediction, in

turn, provide the basis for controlling outcomes. A thorough understanding

of how agricultural systems operate is crucial to helping farmers and

government planners control outcomes in desirable and predictable ways.



Data Bases, Expert Systems and Simulation Models

Data bases on the one hand and expert systems and simulation models on the

other are the critical ingredients of a systems-based research strategy.

Figure 1 illustrates how data bases and experts system/simulation models relate

to predicting and controlling outcomes of alternative choices. Predicted








Program Goal L2 c- ( J

The principal goal of the TropSoils program is to uncover principle which

will enable resource-poor farmers to adopt soil management practice that will

increase farm productivity and-family--income and, at-the-same time preserve

land quality for future generations. The research strategy is designed to

ensure that social, cultural, economic and environmental factors that enhance

adoption of soil management innovations are made an integral part of the

research plan. To achieve its goal, the program conducts a significant part of

its research in farmers' fields, using systems-based research.



Systems-Based Research

The basic aim of systems-based research is to enable users to apply

knowledge of specific processes in complex agricultural systems and then to be

able to utilize this knowledge to obtain a comprehensive understanding of

the way the systems operates as a whole. A thorough understanding of systems

operation is the basis for developing expert systems and simulation models that

mimic, and therefore, predict the behavior and.performance of each part as

it interacts with other parts in the system. Understanding and prediction, in

turn, provide the basis for controlling outcomes. A thorough understanding

of how agricultural systems operate is crucial to helping farmers and

government planners control outcomes in desirable and predictable ways.



Data Bases, Expert Systems and Simulation Models

Data bases on the one hand and expert systems and simulation models on the

other are the critical ingredients of a systems-based research strategy.

Figure 1 illustrates how data bases and experts system/simulation models relate

to predicting and controlling outcomes of alternative choices. Predicted








Program Goal L2 c- ( J

The principal goal of the TropSoils program is to uncover principle which

will enable resource-poor farmers to adopt soil management practice that will

increase farm productivity and-family--income and, at-the-same time preserve

land quality for future generations. The research strategy is designed to

ensure that social, cultural, economic and environmental factors that enhance

adoption of soil management innovations are made an integral part of the

research plan. To achieve its goal, the program conducts a significant part of

its research in farmers' fields, using systems-based research.



Systems-Based Research

The basic aim of systems-based research is to enable users to apply

knowledge of specific processes in complex agricultural systems and then to be

able to utilize this knowledge to obtain a comprehensive understanding of

the way the systems operates as a whole. A thorough understanding of systems

operation is the basis for developing expert systems and simulation models that

mimic, and therefore, predict the behavior and.performance of each part as

it interacts with other parts in the system. Understanding and prediction, in

turn, provide the basis for controlling outcomes. A thorough understanding

of how agricultural systems operate is crucial to helping farmers and

government planners control outcomes in desirable and predictable ways.



Data Bases, Expert Systems and Simulation Models

Data bases on the one hand and expert systems and simulation models on the

other are the critical ingredients of a systems-based research strategy.

Figure 1 illustrates how data bases and experts system/simulation models relate

to predicting and controlling outcomes of alternative choices. Predicted









outcomes have no credibility with users until the models that produce them have

been tested and validated in a number of environmentally different locations.

A simulation model is considered to be validated if it can reliably predict the

performance and yield of a crop anywhere in Indonesia, at any. time of the year

and for a wide range of soil management options. To do so, the model must have

access to a data base that contain (1) a genetic resource data base, (2) a soil

resource data base and (3) a weather data base.

Simulation models are particularly useful for strategic planning by

government planners. With the proper natural resource data base for a country,

a planner can simulate the performance of a crop, product or practice for any

location and for as many years as one wishes. Simulation models are

indispensable not only because they provide answers cheaply, but because they

can do what no researcher can do experimentally. Models can simulate processes

for 25, 50, or 100 years. It is now possible to execute one year of simulation

in two or three seconds on a mainframe computer or two to three minutes on a

personal computer. Long term simulations require long term weather data.

Since few countries have long term weather data, modelers have developed

weather generators to extend 15-20 year weather data into long term

probabilistic weather patterns. The long term simulated results are

particularly valuable for risk analysis and will show the distribution of good

and bad years for a particular location and crop.

Expert systems are more useful for tactical planning at the farm level.

They can be used to identify and control plant diseases, schedule

irrigation, fertilizer application or spraying. The TropSoils Project, for

example, has produced an expert system to assist users to make lime

recommendation to correct soil acidity. By distilling and condensing the

knowledge of human experts into a set of interlinked rules, knowledge engineers









are able to develop computerized expert systems that enable young and

inexperienced extension agents to shift through a large knowledge base in a

matter of few minutes to obtain expert recommendations for specific problems

for a specific location and situation.



Using Farmer Knowledge in Decision Making

The power of expert systems lies in the system's capacity to capture and

mimic knowledge known not only to scientists, but to farmers as well.

Scientists generate quantitative, mechanistic knowledge that can be expressed

mathematically. This type of knowledge has been called algorithmic knowledge.

