37

Improving Phosphorus Fertility in Tropical Soils

through Biological Interventions
Astrid Oberson,
1
Else K. Bu nemann,
2
Dennis K. Friesen,
3
I.M. Rao,
4
Paul C. Smithson,
5
Benjamin L. Turner,
6
and Emmanuel Frossard,
1
1
Institute of Plant Sciences, Swiss Federal Institute of Technology, Zurich, Switzerland
2
School of Earth and Environmental Sciences, University of Adelaide, Adelaide, Australia
3
International Center for the Improvement of Maize and Wheat (CIMMYT) and International
Q1
Fertilizer Development Center (IFDC), Ethiopia
4
Tropical Soil Biology and Fertility Institute, International Center for Tropical
Agriculture (CIAT), Colombia
Q1
5
Berea College, USA
Q1
6
Smithsonian Tropical Research Institute, Panama
CONTENTS
37.1 Introduced Pastures and Cropping Systems in Eastern Colombia........................ 532
37.1.1 Measuring and Explaining Differences in Phosphorus
Availability ........................................................................................................ 533
37.1.2 Assessing Phosphorus Turnover ................................................................... 535
37.1.3 Implications of Experience from Colombia ................................................. 535
37.2 Managed Short-Term Leguminous Fallows in Western Kenya.............................. 536
37.2.1 Planted Fallow Alternatives in the Crop Rotation ..................................... 537
37.2.2 Soil Microbial Involvement in CropFallow Systems ............................... 537
37.2.3 Effects of Fallows on the Overall System..................................................... 539
37.2.4 Implications of Experience from Western Kenya ....................................... 540
37.3 Some Further Examples of Improved Soil Phosphorus Availability
by Enhancing Soil Biological Activity......................................................................... 540
37.4 Assessing and Improving Phosphorus Fertility in Tropical Soils.......................... 541
37.5 Discussion........................................................................................................................ 542
References ................................................................................................................................... 543
Low availability of phosphorus is a major constraint on agricultural productivity in highly
weathered tropical soils. Such soils have a signicant capacity to sorb large amounts of
phosphorus, taking them out of the soil solution. This limits the availability of inorganic
phosphorus for plants, whether it is already contained in the soil or added as fertilizer.
Further, some tropical soils contain only small amounts of total phosphorus, with
a relatively large proportion of this present in organic forms. This makes biological
531
processes vitally important for enhancing phosphorus availability to crops in tropical
soils, especially those that are receiving organic amendments as their major nutrient
source (Nziguheba and Bu nemann, 2005; Oberson and Joner, 2005).
Soil microbes play a central role in enhancing phosphorus availability to plants because
they mediate the turnover of phosphorus contained in organic amendments and in soil
organic matter. At the same time, the incorporation of phosphorus into microbial cells
prevents its strong sorption to soil constituents, thereby maintaining it in a form that can
be released subsequently into the soil solution following microbial turnover. Microbial
processes are driven by the availability of decomposable organic carbon, which highlights
the importance of sustaining and improving soil organic matter concentrations if large
populations of microbes are to be active in the soil.
Organic amendments such as manures and plant residues are a major source of organic
carbon to the soil. Plants also promote microbial populations and subsequent turnover by
exuding organic carbon from their roots (Merckx et al., 1985). Some plants are well-
adapted to low phosphorus availability owing to their rooting pattern and
root characteristics (Rao et al., 1996), their associations with mycorrhizal fungi (Sieverding,
1991), and/or their ability to take up soil phosphorus from recalcitrant compounds
(George et al., 2002b). When the selection and breeding of germplasm is adapted for these
traits, plants can further enhance the biological cycling of soil phosphorus.
Biological processes that enhance soil phosphorus availability are discussed in detail in
Chapter 13, detailing the turnover of phosphorus through the microbial biomass. Such
processes can be difcult to assess because they are subject to subtle and complex
interactions with various other processes. It is also difcult to synchronize nutrient release
by microbial processes with plant demand, an issue that requires further study.
Consequently, it is difcult to make broad predictions or generalizations about the impact
that biological improvements will have on soil nutrient availability. However, there is
sufcient evidence of positive effects in eld trials from a variety of agroecosystems that
such practices merit serious consideration, even if the scientic basis for understanding the
whole process is incomplete.
In this chapter we report on cases where improvements in soil phosphorus availability
and turnover were achieved by enhancing soil biological activity. We focus on examples
from Colombia and Kenya that combined organic matter management with the use of
adaptedgermplasmandstrategic inputs of lowdoses of phosphorus fertilizers to raise crop
productivity and enhance soil fertility. Several additional examples of similar improve-
ments using different biological interventions are considered briey in the third section.
37.1 Introduced Pastures and Cropping Systems in Eastern Colombia
The eastern plains of Colombia include 17 million ha of land dominated by acidic,
nutrient-poor Oxisols with a well-dened dry season of 3 to 4 months. The herbaceous
native savanna vegetation is of lownutritional value and is used traditionally for extensive
beef production. The introduction into pastures of exotic tropical grasses that are adapted
to acidic soils, mostly Brachiaria spp., either alone or in conjunction with tropical forage
legumes, can increase the productivity of the grazing animals by between 10- and 15-fold
(Lascano and Estrada, 1989).
After cultivars adapted to acidic soils were identied by testing, new alternatives for
agricultural production were introduced into the area. A major soil chemical constraint on
agricultural productivity on the Colombian savannas is the low concentration of total and
plant-available phosphorus (Friesen et al., 1997). To understand and improve the efciency
Biological Approaches to Sustainable Soil Systems 532
of phosphorus use and cycling in production systems in this region, a series of eld
evaluations was undertaken in the eastern plains of Colombia to assess the interactions
among soil phosphorus status (including organic phosphorus), the introduction of
pastures or different cropping systems, and microbial phosphorus turnover (Rao et al.,
2004). Data from these studies are presented in Table 37.1.
37.1.1 Measuring and Explaining Differences in Phosphorus Availability
Improved pastures with new grass species yielded a remarkable increase in beef
production with only modest inputs of mineral phosphate fertilizer. This indicated that the
efciency of fertilization was greater than expected on these soils that otherwise strongly
sorbed phosphorus. To understand the effect of introduced pastures on phosphorus
cycling and availability, phosphorus budgets were estimated, and soil phosphorus status
was characterized for soils from (a) unfertilized native savanna pastures, or (b) fertilized,
introduced pastures. The latter were either (b1) grass-only (Brachiaria decumbens cv.
