Dwi Putro Tejo , Baskoro

1. International · ISSN0853-7674
Agricultural ScientificJournal
/ :)]I I ·
+ .
GAKURYOKU
Volume XII, 3, 2006 October 2006
1. The use of Fuzzy Set Functions For Assessing Land Suitability Index
as the Basis for Environmental Management (Owi Putro Tejo
Baskoro and Eduardo P. Paningbatan) .............. ....................... .. ..... 1
2. Mineral composition of Shallots Bulbs (Allium ascalonicum. L) (June
Mellawati and T.M. Nakanishi) .. .. ............... ... ......................................... 6
3. Handling of Arenga pinnata (Wurmb.) Merr. Seed to Provide High ·
Quality Seedling (Muhammad Salim Saleh, Tati Budiarti, H.M.H.
Bintoro, E. Murniati and Y. Yamamoto) ................................,........... ..... 9
4. The Influence of Sauropus androgynus (L.) Merr. Leaves on the Body
Weight Change and Calcium - Phosphorus Absorption in Lactating
Sheep (Agik Suprayogi and Udo ter Meulen) ........................................ 13
5. Morphological and Essential Oil Characterization of The Banda
Nutmeg (Myristica fragrans H'outt.) of Moluccas and North Moluccas
Ecotypes (llyas Marzuki, M.H. Bintoro Djoefrie, Sandra A. Aziz,
Herdhata Agusta, Memen Surahman and H. Ehara) .............. ......... :..... 17
6. The Effect of Green Tea and Ginger on Homogeneous, Acidity, and
Protease Activities of Skim and Whole Milks (Tatik Khusniati, Herlina
Purwiyanti, Nunuk Widyastuti and Zainal Alim Mas'ud) ................. ..... 22
7. The More Precise Medium Modification and Sterilization Level on In Vitro
Propagation of Mushroom (Syammiah, Efendi, and Laura S
MeitznerYoder) ................. ...................................................................... . 27
8. .Grazing Tolerance of Normal and Dwarf Elephantgrasses Pasture at the
First and Second Years after Establishment (Muhammad Mukhtar
and Yasuyuki Ishii) .. ................... .......................... ...... ............ .. .... .... ...... 31
Published by
The Association of Indonesian Alumni from Japan
(PERSADA)

2. Chief Editor
Prof. Dr. H. M. H. Bintoro
Editors
Prof. Dr. Yoshinori Yamamoto
(Kochi University)
Prof. Dr. Hiroshi Ehara
(Mie University)
Prof. Dr. Kopli Bujang
(University Malaysia at Serawak)
Prof. Dr. N. Kubota
(Okayama University)
Reviewers
Dr. Suwardi
(Soil Science)
Dr. Upik Kesumawati Hadi
(Veterinary Medicine)
Dr. Ari Purbayanto
.(Merine Science)
Dr. Asep Sudarman
(Animal Science)
Dr. Lilik Budi Prasetyo
(Forestry)
Dr. Erizal
(Agriculture Engeneering)
Dr. Yoni Koesmaryono
(Agroclimate)
From The Editor
The Gakuryoku Agricultural Scientific Journal is
published by the Association of Indonesian Alumny
from Japan. The articles in th is Journal are the result
of Agricultural and life Sciences. This Journal is In­
ternational Journal.
In this edition, the articles consist of soil science,
mineral composition, seedling quality, tissue culture,
and animal science.
The editor invite scientists all over the world to
. contribute and publish articles in this journal
Bogor, October 2006
Editor Address
Sekretariat PERSADA Cabang ·Bogor
JI. Raya Pajajaran, Kampus IPB Gunung Gede Bogor .
Telp. 02S1-3S6833,Fax. 316178, Telp'/Fax. 623410
E-mail: uptbsipb@indo.net.id; editor_gak@yahoo.com

3. Gakuryoku Volume X II No.3, Th. 2006
THE USE OF FUZZY SET FUNCTIONS FOR ASSESSING LAND SUITABILITY
INDEX AS THE BASIS FOR ENVIRONMENTAL MANAGEMENT
Dwi Putro Tejo Baskoro
1
and Eduardo P. Paningbatan
2
I Department of Soil Sciences and Land Resources, Faculty of Agriculture -
Bogor Agricultural University. Jl Meranti, Kampus IPB Darmaga, Bogor
Email: tejo2baskoro@Yahoo.com
2 Department of Soil Sciences UPLB, Philippines
ABSTRACT
A study is conducted to evaluate the potential use of Geographical Information System (GIS) for land
suitability assessment. PC-based GIS softwares are used in analysis, processing and mapping spatial data.