Farmers, on the other hand, employ rules of thumb, educated guesses, intuitive

judgements or what we call, plain common sense. This type of knowledge which

is receiving increasing attention in the field of artificial intelligence is

known as heuristic knowledge.

One reason for the poor communication between researchers and farmers is

that researchers think algorithmically whereas, farmers think heuristically.

In systems-based research, both types of knowledge and thinking process are

needed to enable both researcher and farmer to make responsible choices which

result in desired outcomes. But with few exceptions, researchers rarely take

time to ask the crucial questions of farmers. It is here that a social

scientists can make a difference by coupling algorithmic to heuristic knowledge

by serving as the "cultural broker" between farmer and agricultural scientist.

It is the social scientist's responsibility to see that the relevant farm data

is collected and that the chain of rules that link a given set of input

conditions to an appropriate output is uncovered.






6


Prototype Decision Support System

When the TropSoils program comes to an end, a computerized, interactive,

decision support system that provides users with easy access to data bases and

decision models to support decision making tasks for strategic planning by

government agencies and tactical planning for farmers by extension agents

should be in place. Like all decision support systems, this prototype is

designed to answer "what if" questions. What if a new cultivar of a soybean

were introduced into an area where soybean had never been grown? How will it

perform there? Will it do better in certain months? What would be the best

seeding rate? Do soils of that region pose special problems to soybean

production? Would lime application benefit the crop? What diseases and

insects can a grower expect to encounter and what measures would one need to

take to protect the crop from them? Are there crops other than soybean that

would do as well, be more profitable or present less risk?

To answer these questions and many more like them for different crops,

different locations and different situation, three essential elements of the

decision support system must be in place at project's end. These elements are

(1) the natural resource (soil, climate, plant genetics) data base, (2) the

decision models (crop simulation models and expert system), and (3) a dialog

generator (means through which users communicate with decision support

system).

It is not the intent of the TropSoils Project to develop a decision

support system from scratch. Software for such systems are commercially

available and a suitable system currently under development by another AID

funded projects is available for use by the TropSoils program.

The value of a decision support system is that it serves as an instrument

for integrating the various components of a national agricultural research into

a coherent whole. In this way the TropSoils program is able to conduct its









research in concert with other activities of the national research system.

TropSoils considers this integration to be an essential part of systems-based

research.

Figure 2 illustrates the connection between expert systems, crop models,

soils, weather, genetic coefficients and experimental data, and Table 1

describes each component of Figure 2. A more detailed description of what is

entered into each file is given in an attachment to this report.

The contents of the attachments is critical to TropSoils team members

because it specifies the minimum data set that must be collected from each

experiment to enable the results to be used in national or farm level decision

making tasks.

The principal research hypothesis of the Indonesia TropSoils Project is

that soil management research conducted in the manner outlined will meet the

challenge of providing more services, of higher quality, with fewer resources.



Objective

The objective of the Indonesia TropSoils Project is to establish a

decision support system that will:

integrate TropSoils research with the national agricultural development

plan;

produce user-oriented decision models by adding farmer knowledge to the

knowledge base;

render soil management experimental results directly useful for decision

making; and

increase efficiency of soils-related research.









Project Outputs

The expected project outputs by 1989 are:

an interactive decision support system that provides users with easy

access to data bases and decision models to support decision making

tasks; and

personnel trained to validate decision models and operate, maintain and

upgrade decision support systems.



Work Plan by Objectives/Schedules

I. Integrade TropSoils research into national agriculture development plan

Distribute strategic plan to Management Entity, Indonesian Agency and

USAID, April 1, 1986.

Present strategic plan to External Evaluation Panel and Management

Entity, April 21-22, 1986.

Meet with Indonesian and USAID administrators to review strategic plan

for TropSoils Project, May 29-30, 1986.

Research planning meeting in Sitiung by team members and Drs. Lalit Arya

and Russel Yost of the University of Hawaii April 1986.

Training workshop on expert systems and decision support models

conducted by Drs. Russel Yost and Stephen Itago of University of Hawaii

in Bogor, August 1986.

II. Produce user-oriented decision models by adding farmer knowledge to the

knowledge base

Stephenie Ran, University of Florida, arrives in Indonesia to work with

Dr. Carol Colfer.

Dr. Carol Colfer leaves TropSoils program June 1986.

Position announcement for Dr. Carol Colfer's position May 1986.









Project Outputs

The expected project outputs by 1989 are:

an interactive decision support system that provides users with easy

access to data bases and decision models to support decision making

tasks; and

personnel trained to validate decision models and operate, maintain and

upgrade decision support systems.



Work Plan by Objectives/Schedules

I. Integrade TropSoils research into national agriculture development plan

Distribute strategic plan to Management Entity, Indonesian Agency and

USAID, April 1, 1986.

Present strategic plan to External Evaluation Panel and Management

Entity, April 21-22, 1986.

Meet with Indonesian and USAID administrators to review strategic plan

for TropSoils Project, May 29-30, 1986.

