Agricultural Information Technology in Asia and Oceania 1998
© 1998 by The Asian Federation for Information Technology in Agriculture
Decision Support System for Management of Upland
Farming with Special Consideration on Soil Conservation
Setyo Pertiwi*1), Toshio Konaka2) and Masayuki Koike2)
1) Faculty of Agricultural Technology, Bogor Agricultural University PO Box 220, Bogor, Indonesia
2) Institute of Agricultural and Forest Engineering, University of Tsukuba 1-1-1 Tennodai, Tsukuba, Japan
Abstract
This paper describes the development of a Decision
Support System (DSS) for management of upland
farming with special consideration on soil
conservation. It is intended for use by farm advisors
and others when consulting with a farmer on adoption of
conservation cropping system.
After having user input data of certain field and
farm practice, the DSS examines rules and databases and
then estimates the potential annual soil loss, in terms of
soil water erosion. When predicted soil loss is beyond the
tolerable soil loss the DSS will find some better
alternatives of farm practice for conserving soil with its
financial consequences. Furthermore, the DSS will also
suggest the most suitable soil tillage machinery for each
derived alternative.
The computer program was developed by using
Visual Basic programming language under Windows 95
environment. The databases were organized by using
Microsoft Access. Sylvanmaps, an OCX control, was
also installed to the computer system to enable data
access through maps visualization and operations as in
the Geographical Information System.
At present, the DSS is loaded with databases suitable
for Japan condition. The program test with some study
cases provided a reasonable results.
Introduction
At a time when agricultural efforts are focused on
increasing food production, soil degradation worldwide
is increasing. Soil erosion is one of the most serious
environmental problems in the world today, because it
seriously threatens agriculture and the natural
environment (Pimentel, 1993). It is noted that 84% of
the soil degradation in the world is due to soil erosion
(56% for water erosion and 28% for wind erosion),
leaving chemical deterioration (12%) and 4 % of
physical deterioration (UNEP, 1992 in Takase, 1995).
Soil erosion in agricultural land causes the
*
To whom correspondence should be addressed. Fax.: +62-251-623203;
E-mail: ftetaipb@indo.net.id
- 73 -
73-73
deterioration in the quality of land that brings about
decreased productivity and increased expenditure on
fertilizers to maintain fertility. In extreme cases yield
become so poor so that land has to be taken out of
cultivation. Besides it, the effects of soil erosion are felt
also in the areas down valley or downwind where the
ground is covered with sand and silt deposits, ditches
and canals are clogged with sediment and reservoirs silt
up. The siltation of reservoirs and rivers reduces their
capacity, furthermore creates flood hazard, and the
sediment is a major pollutant that lowering water
quality.
Many factors control the working of the soil erosion
system. A considerable number of researches on soil
conservation against water erosion have been carried out.
However the researches usually investigate only a partial
factor from which a practical implication for soil
conservation is difficult to be derived. The integration of
such agricultural research results into useful software-
driven systems has potential to aid managers of
agricultural resources. The objective of this research is to
develop a Decision Support System with which farm
management plan, especially in upland farming can be
suggested with special consideration on soil
conservation.
Theoretical Background
Wischmeier and Smith (1978) proposed a Universal
Soil Loss Equation (USLE) for estimating sheet and rill
erosion losses from cultivated fields. The equation is :
A = R x K x LS x C x P
where
A : the computed annual soil loss per unit area, t/ha
R : the rainfall and runoff factor, j/ha
K : the soil erodibility factor, t/j
LS : the slope length and steepness factor,
dimensionless
C : the cropping management factor, dimensionless
P : the erosion control supporting practice factor,
dimensionless.
Rainfall and runoff factor is equal to the mean annual
erosivity factor (EI30) divided by 100, while K is defined
as mean annual soil loss per unit of rainfall erosivity
from a standard condition of bare fallow on a 9% slope
of 22.1 m length with no conservation practice. K value
 Pertiwi, et. al.
Start
%silt+very fine sand, % sand,
rain fall;
% organic matter, soil structure
temperature
soil permeability
R K
Potential Annual Soil Loss (Ea)
∨
no
Ea > tolerable limit
Determine conservation practice Change crop
Economic analysis
yes (1)
find other alternative
no
Figure 1. Flowchart of the computer program
is influenced by physical properties of the soil, i.e. silt,
sand and clay fractions in the soil, soil structure and soil
permeability. A nomograf has been developed for getting
the K value from a certain soil properties. If the silt
fraction does not exceed 70% the K value can also
quantitatively predicted by the following equation.
K = (2.1 x 10-6) (12- Om) M1.14 + 0.0325 (S – 2)
+ 0.025 (P – 3),
where Om is percent organic matter, M is the product
of %-silt times the quantity of %-sand plus %-silt, S is
the soil structure index, and P is the profile-permeability
class (Wischmeier, 1978).
The factor of slope length (L) and slope steepness (S)
are combined in a single LS factor which is :
LS = (field slope length/22.1)m (0.065 + 0.045s
+ 0.0065 s2),
where s is the vertical fall per 100 units along the land
surface, m has a value ranging from 0.2 to 0.5, depends
on steepness of the slope.
The cropping management factor C represents the
ratio of soil loss under a given crop to that from bare soil.
Similarly, the P factor indicates the fractional amount of
erosion that occurs with the special practice, such as
contouring, strip cropping or terracing, as compared to
- 74 -
slope and Erosion control Cropping pattern,
slope length practice yield, residue &
soil management
LS P C
yes
Select crop
yes (2)
Derive conclusion
Machinery ＞ End
selection
what it would be without them.
The USLE was originally proposed for use on
cropland in the area of the United State east of the Rocky
Mountains. It has, however, been tested and used in
other section of the United States, in Europe, and in the
tropics, especially in Hawaii and West Africa. It has
also been tested for use on rangeland and in forest area.
The equation has been useful wherever tested, although
some factors have occasionally had to be modified for
effective prediction (Troeh et al, 1980).
The soil loss prediction procedure may provide farm
managers with a concise reference from which he can
investigate, for each particular situation encountered,
which specific land use and management combination
will provide the desired level of erosion control. A
number of possible alternatives are usually indicated,
mostly by affecting the C or P factor in the USLE. This
can be done for example by changing the cropping
management practice, changing the conservation support
practice, or both. L factor can be modified by
constructing terraces. Adding a certain amount of
organic matter to the soil may also considered to affect
the soil erodibility factor K. From these, the farm
managers will be able to make a choice in line with his
 Pertiwi, et. al.