Basilisk) or (b2) a grasslegume mixture (B. decumbens with Pueraria phaseoloides,
commonly known as kudzu) (Oberson et al., 1999).
The activity of soil phosphatase enzymes and the amount of phosphorus in the soil
microbial biomass were found to be greater in grasslegume than in grass-only or native
savanna pastures. Seasonal analysis of soil phosphorus indicated that the grasslegume
system maintained greater concentrations of organic and plant-available phosphorus with
less temporal variation than in the two other systems. Analysis of the phosphorus
associated with humic and fulvic acids by solution
31
P nuclear magnetic resonance
spectroscopy revealed in the grasslegume soils greater reserves of phosphate diesters
(Guggenberger et al., 1996), which are usually assumed to contribute substantially to plant
nutrition (see Chapter 13). Thus the improvement in soil phosphorus availability, as
determined by several methods, was clearly greater in grasslegume than in grass-only
pastures (Oberson et al., 1999). What could be the reasons for this?
The improved phosphorus availability in grasslegume pastures was not due
to differences in fertilizer inputs or exports from the system; alternatively, it could be
attributed to changes in the overall biological activity in the soil plant system caused
by the presence of legumes in the vegetation cover. Total carbon and organic phosphorus
concentrations (Oberson et al., 1999) as well as macrofaunal activity (Decaens et al., 1994)
were all found to be signicantly greater in grasslegume soils. Roots from legumes
decomposed faster than the roots from Brachiaria grasses (Gijsman et al., 1997a), and
annual root production, in terms of both biomass and length, was measured to be
signicantly greater in the introduced pastures (grass-only and grasslegume) compared
with native pasture (Rao et al., 2001).
Mean aboveground litter production in the introduced pastures during the wet season
was about 40 g m
22
month
21
(Thomas et al., 1993). Greater pasture productivity with
legumes, associated with greater inputs of plant litter and excrement from the grazing
animals in grasslegume pastures, probably provided more consistent organic phos-
phorus inputs and greater phosphorus cycling and availability, which was in agreement
with the results of a simulation of the nitrogen cycle in grasslegume vs. grass-only
pastures (Thomas, 1992).
The increased activity of macrofauna in the introduced pastures interacted positively
with the increased phosphorus cycling, since total phosphorus was markedly higher in
earthworm casts than in the surrounding soil (Jimenez et al., 2003). Increases in microbial
phosphorus and plant-available phosphate in the casts were even more pronounced. This
suggested that greater concentrations of labile organic phosphorus in pasture soils were
linked to a greater abundance of earthworms. Levels of labile organic phosphorus,
Improving Phosphorus Fertility in Tropical Soils 533
TABLE 37.1
Characteristics of Colombian Oxisols Evaluated
Plant-Available P
(mg kg
21
)
Location and
Grassland Type
Soil Type and
Texture
Depth
(cm) pH
Organic C
(g kg
21
)
Total P
(mg kg
21
)
Organic P
(mg kg
21
)
Microbial P
a
(mg kg
21
) (Bray-II) Resin-extractable P Reference
Introduced pastures
replacing native savanna
Native savanna Oxisol: 010 4.8 23.5 179 64 5.2 1.3 3.5
Grass-only 39% clay, 4.85 23.2 228 68 5.9 1.4 4.0 Oberson et al. (1999)
Grasslegume 19% sand 4.96 24.9 226 77 7.3 2.2 5.2
Ricepasture rotations and
rice monocrop replacing
native savanna
Native savanna Oxisol: 010 4.7 26.5 178 85 3.9 5.7 2.1 Gijsman et al. (1997b)
Ricegrasslegume 40% clay, 4.8 31.7 209 81 5.2 7.8 3.0
Rice-only 30% sand 4.7 29.2 221 95 4.6 6.4 3.1 A. Oberson (unpublished)
Ricegrasslegumerice 4.6 27.5 250 101 5.2 21.2 8.3
Ricegrassrice 4.6 30.7 255 95 5.4 23.5 9.4
Continuous rice 4.8 26.8 307 107 3.8 32.2 11.0
Bulk soil and earthworm casts
from native savanna and
grasslegume pasture
b
Native savanna Oxisol: Jimenez et al. (2003)
Soil 39% clay, 015 5.1 21 179 73 2.5 1.0 0.8
Casts 19% sand Surface 5.4 41 267 108 4.0 6.3 7.8
Grasslegume
Soil 015 5.2 21 194 62 4.1 2.5 1.4
Casts Surface 5.8 52 396 136 10.9 11.0 12.8
Introduced pastures and
rice replacing native savanna
Savanna Oxisol: 010 4.9 26.1 216 82 5.4 1.4 2.6 Oberson et al. (2001)
Grasslegume silty clay 4.8 28.4 272 98 6.6 3.4 4.8
Continuous rice 4.7 24.7 354 98 2.6 15.6 14.3
a
All values present microbial phosphorus extracted after soil fumigation (i.e., no conversion factors (k
P
) applied).
b
Same site as in Oberson et al. (1999).
B
i
o
l
o
g
i
c
a
l
A
p
p
r
o
a
c
h
e
s
t
o
S
u
s
t
a
i
n
a
b
l
e
S
o
i
l
S
y
s
t
e
m
s
5
3
4
including phosphate diesters such as DNA, were found to be enriched in earthworm casts
(Guggenberger et al., 1996), which could in turn be linked to the greater phosphorus
availability in pasture soils.
Earthworms, which play a fundamental role in promoting soil nutrient availability
(Jimenez and Decaens, 2004), can be greatly inuenced by changes in land use (Decaens
et al., 1994). Since earthworm species differ in their nutritional behavior and inhabit
different soil layers, the composition of the earthworm community will affect the impact
that these organisms have on phosphorus availability. For instance, there is greater
phosphate availability in earthworm casts of the geophagous earthworm Pontoscolex
corethrurus compared with the bulk soil of an Oxisol, which can be attributed to
a combination of selective ingestion of small particles and the partial mineralization
of organic phosphorus (Chapuis-Lardy et al., 1998).