This paper presents a spatial modeling procedure for the assessment of land suitability by using Fuzzy Set
Functions in Geographical information Systems (GiS) environment. This technique involves the creation of
membership maps for every land attribute used as the basis for evaluation criteria adoptedfrom FAG. The
study shows that Fuzzy method results in aflexible land-suitability index for identifying and delineating land
management units and the suitability of the unit for specific land use. Land suitability map for rice, upland
annual crops and coconut derivedfrom Fuzzy method shows that the suitable areafor rice and other annual
crops with suitability index 0.5- 1 is about 57 % ofthe study area. The highly suitable areas for the two land
use types are especially found in level to gentle slope. Suitable area for coconut covers about 68 % of the
study area.
Key words: GIS, Fuzzy set/unction, Suitability Index, Environmental management
INTRODUCTION
A proper land use decisions is of a great impor­
tance to achieve optimum productivity of the land
and to ensure environmental sustainability. This re­
quires spatially accurate information of land re­
sources and effective management of the informa­
tion on which such decisions should be based. Land
suitability evaluation is one of the effective tools for
such purposes [I].
Land evaluation is a tool to predict land perform­
ance, both in terms of the expected benefits from·
and constraints to productive land use, as well as the
expected environmental degradation due to these
uses and can employe as a preliminary to more de­
tailed investigations [2;3]. Therefore, for land to be
suitable (for a given purpose) and for the use to be
sustainable, it must address the values that are re­
lated to both aspects: degree of suitability and poten­
tial degradation (from long-term perspective) result­
ing from land management practices. Land suitabil­
ity analysis shou Id consider' both aspects of land
evaluation, and a GIS may be used to facilitate every
stage of the evaluation processes [4]. The paper
demonstrates the use of GIS for land evaluation
processes. The main purpose is to establish indices
of land suitability for cropping land use systems A
fuzzy set methodology is employed in the modeling
procedures where the classes do not have sharply
defined boundaries [5;6].
MATERIALS AND METHOD
The area selected for the study include some wa­
tersheds in North-Eastern Laguna covering an area
of approximately 90.000 Hectares, located about 60
km, to the east of Manila-the Capital of the Philip­
pines.
In the application of fuzzy set functions, soil at­
tributes maps are used as the basis for generating
membership function of every land attributes to rep­
resent their suitability rating. Model parameters used
for generating the membership function for three
selected land use are based on FAO [7]. Once the
model parameters have been determined, a member­
ship function for each land attributes is generated. A
joint membership function (JMF) of combined land
attributes is then determined by using the minimum
operation as follows [8]:
Result (JMF) = Min ( MFa, MFb, MFc, ....).
where Mfa, MFb, MFc are memberships function of
land attributes.
Memberships functions used for fuzzy member­
ship classification consists of two basic models:
symmetric and asymmetric [9]. The symmetric mod­
els consist of two types: one using a single point as
the central concept (Type I), while the other em­
ploys a range of ideal points (Type 2). The asym­
metric model uses the lower and upper boundaries of
a class. This model also consists of two types:
asymmetric left (Type J) and asymmetric right
(Type 2). The computation of criterion membership
functions is based on the following equation:
Type I : MF(xi) = [J/(i + {(ri - b)/e/)]
Type2 : MF(xi)= I if(b+eJ) < xi«c - d2)
Type 3 : MF(xi)=[J/(J + {(xi-b-ei)/el/)]ifxi
< (b + el)
Type 4: MF(xi) = [I/(l + {(xi - c + e2)/e2}2)] ifxi
> (c - e2)
The choice of membership function to suit deci­
sion criterion depends on the trend of performance
of a land attribute in creating a favorable condition

4. Volume XII No.3, Tb. 2006 Gakuryoku
for selected land use. Model parameters used include
ideal point (b and c) where the MF is set to be 1,
zero points (a and d), and e (width of transition zone
which is a distance to cross over point-CP, the point
representing situation where a land attribute exam­
ined is at the marginal level for a given purpose).