Research planning meeting in Sitiung by team members and Drs. Lalit Arya

and Russel Yost of the University of Hawaii April 1986.

Training workshop on expert systems and decision support models

conducted by Drs. Russel Yost and Stephen Itago of University of Hawaii

in Bogor, August 1986.

II. Produce user-oriented decision models by adding farmer knowledge to the

knowledge base

Stephenie Ran, University of Florida, arrives in Indonesia to work with

Dr. Carol Colfer.

Dr. Carol Colfer leaves TropSoils program June 1986.

Position announcement for Dr. Carol Colfer's position May 1986.









Farming systems specialist hired September 1986. Arrives in Hawaii to

learn Bahasa Indonesian and expert systems.

Individual arrives in Indonesia December 1986 to continue Carol Colfer's

work.

III. Render experimental results directly useful for decision making

Dr. Upendra Singh travels to Sitiung February 1986 to install maize and

rice modeling experiment with Dr. Putu.

Drs. Ron Guyton and Lalit Arya arrive in Indonesia.

An Indonesian comes to Hawaii for training in weather station

maintenance in preparation for installation of several new weather

station by the Center for Soil Research in June 1986.

Dr. Singh travels to Sitiung to install phosphorus-lime modeling

experiment September 1986.

IV. Increase research efficiency

Dr. Gordon Tsuji and Mr. Clement Chan travel to Bogor to demonstrate use

of prototype decision support system August 1986.

Indonesian scientists provided with manual on the minimum data set to

collect from field experiments. (July 1986)

Two Indonesian scientists invited to rice modeling workshop in Malaysia

June 22-30, 1986.

One Indonesian scientists invited to participate in groundnut modeling

workshop scheduled for February 1987 in ICRISAT.

Soil physics and crop modeling experiments installed with first heavy

rains of monsoon September-November 1986.

Minimum data set for decision support system collected from all

TropSoils experiments starting June 1986.






10


Figure 1. The fundamental responsibility of the TropSoils program is to
provide its clients with knowledge they can use to make responsible choices
that results in desired outcomes.





Figure 2.


I I

TROPSOILS SOIL PROFILE GENETIC CROP-SPECIFIC
EXPERIMENTAL DESCRIPTION COEFFICIENTS MODEL
DATA DATA COEFFICIENTS


PROGRAM TO
EXTRACT MODEL
INPUT DATA


FILE-1 FILE-2 FILE-3 FILE-4 FILFILE-ILE-6 FILE-7 FILE-8 FILE-9





..-J
-- E J, 1 J. j,



EXPERT SYSTEMS
CROP MODEL -



IMULATED WEATHER AND
IP VARIABLE SOIL WATER SUMMARY
VS TIME SUMMARY OUTPUTS
(OUT 2) (OUT 3) (OUT 1)



PROGRAM FOR PROGRAM FOR
GRAPHICAL DATA ANALYSIS
OUTPUT ,



USERS


Schematic of the linkage between crop models and the TropSoils experiments, weather,
and soil data and the programs for providing graphical outputs and data analysis
results. The dashed line encompasses the standard data files and crop model.








ATTACHMENT



Brief descriptions of each of the input and output files and variable

descriptions for each file are presented in this attachment. Source code, in

FORTRAN for the IBM-PC, is available on request for reading each file and for

formating and writing each output file.

I. Structure for Model Input Data Files

A. FILE-1: Daily Weather Data

1) Description:

Daily weather data must be available in FILE-1 for all days of the

growing season at a minimum, starting with planting and ending

with crop maturity. The file should contain more weather data

both before planting and after crop maturity. Then, the

simulation would start before planting so that the soil processes

would be simulated. Initial conditions for the soil should

coincide with the first day of simulation which could occur before

planting. Additional weather data would also allow users to

select alternate planting dates for model sensitivity analysis.

Eight variables, listed below, are recorded on each line of the

weather data file. The institute ID and weather station ID are

necessary to relate specific experiment sites to the weather data

from a particular station. The variables and format for each line

of data for FILE-1 are given below.









2) Format for each line of weather data:

VARIABLE FORTRAN
NAME FORMAT DESCRIPTION
INST ID A2 code for institute ID
STAT ID A2 code for weather station ID
IYR 13 last two digits of the year
JUL 14 day of year
SOLRAD F6.2 daily values of solar radiation, MJ/m2
XTMAX F5.1 daily value of maximum temperature,
centigrade
XTMIN F5.11 daily value of minimum temperature,
centigrade
XRAIN F6.1 daily value of rainfall, mm/d



3) Example:


111111111122222222223333 <-- Column Numbers
123456789012345678901234567890123

IBWA 83 340 17.24 29.5 18.0 0.0
IBWA 83 341 14.56 30.5 17.5 0.0
IBWA 83 342 14.73 29.0 17.5 0.0
IBWA 83 343 14.73 31.0 17.0 0.0
IBWA 83 344 14.73 29.0 17.0 0.0
IBWA 83 345 14.73 20.5 17.5 10.0
IBWA 83 346 14.73 27.0 15.5 0.0
IBWA 83 347 5.27 27.5 17.5 0.0
IBWA 83 348 13.43 22.5 13.5 0.0
IBWA 83 349 10.04 27.0 12.5 0.0
IBWA 83 350 13.51 27.5 12.5 0.0
IBWA 83 351 13.51 25.0 13.0 0.0
IBWA 83 352 13.51 26.5 12.0 0.0
IBWA 83 353 17.15 25.0 13.0 10.0
IBWA 83 354 15.15 26.0 12.0 0.0
IBWA 83 355 12.55 27.5 11.5 0.0
IBWA 83 356 16.57 27.5 14.0 17.0
IBWA 83 357 11.34 25.0 20.0 8.0
IBWA 83 358 11.34 28.0 18.0 1.0
IBWA 83 359 11.34 27.5 18.0 0.0
IBWA 83 360 11.34 28.0 17.5 0.0









B. FILE-2: Soil Profile Properties

1) Description:

Soil profile characteristics recorded in FILE-2 are listed below.

These properties are used in soil water, nitrogen and root growth

sections of the crop models. One line of data is used for each

layer in the profile and the number of lines in this file

determine the number of layers to use in the crop model. The

institute ID and site ID record the location which may ultimately

be used for several experiments. Note that layer thickness is

also specified in this file. These number of layers in this file

and the thickness of each must be consistent with the initial

conditions in FILE-5. Model developers can use this format and

manually input their own values for a soil. In the decision

support system under development, a program will estimate these

values from soil pedon data available from the Center for Soil

Research.


2) Format for

VARIABLE
NAME
INST ID
SITE ID
DLAYR(L)
LL(L)

DUL(L)

SAT(L)

WR(L)


BD(L)

OC(L)


each line

FORTRAN
FORMAT
A2
A2
F6.0
F7.3

F7.3

F7.3

F7.3



F6.2

F6.2


of soil profile data:


DESCRIPTION
code for institute ID
code for site ID
thickness of soil layer L, cm
lower limit of plant-extractable soil
water for soil layer L, cm/cm
drained upper limit soil water content
for soil layer L, cm/cm
saturated water content for layer L,
cm/cm
weighting factor for soil depth L to
determine new root growth distribution,
unitless
moist bulk density of soil in layer L,
g/cm3
organic carbon concentration in layer L, %









3) Example:


11111111112222222222333333333344444444445
12345678901234567890123456789012345678901234567890

IBWA 10 0.220 0.350 0.550 1.000 1.00 2.27
IBWA 20 0.240 0.350 0.550 0.800 1.05 1.10
IBWA 20 0.250 0.370 0.480 0.400 1.17 1.41
IBWA 20 0.260 0.380 0.460 0.200 1.22 0.59
IBWA 20 0.250 0.380 0.460 0.050 1.22 0.36
IBWA 20 0.260 0.400 0.480 0.020 1.17 0.27


<-- Column
Number


C. FILE-3: Soil and Site Characteristics

1) Description:

These parameters are additional soil and site characteristics that

must be specified for simulating each crop. Users must input

values for sites used in model development. The decision support

system will create this file to allow users to simulate crop

growth and yield for their sites where soil pedon data are

available using a combination of profile data and weather, genetic

and management data.


2) Format

VARIABLE
NAME
INST ID
SITE ID
NLAYR
SALB
U

SWCON

CN2


FORTRAN
FORMAT
A2
A2
13
F7.2
F7.2

F7.2

F7.2


DESCRIPTION
code for institute ID
code for site ID
number of layers in soil
bare soil albedo, unitless
upper limit of stage 1 soil
evaporation, mm
soil water drainage constant,
fraction drained per day
curve number used to calculate
daily runoff









VARIABLE FORTRAN
NAME FORMAT DESCRIPTION
TAV F6.1 annual average ambient temperature, C
AMP F6.1 annual amplitude in mean monthly
temperature, C
DMOD 13 zero-to-unity factor which reduces the
rate constant for mineralization of the
humus pool for soils which are poor
mineralizers due to chemical or physical
protection of the organic matter default
value = 1
SITE A10 site name (as described in the MDS files)
PEDON A12 Reference Pedon Number
TAXON A70 Soil family according to USDA/SCS
Soil Taxonomy
SWCON12 F10.3 coefficient in the steady-state
solution to the radial flow, root uptake
equation, cm3/cm root-day
(default = 0.00267)
SWCON22 F7.1 coefficient in the steady-state
solution to the radial flow, root uptake
equation (default = 58)
SWCON32 F8.2 coefficient in the steady-state
solution to the radial flow, root uptake
equation (default = 6.68)
RWUMX F5.2 maximum daily root water uptake per unit
root length, cm3/cm root-day
(default = 0.03)



3) Examples

Column
Numbers--> 111111111122222222223333333333444444444455555555556666666666
123456789012345678901234567890123456789012345678901234567890123456789