Figure 2. Examples of data input dialog

desire and financial resources.
The Decision Support System
The Decision Support System for management of
upland farming is developed in the form of computer
package program. The DSS is intended for use by farm
advisors and others when consulting with a farmer on
adoption of conservation cropping system. After
having the necessary input data from the user, the DSS
will examine rules and databases and then estimates the
potential annual soil loss, in terms of soil water erosion.
When predicted soil loss is beyond the tolerable soil loss
the DSS will find some better alternatives of farm
practice for conserving soil with its financial

- 75 -
 Pertiwi, et. al.
consequences. The program also be able to help the
user finding the most suitable tillage machinery. Figure
1 shows the flowchart of the computer program.
Obviously, the use and treatment of a given area of
land must be determined not only by its physical
characteristics but also by such considerations as
available facilities, implements, power, and financial
means and even by the preference of the farmer, his
ability to learn and willingness to try new methods. In
other words, the treatment must fit not only the needs
and adaptability of the land but the facilities and
adaptability of the farmer as well. Considering this fact,
the program is designed to find better alternatives for
soil conservation step by step so that any consequences of
any change to the practice can be considered by the user.
The intention of the user whether to change crop or not
is also elaborated. On the final, the user has a freedom
to choose the most suitable practice to his/her situation.
Result and Discussion
The computer program was developed by using
Visual Basic programming language under Windows 95
environment for use on PC with an 80386 processor or
higher, a minimum of 16 MB RAM, and VGA or
higher-resolution screen supported by Microsoft
Windows. The databases were organized by using
Microsoft Access. It includes area database, climate
database and crops database, specifically for Japan area.
Sylvanmaps, an OCX control, was also installed to the
computer to enable data access through maps
visualization as in the Geographical Information System.
The computer program requires input data such as
farm location, percent silt plus very fine sand and
percent sand of the soil, percent organic matter, soil
structure, soil permeability, field shape in terms of slope
length and slope height, and cropping system as well as
conservation currently practiced. Figure 2 shows the
examples of input data dialog. It is assumed that
farmer’s field is an area of uniform soil properties,
cropping and support practices.
With farm location identification, the rainfall
database will be accessed to calculate R factor. However,
to keep the database size manageable, calculated R
values, one for each region, are stored in the database
instead of rainfall database. With the other input data the
necessary USLE parameters are determined by the way it
is theoretically described, and the potential annual soil
loss is calculated. It is then compared with the tolerable
soil loss value for the region. Warning message is given
to the user if the intended cropping practice may produce
excessive soil loss. In this case, the program then will
search the alternatives of cropping practice that reduce
the amount of annual soil loss.
The tolerable soil loss (soil loss tolerance) refers to
the maximum rate of soil loss at which there are still no
- 76 -
significant loss of soil productivity. Four major factors
affect the soil loss tolerance : depth of soil, type of parent
materials, relative productivity of top soil and subsoil,
and amount of previous erosion (Troeh et al, 1980). Due
to lack of data collected, no specific value of tolerable
soil loss assigned for each region. Here, the compromise
value of about 10-15 t/ha annual soil loss (MAFF,1993)
is considered to be tolerable.
Alternatives of farming practice that reduce the
amount of annual soil loss are searched step by step
forward, start from the practice currently being carried
out in the farm to the more conservative ones. For
example, from removing plant residue to leaving plant
residue in the field, from up-and-down row crop to
contour wise row crop, and so on. In the same way,
when it is intended to do, crop will also be changed only
with other crop having less soil loss ratio. No back step
will be considered at all.
Conservation practices vary greatly in effectiveness,
costs, and returns. The effectiveness of conservation
practiced can be predicted using USLE. The costs of soil
conservation are usually obvious, but the returns (profits)
are less identifiable. Some conservation practice produce
an immediate return, some lead to a delayed profit in
exchange for immediate costs, and some produce no
monetary profit but are used for non monetary reasons
