Scientific registration no: 1394
Symposium no: 12
Soil fertility recapitalization in acid upland soils in
Southeast Asia: the example of Indonesia
Recapitalisation de la fertilité des sols acides en Asie du
Sud-Est: l’exemple de l’Indonésie
FAIRHURST Thomas H. (1), LEFROY Rod D.B. (2), MUTERT Ernst W. (1),
SRI ADININGSIH J (3), SANTOSO, Djoko (3)
(1) Potash & Phosphate Institute, 126 Watten Estate Road, Singapore 287599.
(2) IBSRAM, PO Box 9-109, Chatuchak, Bangkok 10900, Thailand.
(3) Centre for Soil and Agroclimate Research, Jln. Juanda 98, Bogor, 16123, Indonesia.
Productive and relatively sustainable flooded rice and plantation crop systems have been
developed over the past forty years in Indonesia with large investments in infrastructure and
in soil fertility, partly through subsidised fertilizers, particularly phosphorus. In contrast, the
soil fertility reserves of annual and mixed cropping systems of the acid uplands have been
continuously degraded. As populations continue to grow, competition for use of agricultural
land increases, and arable land reserves are increasingly marginal, Indonesia, and other
countries in Southeast Asian, must increasingly turn to the infertile acid upland soils for
future supplies of food, fibre and industrial crops (von Uexküll and Mutert, 1995).
Various strategies have been outlined for soil fertility recapitalization in Africa
(Kuyvenhoven et al., 1996; Sanchez et al., In press; The World Bank, 1994) since
comprehensive surveys revealed the extent of soil nutrient mining (Stoorvogel and Smaling,
1990). However, little attention has been paid to Southeast Asia, despite the fact that 60%
of soils in the region are acid and the total area occupied by acid soils is almost twice that
found in Africa (von Uexküll and Mutert, 1995). Part of the greater interest in soil fertility
recapitalization in Africa derives from the dual benefits of investing in rock phosphate
mining operations while increasing soil fertility in farmers’ fields. In Southeast Asia, local
deposits are not sufficient in terms of either quality or quantity and the region depends upon
RPRs imported from other parts of Asia, North America and North Africa.
Soil fertility parameters for recapitalization
All soil properties which are amenable to change through investment and which affect
long term crop productivity may be regarded as assets which can be recapitalized. However,
particular properties may be more limiting than others and returns on investment may differ
between strategies aimed at different soil properties.
Soil organic matter (SOM) is an important part of soil capital, but it is difficult to increase
in tropical soils (Sanchez et al., 1989). Several technically feasible approaches have been
suggested (Giller et al., In press), but most are impractical for poverty stricken upland
farmers. Limited availability of residue and manure restricts large inputs of organic materials,
and poor farmers cannot afford legume fallows that produce no marketable product. Grain
legume crops are an option, but frequently they result in little or no gain in N due to large N
A more practical approach, albeit more gradual, is to increase the amount of crop
residues produced in situ and returned to the soil, either through increasing total biomass
production, or by selecting crop varieties with narrower harvest indices.
It is frequently asserted that the two most limiting nutrients for crop production are N
and P (Sanchez et al., In press). However, if biological N2 fixation is to supply part, or most,
of the N required in upland farming systems, P becomes the more limiting amendment. In
addition, P is considered the most widespread and economically important mineral
deficiency of livestock (McDonald et al., 1987). The potential for livestock to generate
income and transfer nutrients from crop residue and fodder into soil fertility, requires that P
deficiency is corrected by directly supplying P to livestock, or by increasing P in the soil.
K and Mg are not good candidates for recapitalization of acid upland soils as their
chemical and physical properties may result in large losses by leaching and erosion due to
poor retention. However, there must be balanced additions of these and other nutrients so
they do not become major limiting factors as productivity increases with soil recapitalization.