IBWA 0.14 5.00 0.60 60.00 22.0 7.0 1WAIPIO-F 77HA-7-1


1111111111111111111111111111111111111111
7777777777888888888899999999990000000000011111111122222222223333333333
0123456789012345678901234567890123456789012345678901234567890123456789

Tropeptic Eutrustox, clayey, kaolinitic, isohyperthermic



2 The equation for radial flow, as given below, is described by Ritchie (198 ):

Uptake = SWCON1 EXP(SWCON2(SW(L)-LL(L)))(SWCON3-ALOG(RLV(L)))









------------------------------
111111111111111111111111111111
444444444455555555556666666666
012345678901234567890123456789

0.00267 58 6.68 0.03



D. FILE-4: Soil Nitrogen Balance Parameters

1) Description:

These are site-specific parameters required by all crop models that

use the nitrogen dynamics component. For this file, the parameters

may also depend on the experiment, and thus also include EXPT-NO

from the data sets. For models that do not use the nitrogen

components, e.g. nitrogen fixing crops or fully fertilizer plots,

this file is not needed and can be ignored. A program will be

written to obtain these parameters from a combination of profile

data, other data sets, and user input.


2) Format

VARIABLE
NAME
INST ID
SITE ID
EXPT NO
TRT NO
STRAW

SDEP
SCN


ROOT


FORTRAN
FORMAT
A2
A2
12
12
F6.0

F6.0
F6.0

F6.0


F6.0


DESCRIPTION
code for institute ID
code for site ID
experiment number
treatment number
weight of organic residue of previous crop
and/or added green manure, kg/ha
depth of surface residue incorporation, cm
C:N ratio of surface residue of previous
crop, kg C/kg N
dry weight of root residue of previous
crop, kg/ha
root residue C:N ratio (kg C/kg N)


RCN









3) Example:


111111111122222222223333333 <-- Column Numbers
123456789012345678901234567890123456

IBWA0101 500 30 80 300 40



E. FILE-5: Soil Profile Initial Conditions

1) Description:

FILE-5 contains initial conditions for soil profile water and

nitrogen dynamics submodels. These initial conditions specify the

values of water content, ammonium, nitrate, and pH in each vertical

layer at the start of the first day of the simulation. Thus, the

simulation must be started on the day for which the initial

conditions are specified, even if the planting date is later. Soil

profile initial conditions must be specified for a date before

planting, or at the latest, on the date of planting which is input

in FILE-8. The thickness of each layer and the number of layers in

this file must correspond exactly with those in FILE-2. There will

be one line of data for each soil layer. In addition, this file is

specific to each treatment of an experiment at a site and is

identified accordingly on each line of data.


2) Format for

VARIABLE
NAME
INST ID
SITE ID
EXPT NO
TRT NO
SW(L)
NH4(L)


each line of soil profile data:

FORTRAN
FORMAT DESCRIPTION
A2 code for institute ID
A2 code for site ID
12 experiment number
12 treatment number
F10.3 soil water content of layer L, cm/cm
F5.1 soil ammonium in layer L,
mg elemental N/kg soil










DESCRIPTION
soil nitrate in layer L,
mg elemental N kg/soil
pH of soil in layer L in a 1:1
soil:water slurry


3) Example:


1111111111222222222233333
1234567890123456789012345678901234

IBWA 105 0.260 2.8 2.8 5.2
IBWA 105 0.300 2.0 2.0 5.0
IBWA 105 0.370 1.4 1.4 4.8
IBWA 105 0.320 1.0 1.0 4.8
IBWA 105 0.290 0.5 0.5 4.5
IBWA 105 0.320 0.5 0.5 4.5


<-- Column Numbers


F. FILE-6: Irrigation Management Data

1) Description:

For each experiment and treatment at a site, the dates and depths

of irrigation are contained in FILE-6. These data are to be

retrieved from the data bank in the decision support system, but

this file can also be created manually during model development.

One line of data is required for each irrigation event.


2) Format:

VARIABLE
NAME
INST ID
SITE ID
EXPT NO
TRT NO
JDLAPL(J)
AMT(J)


FORTRAN
FORMAT
A2
A2
12
12
14
F4.0


DESCRIPTION
code for institute ID
code for site ID
experiment number
treatment number
day of year of irrigation
amount of irrigation added on JDLAPL(J), mm


VARIABLE
NAME
N03(L)


PH(L)


FORTRAN
FORMAT
F5.1


F5.1









3) Example:


1111111
1234567890123456


IBWA
IBWA
IBWA
IBWA
IBWA
IBWA
IBWA
IBWA
IBWA
IBWA


<-- Column Numbers


105
105
105
105
105
105
105
105
105
105


G. FILE-7: Fertilizer Management Data

1) Description:

For each fertilizer application, one line of data with the eight

variables listed below must be supplied in FILE-7. Since

fertilizer applications may vary among treatments, a separate file

will be required for each treatment of each experiment.