The endogeic earthworm, Polypheretima elongate, on the other hand, has been found to
increase phosphorus availability by digestive and microbial processes during gut transit
(Brossard et al., 1996). However, in both cases, total phosphorus content did not differ
between earthworm casts and bulk soil. Still different, the anecic species, Martiodrilus
carimaguensis, increases available and total phosphorus content in its casts compared with
that in the bulk soil, presumably through the ingestion of phosphorus from plant residues
or other organic sources (Jimenez et al., 2003). Earthworms play critical roles within soil
food webs and in improving soil physical structure, as discussed in Chapters 5 and 11.
37.1.2 Assessing Phosphorus Turnover
The benecial effect that tropical pastures have on phosphorus cycling through the microbial
biomass was evident when ricepasture rotations replaced native savanna (Gijsman et al.,
1997b). Microbial phosphorus, when measured, was lower under monocropped rice and
savanna soils than in croppasture rotations. In the latter, the activity of the microbial
biomass increased in parallel with the overall improvement in soil fertility.
The relationship between microbial phosphorus cycling and the dynamics of extractable
organic phosphorus fractions has been studied by Oberson et al. (2001), Bu hler et al.
(2002). Grasslegume pastures with greater soil organic phosphorus concentrations were
associated with greater soil biological activity than under native savanna or soils cropped
continuously with rice. Isotopic labeling indicated a rapid microbial phosphorus turnover
that was greatest in grasslegume soils. Phosphorus mineralization rates appeared to be
greater in grasslegume than in continuous-rice soils, suggesting that the former had a
greater potential to render organic phosphorus available to plants (Oberson et al., 2001).
Studies using radio-labeled phosphate have documented the rapid turnover of
phosphorus through both the inorganic and organic pools. In one example, a signicant
proportion of radio-labeled phosphate was recovered in organic phosphorus fractions
2 weeks after labeling, demonstrating that the dynamics of organic phosphorus are
important when soil phosphate availability is limited (Bu hler et al., 2002). Isotopic labeling
has also shownthat phosphorus owedrapidlythrough, but didnot accumulate in, organic
phosphorus pools, at least in the short term (Rao et al., 2004). In an experiment with crop
rotation and ley pasture, chemical fractionation of soil phosphorus indicated that applied
phosphorus fertilizer moved preferentially into labile phosphate pools and then slowly, via
biomass production and microbes, into organic phosphorus pools (Friesen et al., 1997).
37.1.3 Implications of Experience from Colombia
Overall, the results obtained on low-phosphorus, acidic soils in the eastern plains of
Colombia have demonstrated that the combined use of improved germplasmand modest,
Improving Phosphorus Fertility in Tropical Soils 535
strategic inputs of fertilizer resulted in signicantly increased performance of both
pastoral and arable production systems, provided that application to arable systems
is combined with appropriate tillage and crop rotations (Phiri et al., 2003). The increase
in system performance, whether in terms of livestock or crop yields, was clearly related
to enhanced soil biological activity and microbial phosphorus turnover.
However, the widespread adoption of agropastoral systems by farmers requires some
investment in germplasm and fertilizers as well as knowledge of pasture and crop
management techniques. These investments, which occur in conjunction with an
intensication and diversication of the overall farm operations, are likely to be made
only when remunerative markets for the products are accessible. In the Colombian llanos,
local institutions and the Colombian government have restricted their activities and
investments, mainly because of social insecurity (Guimaraes et al., 2004). Even though the
infrastructure is insufcient there, ofcial efforts continue to make this region more
productive and its residents better off.
37.2 Managed Short-Term Leguminous Fallows in Western Kenya
Soils in the West Kenyan highlands are predominantly ne-textured, phosphorus-sorbing
Oxisols and Alsols. The rural population there, with 500 to 1200 people km
22
, is one of the
densest in the world. In the traditional system of shifting cultivation, cropping periods of
1 to 4 years duration were alternated with fallow periods of up to 15 years. This allowed
restoration of soil organic matter by above- and belowground biomass production of
woody secondary vegetation. However, the duration of fallow periods has now declined
to only one to two growing seasons, i.e., one half to one year, due to the high population
density and scarcity of arable land. The biomass production of short, weedy fallows is
usually insufcient to replenish soil organic matter lost during the cropping period. Many
farmers do not fallow their land at all, or only infrequently, with the poorest farmers who
have the least amount of land doing the least fallowing if any. This leads to the further
impoverishment of both their soils and themselves.
Depletion of soil organic matter leads to a corresponding depletion of nutrients, in part,
because mineral fertilizers are too expensive for the rural poor. To address the need to
enhance soil system fertility among poor farmers, the World Agroforestry Centre (ICRAF)
has developed crop rotations that include more productive fallows, often involving
planted legumes, which restore soil fertility and organic matter (Sanchez, 1999). Results of
this research program are reported from Zambia and some other countries in Chapter 19
under the rubric of fertilizer trees. Here, we examine one application of this concept
with particular interest in phosphorus mobilization.
Leguminous fallow plants can x signicant amounts of nitrogen from the atmosphere,
although the proportion of xed nitrogen that builds up in plant biomass as well as the
total amount of xed nitrogen will vary according to the species. Gathumbi et al. (2002)
reported that in the absence of phosphorus and potassium limitation, the proportions of
total plant nitrogen that were xed from atmospheric nitrogen ranged between 35 and
83%, with the amount of xed nitrogen being between 8 and 140 kg ha
21
. Of the seven
leguminous species studied, Crotalaria grahamiana yielded the highest values of xed
nitrogen. The leguminous fallow tree Sesbania sesban is particularly effective in obtaining
nitrogen from the subsoil (Hartemink et al., 1996).
However, despite its overall importance, nitrogen may not be the most limiting nutrient
in many situations. On many soils in western Kenya, phosphorus availability is the main
Biological Approaches to Sustainable Soil Systems 536
constraint, as seen from the fact that nitrogen fertilization without the simultaneous
addition of phosphorus often fails to increase maize yields (Jama et al., 1997). Even in the
absence of external phosphorus inputs, annual maize yields were signicantly higher in
maizelegume fallow rotations than in continuous maize or maize-natural fallow
rotations (Niang et al., 2002; Smestad et al., 2002). This suggests that phosphorus
availability was enhanced by alternation with a legume fallow. Secondary effects of fallows
on maize growth, e.g., modied soil moisture conditions, may also play a role.