The process of P capitalization
In addition to increased available and total soil P, many benefits accrue from the
increased productivity that can result from soil P capitalization (The World Bank, 1994):
• increased above ground biomass, which increases residue return and reduces erosion risk
• increased root biomass, providing in situ organic inputs and efficient use of soil resources
• increased cation exchange capacity (CEC), particularly in soils with low effective CEC
• improved soil aggregation, aggregate stability, and water holding capacity
• increased soil microbial activity and biological fertility
• increased biological N2 fixation.
Soil P recapitalization takes advantage of the comparative immobility of P in soil. Losses
of P through leaching are small, except on sandy textured soils with limited P sorption
capacity. However, the immobility of P in soil may render added P susceptible to loss
through erosion and surface run-off. Incorporation of P within the rooting depth of the
crops reduces losses by erosion and increases reactions with the soil, which is of particular
benefit for rock P sources. Incorporation of P may be difficult in no-till slash-and-burn
systems. An alternative is to incorporate added P in the planting hole of tree crops, which
are frequently an integral part of these farming systems. Vegetative or physical barrier
systems can significantly reduce P losses in surface run off and eroded soil on sloping land.
Once soil fertility has been recapitalized, crop varieties must be selected to exploit the
potential for greater crop and residue production and enhanced biological N2 fixation.
Source of P fertilizer
The gradual manner in which RPRs release plant available P make them ideally suited to
long term plantation crops and, where soils are sufficiently acid, for seasonal crops (Rajan et
al., 1996). The relative agronomic effectiveness (RAE) of different rock P sources have
been evaluated in a range of cropping systems in Indonesia and the RAE of quality RPR was
found to be equal or even greater than TSP (Partohardjono and Sri Adiningsih, 1991).
Rock phosphate is a less costly source of P fertilizer when compared with triple super
phosphate and other locally available P fertilizers (Table 1), which, in most cases, more than
compensates for the greater transport and handling costs incurred per unit of P for RPR.
Table 1 Cost of P fertilizer in Indonesia (1997 farmgate prices)
TSP SP 36 Rock P
P content (%) 20.0 15.7 13.1
Cost (US$ kg-1 P) 1.05 1.02 0.61
Soil P recapitalization in Indonesia
Results from a soil fertility recapitalization trial on a degraded Kandiudult in Terbanggi,
Lampung Province, Indonesia, illustrate a number of important economic and agronomic
issues (Sri Adiningsih and Fairhurst, 1996; Fairhurst et al., In prep.). The experiment was
carried out over 11 cropping seasons, between 1983 and 1994, on land formerly covered
with Imperata cylindrica. Two RPR sources (North Carolina and Morocco) were applied at
1 t ha-1 and compared with 400 kg TSP ha-1 applied with 1 t ha-1 lime and a zero P control.
Although the amounts of P applied differed between the sources (131 kg P ha-1 as RPR and
80 kg P ha-1 as TSP), the cost for P inputs were similar. Additional applications of 200 kg
TSP and 500 kg RPR were applied to the respective treatments after the seventh cropping
season. Three further fertilizer treatments were imposed on the P treatments: no additional
fertilizer; added N (46 kg N ha-1, applied as urea); and N plus K (46 kg N ha-1 and 25 kg K
ha-1 applied as KCl). In addition, three cropping systems were used: the farmers’ practice of
an annual rotation of rice/soybean/mungbean; an intercrop rotation of
maize+rice/maize+soybean/ mungbean between alleys of Flemingia macrophylla; and an
intercrop rotation of maize+rice/maize+soybean/mucuna cover crop. There was no
replication, but the factorial design allowed some assessment of experimental error across
(i) Changes in soil properties
Although there was variation between P treatments and with time, in general, the amount
of available soil P increased and was sustained following the initial applications of P and
after the second application in the eighth cropping season (Figure 1). The increased available
P with all P treatments in the tenth cropping season may have resulted from a combination
of the recent reapplication of P, in the eighth season, and mineralization of organic P
following the drought in the ninth season.