like reduction in sediment damage and environmental
benefits. Due to this reason, only analysis of
conservation costs are being carried out in the financial
analysis section of the program.
The costs calculated in the program are nutrient cost,
depreciation cost and conservation practice cost.
Nutrient cost refers to the cost incurred due to nutrient
losses together with soil loss. In practical situation it
could be the amount of expenditure on fertilizers to
maintain soil fertility. Depreciation cost represents the
diminishing value of agricultural land caused by erosion
damages. This refers to Ely and Wehrwein (1964) who
stated that every acre of land is subject to the influence of
both location and productivity, and rent and value are the
result or composite of both. Conservation practice cost
represents direct cost incurred by conservation practice
such as construction cost, additional labor cost or reduce
of return due to a less intensive land use.
Machinery selection by the rule of thumb is limited to
the selection of tractor and implement type. Factors to be
considered are including slope steepness, soil and
residue management, crop to be planted and land
acreage. The suggestion can be a hand tractor with rotary,
two wheel drive riding type tractor with blade, four
wheel drive tractor with rotary and ridger, or some other
combinations.
To illustrate the use of the DSS, a case of an upland
farm in Otoyo town, Kochi prefecture was conferred.
Otoyo town is a place with a fairly high rainfall. A 10
 Pertiwi, et. al.
Figure 3. Output showing the suggested cropping practice alternatives
acre of field in 30 m length and 10% slope, are planted
with water melon every year in an up-and-down slope
row, start in May and end in September. Before, no
conservation support practice at all, but after the farmer
realizing a severe field damage, currently the field is
mulched with wheat straw. The farmer claimed that
without mulching, about 10 trips of mini (light) truck is
needed to carry the deposited soil in the down field back
to the field while no significant soil deposit seems after
mulching. Conferring the previous practice, the DSS
predicted an annual soil loss of about 49 t/ha.
Furthermore, the DSS suggested alternatives are as
shown in Figure 3. It is considered that at least two of the
values shown in Figure 3 are verified by the farmer claim.
It is also understood that mulching practice will provide
soil protection with the same effectiveness, regardless of
the crops to be cultivated.
Compared with Otoyo town farm case, the potato
farmer in Memuro town, Hokkaido has less soil erosion
problem. Although the field is considerably long, but
because of gentle steepness and low annual rainfall the
predicted annual soil less is under the tolerable soil loss.
In fact, no sign of field damage in the site.
Despite the progress that has been made on the DSS
development as described above, to make the DSS
available for public use, many works still need to be done.
- 77 -
For example, soil loss tolerance needs to be determined
specifically for each region, crop data base needs to be
enhanced with various crops. Besides it, economic
analysis needs to be carried out more intensively,
including analysis of externality.
Conclusions
The Decision Support System (DSS) has been
developed to assist farm managers making plan of their
conservative cropping system on upland farming. It can
be used first, to predict annual soil water erosion risk, for
both the existing situation as well as for the projected
effects of proposed land use practices, and next to find
the more conservative cropping practice with its
financial consequences.
Acknowledgments
The author wishes to express her sincere thank to the
Japan Society for the Promotion of Science (JSPS) for
providing a one year postdoctoral research fellowship in
Japan, from December 1996 to November 1997. This
research was funded by the Monbusho’s Grant-in -Aid
Fund for JSPS Fellows (Tokubetsu Kenkyuin Shore-hi).
 Pertiwi, et. al.

References

1. Ely R.T., and G.S. Wehrwein. 1964. Land

Economics. The University of Wisconsin Press.
2. Ministry of Agriculture, Forestry and Fisheries;
Planning Department; Japan. 1993. The Planning
Policy of Soil Improvement Activities (in Japanese).
3. Pimentel, D. 1993. World Soil Erosion and
Conservation. Cambridge University Press, Great
Britain.
4. Takase, K. 1995. Global Environment and
Agricultural Resource Management (IV) - With
Special Emphasis on Land Degradation by
Salinization and Soil Erosion -. International
Development Center of Japan.
5. Troeh, F.R., J.A. Hobbs and R.L. Donahue. 1980.
Soil and Water Conservation for Productivity and
Environmental Protection. Prentice-Hall, Inc.,
Englewood Cliffs, New Jersey 07632.
6. Wischmeier, W.H. and Smith, S.D. 1978. Predicting
rainfall erosion losses – a guide to conservation
planning. U.S. Department of Agriculture,
Agriculture Handbook No. 537.

- 78 -