2) Format for each nitrogen fertilizer application:


FORTRAN
FORMAT
A2
A2
12
12
14

F6.1

F6.1

13


DESCRIPTION
code for institute ID
code for site ID
experiment number
treatment number
day of year of nitrogen fertilizer
application J
amount of fertilizer nitrogen added on
JFDAY(J), kg N/ha
depth of incorporation of fertilizer
application on Julian day (JFDAY), cm
code number for type of fertilizer as
specified in IBSNAT Technical Report No. 1


VARIABLE
NAME
INST ID
SITE ID
EXPT NO
TRT NO
JFDAY(J)

AFERT(J)

DFERT(J)

IFTYPE(J)









3) Example


111111111122222222
123456789012345678901234567

IBWA 105 333 17.0 15.0 5
IBWA 105 6 17.0 15.0 5
IBWA 105 41 17.0 15.0 5


<-- Column Numbers


H. FILE-8: Crop Management and Summary Measurement Data
(May vary by crop)

1) Description:

For each treatment of each experiment, crop management data and

experimental data may differ. FILE-8 contains crop management data

and measured field data for each treatment averaged over all

replications. The management data are inputs to the model whereas

the measured field data are needed for the standard outputs which

list simulated and measured data side-by-side. FILE-8 may vary

slightly among the different crops. For that reason, we have

listed FILE-8 format for maize, wheat, and soybean. Note that the

files are very similar, and modelers are asked to conform to these

formats as closely as possible to facilitate model linkage into the

decision support system once it is available.


2) Format:

a. Maize

VARIABLE
NAME
INST ID
SITE ID
EXPT NO
TRT NO
ISOW
ISIM


FORTRAN
FORMAT
A2
A2
12
12
14
14


DESCRIPTION
code for institute ID
code for site ID
experiment number
treatment number
sowing date, day of the year
date simulation begin








VARIABLE FORTRAN
NAME FORMAT DESCRIPTION
PLANTS F6.2 plant population, plants/m2
SDEPTH F5.1 sowing depth (cm)
LAT F7.2 latitude (deg., negative for southern
hemisphere
IIRR 12 switch describing irrigation
(default value = 1)
0 : no irrigation applied
1 : irrigation applied
99 : automatically irrigated when water
stress occurs (nitrogen version only)
ISWSWB 12 switch to indicate if water balance is used
(default value = 1)
0 : water balance is not used,
assumes adequate water
1 : water balance is used
ISWNIT 12 switch to indicate if nitrogen routines are
used (default value = 1)
0 : nitrogen subroutines are not used,
assumes adequate nitrogen
1 : nitrogen subroutines are used
XYIELD F6.0 actual field-measured grain yield at 15.5%
moisture, kg/ha
XGRWT F6.4 field-measured kernel dry weight, g/kernel
XGPSM F5.0 field-measured grain number, grains/m2
XGPE F4.0 field-measured grain number, grains/ear
XLAI F5.2 field-measured leaf area index at silking
XBIOM F6.0 field-measured above-ground dry biomass at
maturity, kg/ha
XSTRAW F7.1 measured stover dry weight at maturity,
kg/ha
ISLKJD 14 field-measured silking date, day of year
MATJD 14 field-measured physiological maturity date,
day of year
GRPCTN F6.2 measured nitrogen concentration in grain at
maturity, %
XTOTNP F6.1 measured crop nitrogen content at maturity,
kg/ha
XAPTNP F6.1 measured stover nitrogen content at
maturity, kg/ha
XGNUP F6.1 measured grain nitrogen content at
maturity, kg/ha









b. Wheat

VARIABLE
NAME
INST ID
SITE ID
EXPT NO
TRT NO
ISOW
ISIM
PLANTS
SDEPTH
LAT


PHINT
IIRR








ISWSWB






ISWNIT






XYIELD


XGRWT

XGPSM

XGPE


XLAI


FORTRAN
FORMAT
A2
A2
12
12
14
14
F6.2
F5.1
F7.2

12
12








12






12






F6.0


F6.4

F5.0

F4.0


F5.2


DESCRIPTION
code for institute ID
code for site ID
experiment number
treatment number
sowing date, day of the year
date simulation begin
plant population, plants/m2
sowing depth, cm
latitude (deg., negative for southern
hemisphere)


switch describing irrigation
(default value = 1)
0 : no irrigation applied
1 : irrigation applied
99 : automatically irrigated when water
stress occurs
switch to indicate if water balance
is used (default value = 1)
0 : water balance is not used,
assumes adequate water
1 : water balance is used
switch to indicate if nitrogen routines
are used (default value = 1)
0 : nitrogen subroutines are not used,
assumes adequate nitrogen
1 : nitrogen subroutines are used
actual field-measured grain yield
at 15.5% moisture, kg/ha
field-measured kernel dry weight,
g/kernel
field-measured grain number,
grain/m2
field-measured grain number,
grains/ear
field-measured leaf area index at
silking









