TROPSOILS-INDONESIA/UNIV. OF HAWAII
Peter E. Hildebrand
May 12-26, 1985
Sunday, May 12
LV Gainesville 7:30 AM
AR Honolulu 3:30 pm
Met by Hal McArthur.
Monday, May 13
Picked-up at 8:30 by McArthur and taken to the University
of Hawaii. I met briefly with McArthur and Vicky Sigman and then
for an hour with Gordon Tsuji and Goro Uehara.
From 10:30 to noon gave a seminar on the Value of On-Farm
Research. It was attended by approximately 25 people including
Foster Cady who was there on TDY and John Malone, former Chairman
of Agricultural Economics in Penn. State. I understand that both
extension and research people were there as well as people from
several departments. Following lunch I met with a group of
people who had taken the farming systems course and were
interested in general discussions and follow-up to the seminar.
Later I met with Vicky Sigman, Russ Yost and Perry Philipp to
discuss the case study they are doing for the FSSP with the
Population Council. Finally in the afternoon I met with Ada Demb
and discussed the Soils-CRSP in general.
Tuesday/Wednesday May 14/15
Left Honolulu at 3:20 a.m. on Tuesday arriving Jakarta at
3:30 p.m. on Wednesday. Met by Carol Colfer and Richard Dudley.
Also had dinner with them. Met Russ Brannon and Kurt Anschel who
were there on a University of Kentucky Project. Anschel is on
TDY and Brannon is resident in Indonesia.
Thursday/Friday, May 16/17
The schedule was to go to Sitiung on Thursday. However,
John Thompson had been asked by Sudjadi and/or AID to schedule
time for them to see me before I left for Sitiung. Because
Thursday was an Indonesian holiday, we were not able to meet with
AID and Sudjadi until Friday. Attending at that meeting were
Alan Hurdus, Dick Cobb, and Sudjadi. Carol Colfer also attended.
They indicated they definitely wanted a debriefing on Thursday,
May 23 as originally scheduled. We discussed the priorities of
the Project to Indonesia. Cobb indicated that because of reduced
budget and personnel in AID they were in a situation of needing
to determine which of their activities were most important. He
was interested in learning if the problems that the CRSP was
working on were important and less interested in how they are
being attacked. He had no concern about the modeling aspects of
the project but was definitely interested more in research and
Sudjadi had wanted me to go to Jambi (on Sumatra) to look
at a new project there. He was interested in creating some sort
of model farm which they could use in the area that would make
the people from Java feel at home and provide them sufficient
plant and animal protein and other necessities in their diet. He
raised the general question of how can they make the farmers from
Java feel at home on Sumatra. I indicated that because of the
delay in meeting with them before going to Sumatra I felt it
would not be possible to take two days to go to Jambi. However,
Carol indicated the CRSP people on Sumatra were familiar with
Jambi and I thought I would be able to discuss the situation with
them. This satisfied him and he planned to meet us on Thursday,
May 23 at AID at 1 p.m. In general, Thursday and Friday were
utilized in discussions with Carol about the Project in general
and about the farming systems aspects in particular. It was a
useful time and provided me excellent background prior to arrival
Saturday, 18 May
Got up at 4 in the morning to catch the first flight to
Padang. We arrived at Sitiung at about 4 p.m. The Wades hosted a
Bar-B-Q dinner. Carol did not go because her throat was
bothering her and she was losing her voice. There was a
pre-Ramadan party at Carol's neighbors most of the night.
Sunday, May 19
With Carol and Herman Agus, went to Koto Padang in the
morning. This is a Manangkabau community which is a well
established village with tree farming including rubber and fruit
trees. Team members utilize it as a source of indigenous
knowledge in the area. In the early afternoon visited the
Sitiung 1 station with Mike Wade and John Thompson. Later in the
afternoon Carol hosted an open house at which all members of the
team were present. We discussed wide ranging topics including a
program for the next three days.
Monday, May 20
Traveled to Sitiung 4 and 5 with Mike Wade visiting some of
the on-farm work being undertaken. In Sitiung 5, saw Carl
Evensen who was harvesting one of his experiments. In the
afternoon I continued general discussions with Carol and working
on my summarization of the various projects. I had dinner with
Carl and Stacy Evensen and spent the night at their house because
there was another all night party at another neighbor of Carol.
Tuesday, May 21
Devoted to an all day discussion with the team at Evensen's
house and ended in the afternoon with a Seminar at Thompsons. I
was also given the all-clear to move back to Carol's house for
Wednesday, May 22
Morning was utilized to rewrite the farming systems
component of the project with Carol and in working with Mike on a
summary of all the experiments underway at the present time.
Draft copies of these documents are attached. Final versions
will be sent from Sitiung within a few days. In the afternoon,
Carol and I returned to Padang where we met with Russ Yost and
Samir El-Swaify. I had dinner with Russ and Samir that night and
breakfast with them in the morning.
Thursday, May 23
I returned to Jakarta and met in AID with Dick Cobb, Alan
Hurdus and Jim Gingerich. Sudjadi did not make the meeting
because he had some other visitor at Bogor. John Coulter did not
spend the night in Jakarta, so I was unable to meet with him as
we had formerly programmed.
Friday, May 24
In the morning I again met Brannon who was with Ed Vickery,
Director of SECID. I visited with them a short while before
leaving for the Airport. I caught the flight to Singapore and
Saturday morning early caught the flight to Tokyo and New York.
Returned to Gainesville on Sunday, May 26 at 10:07 P.M.
This is an exciting project being conducted by highly
motivated scientists. These scientists and their families are to
be highly commended for the dedication they are demonstrating.
They have integrated themselves into village life in Sitiung 1
and are apparently considered as valuable (although at times
somewhat odd) members of the village. Their story warrants wide
In overall terms, I found the project very satisfactory.
The team has been spending time trying to improve ways to convey
their focus which has been part of the criticism from previous
reviews. As will be noticed in the summary of the projects,
attached, they have divided their work into four categories:
Soil Chemistry, Soil Physics, Soil Biology, and Soil People. The
Soil People category is basically the farming systems aspect and
includes the on-farm research when that research has a heavy
component of farmer evaluation. Other on-farm research which is
researcher managed and evaluated is listed under the categories
referring to the specific nature of the project.