Some examples for deriving model parameters are
illustrated at Figure 1.
In calculation joint membership function, the
land attributes are assumed to have equal weight.
The idea is that these land properties represent land
variables having an equal importance to each other.
There is no tradeoff between the land variables. For
example, very steep slope of a land area (or block)
cannot be compensated for by the excellent quality
of soil profile, or vice versa. Results of the minimum
operation are Land Suitability Index (LSls), ex­
pressed as continuous values ranging from 0 (very
poor or fully not suitable) to 1.0 (excellent or highly
suitable). These land suitability indices could then
be reclassified to generate land suitability rating map
based on the user-defined criteria.
RESULT AND DISCUSSION
An example of land suitabi Iity index of the study
area for annual crops resulted from Fuzzy Approach
is given in Figure I.
The figure shows that land su itab iIity of the
study area for upland annual crops is expressed as
continuous values ranging from 0 (fully not suitable)
to 1.0 (excellent or highly suitable). This continuous
expression provides that the fuzzy approach is more
flexible than the traditional approach which usually
results in classes with strict border (highly suitable,
moderately suitable, marginally suitable and not
suitable). The flexibility presented especially in term
of the decision to what level of suitability index the
suitability rating should be set. As the acceptance
level of the suitability index is lowered, the propor­
tion of the total area assigned as the desired suitabil­
ity class increases. This imp lies that if, for example,
the suitable area obtained for a give land use is lim­
ited and not sufficient to satisfy the basic minimum
area then the land suitability index can be justified to
increase suitable area obtained within the acceptable
range. Nevertheless, it does not necessarily mean
that contiguous regions of 'l given minimum size
larger than the area sized used for mapping can be
located anywhere. .
Assuming that the highly suitable area in fuzzy
approach corresponds to 0.8 to 1 land suitability
index value, the result also shows that the suitable
area obtained from Fuzzy Approach is much larger
than that obtained from Trad itional approach (Table
3). In Traditional approach, the intersection only
accepted sites that match all the strict requirements.
Many sites would be rejected in this manner, espe­
cially if there are many sites having a value of one or
more attributes just lower or higher than the class
boundary.
By accepting sites that are just outside the ideal
strict matching, the coverage area that can be judged
2
as highly suitable can be increased substantially
without any change in the original class limits ap­
pi ied to the observations.
The evaluation of the result obtained by fuzzy
approach is also undertaken by using cross compari­
son analysis. An analysis was made by establishing
fuzzy membership function (MF) values of land
suitability rating obtained from Traditional approach
(LST), then comparing these values, based on their
corresponding position to MF values of fuzzy ap­
proach outputs. Membership values of LST are gen­
erated as follows:
(Category, MF value) = {(I; 0.9), (2; 0.70), (3;
0.50), (4; 0.30), (5; O.I)}
There are about 500 points for this validation ex­
ercise, distributed randomly over the study area,
using stratified random sampling technique. The root
mean square error (RMSE) for the corresponding
points is then calcu lated, and the resu Its were pre­
sented in Table 4.
Based on spatial comparison analysis, it is found
that the variations of land suitability index tend to
increase from the area with high suitability classes to
medium suitability classes. This is in accordance to
land suitability classification study conducted by
Burrough et al. [10], who found that the use of Tra­
ditional-based categorical system of land suitability
analysis had resulted in the rejection of considerable
suitable areas. As the result, the land suitability in­
dex variations of the lower class increase. The fuzzy
approach used thus demonstrates its significance for
fine discriminations of land qual ity in the area of
interest. As has also been demonstrated by Burrough
et aI., [10], fuzzy classification approach allows a
user to retrieve data and classify them as normal, but
includes the advantage of user-defined criteria (with
a certain tolerance level) the class limits in the form
of the transition zone. Since the information is not
thrown away by strict classification early in the
analysis, the user can see which attributes contribute
most to the final result.
It has been demonstrated that fuzzy approach
used integrates various land attributes to establish
land suitability index for the selected crops. The
fuzzy approach utilizes membership function-fuzzy
set methodology in the assessment procedures, and
the results presented in a continuous pattern. Land
suitability index is generated by using actually three
basic land variables climate, soils and topography.