37.2.1 Planted Fallow Alternatives in the Crop Rotation
We consider here data on the effects of a leguminous fallow on soil phosphorus dynamics
from a case study that lasted 5.5 years (Bu nemann, 2003) (Table 37.2). It investigated the
relationships among organic matter inputs made through the fallow plant C. grahamiana,
soil microbial activity, phosphorus availability, and maize crop performance (Bu nemann
et al., 2004b). The eld experiment, established on an Oxisol, included three maize-based
rotations (continuous maize, maizeCrotalaria fallow, and maize-natural fallow rotation)
with two levels of phosphorus fertilization (0 and 50 kg P ha
21
, applied as triple
superphosphate). The maize in each rotation was grown during the long rainy season
lastingfromMarchtoAugust. The fallows were maintainedinthe rotations duringthe short
rainy season lasting from September to February, while in the continuous maize system
a second maize crop was planted during the short rains.
Seasonal grain yield in the continuous maize system without phosphorus fertilization
was low, varying between 0.1 and 1.5 t ha
21
. Yields during the short rains are usually poor
because the rain is unreliable, and there is little or no break between long-rain harvest and
short-rain planting. This increases pest pressure, which in turn spreads the maize streak
virus that reduces yields. In all rotations, maize production was roughly doubled by
phosphorus fertilization, indicating signicant phosphorus deciency at this site. Maize
grain yields reported from the same region by other authors have averaged about 1 t ha
21
without and 2 t ha
21
with phosphorus fertilization (Maroko et al., 1999; Nziguheba et al.,
2000; Kwabiah et al., 2003). This low productivity can be due to a variety of factors, such as
limited water availability, deciency in nutrients other than phosphorus, and weed,
disease, and pest problems.
Cumulative maize production over 5.5 years did not differ signicantly among the three
rotations, indicating that the maize yield forgone during the fallow season was
compensated for by higher postfallow yields (Figure 37.1). During the rst fallow season,
the Crotalaria fallow was more productive than the natural fallow, recycling 5.3 t dry
matter ha
21
compared with 3.2 t ha
21
for the natural fallow. The subsequent maize yield
was doubled after the Crotalaria fallow at both levels of phosphorus fertilization.
However, Crotalaria growth decreased during the course of the experiment due to pest
problems, such as infestation with planthopper and subsequent colonization of stem
lesions by saprophytic fungi (Girma, 2002). As a consequence, with longer duration, the
introduced fallow was no longer benecial for the subsequent maize crop compared with
continuous maize, and at the end of the eld experiment, the cumulative maize yield
was similar between the three rotations (Figure 37.1).
37.2.2 Soil Microbial Involvement in CropFallow Systems
In terms of changes in the soil system, both fallow types reversed the trend of soil organic
matter depletion observed under continuous maize. The highest concentrations of soil
organic matter and microbial nutrients were found in the maizeCrotalaria rotation, while
plant-available soil phosphate concentrations were similar in all three rotations, being
Improving Phosphorus Fertility in Tropical Soils 537
TABLE 37.2
Characteristics of Kenyan Oxisols Evaluated
Location and Grassland
Type
Soil Type and
Texture
Depth
(cm) pH
Organic C
(g kg
21
)
Total P
(mg kg
21
)
Organic P
(mg kg
21
)
Microbial P
a
(mg kg
21
)
Plant-Available P
(Resin-P) (mg kg
21
) Reference
Maize cropping systems, zero
phosphorus treatment
Continuous maize Oxisol: 015 5.0 24.0 720 271 3.4 1.8 Bu nemann (2003),
Bu nemann et al.
(2004b)
Weed fallow 39% clay, 5.0 25.8 703 267 5.2 1.7
Crotalaria fallow 37% sand 5.0 25.6 721 282 6.2 1.7
Maize cropping systems,
fertilized treatment
(50 kg P ha
21
year
21
)
Same
Continuous maize Oxisol: 015 4.8 23.7 838 257 3.5 6.9
Natural fallow 39% clay, 5.1 25.2 837 278 5.4 6.7
Crotalaria fallow 37% sand 5.0 26.3 829 289 6.6 6.3
a
All values present microbial phosphorus extracted after soil fumigation (i.e., no conversion factors (k
P
) applied).
B
i
o
l
o
g
i
c
a
l
A
p
p
r
o
a
c
h
e
s
t
o
S
u
s
t
a
i
n
a
b
l
e
S
o
i
l
S
y
s
t
e
m
s
5
3
8
signicantly increased by phosphorus fertilization (Bu nemann et al., 2004b) (Table 37.2).
The composition of the microbial community, as indicated by phospholipid fatty acid
analysis, differed somewhat between soils under continuous maize and in maize
Crotalaria rotation (Bu nemann et al., 2004a). In particular, soils with greater microbial
biomass concentrations contained a greater relative abundance of fungi and Gram-
negative bacteria.
An incubation study with soils fromthis eld experiment indicated that microorganisms
took up phosphorus from stable pools in the soil following the addition of C. grahamiana
residues labeled with radioactive phosphate (Bu nemann et al., 2004c). Some incorporation
of labeled phosphate into stable soil organic phosphorus supported the shift toward
organic phosphorus observed in fallowed vs. monocropped soils. However, during the 9-
week incubation period, a large proportion of phosphorus from the residues remained in
the microbial biomass. This prevented sorption of phosphorus to the soil matrix, but
further work is needed to understand how microbial immobilization and remineralization
cycles can be managed in order to render microbial phosphorus more available to eld
crops grown after fallow periods. Environmental alternations, such as drying and
rewetting of soils, may solubilize phosphorus immobilized in the soil microbial biomass
(Chapter 13).
The shift toward organic and microbial nutrients in maizefallow rotations was also
observed by Smestad et al. (2002), who measured increases in phosphorus contained in the
soil microbial biomass, particulate soil organic matter, and stable soil organic phosphorus.
The activation of native soil phosphorus by microbial uptake in response to organic matter
addition to highly weathered soils has also been reported by Guppy (2003), but cannot be
expected if total soil phosphorus contents are very low (Compaore et al., 2003).
37.2.3 Effects of Fallows on the Overall System
Studies in southern Cameroon have observed no benecial effect of Calliandra tree fallows
on the yield of the main crops (Nolte et al., 2005). This indicated that short-term fallow
plants, which provide an immediate benet to the farmer, e.g., through the production
of rewood, even if readily adopted can lead to substantial nutrient removal from the
c
u
m
u
l
a
t
i
v
e

g
r
a
i
n

y
i
e
l
d

(
M
g

h
a

1
)
0
5
10
15
20
c
c
bc
ab
a
bc
COM MCF MNF COM MCF MNF
P + P
Short rain
Long rain
FIGURE 37.1
Cumulative maize grain yields during 5.5 years of eld experimentation. COM, continuous maize; MCF,
maizecrotalaria fallow, MNF, maize-natural fallow. Source: Authors data.