Other than some work that Carol is doing with "expert
systems" (with Russ Yost) the team had very little information on
the modeling aspects of the project. They are unaware of what
additional work may need to be done by them with respect to
modeling. We discussed the possibility of including all modeling
activities under a fifth category, Soil models.
The team is very much oriented toward solving soils problems
in the Sitiung area. They work closely with each other and
depend upon each other for scientific support. This makes up for
some lack of research experience in the younger members of the
team. They look to Carol for help in understanding the response
of the people and the possibilities for implementing technology.
She understands and follows through on this role.
At the present time they have several things that can
probably be moved into more complete farmer evaluation:
1) Because they have noticed significant responses to green
manure they are excited about the possibility of establishing
tree crops on the steeper slopes which would provide a source of
green manure from the cover crop. The tree crops could be fruit
or rubber (both important in Koto Padang, the Manankabau
village). However, at the present time there is a restriction on
planting trees in the area. This is some sort of bureaucratic
decree which must have been based on faulty assumptions and
obviously should be changed.
2) Although organic matter may substitute for lime,
phosphorus and potassium, they have found significant responses
to low levels of all three of these amendments. Thus, if
chemicals are needed, it may be more efficient to use low levels
than the high levels currently being recommended.
3) Some of the terracing work being done by CSR with bahia
grass on the terraces also looks promising. The grass can
provide forage for livestock and appears to be well adapted to
4) The team is planning a new emphasis on home gardens
which produce 30 to 60% of the food produced on the entire farm.
This concern to include gardens as well as upland crops appears
warranted to me.
Sub-systems for Jambi
While moving some of the above sub-systems into farmer
evaluation, they can also be considered by Sudjadi as alternative
sub-systems for the Jambi area and evaluated there by CSR with
CRSP support. In addition to providing information on these
sub-systems for Sudjadi, the team can also provide him with other
information on how to make the transmigrants "feel more at home",
as expressed by Sudjadi. As an example, Carol indicates two
simple things that could be done. One is to leave some of the
more beneficial or ornamental trees standing when the areas are
cleared. This is normally done by slash and burn farmers when
they are clearing new lands. Another suggestion is to bring not
only farmers into the area but also artisans, musicians,
merchants, and others who are required to create a whole village
and cultural situation.
The team expressed a series of felt needs. One is the need
for economic analysis. This involves not only the analysis of
the individual technologies but also, particularly Carol, feels a
need to have more information on the economic status of the
families and on potential sources of cash for investment in
alternative inputs. I personally feel that the knowledge of all
income and expenses of the family is not necessary. If farmers
say they do not have cash for a particular practice, that
satisfies me that they will not use cash for it whether or not
they have some cash available. However, Carol has some very
reasonable objectives and justification for carrying out this
type of research and I would not suggest it be deleted. I
believe the addition of a farming systems economist to the team
would be useful.
Another felt need is to study the chemistry and physics of
the organic matter response. With just organic matter, responses
are equal to those of the highest levels of inputs of lime,
phosphorus and potassium. Probably a study should be designed to
more accurately interpret any interactions between lime, P, K and
Another need expressed by the team is to understand why
these soils are predicted to be high P fixers when they are not.
Another topic of interest would be soil moisture
characterization to help understand what might be done in this
soil which has a 60% clay content but which is also prone to
short term drought stress.
Some additional work in agro-forestry was also expressed as
a need. This is in addition to the alley cropping being
conducted by Evensen. It would include fruit, spices and/or
rubber trees with cover crops for green manure.
Even after fully studying the 1984/89 Program Plan for the
Hawaii/North Carolina/Indonesia Project, I still found a great
deal of difficulty in relating it to the work being done in
Sitiung. This apparently has been a complaint by other
reviewers. The team has made an effort to create a structure
which would help those outside the team understand what they are
doing and this is attached to this report. The team, itself,
does not feel the same confusion as those of us who come from the
outside. This is not surprising nor alarming. However, what is
of concern is the nature and quality of the proposals published
in the plan. It is alarming that the team on Sumatra had
virtually no input into the project proposals. This is
particularly so since they knew as early as June, 1983 that the
proposals would need to be written for the extension proposal.
Even more alarming is the apparent lack of communication among
the US based scientists with respect to these proposals.
Another problem which is marginally satisfactory at the
present time, but which will become critical if three to six new
families move to Sitiung, is that of logistics, planning and
expediting. The Chief of Party presently spends a great deal of
time in Jakarta and as a result, Carol is doing a significant
share of the administrative duties at Sitiung. The system
functions at the present time but utilizes a significant amount
of professional time that could better be devoted to scientific
endeavors. If more families are going to be moved to Sitiung it
is critical that, as a minimum, an expediter be hired to be
stationed in Jakarta in order to minimize the number of trips now
required by the Chief of Party.
In planning for new families, housing should be developed
starting immediately and vehicles and other equipment should be
ordered for these new families immediately. I would also
recommend that radios be ordered not only for the new families
but also for all the jeeps in the project. The team has been
very conservative in their requests for equipment and support and
I think all efforts should be made to improve their efficiency as
much as reasonably possible. I find it rather odd that with the
exception of Mike Wade, no one at Sitiung has any idea how much
budget they have to work with. There apparently is plenty of
budget available to them, as they have never been refused a
request, but the total amount they have to work with is unknown.
Because the External Panel is also responsible for evaluating
general budgetary and accounting procedures (but not auditing) I
think we need to have a more detailed breakdown of the budgets.
It is also probably a requirement that the budgets indicate the
amount to be spent in the host country and the amount to be spent
at the US institutionss.
Communications with Hawaii always will be slow, but
improvements can still be made. Every effort should be made to
assure that telexes are not garbled and that the information
contained in them is correct.
The team in Sitiung has little knowledge of what the
modeling projects will involve. According to Yost, they probably
will involve very little time from the Sitiung team. In order
to be important to Sitiung, rice and peanuts should be the crops
that are modeled first rather than maize and soybeans which are
much less important in the area.