The three land variables are combined by using
minimum function assuming that they have equal
importance to avoid tradeoffs between them.
Land suitability index produced in this study
may serve as a guide to determine what areas can be
opened up for agricultural land and what land use
management practices is the most appropriate for the
areas. It is related to the importance of assessing and
identifying the spatial pattern of land suitabil ity of
the planned area for selected land use. Nevertheless,
the land suitability index resulted from Fuzzy Ap­
proach does not indicate the limiting factor acting in
the area for the selected crop as it is in land suitabil­
ity class obtained from conventional approach.

5. I Sf,: Mana 'I
.,,-,,~'i
"
121 "00' 1~ r..lJ·
Land Suitability Index
LAGUNA PROVINCE
U...~
/.L...-
114030'I .
• Milllil,l
,
Lft.GUNA DE BAY
,- ' l...
~S' )~/ ,./' ~ ~ .
L--_____-, 7") PRO. ICE ~
w
San Pablo CI~-~LEG END d!t.nr ,_
Province 80un -~..,....V--
14°00'
Main road
B Studl' A rea
N
A
Meters
10000.00
{-1. REPUBLIC OF THE
,rI I PHILIPPINES
'v.~~ LAGUNA PROVINCE
," -~O
,P~S:i
~ "~~ c::It' /" J "!-:;
,/': '1!f~:.''~"
• <>
!,.,s " ~
Figure I. Land suitability index of the study area for upland annual crops obtained from Fuzzy approach
~~
:;;:
~
c
~
:;;:
~
~
"=
~
'"
~>-;
'"
~
~;."
~
f'..,,)
C)
C)
C'

6. Volume XII No. 3, Th. 2006 Gakuryoku
MF MF
UCP
0
bLCP
1.0 1.0
.~
a
6 0
0 30 60
Drainage rating
Soil effective depth (cm)
a) MF for soil drainage b) MF for soil depth
MF
MF
b c uCPLCP c UCP
1.0
..........~ ......;.
1.0
0.5 0.5 . ·· ·
o o3.0 4.5 6.0 7.0 8.5 10.0 2 15 30
Soil pH Slope Gradient ('!oj
c) MF for soil pH d) MF for slope grad ient
Figure 2. Membership function of selected land properties: drainage (a ), soil depth (b), soil pH (c), and
slope gradient (d)
Table 3. Coverage area of land suitability rating for selected crops resulted from Boolean approach and fuzzy
approach
SUITABILITY RATING COVERAGE AREA
Rice Annual crops Coconut
Hectares Percent Hectares Percent Hectares Percent
Tradilional approach
Highly suitable 7751 8.5 65 0.1 7393 8.1
Moderately suitable 23734 26.1 10800 I1.9 15066 16.6
Marginally suitable 19378 2lJ 37444 41.2 38903 42.8
Sub 10lal 50863 55.9 48309 53 .2 61363 67.5
Currently not suitable 16567 18.2 19122 21.0 18343 20.2
Permanently not suitable 21846 24.0 21845 24.0 9569 10.5
Sub lolal 38413 42.2 40967 45.0 27912 30.7
Suilability index from Fuzzy approach
0.9 ­ 1.0 18590 20.5 10340 11.4 16665 18.3
0.8 - 0.9 16516 18.2 7830 8.6 24813 27J
0.7 - 0.8 6053 6.7 16745 18.4 8401 9.2
0.6 - 0.7 5273 5.8 6798 7.5 5575 6.1
0.5 - 0.6 5227 5.8 9357 10.3 6581 7.2
Sub lolal 51658 56.8 51070 56.2 62035 68.3
0.4 - 0.5 3706 4.1 3706 4.1 3361 3.7
OJ - 0.4 3211 3.5 3211 3.5 2845 3.1
0.2 - OJ 3067 3.4 3655 4.0 2361 2.6
0.1 - 0.2 4198 4.6 4198 4.6 8228 9.1
0.0-0.1 23434 25 .8 23434 25.8 10444 11.5
Sub lOlal 37616 41.4 38204 42.0 27240 30.0
Therefore, the combination of land suitability index
and land suitability class may be used as a good ba­
sis for identifying and defining land management
units, and serves as a spatially-based
making parameter in land use planning.
decision­
4

7. ...