Improving Phosphorus Fertility in Tropical Soils 539
system (Nolte et al., 2003). Managed fallows may therefore lead to more negative
phosphorus balances than those reported generally for sub-Saharan Africa (Smaling et al.,
1997) if there are no corresponding phosphorus inputs (Bu nemann et al., 2004b).
Problems arising from pests when fallow systems lack biodiversity were observed in
other experiments as well as on farms (Drechsel et al., 1996; Smestad et al., 2002).
Of particular concern is an increase in pathogenic nematodes (Desaeger and Rao, 2000).
These examples demonstrate the need to test carefully the various biophysical effects
of inserting particular fallow plants into cropping systems. They also conrm the value of
participatory approaches to improve adoption by farmers. This does not diminish the
value of systematic screening efforts that assess which of many leguminous species can,
from a soil fertility standpoint, best serve as managed fallows (Niang et al., 2002).
Besides leguminous fallows, a nonleguminous shrub in the family Asteraceae,
Tithonia diversifolia, has been widely tested as a fallow and green manure plant (Jama
et al., 2000). It can improve phosphorus availability to crops through the acquisition of
stable soil phosphorus, including organic phosphorus (George et al., 2002a), with
subsequent rapid release of this phosphorus following residue amendment (Kwabiah
et al., 2003). Being a nonlegume, it makes no net nutrient input of any element except
carbon, although it may transfer nutrients from hedgerows or eld boundaries onto arable
areas (Jama et al., 2000).
37.2.4 Implications of Experience from Western Kenya
Short-term fallows have the potential to reverse soil organic matter depletion and to
increase phosphorus held in the soil microbial biomass and in more stable soil organic
phosphorus pools. They can mobilize phosphorus held in recalcitrant soil phosphorus
pools through the effect of plant roots (George et al., 2002b) and enhanced microbial
activity (Bu nemann et al., 2004c). They may also reduce the phosphorus sorption capacity
of soils during decomposition (Nziguheba et al., 1998).
It is noteworthy that these effects, which indicate enhanced soil organic phosphorus
turnover, are not necessarily measurable as an increase in plant-available phosphate
(next section). All these processes increase the availability of soil phosphorus, but must be
complemented by balanced addition of mineral or organic fertilizer to prevent nutrient
depletion of the soil (Smithson and Giller, 2002). Furthermore, attention has to be paid to
the diversity of planted fallow species in order to avoid pest problems.
37.3 Some Further Examples of Improved Soil Phosphorus
Availability by Enhancing Soil Biological Activity
The two experiences considered above focused on combining organic matter management
with the use of selected germplasm and strategic inputs of low doses of phosphorus
fertilizer. However, there are some good examples of similar improvements made by
enhancing soil phosphorus availability through other kinds of biological interventions.
Planted tree or shrub fallows of Calliandra calothyrsus, Indigofera constricta, and
T. diversifolia increased the amounts of carbon, nitrogen, and phosphorus in the sand-
sized soil organic matter in a volcanic-ash soil of the hillsides in southwestern Colombia
(Phiri et al., 2001). Similarly, tree crops improved soil phosphorus availability to perennial
crops in central Amazonian Oxisols (Lehmann et al., 2001). In the Amazon research
reported in the preceding chapter, enhanced phosphorus cycling through the soil
microbial biomass and between plants and soil meant that the need for phosphorus
Biological Approaches to Sustainable Soil Systems 540
fertilizer application could be reduced to only as much as would replenish the phosphorus
exported in harvested crops.
Agroforestry systems in Brazil have improved phosphorus availability in a shaded
coffee cultivation system compared with the conventional unshaded system, apparently
by promoting the turnover of organic phosphorus (Cardoso et al., 2003). In agroforestry
production systems established on sandy savanna soils in Venezuela, the continued
addition of animal manure improved soil physical and chemical characteristics
and enhanced soil microbial activity (Lopez-Hernandez et al., 2004). Enhanced
earthworm abundance was related to greater enzyme activities and microbial biomass.
The concentrations of both inorganic and organic phosphorus in the soil were increased.
Leguminous cover crops grown in the interspaces of coconut plantations have increased
plant-available phosphorus and the rates of biochemical processes in a sandy clayloam
soil in a humid tropical region of India (Dinesh et al., 2004).
Soil microbiological and biochemical parameters were sensitive indicators of soil quality
under managed fallow systems with herbaceous or shrubby legumes established on sites
with varying degrees of soil degradation in southwestern Nigeria (Wick et al., 1998).
In particular, changes in alkaline phosphatase activity can highlight interactions between
organic phosphorus dynamics and overall soil fertility in tropical agroecosystems.
Greater soil microbial activity by itself may be benecial only if the system is modied.
Fire-free alternatives to smallholders slash-and-burn agriculture have been evaluated
in the eastern Amazon by Denich et al. (2004). In this region, a 2-year cropping period is
alternated with 3 to 7 years of fallow, during which time a woody secondary-forest
vegetation regenerates by resprouting from the roots. Traditional burning of fallow
vegetation causes most of the carbon, nitrogen, and phosphorus stocks in the aboveground
biomass to be lost by volatilization or ash-particle transfer (Sommer et al., 2004). In contrast,
the use of mechanized chop-and-mulch technology for land preparation avoids the
nutrient losses caused by burning, yet crops grown in the mulch layer do not immediately
benet. This occurs because the availability of nutrients, especially phosphorus, is reduced
by microbial immobilization (Bu nemann et al., 1998). This probably explains, at least in
some cases, the lower yields in mulched compared with burned elds (Kato et al., 1999).
With moderate quantities of NPK fertilizer, the rst two crops (rice and cowpea)
performed equally well in burned and mulched areas. Only the last crop of the sequence,
cassava, did not require fertilizer to yield as well in mulched as in burned areas.
The change from burning to mulching certainly stimulates soil biological activity and
preserves soil organic matter and associated nutrients. To be successful, however, there
may need to be other modications of the system. These can include selection of cultivars
adapted to the more acidic surface soils in mulched as compared with burned areas, and
inversion of the cropping sequence by planting rst the crop with the lowest nutrient
requirements. This cascade of required changes poses a challenge to farmers and
researchers. The mulch system, compared with traditional burning, requires additional
labor and changes in the cropping system as well as certain agronomic changes. However,
it offers opportunities such as more exible planting dates and an extended cropping
phase that allows a maximization of benets from the slowly decomposing mulch layer
(Denich et al., 2004).