The relationship with the Yurimaguas Project is a serious
problem. Mike has not been able to obtain any materials from
Yurimaguas even though he has asked on more than one occasion
over a period of several months. I can find no excuse for the
materials available in Yurimaguas not to be available to the team
Although there is some contact with the Indonesian Crops
Program and the IADS project at Sukarama on Sumatra, there is
little scientific interchange between the two groups. This
situation should definitely be changed. I would recommend
periodic, perhaps every other month, meetings to exchange ideas
and information and a formal meeting once a year approximately at
this time of year, to discuss results and plan future projects.
This would help assure that current findings are used in the
following year's work and should provide a means for exchanging
materials and other technologies. I mentioned the same point to
AID because both of the projects are AID sponsored.
In addition to those recommendations made in the preceding
paragraph or to emphasize those which were made, the following
recommendations are given special emphasis.
1) Thesis or dissertation research being conducted at the
Sitiung site must have the capacity for flexibility in order to
allow the researchers to adjust to the conditions and the needs
found in the course of their work. Because the team in Sitiung
works together very closely, they produce a synergistic effect
with respect to interpretation of both formal and informal
results. Obviously, this requires constant communication with
the home campus and major professors. But I do encourage the
major professors and principal investigators to respond to
requests for modifications from the team in the field. This has
not been indicated as a particular problem so far, but it could
arise in cases when major professors are not fully and currently
advised of the nature of the work on site.
2) I would encourage collaboration with other CRSPs. The
Management Office could perhaps facilitate exchange of materials,
for example, with the peanut and the bean/cowpea CRSPs. The team
should also make an attempt to obtain information from Malasia on
the work done there on trees and cover crops.
3) Requests from the field for materials, supplies or
equipment should be given top priority by the home campuses. Not
only should action be initiated, but frequent checks should be
made to assure that unnecessary delays are not encountered.
Furthermore, the directions of the field team should be followed
with respect to methods of shipment.
4) CSR and/or AID/Jakarta should see what could be done to
modify the ruling that trees cannot be planted in the Sitiung
area. It should be obvious that this is one useful alternative
for stabilizing the soils on the steeper slopes. It provides a
source of organic matter and a cash crop.
5) The budget available to the field team should be
6) This list of problems and recommendations should in no
way be construed as indicating the project is in trouble. On the
contrary, I consider it to be an exceptionally good project and
one which could serve as a model for many other similar projects
around the World. Hence, my final recommendation is that efforts
be made to help this project receive the recognition it merits.
A. Collaborative Soil Management Research with farmers (Farming
Systems Package). Draft Outline.
B. Conversion of 1984-89 Program Plan Experimental Numbers to
C. Field experiments 1984-85, Sitiung. Classification by 1)
Soil Chemistry/fertility, 2) Soil physics/conservation, 3)
Soil biology/pasture, 4) Soil people/FSR.
D. Statement by team of status of project as of May 1985. Has
some of my comments on it.
E. April/May 1985. Status of field experiments. This will be
written periodically by field team to communicate with home
campus, PIs and others.
COLLABORATIVE SOIL MANAGEMENT RESEARCH WITH FARMERS
(FARMING SYSTEMS PACKAGE)
PROJECT I. Interpretation of existing soil management systems.
a. Cognitive mapping of soil related concepts and indigenous
soil classification systems. Herman Agus.
b. Comparative study on Java. Evensens.
c. Intrahousehold decision making/allocation. Vicky Sigman.
d. Time allocation (final). Colfer.
e. Nutrition/income. Colfer.
f. Economic evaluation of soil management technology.
1. Preliminary evaluation of technologies or components.
2. Evaluation of components in the farming system.
PROJECT II. Modifications of soil management systems.
a. Home gardening
1. Utiliation of waste materials. Joniarta.
2. Pasture. Carl Evensen
3. Nutrition. Stacy Evensen.
4. Fish ponds. Dudley.
b. Collaborative research on farmers upland fields. Wade.
c. Land management/small catchments. Soleh
TROPSOILS Program Plan
Conversion of Program Experimental Numbers to Sitiung System
Eragrlam UnD. Siliing. Un.
S-1006 1102, 1204
S-1007 1001, 1205
S-1008 not initiated
S-1009 proposed for 9/85
AddiiLonal wLrk nni lised in .Euroram Elan
1002, 1103, 1404, 1403, 2001, 3001, 32001, 3501
lonpic N ECoaiei NMo. Siliuag Un.
2.3 4101, 4501, 4601, 4201, 4502
3.2 not initiated
3.4 not initiated
4.1 redefined, new staff to do
4.3 3502, 3503
4.4 proposed for 9/85
Addi ional iok. awua ila Eraogam Elan 1104, 3401
Projects funded and/or directed by CSR
Field Experiments 1984-85 Planting Season
- - - 1000's: Soil Chemistry - - -
1001. Sulfur LC
1002. Lime Reaction
1103. Lime Reaction
1104. Ca and Mo
1201. Reclamation lid
a. Initial-0,10,20,40 kg S/ha
b. Per crop-0 or repeat above
a. Lime-0,1,2,3,4,6,8 meq/100m1
b. very frequent sampling to
monitor acidity changes
a. Initial rates-0,20,40,80
160,320 kg P/ha
b. Maintenance-0,20 band,20
a. Lime-0.4,1.5,5.0 T/ha
b. K-0,25,50,100,150,250 kg K/ha
same as 1002
D. oval ofol i um,
b. Ca-none, 1.5, 10t/ha
c. Mo-0,1,2,4 kg/ha
a. Fertility-low lime and fert Karim
b. Tillage-hoe, mulch, green manure,
rototil, deep, deep strip
1202. 'ariability lie
1203. Lime lie
1204. Potassium lid
a. No treatment--intense measure-
ment of soil properties and
crop growth to identify
leading causes) of
a. Initial-0,0.5,1,2,4 T/ha
b. Annual-none or maintain
60,40,20,10,0 Al sat
a. and b. same as 1102
1205. Sulfur lid
1301. Lime IIIc
1401. Magnesium IVb
1402. Rice X P IVb
1403. Rice IVb
1404. Lime Reaction
same as 1001
a. Rates-0,30,60,90,120 MgO/ha
a. 6 lines/varieties
b. +/- lime
c. 4 rates of P
IRRI upland, acid screening
trial +/- lime as well as
same as 1103
- - 2000's:
2401. Erosion IV
2402. Conservation IVb
a. Rates-0,20,40,80 kg P/ha Wade
b. Method of Application NCSU
Soil Physics (conservation) - - -
a. Lime-0 or 3 T/ha
b. Burn after clearing-none,
iX (flash), 2X (complete)
c. Tillage-none or hoe
a.burn and no burn site
b.till and no till
recycled as mulch,
recycled + compost
b. Terrace vegetation-none, grass
c. Slot vegetation o-none,mulch
a. No control, grass strip only,
sloping bench, level bench
b. Vegetation-crop or cover legume
- - - 3000's: Soil Biology - - -
3001. Mo X Inoculum
(soy, cwp, pnt)