No P TSP/L NCRP MRP
Figure 1 The effect of P treatment on available soil P (Bray 1) over eleven cropping
seasons at Terbanggi, Indonesia. (No samples in season 9 due to drought.)
Exchangeable K in soil was only maintained when K fertilizer was applied (Figure 2a).
Exchangeable Mg in soil declined with time, and depletion was more rapid when N and K
fertilizers were applied, due to greater Mg removal from higher yields (Figure 2b).
Nutrient balances, calculated from nutrient inputs and product removal, but with no
allowance for leaching or erosion, and assuming that all the nutrients in residues remained on
the plots, provide a crude measure of sustainability. An initial large application of P resulted
in a large positive P balance, whilst under the control treatment, a negative P balance
accumulated over five years (Figure 3a).
Large negative K balances were calculated when no K fertilizer was applied, whereas a
very large positive balance resulted from application of K (Figure 3b).
1 2 3 4 5 6 7 8 9 10 11
1 2 3 4 5 6 7 8 9 10 11
Figure 2 The effect of N/K treatments on exchangeable (a) K and (b) Mg over eleven
cropping seasons at Terbanggi, Indonesia. (K data for season 6 not available.)
Cum.balance (kg ha
MRP NCRP TSP/L No P
Year NK 0N
Figure 3 Cumulative balance for (a) P and (b) K over five years at Terbanggi, Indonesia
(ii) Economic assessment
Incremental net present values (INPV) were calculated for each treatment after
accounting for differences between the treatments in input costs, including fertilizer, seed
and labour, and returns. The large yield responses to added P resulted in INPV which were
twice those for the control, with similar economic returns for the three P sources (Figure 4).
No P MRP NCRP TSP/L
Figure 4 Effect of P and N/K treatments on INPV over five years at Terbanggi, Indonesia.
The very positive K balance when K fertilizer was applied (Figure 3b) suggests excess K
was applied, while the soil analyses (Figure 2a) suggest that much of the added K may have
been leached. Split applications of less K may have resulted in greater returns on investment.
There is evidence of a larger return to investment in K when P was applied as RPR.
These simple economic analyses ignore the value of longer-term improvements in soil
fertility. As P, other nutreints, and SOM build up, and the resource base improves, longer-
term increases in soil fertility should produce even greater INPV and sustainability. These
benefits are likely to be greater on economic and agronomic bases with an RPR source.
The farming systems and soil types for which these recapitalization strategies are likely to
succeed can be identified by measurement of soil pH, soil organic matter, available P, P
sorption characteristics, exchangeable bases, aluminium saturation and the nutrient demand
of the system. However, extension workers do not usually have recourse to detailed soil
analyses. Whilst improved analytical services are an important part of the strategies to
improve productivity in Asia’s uplands, simple surrogate measurements are required so that
extension workers can more easily identify areas where these strategies are likely to succeed.
Crop plants are good indicators of nutrient deficiencies and many non-crop/weed species
are indicative of impoverished soil (e.g. Melastoma malabathricum, Dicranopteris linearis,
Imperata cylindrica). Simple low cost pH field kits (Hellige™ pH kit) may be used to
determine whether soil is sufficiently acid to provide a reaction with rock phosphate. Soil
texture can be used to indicate potential adsorption and leaching capacities. Soil colour and
change in colour with depth are indicative of the amount of SOM and, in conjunction with
observation of rooting behaviour, an indication of available nutrients and potential toxicities.
Crude estimates of nutrient balance can help establish appropriately balanced initial and
maintenance fertilizer programmes. While these parameters are too imprecise for many
scientific uses, experienced field agronomists and soil scientists use many of them, often
unwittingly, in conjunction with more scientific measures. The challenge is to develop
understanding of these surrogate measurements to the extent that extension workers and
farmers can be trained to make sufficiently accurate assessments of land capability. To this
end, technical field kits need to be developed.