37.4 Assessing and Improving Phosphorus Fertility in Tropical Soils
Assessing the success of biological approaches to improving phosphorus availability in
tropical soils remains a challenge. In current agricultural practice, phosphorus availability
Improving Phosphorus Fertility in Tropical Soils 541
in soils is determined by measuring the size of a phosphate pool, for example, by
extraction with bicarbonate (e.g., Olsen et al., 1954), by dilute mineral acid (e.g., Bray and
Kurtz, 1945), or by anion exchange resins (Amer et al., 1955). Such snapshot
measurements of phosphate availability are related empirically to observed crop growth
using eld experiments, and these relationships are then used to determine fertilizer
requirements.
Such methods adapted and calibrated for the specic soil under study can provide
relevant information on the amount of phosphate that is available to the plant in a tropical
soil (Bu hler et al., 2003). However, they do not provide information on the turnover of
(organic) phosphorus, even though this is closely related to the actual phosphorus
availability in tropical soils (Tiessen and Shang, 1998). In fact, a low extractable phosphate
concentration may conceal a rapid rate of organic phosphorus turnover that contributes
signicantly to the supply of phosphorus to plants. This is apparent when conventionally
managed farmland is rst brought under organic cultivation (Oberson and Frossard, 2005).
The dependence of phosphorus fertility on organic matter turnover leads to feedback
mechanisms that confound further the assessment of phosphorus availability in tropical
soils by chemical tests. For example, if it is organic phosphorus turnover that regulates the
availability of phosphate, then a decrease in the rate of turnover may place a nutrient
limitation on organic matter decomposition. This, in turn, will constrain the turnover
of organic phosphorus (Tiessen and Shang, 1998).
Such complexity means that phosphorus availability in tropical cropping systems based
on organic matter inputs cannot be assessed accurately by conventional soil testing
procedures. These were developed in industrialized countries of the temperate zone
where mineral fertilizer is readily accessible to farmers, and the soils have generally less
capacity to sorb phosphorus compared with those in many tropical regions. In temperate
soils, a single measurement of extractable phosphate can yield results with some relevance
to crop uptake, whereas in tropical soils there is a clear need also to assess continually the
rates of phosphorus turnover (Organic Phosphorus Workshop, 2005). Unfortunately, there
are currently no straightforward methods for assessing organic phosphorus turnover
in tropical soils.
37.5 Discussion
Productivity of many tropical agricultural systems must be urgently increased to meet the
needs of a growing and often malnourished world population. Phosphorus is a crop-
nutrient element that is fundamental in any attempt to achieve this objective because
crop production in the tropics is so often limited by the availability of soil phosphorus.
For most farmers in tropical regions, mineral phosphate fertilizer is an expensive and often
inaccessible means of improving phosphorus fertility. The case studies from the eastern
plains of Colombia and western Kenya show that biological interventions are available
that are relatively simple and low cost and can improve phosphorus fertility in tropical
agroecosystems. The common feature of these interventions is the input of organic matter
into the soil, mainly through the residues from leguminous pasture or fallow plants that
are adapted to low-phosphorus tropical soils.
The use of well-adapted germplasm together with low doses of phosphorus fertilizers
can enhance system productivity, in turn adding to the supply of organic residues above-
and belowground. Phosphorus acquisition varies according to germplasm much as
symbiotic nitrogen xation does in legumes, with the quality as well as the amount of
residues returned to the soil being important. Since organic matter boosts soil microbial
Biological Approaches to Sustainable Soil Systems 542
activity, the microbiologically-driven processes in soil phosphorus dynamics are
enhanced, and the microbial phosphorus pool is increased when greater amounts of
organic matter are made available. Phosphorus recycles more efciently among plants,
microorganisms, and organic forms of phosphorus in the soil where it is protected from
strong sorption in highly weathered tropical soils. Enhanced turnover can increase
phosphorus availability for crops.
This said, the widespread adoption of biological interventions by farmers requires
investments in germplasm, fertilizers, and knowledge of pasture and/or crop manage-
ment techniques. Such investments, which must occur alongside an intensication and
diversication of the overall farm operations, are made only when farmers perceive
and expect tangible benets from self-supply and when favorable markets for their
produce are accessible. Finally, sustainable agricultural systems require that phosphorus
inputs and outputs be in balance over the long term, in order to avoid an excessive
depletion of soil phosphorus stocks.
References
Amer, F. et al., Characterisation of soil phosphorus by anion exchange resin adsorption and 32P
equilibration, Plant Soil, 6, 391408 (1955).
Bray, R.H. and Kurtz, L.T., Determination of total, organic and available forms of phosphorus in
soils, Soil Sci., 59, 3945 (1945).
Brossard, M., Lavelle, P., and Laurent, J.Y., Digestion of a vertisol by the endogeic earthworm
Polypheretima elongata, Megascolecidae, increases soil phosphate extractability, Eur. J. Soil Biol.,
32, 107111 (1996).
Bu hler, S. et al., Sequential phosphorus extraction of a P-33-labeled oxisol under contrasting
agricultural systems, Soil Sci. Soc. Am. J., 66, 868877 (2002).
Bu hler, S. et al., Isotope methods for assessing plant available phosphorus in acid tropical soils,
Eur. J. Soil Sci., 54, 605616 (2003).
Bu nemann, E., Phosphorus dynamics in a Ferralsol under maizefallow rotations: The role of
the soil microbial biomass. Ph.D. Thesis, Swiss Federal Institute of Technology, Zurich (2003).
http://e-collection.ethbib.ethz.ch/cgi-bin/show.pl?type diss&nr 15207.
Bu nemann, E. et al., Fertilizer response of maize and cowpea under conditions of re-free land
preparation in NE Para, In: Proceedings of the Third SHIFT-Workshop, Manaus, March 1519, 1998,
Lieberei, R., Vo, K., and Bianchi, H., Eds., BMBF/GKSS, Hamburg, 157159 (1998).
Bu nemann, E. et al., Microbial community composition and substrate use in a highly weathered
soil as affected by crop rotation and P fertilization, Soil Biol. Biochem., 36, 889901 (2004a).