3101. Pasture la
3201. Lime-P-Green Man.
3502. Alley cropping
3503. Cover management a.
collection of vining, bush
and tree leguminous crops
CIAT Trial B collection
of legumes and grasses for
forage/cover/ and green
a. inoculum-none, forest debris,
pot culture VAM
a. Lime-0, 1, 2, 3 T/ha
b. P-0,10,25,50 kg P/ha/crop
c. GM-0 or 20 T fresh/ha
returned + green manure
Lime-0, 0.37, 2.0 T/ha
Cal l iandra calothyrus,
G1 iricidia sepium,
Fertili ty-4 rates N,
and 4 rates lime
cut and carry, grown in-situ
Source of G.M.-Crotolaria
- - - 4000's: Soil People - - -
open-ended interviews on
relationship between soil
and people (Javanese)
Ten indepth interviews on
major constraints to produc-
tion, as perceived by farmers
open-ended interviews on
4502. Land management
4510. Farmer Systems
4602. Minang Tree
relationship between soil
and people (Sundanese)
a. combinations of fertility
inputs and specie selection
depending on landscape
Ten indepth interviews on
major constraints to produc-
tion, as perceived by farmers
2)Gov. rate of lime
according to the
b.+/- green manure
construction of questionnaires
and preparation for survey of
100 respondents (Minang)
Survey of local farmer
useage of tree crops in
their farming system
Note: Numbering system is as follows:
hundreds' digit is location
600=Koto Padang (Minang)
tens and ones are just for sequential numbering.
TROPSOILS PROJECT, CURRENT STATUS,
AND PROPOSAL FOR EXTENSION
The TROPSOILS project is operated through a Memorandum of
Understanding between the Government of Indonesia (GOI) and the
United States Agency for International Development (USAID). The
agencies responsible for conducting the intended research are the
Agency for Agricultural Research and Development (AARD) through
the Center for Soil Research (CSR), the University of Hawaii (UH)
and North Carolina State University (NCSU), through Management
Entity, Soils Management (CRSP).
The original agreements were signed in 1982, and project
duration was intended for five years. This period will terminate
in 1986. However, due to circumstances beyond the control of the
operating agencies, the approval for starting the project was not
finalized until April of 1983 and personnel to operate the
project did not come on board until July 1983. So the project
has been in operation for less than two years. As the necessary
time for obtaining conclusive results will take at least five
years, an extension of project duration has been applied for
through USAID. The project has been viewed by personnel
responsible for an extension of project funding, and favorable
approval of an extension to 1989 is expected.
ihis soil management project has included from ten to twenty
Indonesian personnel from CSR, as well as three to five American
families in the area, and several short term consultancy
specialists all involved in the research. Numerous experiments
and surveys have been conducted, but the documentation of
TROPSOILS results is just beginning, For this reason, the
following information is provided, to indicate the kinds of
results that have been obtained as well as some sense of the
pressing research problems that remain.
The presentation is divided into four categories, to capture
the diversity of topics that we have identified as critical to
improving and sustaining good crop production in Sitiung, and
that we are addressing. These are, briefly, soil fertility, soil
physics, soil biology, and soil people.
SOIL FERTILITY LC WR 1
In Sitiung, and in many transmigration areas of the Outer
Islands, soil acidity and soil nutrient infertility, especially
P, are prevalent. Therefore the majority of our field research
has been directed toward identifying and efficiently correcting
these soil constraints.
Liming is an agricultural practice that has been done
throughout the world for many years. We are not trying to re-
discover the basic established principles of liming acid soils;
rather, we are trying to test and quantify those principles under
the soil and climatic conditions of Sitiung. We have conducted,
and are continuing to conduct, various experiments involving lime
rates, methods of application, residual effects and interaction
with other soil amendments such as phosphorous and potassium
fertilizer and green manure. We have found a relative tolerance
of upland food crops as follows: Upland rice > peanuts > corn >
soybeans > mungbeans. The liming criteria are very dependent on
the crop to be grown.
Finding site specific crop responses to lime is useful, of
course, but it is not efficient nor feasible for all agricultural
areas. Thus our research must develop ways to,predict (a) which
soil and crop combinations need lime, and, for those needing
lime, (b) what rates should be applied. We have usually found
good correlation between crop yield and Al saturation of the soil
(Al sat = exch Al by KC1 extract/ cations (A1+H+Ca+Mg+K) for the
crops listed above at a given location.
However, the nature of the relationship (critical levels and
efficiency of lime rates) has not always been consistent between
locations. We feel that organic matter and/or soil mineralogy
may have an important impact on this relationship. Thus our
current studies include both (a) long term experiments that can
study liming criteria over time and (b) specific field and lab
experiments to determine and quantify soil parameters (e.g.,
organic matter, mineralogy) that can.unify lime requirement
predictions for extrapolation to other food crop production
The management of lime and its use in agricultural
production is and will be a critical issue in Indonesia's quest
for food self-sufficiency. Research information being produced
and sought by the TROPSOILS research team can make significant
contributions to policymakers regarding the use of lime for food
crop production in the nation.
Liming acid and infertile soil is not a complete program
that guarantees successful crop production. Liming will reduce
or eliminate soil acidity (usually toxic Al) as a limiting
factor, but will not correct soil nutrient deficiencies. Many of
the acid soils of the Outer Islands (oxisols and ultisols by U.S.
taxonomy) are very deficient in available phosphorous. Just as
prediction of deficiency and correction rates is vital to a
liming program, it is equally important for the major nutrients,
such as P. Thus TROPSOILS is also pursuing field work to
establish soil analysis procedures for makingthese predictions.