Gakuryoku VO/llme XJJ No.3, TIJ. 2006
Table 4. Root mean square error (RMSE) values for land suitability index obtained from Fuzzy approach
against land suitability from Boolean Approach
BOOLEAN NUMBER AVERAGE RMSE
SUITABILITY CLASSES OF POINTS LAND SUITABILITY INDEX
Rice
I 40 0.937 0.097
2 147 0.873 0.177
3 101 0.698 0.227
4 93 0.3 12 0.149
5 I 15 0.003 0.097
___O~ve~r~al~I________________________4_9~6_______________0.53~6~_________________~0~.1~6~4_____
. Annual Crops
I
2 56 0.911 0.228
3 209 0.724 0.259
4 115 0.354 0.243
5 117 0.003 0.098
Overall 497 0.488 0.224
COCOllut
I 40 0.948 0.104
2 79 0~09 0212
3 221 0756 0.280
4 114 0.304 0.137
5 48 0.021 0.107
Overall 502 0.622 0.219
CONCLUSION
I. The use of fuzzy approach in land suitability
assessment provides more flexibility to identify
the suitable area for certain crop, depending on
the suitability index value preferred by the user
defined as suitable. The method developed
shows the significance of the fuzzy approach for
refining available land suitability evaluation sys­
tems. Land suitability map for rice, upland an­
nual crops and coconut derived from Fuzzy
method shows that suitable area for rice with
suitability index 0.5 - 1 is about 57 percent of
the study area. Suitable area for upland annual
crops covers about 56 percent. The suitable areas
for the two land use types are especially found in
level to gently slope. Suitable area for coconut
covers about 68 percent of the study area. The
dominant lim iting factors are slope steepness,
erosion hazard and soil pH.
2. The study confirms the capability of remote
sensing and GIS to integrate spatial and attribute
data and offers a quick and reliable method of
natural resources appraisal. The spatial relation­
ship between different geograph ically referenced
data can be established by using GIS. In the
study, RS has been used to derive land use map
and GIS has been used for digitization of base
map, digitization and delineation of land re­
source boundary and also for generation of soil
erosion and land suitability map. The result pre­
sented shows the potentialities and constraints of
the area with regard to its land resources and will
be a useful tool for any planning in the study
area.
REFERENCES
(J J FAO. (! 976). A Framework for ,Land Evalua­
tion. Sods Bulletin No. 32, FAO: Rome.
(2J Rossiter, D.G ., 1996. A theoretical framework
for land evaluation. Geoderma, 72: 165-190.
[3] Dent, D. and A. Young. 1981. Soil Survey and
Land Evaluation. George Allen & Unwin: Bos­
~n . .
[4] Burrough, P A, 1986. Principles ofgeographi­
cal information systems for land resources as­
sessment. ivlonograohs on soil and resources
survey no 12. Oxj~rd University Press Inc.,
New York. 194 p.
[5] Burrough, P.A., P. van Gaans, and R..Hoots­
mans. 1997. Continuous classification in soil
survey: Spatial correlation, confusion, and
boundaries. Geoderma 77: IJ5 -- 135.
[6] Zhu, A. X., B. Hudson, J. Burt, K. Lubich, and
D. Simonson. 2001. Soil mapping using GIS,
expert knowledge, and fuzzy 'logic. Soil Sci.
Soc. Am. J. 65:1463 - 1472
[7J FAO. (1985) Guidelines: Land Evaluation for
Irrigated Agriculture. Soi I Bu lIetin No.55, FAO:
Rome.
[8] Burrough, PA (1989) "Fuzzy mathematical
methods for soil survey and land evaluation".
Journal of Soil Science, 40:477-492.
[9] McBratney, A.B ., and 1.0A ODEH. 1997. Ap­
plication of fuzzy sets in soil science: fuzzy
logic, fuzzy measurements and fuzzy decisions.
Geoderma, 77:85-! 13
(IOJ Burrough, P.A., R.A. Macmillan, and W. van
DeUl·sen. 1992. Fuzzy classification methods for
determining land suitnbility from soil profile
observation and topography. Journal of Soil Sci­
ence. 43: 93-21 O.
5