Bu nemann, E. et al., Maize productivity and nutrient dynamics in maizefallowrotations in western
Kenya, Plant Soil, 264, 195208 (2004b).
Bu nemann, E. et al., Phosphorus dynamics in a highly weathered soil as revealed by isotopic labeling
techniques, Soil Sci. Soc. Am. J., 68, 16451655 (2004c).
Cardoso, I.M. et al., Analysis of phosphorus by (PNMR)-P-31 in Oxisols under agroforestry and
conventional coffee systems in Brazil, Geoderma, 112, 5170 (2003).
Chapuis-Lardy, L. et al., Phosphorus transformations in a ferralsol through ingestion by Pontoscolex
corethrurus, a geophagous earthworm, Eur. J. Soil Biol., 34, 6167 (1998).
Compaore, E. et al., Inuence of land-use management on isotopically exchangeable phosphate
in soils from Burkina Faso, Commun. Soil Sci. Plant Anal., 34, 201223 (2003).
Decaens, T. et al., Impact of land management on soil macrofauna in the Oriental Llanos of
Colombia, Eur. J. Soil Biol., 30, 157168 (1994).
Denich, M. et al., Mechanized land preparation in forest-based fallow systems: The experience from
Eastern Amazonia, Agroforest. Syst., 61, 91106 (2004).
Improving Phosphorus Fertility in Tropical Soils 543
Desaeger, J. and Rao, M.R., Parasitic nematode populations in natural fallows and improved cover
crops and their effects on subsequent crops in Kenya, Field Crops Res., 65, 4156 (2000).
Dinesh, R. et al., Long-term inuence of leguminous cover crops on the biochemical properties of
a sandy clay loam Fluventic Sulfaquent in a humid tropical region of India, Soil Till. Res., 77,
6977 (2004).
Drechsel, P., Steiner, K.G., and Hagedorn, F., A review on the potential of improved fallows and
green manure in Rwanda, Agroforest. Syst., 33, 109136 (1996).
Friesen, D.K. et al., Phosphorus acquisition and cycling in crop and pasture systems in low fertility
tropical soils, Plant Soil, 196, 289294 (1997).
Gathumbi, S.M., Cadisch, G., and Giller, K.E., N-15 natural abundance as a tool for assessing
N-2-xation of herbaceous, shrub and tree legumes in improved fallows, Soil Biol. Biochem., 34,
10591071 (2002).
George, T.S. et al., Changes in phosphorus concentrations and pH in the rhizosphere of some
agroforestry and crop species, Plant Soil, 246, 6573 (2002a).
George, T.S. et al., Utilisation of soil organic P by agroforestry and crop species in the eld, western
Kenya, Plant Soil, 246, 5363 (2002b).
Gijsman, A.J., Alcaron, H.F., and Thomas, R.T., Root decomposition in tropical grasses and legumes,
as affected by soil texture and season, Soil Biol. Biochem., 29, 14431450 (1997a).
Gijsman, A.J. et al., Nutrient cycling through microbial biomass under ricepasture rotations
replacing native savanna, Soil Biol. Biochem., 29, 14331441 (1997b).
Girma, H., Insect pest problems and their management in planted fallow crop rotation in western
Kenya. Ph.D. Thesis, Kenyatta University, Nairobi, Kenya (2002).
Guggenberger, G. et al., Assessing the organic phosphorus status of an Oxisol under tropical
pastures following native savanna using 31P NMR spectroscopy, Biol. Fertil. Soils, 23, 332339
(1996).
Guimaraes, E.P. et al., Research on agropastoral systems: What we have learned and what we should
do, In: Agropastoral Systems for the Tropical Savannas of Latin America, Guimaraes, E.P. et al., Eds.,
CIAT/EMBRAPA, Cali, Colombia/Brazil, 326336 (2004).
Guppy, C.N., Phosphorus and organic matter interactions in highly weathered soils. Ph.D. Thesis,
University of Queensland, Australia (2003).
Hartemink, A.E. et al., Soil nitrate and water dynamics in sesbania fallows, weed fallows and maize,
Soil Sci. Soc. Am. J., 60, 568574 (1996).
Jama, B., Swinkels, R.A., and Buresh, R.J., Agronomic and economic evaluation of organic and
inorganic sources of phosphorus in Western Kenya, Agron. J., 89, 597604 (1997).
Jama, B. et al., Tithonia diversifolia as a green manure for soil fertility improvement in western Kenya:
A review, Agroforest. Syst., 49, 201221 (2000).
Jimenez, J.J. and Decaens, T., The impact of soil organisms on soil functioning under neotropical
pastures: A case study of a tropical anecic earthworm species, Agric. Ecosyst. Environ., 103,
329342 (2004).
Jimenez, J.J. et al., Phosphorus fractions and dynamics in surface earthworm casts under native
and improved grasslands in a Colombian savanna oxisol, Soil Biol. Biochem., 35, 715727
(2003).
Kato, M.S.A. et al., Fire-free alternatives to slash-and-burn for shifting cultivation in the eastern
Amazon region: The role of fertilizers, Field Crops Res., 62, 225237 (1999).
Kwabiah, A.B. et al., Phosphorus availability and maize response to organic and inorganic
fertilizer inputs in a short term study in western Kenya, Agric. Ecosyst. Environ., 95, 4959
(2003).
Lascano, C. and Estrada, J., Long-term productivity of legume-based and pure grass pastures in
the eastern plains of Colombia, Proceedings of the XVI International Grassland Conference, Nice,
France, 11791180 (1989).
Lehmann, J. et al., Phosphorus management for perennial crops in central Amazonian upland soils,
Plant Soil, 237, 309319 (2001).
Lopez-Hernandez, D. et al., Changes in soil properties and earthworm populations induced by long-
term organic fertilization of a sandy soil in the Venezuelan Amazonia, Soil Sci., 169, 188194
(2004).
Biological Approaches to Sustainable Soil Systems 544
Maroko, J.B., Buresh, R.J., and Smithson, P.C., Soil phosphorus fractions in unfertilized fallowmaize
systems on two tropical soils, Soil Sci. Soc. Am. J., 63, 320326 (1999).
Merckx, R., den Hartog, A., and Van Veen, J.A., Turnover of root-derived material and
related microbial biomass formation in soils of different texture, Soil Biol. Biochem., 17,
565569 (1985).