Information regarding the frequency and methodology of applng
fertilizer is important in management decisions, and is also
being studied in Sitiung. Our results to date indicate that
frequent application of low rates of P fertilizer is sufficient
to overcome P deficiency. Either broadcasting or incorporating
prior to planting, or placing the fertilizer in the dibble hole
SJ with the seed is equally effective with other application methods
S ^ such as banding, but is far less time and labor consuming for the
Farmers. As with liming, we find variation among crops in their
tolerance to low soil P as follows: rice > corn = soybeans >
peanuts. Again, long-term studies have been initiated to clarify
the best agronomic and economic management of P fertilizer.
TSP fertilizer is readily available and relatively
inexpensive to the farmer. However, this is possible only
because of considerable subsidy by the Government. Therefore
improved efficiency in fertilizer use could not only save the
Government a considerable amount of money, but should help insure
improved farmer yields as well.
Additional studies investigating rates and management of
potassium, magnesium, and sulfur are also being conducted.
Substantial responses to both K and Mg have been shown on rice,
soybeans, corn and peanuts.
SOIL CONSERVATION (1 )
The presence of severe erosion in Sitiung has been apparent
from the beginning of CSR interest in this area. One of our
first research activities was to plan and implement a longterm
erosion experiment, to determine the rate of erosion under
various conditions. An Ultisol soil was chosen in Sitiung IV,
with a slope ranging from 9% to 13%, for our experimental plots
(22 m along the slope and 2 m across the slope; with 8 different
Between February 1984 and February 1985, the amount of soil
loss on bare soil was 246 tons/ha, equivalent to the loss of 2.5
cm of topsoil. Under food crops without soil conservation, the /
amount of soil loss was still quite high: 48 tons/ha. Both slot
mulching and bench terracing were found to provide excellent
protection for the soil, with losses of less than 5 tons/ha.
Grass strips and ridge treatments resulted in soil losses of 13
and 20 tons/ha, respectively. The amount of rainfall was 3888
mm. This study is planned to continue for at least five years.
The effectiveness of bench terraces as an erosion control
method is well known. However, on newly constructed bench
terraces, soil productivity on the back part of the land surface
(+/- 50% of the total surface) is generally low due to the
removal of the topsoil. In an attempt to assess and overcome
this problem, a study using green manure and crop residues has
been carried out on an Ultisol in Sitiung IV, beginning in August
1984. The effects of grass strips on the formation of bench
terraces is being studied, as well as changes in soil physics and
Preliminary results showed a pronounced difference in yield
(both rice and crotalaria) between the scraped part and the
filled part, especially on the sloping bench terraces. Grain
yields remain quite low, ranging from 6.2 to 9.1 quintal/ha.
Crotalaria usaramoensis produced 2.2 to 9.9tons/ha, averaging
4.9 tons/ha, of green manure within three months. This
experiment is planned to continue until significant soil
improvement is reached and terrace formation by grass strips is
Transmigrants are given 1 ha per family to be used for food
crop production. Many of these lands are located on undulating
to rolling topography, having a gentle slope at the top and steep
slopes leading down toward streambeds. Slopes reach 50% in some
areas, and slope lengths of 25 m are not uncommon. Severe
erosion is undoubtedly taking place on such slopes.
More effective soil/land management and conservation are
important in such circumstances. Assuming that 0.5 to 0.75 ha of
land is sufficient to provide food for farm families, it would
seem possible and desirable to integrate some effective soil
conservation practices, such as tree crops or green manure crops,
into the cropping system. A land management project, in
collaboration with four farming families in one catchment area in
Sitiung V has been initiated, and is planned to continue for at
least five years. Bench terraces, two kinds of grass strips, and
no conservation are being compared, and coconuts are being
incorporated into the cropping system, planted along the
contours. Soil characteristics, yields, and socioeconomic
aspects of the households will be monitored.
To study the value of grasses and legumes for forage
production (for use by livestock, as green manure, or for erosion
control), several species and cultivars of tropical forages have
been introduced from CIAT and other tropical research agencies.
These have been planted with low inputs of lime and fertilizer as
it is not expected that local farmers will have resources for
high inputs into a crop which does not have an immediate cash
return. Fortunately several species have produced very well
under the low inputs and will undoubtedly be able to have an
impact on the restoration of badly eroded soil, to contribute to
a more nutritious ration for livestock, and provide better
yielding forages for green manure.
Among the important new species are several varieties of
centrosema, desmodium and brachiaria. These species are actually
performing much better than calopagonium or pueraria which have
been the primary species utilized previously.
An expanded trial with several new introductions is
currently being evaluated. Further studies should include the
use of grasses and legumes in combination to promote better use
of the biological nitrogen fixing capacity of the legumes with
the superior erosion control capabilities of the grasses. To
determine the best combinations and species for all these
purposes will take time, but in the final analysis they can
contribute greatly to a profitable agriculture for Sitiung and
SOIL BIOLOGY N
Recent studies by TROPSOILS with green manuring have show.
-A amazing yield increases. Even on plots with high lime and
ipr fertility applications, yields of rice and soybeans have been
increased with green manuring. Also on soils not limed, we have
found that green manuring produces good yields of rice,
apparently replacing or at least minimizing the need for lime.
This is a very recent result and needs to be substantiated with.
further investigations on additional crops. The effect is g
currently being tested on soybeans. A similar replacement
effect for phosphorous and potassium fertilizer as been observed
in other trials. Our soil analyses do not reflect this a rnt
enhanced nutrient availability and decreased soil acidity.'
therefore more research is needed to develop laboratory
L\ techniques that can detect and therefore predict this beneficial
effect on soil properties and crop yield.
The implications of this initial research are great. If
indeed farmers can grow tree or forage legumes in their cropping
systems to produce green manure that can replace or minimize the
need for lime and inorganic fertilizers, then a major step toward
self-sufficiency will have been made. These legumes can also be
used for high quality forage for farmer's livestock. Field
experiments have been started to test both sources and management
of green manure for optimum utilization by farmers.
Research on this biological aspect of enhancing soil
fertility is just beginning. Because of its potential, this work
must be continued and expanded. The use and benefits of green
manuring have been demonstrated around the world for many years.