VARIABLE FORTRAN
NAME FORMAT DESCRIPTION
XBIOM F6.0 field-measured above-ground dry
biomass at maturity, kg/ha
XSTRAW F7.1 measured stover dry weight at
maturity kg/ha
ISLKJD 14 field-measured silking date, day of year
MATJD 14 field-measured physiological maturity
date, day of year
GRPCTN F6.2 measured nitrogen concentration in grain
at maturity, %
XTOTNP F6.1 measured crop nitrogen content at
maturity, kg/ha
XAPTNP F6.1 measured stover nitrogen content at
maturity, kg/ha
XGNUP F6.1 measured grain nitrogen content at
maturity, kg/ha

c. Soybean

VARIABLE FORTRAN
NAME FORMAT DESCRIPTION
INST ID A2 code for institute ID
SITE ID A2 code for site ID
EXPT NO 12 experiment number
TRT NO 12 treatment number
IPLT 14 planting date entered as julian date
and converted to days from beginning of
simulation
ISIM 14 date simulation begin
PLTPOP F6.2 plant population, plants/m2
SDEPTH F5.1 sowing depth, cm
XLAT F7.2 latitude (deg., negative for southern
hemisphere)
XLONG F7.2 longitude
ROWSPC F6.1 spacing between adjacent rows of
soybeans, cm
IIRR 12 switch describing irrigation
(default value = 1)
0 : no irrigation applied
1 : irrigation applied
99 : automatically irrigated when
water stress occurs








VARIABLE FORTRAN
NAME FORMAT DESCRIPTION
ISWSWB 12 switch to indicate if water balance is used
(default value = 1)
0 : water balance is not used, assume
adequate water
1 : water balance is used
ISWNIT 12 switch to indicate if nitrogen routines are
used (default value = 1)
0 : nitrogen subroutines are not used,
assumes adequate nitrogen
1 : nitrogen subroutines are used
XYIELD F6.0 actual field-measured grain yield
at 15.5% moisture, kg/ha
XSDWT F6.4 field-measured seed dry weight, g/seed
XSDSM F5.0 field-measured seed number, seed/m2
XSPP F4.0 field-measured seed per pod, seed/pod
XLAIR4 F5.2 leaf area index of all leaves
at R4 stage
XBIOM F6.0 field-measured above-ground dry biomass
at maturity, kg/ha
XSTALK F7.1 measured stem dry weight at maturity,
kg/ha
IFLRJD 14 field-measured flowering date (Rl
stage), day of year
MATJD 14 field-measured physiological maturity
date (R7 stage), day of year



3) Example for CERES-MAIZE Model


Column -> 111111111122222222223333333333444444444455555555
Numbers 123456789012345678901234567890123456789012345678901234567

IBWA 105 334 334 5.79 5.00 21.25 1 1 1 5695 0.220 2351

Continued from previous line:


1111111111
556666666666777777777788888888889999999999000000000011
890123456789012345678901234567890123456789012345678901

400 4.30 11016 6769. 48 104 1.00 68.4 20.2 48.2








I. FILE-9: Genetic Coefficient Data

1) Description:

FILE-9 contains crop genetic coefficients which vary considerably

among crops. For a given crop model, this file may have

coefficients for any number of varieties. These coefficients are

now supplied by the modelers but the TropSoils program will

initiate training of host country scientists to enable them to

measure genetic coefficients from field experiments. Users should

consult crop model documentation to obtain more detailed

information on these coefficients. Modelers should store genetic

coefficients in FILE-9 and when a new model is integrated into the

DSSAT, a unique crop identifier will be used to distinguish genetic

coefficients for the new crop from those for existing crops.


2) Format for

a. Maize

VARIABLE
NAME
IVARTY

VARTY
PI



P2
P5


G2
G3


each variety in the file:


FORTRAN
FORMAT
14

A16
F7.2



F7.4
F7.2


F7.2
F7.3


DESCRIPTION
number assigned to a cultivar in the
genetics file
variety name
growing degree days (using 8*C as the base
temperature) from seedling emergency, to
the end of the juvenile phase
photoperiod sensitivity coefficient, 1/hr
cumulative growing degree days (using 8C
as the base temperature) from silking to
physiological maturity
maximum kernel number, kernels/plant
potential kernel growth rate, mg/kernel-d








b. Wheat

VARIABLE
NAME
IVARTY
VARTY
P1V
P1D


c. Soybean

VARIABLE
NAME
IVAR

VRNAME


IVRGRP
VARN1


VARNO



VARTH




THRVAR(1)


THRVAR(2)


VARFRC

VRFRC1


FORTRAN
FORMAT
14

A16


12
F5.1


F5.1



F5.1




F6.2


F6.2


F7.3

F7.3


DESCRIPTION
number for variety for which parameters
are being input
character name for the variety must be
16 characters or less
maturity group for the variety
critical night length for the variety,
below which accumulation of night time
accumulator proceeds at minimum rate
critical night length for the variety,
above which accumulation of night time
accumulator proceeds at maximum rate
physiological days required for floral
induction (or to accumulate one unit of the
night time accumulator for the variety, if
night length is below VARN1)
variety specific threshold for the
accumulator which determines development
from Rl to R7
variety specific threshold for accumulator
which determines development R7 to R8
fraction of development from R-1 to R-7
which must be completed before NDSET occurs
fraction of development from R-1 to R-7
which must be completed before R-4 occurs