Niang, A.I. et al., Species screening for short-termplanted fallows in the highlands of western Kenya,
Agroforest. Syst., 56, 145154 (2002).
Nolte, C. et al., Effects of Calliandra planting pattern on biomass production and nutrient
accumulation in planted fallows of southern Cameroon, Forest Ecol. Manage., 179, 535545
(2003).
Nolte, C. et al., Groundnut, maize and cassava yields in mixed-food crops elds after Calliandra tree
fallow in southern Cameroon, Exp. Agric., 41, 2137 (2005).
Nziguheba, G. and Bu nemann, E., Organic phosphorus dynamics in tropical agroecosystems,
In: Organic Phosphorus in the Environment, Turner, B.L., Frossard, E., and Baldwin, D.S.,
Eds., CAB International, Wallingford, UK, 243268 (2005).
Nziguheba, G. et al., Organic residues affect phosphorus availability and maize yields in a Nitisol
of western Kenya, Biol. Fertil. Soils, 32, 328339 (2000).
Nziguheba, G. et al., Soil phosphorus fractions and adsorption as affected by organic and inorganic
sources, Plant Soil, 198, 159168 (1998).
Oberson, A. and Frossard, E., Phosphorus management for organic agriculture, In: Phosphorus:
Agriculture and the Environment, Sims, J.T. and Sharpley, A.N., Eds., ASA, CSSA and SSSA,
Madison, WI, 761779 (2005).
Oberson, A. and Joner, E.J., Microbial turnover of phosphorus in soil, In: Organic Phosphorus in the
Environment, Turner, B.L., Frossard, E., and Baldwin, D.S., Eds., CAB International, Wallingford,
UK, 133164 (2005).
Oberson, A. et al., Phosphorus status and cycling in native savanna and improved pastures on
an acid low-P Colombian Oxisol, Nutr. Cycl. Agroecosyst., 55, 7788 (1999).
Oberson, A. et al., Phosphorus transformations in an Oxisol under contrasting land-use systems:
The role of the soil microbial biomass, Plant Soil, 237, 197210 (2001).
Olsen, S.R. et al., Estimation of Available Phosphorus in Soils by Extraction with Sodium Bicarbonate,
United States Department of Agriculture, Washington, DC (1954).
Organic Phosphorus Workshop, Synthesis and recommendations for future research, In: Organic
Phosphorus in the Environment, Turner, B.L., Frosssard, E., and Baldwin, D.S., Eds., CAB
International, Wallingford, UK, 377380 (2005).
Phiri, S. et al., Changes in soil organic matter and phosphorus fractions under planted fallows and
a crop rotation system on a Colombian volcanic-ash soil, Plant Soil, 231, 211223 (2001).
Phiri, S. et al., Constructing an arable layer through chisel tillage and agropastoral systems
in tropical savanna soils of the llanos of Colombia, J. Sustain. Agric., 23, 529 (2003).
Rao, I.M., Plazas, C., and Ricaurte, J., Root turnover and nutrient cycling in native and introduced
pastures in tropical savannas, In: Plant Nutrition: Food Security and Sustainability of Agro-
Ecosystems through Basic and Applied Research, Horst, W.J. et al., Eds., Kluwer Academic
Publishers, Dordrecht, Netherlands, 976977 (2001).
Rao, I.M. et al., Adaptive attributes of tropical forage species to acid soils I. Differences in shoot and
root growth responses to varying phosphorus supply and soil type, J. Plant Nutr., 19, 323352
(1996).
Rao, I.M. et al., Soil phosphorus dynamics, acquisition and cycling in croppasturefallow systems
in low fertility tropical soils of Latin America, In: Modelling Nutrient Management in Tropical
Cropping Systems, Delve, R.J. and Probert, M.E., Eds., Australian Center for International
Agricultural Research (ACIAR), Canberra, Australia, 126134 (2004).
Sanchez, P.A., Improved fallows come of age in the tropics, Agroforest. Syst., 47, 312 (1999).
Sieverding, E., VesicularArbuscular Mycorrhiza Management in Tropical Agrosystems. Deutsche
Gesellschaft fu r Technische Zusammenarbeit, Eschborn, Germany (1991).
Smaling, E.M.A., Nandwa, S.M., and Janssen, B.H., Soil fertility in Africa is at stake, In: Replenishing
Soil Fertility in Africa, Buresh, R.J., Sanchez, P.A., and Calhoun, F., Eds., Soil Science Society
of America and American Society of Agronomy, Madison, WI, 4761 (1997).
Improving Phosphorus Fertility in Tropical Soils 545
Smestad, T.B., Tiessen, H., and Buresh, R.J., Short fallows of Tithonia diversifolia and Crotalaria
grahamiana for soil fertility improvement in western Kenya, Agroforest. Syst., 55, 181194 (2002).
Smithson, P.C. and Giller, K.E., Appropriate farm management practices for alleviating N and P
deciencies in low-nutrient soils of the tropics, Plant Soil, 245, 169180 (2002).
Sommer, R. et al., Nutrient balance of shifting cultivation by burning or mulching in the Eastern
Amazon: Evidence for subsoil nutrient accumulation, Nutr. Cycl. Agroecosyst., 68, 257271
(2004).
Thomas, R.J., The role of the legume in the nitrogen cycle of productive and sustainable pastures,
Grass Forage Sci., 47, 133142 (1992).
Thomas, R.J. et al., Nutrient cycling via forage litter in tropical grass/legume pastures. In: XVII
International Grassland Congress, Hodgson J, Ed., NZGA, TGSA, NZSAP, ASAP-Qld. and NZIAS,
Palmerston North, New Zealand, 508509 (1993).
Tiessen, H. andShang, C., Organic matter turnover intropical landuse systems, In: Carbonand Nutrient
Dynamics in Natural and Agricultural Tropical Ecosystems, Bergstro m, L. and Kirchmann, H., Eds.,
CAB International, Wallingford, UK, 114 (1998).
Wick, B., Ku hne, R.F., and Vlek, P.L.G., Soil microbiological parameters as indicators of soil quality
under improved fallow management systems in south-western Nigeria, Plant Soil, 202, 97107
(1998).
Biological Approaches to Sustainable Soil Systems 546
Author Query
JOB NUMBER: BK 12237
TITLE: Improving Phosphorus Fertility in Tropical Soils
Q1 Please provide city for afliations CIMMYT, CIAT, Berea College &
Smithsonian Tropical Research Institute.