However, very often such work is done with rates and management
that is totally unfeasible for the local farmers, and thus is
seldom put into practice. The results obtained in Sitiung are
with reasonable rates and the current management studies
mentioned above are aimed at establishing green manure production
as part of the cropping system in such a way that the farmers can
and will accept it. Consistent with our Farming Systems Research
approach, the researchers try to be acutely aware of reaT-world
farmer situations and activities (with aid mrofm an anthropTogist
an- regular farmer-researcher"interaction). This program has
high potential for rapidly developing improved agronomically and
economically sound technology that is acceptable and practical
for the farmer.
The Tropsoils project in Sitiung is being undertaken within
a Farming Systems framework. That is, soil management
problems are being addressed with consistent attention to the
welfare and interests of the farming families we hope will be
able to use the agricultural technologies we develop.
There have been multiple purposes of the research undertaken
within this category. The most important has been to provide
soil scientists with assistance in determining research
priorities that will have a good chance of being used by people.
This function can perhaps best be clarified by some specific
examples. The team had expertise in pasture crops and saw a
number of soil management uses for improved pasture in the area.
They wanted to know whether people were likely to consider
improved pasture a priority or not. To determine this, a survey
was done on animal ownership in Sitiung I, and the data from "a
time allocation study were examined to learn how much time was
spent in the collection of fodder. Respondents in Sitiung I and
II, the longest settled areas, generally had between one and two
cattle per family, as well as several goats; and residents of
Sitiung I spent about 45 minutes per day per family member
collecting fodder. These Javanese farmers place a high value on
their animals for manure and field labor; and they view the
livestock as an asset which can be readily converted into cash at
any time. All these factors led us to conclude that research on
fodder in the area was warranted.
Team members similarly noted the greater suitability of tree
crops to the soils of Sitiung, and numerous ecological advantages
to maintaining a tree cover. It seemed likely that incorporating
tree crops more meaningfully into the farming system of the
settlers might provide an opportunity for substantially improving
settlers' (and environmental) well-being. The indigenous
population normally gains a considerable amount of their income
from rubber cultivation. For this reason, we viewed the local
people as a source of experiential knowledge, and we undertook a
study of Minang Tree Farming practices. Our findings included
documentation of the tree crops grown, agricultural practices
related to these crops, and income deriving from tree crops.
Although we remain uncertain about the amount of land needed to
undertake profitable tree farming, we are convinced that research
on agroforestr--incorporating food crops and tree crops into one
system---is a suitable objective for the project. Discussion is
currently underway about bringing a graduate student with
experience in agroforestry to the project in June to help plan
appropriate research. We are also incorporating coconut trees
into one of our experiments and rambutan trees into another, both
of which are being conducted collaboratively with farmers in
A second objective of this soil people part of'the project
has been to provide a testing ground for component research
results that look promising. We began working collaboratively
with farmers in September 1983, and have continued. Two
experiments include considerable input from farmers, in an
iterative fashion. The first initiated was with 19 farmers in
Sitiung V. The first year we were particularly interested in the
utility of rock phosphate as a cheaper, more practical source of
phosphorous. We planned the experiment with the farmers, usually
accepting their suggested changes, responding to their problems
with implementing our joint plans, and noting their yields. One
purpose of this activity was to determine the most successful
ways of working collaboratively with farmers. This activity
continues, with a current focus on green manure---a management
strategy that has provided good results in our more controlled
experiments. We have also initiated collaborative work of this
kind with four more families, whose fields comprise a small
catchment area in Sitiung V (discussed under "Soil
Various monitoring activities were conducted last year, to
answer questions that developed within the team as we conducted
our agronomic research. A nutritional study in Sitiung I, II,
and V determined that the nutritional status of the populations
was marinall ade uate, on average meaning that half of the
, -,2popula ion was getting a substandard diet). This confirmed our
suspicion that we needed to attend closely to the nutritional
S contribution of crops we use in our experiments.
I The same study provided data on income (including value of
agricultural and home production and cash) to ascertain whether
or not the people could buy agricultural inputs. Annual incomes
ranged from Rp. 100,000 to Rp. 2,400,000 for the year April 1983-
April 1984. Average income, excluding an average subsidy in
Sitiung V (37 families) of approximately Rp 290,000, was just
over Rp 500,000 for the 77 household surveyed. Almost half of all
the households produced total farm produce worth less than Rp.
100,000 for the year (including what was consumed and what was
sold). There is clearly more non-farm income than farm income
being produced in Sitiung. With total incomes around Rp.
500,000, the availability of cash for agricultural inputs is
questionable, suggesting a team focus on low input technology.
Much work remains to be done. Naturally, the process of
monitoring the work with the collaborating farmers must be
continued, as we develop more finely tuned agricuTtuira-
technologies. The problems that emerge in the course of working
with farmers must be noted, and necessary alterations must be
made to respond to such problems. By this process, we hope to
avoid the fate of many agricultural projects which develop
superior agricultural technologies which only appear to work on
We have already identified a number of important questions
that remain to be answered: How can an agroforestry system be
developed that is possible for the settlers to implement on their
2 hectares of land? And how can they subsist while awaiting the
maturity of trees crops? How can the substantial agricultural
activities of women be incorporated more meaningfully into
TROPSOILS research activities? Who within settlers' households
makes important decisions about the allocation of scarce funds
(particularly to agricultural inputs)? What kinds of factors are
important in these decisions? How do settlers and indigenous
peoples perceive the soil and their relationship to it? Answers
to such questions have important implications for improving soil
A research component has recently been planned, to
incorporate some of the important concerns we have identified: a
home gardening project in collaboration with settlers. This
series of research activities, integrated within the settlers'
home gardens, incorporates nutritional, income producing, women)
---in deelopmen, fisheries, and animal husbandry concerns, as well
as concerns that derive purely from "soil management." We expect
to describe the complexity of the gardens and assess their
contribution to income and nutrition, relative to other important
components of the farming system; to work closely with women as
gardeners in this project; to assess the potential of fishponds
for contributing to soil fertility (using fishpond sludge), as
sources of water during the frequent dry periods, and for
improvements in fish culture; to use these areas as a testing
ground for our forages; and to collaborate with a scientist who
can assess their agroforestry potential.
In addition to the important research questions listed in
the above four categories of activities, the project is
contributing to the strengthening of CSR's institutional
capabilities in the area of personnel.
Looking first at Personnel, the project is providing
assistance in English to the Indonesian project personnel.
Although this is not a direct aim of the project, it is a
critical function, since CSR wants to provide additional
education abroad to many of its personnel. Without additional
English training, many would not be able to pass the TOEFL test.
The purchase of three (possibly four) micro computers, on
site, has created a valuable opportunity for the SR staff
members to learn to use these in data management, data analysis,
and word processing. A training program designed specifically
for CSR staff needs has just begun.
Since the Indonesian staff, except for the site coordinator,
are recent college graduates, the collaboration between the more
senior Americans and the Indonesian scientists is providing a
valuable training function. The system that has developed on
site involves a sort of apprentice program where each American
scientist works closely with one or more Ir. level Indonesian
scientists. This provides both effective collaboration and
strengthening of CSR staff.
Despite a late start, the TROPSOILS project is now in full
swing, with almost two years of on-site experience and results.
Although preliminary answers have been found to some important
research questions, many more pressing research needs have been
identified. A team of collaborating researchers has been
assembled, and effective working relationships have been
established. The'stage has been set for continued work; and the
value of research in Sitiung can be expected to improve, now that
many of the difficult logistical problems have been overcome.
Status of Field Experiments
lj 1001. Suyliur Fert[rli__zation.. Corn harvest has been completed.
Yields were low, on the order of 2T/ha, with no response to S.
The plots are being prepared for a subsequent planting of mung
bean. The original plan calls for soybeans as the second crop in
the sequence but since the dry season is impending the 60-day
mung beans will be grown instead.
il^) 1002. LiM e React.ion Rate and Effecti.veness... IThe soybeans are
nearing harvest, and have only moderate insect damage. There
appears to be little response to lime on this newly cleared area.
1101. E R ate and Maintenance.. Soybeans have been harvested.
A yield plateau of a little under 2 T/ha was obtained. The
response to applied P was very much like the corn; a very sharp
increase in yield with up to 40 kgP/ha with a continued but
moderate increase with rates upto 160 kg P/ha. Plots with >8ppm
extractable P (by modified Olsen) did not respond to additional P
fertilizer. A per-crop maintenance dose of 20 kg P/ha, if
broadcast and incorporated, seems to be adequate in producing
plateau yields, regardless of initial P rates including zero.
Banding of the maintenance dose is less effective than the
broadcast. The residual plots show considerable decline in
yields compared to the maintained plots.
jt 1102. Potassium m Dynamigcs. The peanuts are growing very well and
will be ready to harvest soon. The K response is very strong for
plant top growth, but there is little differences due to the
lime rates. Yields should be quite good and the harvested data
will be most interesting.
1103. Lime Reaction Rate and Effectiveness. These soybeans are
nearly ready to harvest. The no-lime plots are practically
barren. The response to the lime is dramatic, but even the limed
plots will produce only modest yields due to insect damage and dry
weather. The lime reacted very quickly with the soil, as samples
taken only 3 days after application showed nearly as much
reduction in exch. Al as after 7 or 30 days. Samples taken at 15
days after application showed a considerable and consistent
increase in Al as compared with the other sample dates. This
also occurred in the plots at Sitiung IV. The 15-day samples
were taken at a very dry time, near the end of a 3-week drought.
At present we have no explanation for this occurrence.
ha-00104. Response of Foruge Legumes to C a.nd Mgo._ Four species,
Cntcosenma macrocar ens. Desmodium oyalifoll.um. Desmodiym
i-to.ctam.,. ar0d have been planted. Early
germination and stand establishment is good. The Centrosema was
planted from sprigs (2X) and appears to be taking hold this time.
The trial has seven fertility treatments with combinations of
lime and Mo rates.
1201. CEa:iQ.1.r g.:ggided l gIit Ii.:. .m.ali. '.. The soybeans are nearing
harvest, but never fully recovered from the insect damage done
earlier in the season. Yields will be low and it appears
treatment differences will be modest. Final field physical
property measurements are being done, as Karim'is program nears an
end. The water retention apparatus used for measuring macro and
micro pore volumes is now operational, and available for samples
from this trial as well as others dealing with soil physical
,1 .1202. Var.iabl it y Trial. The cassava has been harvested and the
plot replanted to peanuts. Tuber production was remarkably va--
riable and will be compared with the variability of the other
crops (rice and peanuts) as well as soil parameters. Initial
stand of the peanuts is good.
U ~1203. LiMfe Bate and_ Main.t=E'nance. The soybeans were just
harvested, with threshing and data collection in process. As the
dry season is expected to start soon., no additional crop will be
,,=planted this season.
1204.. Potassium Dynamics,. The soybeans are now in pod-fill. and
doing well. The response to both lime and K is very evident..
The no K plots show a strong chlorsis on the leaf margins of the
upper leaves, with a downward curling of the leaf tips on the
most severely effected leaves. The low lime plots have plants
that are stunted with premature leaf senescence yellowingg and
falling powerr 1 eaves).
1205.. SuI1fur Fetil .11tio,.. The corn has been harvested.
Yields were quite low with most plots producing about It/ha.
However the zero S plots yielded less than 0.5T/ha. Despite the
overall poor growth of the corn, a S def:i.ciency is demonstrated.
The next crop will be mung beans. Corn will be grown again next
wet season to foll ow up on this apparent response to S. It
should be noted, as might be expected the S response at this
site (in continuous cultivation for 6 years) is in contrast to
the recently cleared site (1001) which showed no response.
1401. Magnresium SoLr2ce and Meth.od of Ag.1iicatin.. The soybeans
are growing well in this final crop. The primary differences
appear to be due to the lime factor. All +Mg treatments,
regardless of the method of application, are doing well if limed.
)A ,1404. Lime Reaction Rate and Effectiverness. Harvest of the
soybeans will be within a few days. Again the no-lime plots are
nearly bare while the limed plots have done fairly well, although
moisture stress ? has kept production low. Visually the 3 T/ha
rate looks to be about optimum. As in the Sitiung :1 location
(1103), the reaction of the lime seems to have optimized between
3 and 7 days after application. This would indicate that
incubation of applied lime may be an unnecessary practice in this
soil and climatic sitluat:i on.
150:1. Faimint Sst res: e! c... F'g Peanuts are being harvested
from some farmers" fields now.. Centrosema seed has been
distributed to be interplanted between the peanuts of those not
yet or near harvest, otherwise they will pl ant the legume after
the peanut harvest.
1502. P Rate and Placement. Peanut harvest is near. The crop
shows good visual response to P but less striking than last year,
as the zero plots have grown considerably better than last year.
It should be interesting to see the nut yields, as they do not
ne essarily reflect top growth especially on marginal soils.
2001. L.ime X Burn X Ti.l.lage. An improved peanut variety was
*found and therefore planted as the second crop in this trial, not
cowpeas as reported in the March Field Status,, The soybean data
have been compiled. Overall yields were very low, due to both
insect damage and probable P deficiency. However responses were
clear: Tillage reduced yields at all lime and burn levels and
burning reduced the lime response. The 2X burn reduced Al
saturation to nearly zero and did not respond to lime.
S0. ECroi... The soybeans continue to grow erratically both
among and within the plots. The bench terrace treatments
generally are the least uniform due to the drastic soil movement
required to form the terraces, despite efforts to remove and then
return topsoil during the terrace forming activities. Also the
crop growth remains erratic despite substantial blanket rates of
lime and fertilizer.
2402. gil Consernatigo. The most striking aspect of this
experiment at present is difference in growth of the two legumes
used as indicator crops. The various conservation treatments
have been divided in two; 1/2 with cowpeas and 1/2 with soybeans.
The input level is low and the trial is located on a very steep
slope of a poor soil, yet the cowpeas are growing very nicely and
will produce well. The soybeans in sharp contrast are growing
3001-. Mg X IjDg.._g., Treatment revisions are being made as this
trial will be expanded to include more comparisons. Cowpeas will
be planted as soon as the new plots are ready. In case you
wonder why we would do an inoculum trial on something as
promiscuous as cowpeas, our field observations around the Sitiung
area are that none of the common annual legumes, including
cowpeas, nodulate well.
3101-. C.IAT Regi na: Trial B., The first harvest was completed on
April 19. Final data are still being processed but a few of the
species were outstanding. Of the legumes, Centrosema
ma.CocarCEfm and Eubescens and AescGIhynomene hbistr.i were the best.
Androogon gaavanUys was the best of the three grass entries,
although BraciarEi2a dictygneiura and decumbens did very well. The
Andropogon was somewhat over mature and had low palatibility
(according to the farmer who hauled the cuttings home to his cow).
3201,. reen Manur .. !=X Lime X P... The soybeans are growing nicely,
and show marked response to lime, but surprising y little
difference among the 0.5, 1.0 and 1.5 X ex:ch Al rates of lime.
Most of our trials generally show maximum soybean response when
limed to 1-1.5 X exch Al. There is also considerable response to
P. The green manure treatments are much less dramatic for
soybeans than for the previous rice. The green manure was not
reapplied, so this is the residual effect of the manuring. Under
our temperature and moisture conditions it is not surprising that
the benefit of a green manure is indeed short-lived. However
there still appears to be some response, just less remarkable
3501. PFotassium Rates and Stover M anagen et.. The soybeans of
this trial are in the pod fill stage and look good. The response
to green manure (incorporated Calapagonium) is phenomenal at the
zero K level. Despite high lime and P basal applications, the
zero K plots are barely surviving and probably will not produce
grain. However where green manure was added (no fertilizer K)
the beans are lush and vigorous. Even the plots that had rice
straw (from the previous crop) incorporated are looking very
well. They are not quite as good as the green manured ones but
certainly far better than the residue removed treatment. The
response to K is obvious in the non-organic treated plots but the
beans of even the highest K rate are less vigorous than the
manured, 0 K ones. It appears that returning residues and/or use
of green manure may be a critical management practice on these
poor soils even when high inputs of lime and fertilizers are
3502,. Tree Legume A.lley Croei ring.^ The potential for grain yield
from the intercropped upland rice has virtually been destroyed by
blast. Harvest will be for dry matter production only.
Variability is extreme and a fertility map using the rice as
indicator is being made in order to provide data for co-variance
analysis as well as monitoring of uniformity as affected by the
organic matter produced from the tree crops. The trees
themselves are developing much slower than expected. The
11GiriLcidia continues to be the best, with fairly uniform growth
and good development. The Al.bizia is catching up more but is still
quite variable. The Cal QC.liandra is doing poorly and may not
survive well enough to establish "alleys" or produce significant
3503 S9ue and Manage9ment2 of GreenD Manure, The cowpeas are
nearing harvest and will generally produce good yields. However,
as expected variability is high and the cowpeas as indicator crop
will provide critical information for experimental design and
plot layout. As green manuring continues to show sharp and
consistent positive responses, the timeliness and importance of
this trial also continues to grow. If green manuring is indeed
so valuable then it is imperative that we investigate ways to
make green manuring part of the farming system in such a way that
the farmers will accept and use it. 4101 aliileo StLdy. Nine
open ended interviews on the relationship between people and soil
(in Minang) were conducted, transcribed, and the concepts
counted, for use in the Galileo instrument being created, and as
a base for the ethnoscience work: to follow next fall.
45 01 Bt Gali leo S tudy. Four simi.l ar interviews were conducted in
4601 Gal;i.].Jeo Study. The results of the Minang and Javanese open
ended interviews on the relationship between soil and people were
analyzed, and compiled into a single instrument that can be used
for both languages. A decision was made to begin the actual
Galileo interviews with 100 respondents, half male and half
female, in Koto Padang.
4602 Mi.naD.n. ICr E iarlmngl StQtdy. The translation of this
420:1. EaCr.mer! -et .i: s.!'_r. FioMajo giCa l. l.t. a. ConQntraint. Ten
in-depth interviews were conducted with farmers of both sexes.
An attempt was made to visit the randomly selected families
interviewed in the 1984 nutrition/income survey, but a change in
house numbers made that impossible in some cases. These
interviews were taped and are being transcribed, translated, and
shared with other team members.
4502 rg".mers... Per cetti=Qns of. Ma.jr Aigricul.tural] CoC.nstrai:nts
Similar indepth interviews were conducted with ten families,
chosen from the randomly selected "control" farmers interviewed
on three previous occasions. These were also taped and are
being translated, and shared with other team members.