Large-scale recapitalization of acid uplands requires more than identification of the
appropriate soil types and farming systems. Appropriate germplasm must be available and
appropriate soil conservation measures must be identified and implemented. The distribution
of RPR and other essential inputs must be improved. The quality of inputs, particularly RPR,
must be monitored and assured. Monitored demonstrations of the strategies must be
established and used to mobilise the extension services. To this end, there is a joint activity
entitled SebarFos (“to spread phosphorus”), under the auspices of the Agency for
Agricultural Research and Development, which involves assessment strategies using RPR by
farmers in five provinces of Indonesia. These demonstrations need to be extended to village
scale implementation that can monitor changes in, and benefits to, the complete farming
system, encompassing annual crops, perennial crops and livestock, and to gauge the full
effect on, and of, markets. Credit, subsidy, and/or grant schemes that will support the
strategies must be developed. But the greatest challenge is to convince governments and
international agencies to support the strategies to finance poor upland farmers, at an
estimated USD500 ha-1, to allow these farmers to produce their way to a position of secure
food supply, reduced poverty and conserved natural resources.
Giller, K. E., Cadisch, G., Ehaliotis, C., Adams, E. and Sakala, W. D. (In press)
Building-up Soil Nitrogen Capital in Sub-Saharan Africa.
Kuyvenhoven, A., Becht, J. A. and Ruben, R. (1996) Public investment for soil
improvement in West Africa. Economic criteria for phosphate rock use to enhance soil
fertility management. In: IAAE Symposium, Economics of Agro-chemicals, 25-27 April
1996, Wageningen, The Netherlands.
McDonald, P., Edwards, R. A. and Greenhalgh, J. F. D. (1987) Animal Nutirition. 4th
ed. Longman Scientific and Technical, Harlow, Essex, 543pp.
Partohardjono, S. and Sri Adiningsih, J. (1991) Response of food crops to phosphate
rocks. Indonesian Agricultural Research and Development Journal, 13, 46-57.
Rajan, S. S. S., Watkinson, J. H. and Sinclair, A. G. (1996) Phosphate rocks for direct
application to soils. Advances in Agronomy, 57, 77-159.
Sanchez, P. A., Izac, A.-M. N., Buresh, R. J., Shepherd, K. D., Soule, M., Mokwunye,
A. U., Palm, C. A., Woomer, P. L. and Nderitu, C. G. (In press) Soil Fertility
Replenishment in Africa as an Investment in Natural Resource Capital.
Sanchez, P. A., Palm, C. A., Szott, L. T., Cuevas, E. and Lal, R. (1989) Organic input
management in tropical agroecosystems. In: Coleman, D. C., Oades, J. M. and Uehara, G.
(eds.) Dynamics of Soil Organic Matter in Tropical Ecosystems: Result of a Conference
held in Maui, Hawaii, 7-15 October 1988. NifTAL Project, University of Hawaii,
Honolulu, Hawaii, pp.125-152.
Sri Adiningsih, J. and Fairhurst, T. (1996) The use of reactive phosphate rock for the
rehabilitation of anthropic savannah in Indonesia. In: Nutrient Management for Sustainable
Food Production in Asia:IMPHOS-AARD/CSAR International Conference in Asia and
IFA-FADINAP Regional Meeting, Bali, Indonesia 9-12 December 1996. World Phosphate
Stoorvogel, J. J. and Smaling, E. M. A. (1990) Assessment of Soil Nutrient Depletion in
Sub-Saharan Africa: 1983-2000. Volume I: Main Report. (Report 28), The Winand Staring
Centre, Wageningen, The Netherlands, 137pp.
The World Bank (1994) Role of phosphorus in agriculture. In: Feasibility of Phosphate
Rock as a Capital Investment in Sub-Saharan Africa: Issues and Opportunities. World
von Uexküll, H. R. and Mutert, E. (1995) Global extent, development and economic
impact of acid soils. Plant and Soil, 171, 1-15.
Keywords : acid soils, infertility, phosphate, fertility assessment
Mot clés : sol acide, infertilité, phosphate, évaluation de la fertilité