DESCRIPTION


variety name


FORTRAN
FORMAT
14
A16
F8.6
F8.6


F7.1
F6.4
F8.5
F7.5
F7.5
F7.5








VARIABLE FORTRAN
NAME FORMAT DESCRIPTION
VRFRC2 F7.3 fraction of development from R-1 to R-7
which must be completed before NDLEAF
VRFRC3 F7.3 fraction of development from R-l to R-7
which must be completed before NPODO occurs
for the variety (begin pod fill)
SHVAR F6.2 maximum growth rate of an individual pod of
the variety, if temperature is optimal,
mg/shell/day
SDVAR F5.1 maximum growth rate for an individual seed
of the variety, if temperature is optimal,
mg/shel /day
PODVAR F6.1 maximum rate of pod addition for the
variety, number/day
TRIFOL F7.3 number of trifoliolates per physiological
day for the variety
SIZELF F7.2 the size of a normal upper node leaf
(nodes 8 10) of the variety soybeans,
CM2/leaf
SLAVAR F7.2 specific leaf area of new growth during the
vegetative phase after V5 and prior to the
time leaves start to thicken as leaf
expansion slows down, variety specific,
cm /g



3) Example for the CERES-MAIZE Model:

Column
1111111111222222222233333333334444444444555555 --> Numbers
1234567890123456789012345678901234567890123456789012345

42PIO X 304C 390.00 0.5200 860.00 720.00 7.200



J. File-O: Observed Data For Graphics

1) Description:

FILE-0 allows observed data to be plotted with simulated results

for each treatment simulated. One file exists for each treatment.

Replication data for each treatment can be included by using a

different line of data for each replication. These files may vary

slightly among crops. For example, V-stage in soybean replaces









leaf number in corn, but these variables are similar in concept.

In the DSSAT, this file will be created automatically for each

experiment by retrieving the data from the data bank and creating a

FILE-O for each treatment in the experiment.



2) Format for each line of data:

a. Maize


FORTRAN
FORMAT
A2
A2
12
12
14
13
F5.2
F5.1
F7.1
F7.1
F7.1
F6.1


DESCRIPTION
code for institute ID
code for site ID
experiment number
treatment number
julian date
leaf number
leaf area index
root weight, kg/ha
stem weight, kg/ha
grain dry wt, kg/ha
leaf dry wt, kg/ha
biomass, kg/ha


b. Soybean


FORTRAN
FORMAT
A2
A2
12
12
F4.0
F5.1
F5.2
F7.1
F7.1


DESCRIPTION
code for institute ID
code for site ID
experiment number
treatment number
julian date
V-stage
leaf area index
pod number per m2
stem dry wt, kg/ha


VARIABLE
NAME
INST ID
SITE ID
EXPT NO
TRT NO
JULDATE
LEAFNO
LAI
ROOTWT
STEMWT
GRDRYWT
LFDRYWT
XBIOMS


VARIABLE
NAME
INST ID
SITE ID
EXPT NO
TRT NO
JULDATE
STAGE
LAI
PODNUM
STDRYWT








VARIABLE FORTRAN
NAME FORMAT DESCRIPTION
SDDRYWT F7.1 seed dry wt, kg/ha
LFDRYWT F7.1 leaf dry wt, kg/ha
CNDRYWT F8.1 canopy dry wt kg/ha


3) Example for the Maize Model:

1111111111222222222233333333334444444444555
1234567890123456789012345678901234567890123456787901

IBWA 105 33 -9 0.5 -9 -9 -9 3.6 21
IBWA 105 71 -9 3.54 -9 -9 -9 39.9 420
IBWA 105 98 -9 1.74 -9 -9 -9 44.2 700
IBWA 105 105 -9 1.84 -9 -9 -9 50.4 800
IBWA 105 118 -9 1.00 -9 -9 -9 36.4 670









Table 1. Description of Standard Input and Output files
for the TropSoils crop models.


File Name
Input Files:
FILE 1
FILE 2
FILE 3
FILE 4
FILE 5
FILE 6
FILE 7
FILE 8 (by
FILE 9 (by
FILE 0 (by


Description


Daily Weather Data
Soil Profile Properties
Soil and Site Characteristics
Soil Nitrogen Dynamics Properties
Soil Profile Initial Conditions
Irrigation Management Data
Nitrogen Fertilizer Management Data
Crop Management and Summary Measurement Data
Genetic Coefficients
Observed Seasonal Data for Graphics


crop)
crop)
crop)


Output Files:
OUT-1 (by crop)





OUT-2 (by crop)
OUT-3


Output Record of Crop Model Inputs
Simulated Biomass and Water Balance
Components at Selected Phenological Stages
Harvest Summary (Simulated and Observed)
Simulated Crop Variables vs. Time
Weather Variables and Simulated Soil Water
Balance vs. Time




University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs