Biological approaches to sustainable soil systems - Part 3 potx

PART III: STRATEGIES
AND METHODS
q 2006 by Taylor & Francis Group, LLC
18
Integrated Soil Fertility Management in Africa:
From Knowledge to Implementation
Bernard Vanlauwe, Joshua J. Ramisch and Nteranya Sanginga
Tropical Soil Biology and Fertility Institute, CIAT, Nairobi, Kenya
CONTENTS
18.1 Problems Driving Research and Development for Sustainable
Soil Systems in Africa 258
18.2 From an External-Input Paradigm to an Integrated Soil
Fertility Management Paradigm 259
18.2.1 The Search for Less Input-Dependent Agricultural Systems 260
18.2.2 The Search for Optimizing Strategies 260
18.2.2.1 Integrated Soil Fertility Management 260
18.2.2.2 Tropical Soil Biology and Fertility Research 261
18.3 Translating Science into Practice 262
18.3.1 The Organic Resource Quality Concept and
Organic Matter Management 263
18.3.2 Exploring Positive Interactions between Mineral
and Organic Inputs 266
18.4 Challenges and the Way Forward 268
18.4.1 Adjusting to Variability at the Farm and Community Levels 269
18.4.2 Use of Adapted Germplasm to Overcome Abiotic and Biotic
Constraints and Create More Resilient Cropping Systems 269
18.4.3 Market-Led Integrated Soil Fertility Management 269
18.4.4 Scaling Up 270
18.4.5 Policy Changes 270
Acknowledgments 270
References 271

Sustainable management of soil, water, and other natural resources is the most critical
challenge confronting agricultural research and development in sub-Saharan Africa (SSA).
Soil fertility decline is a multi-faceted problem and, in ecological parlance, a “slow
variable,” one that interacts pervasively over time with a wide range of other factors,
biological, and socio-economic. Sustainable agroecosystem management is not just a
matter of remedying deficiencies in soil nutrients. Impediments include mismatched
germplasm and faulty cropping system design, the multiple interactions of crops with
pests and diseases, reinforcing feedback effects between poverty and land degradation,
257
q 2006 by Taylor & Francis Group, LLC
institutional failures, and often perverse incentives that stem from national policies and
global dynamics. Dealing with soil fertility issues in cost-effective and sustainable ways
thus requires a long-term perspective and a holistic approach such as embodied in the
concept of integrated soil fertility management (ISFM).
The concepts of ISFM grew out of a series of paradigm shifts generated through
experience in the field and from changes in the overall socio-economic and political
environments faced by the various stakeholders, in particular, by farmers and researchers.
In retrospect, the need for and elements of this integrated strategy should have been
obvious much sooner than they were, but this is true for many advances in thinking and
practice. We now understand better how the judicious use of mineral fertilizers together
with organic sources of nutrients for plants and soil organisms supported by appropriate
soil and water conservation and land and crop management measures can counteract the
agricultural resource degradation that results from nutrient mining, the exploitation of
fragile lands, and associated losses in biodiversity. Appropriate soil fertility management
will produce benefits that reach beyond the farm, serving whole societies through the
various ecosystem services associated with the soil resource base, e.g., provision of clean
water, erosion control, and support for biodiversity.
Part III of this book presents a series of cases and analyses where new as well as often
old knowledge is being drawn on to inform and formulate improved practices that
can achieve more productive and more sustainable soil systems. In this chapter, after

highlighting some of the problems underlying declining soil fertility in SSA, the region
where we have been working, we briefly review some shifts in paradigms related to
tropical soil fertility management. Several examples are then considered of how science
has been translated into practice, with some discussion in conclusion of the challenges that
persist and how we envisage addressing them.
18.1 Problems Driving Research and Development for
Sustainable Soil Systems in Africa
The fertility status of most soils in SSA is generally poor due to low inherent quality and
inappropriate management practices, the latter being the result of various other secondary
and tertiary causes. This dynamic is seen from a number of observations that have
specified the nature of soil systems’ deficiencies and vulnerabilities in the region:
† Sharply negative soil nutrient balances at the regional and national scale for the
major plant nutrients, with annual losses of NPK estimated at 8 million tons
(Stoorvogel and Smaling, 1990). These negative balances reflect the very low use of
mineral inputs across SSA, although they also show the effects of climatic and
other conditions discussed in Chapter 2. How nutrient limitations can be
mitigated through changes in soil system management is a principal focus of
this and following chapters.
† Average crop yields on smallholder farms in many countries are generally around
30% of the yields obtained on research farms (Tian et al., 1995). Closing this yield
gap is a major challenge to researchers and farmers.
† Moisture stress affects over two-thirds of all soils. While this often reflects adverse
rainfall patterns, much is attributable to the soils’ poor water husbandry. Their low
levels of organic matter (living and dead) and their unfavorable topsoil structure
exacerbate water shortages.
Biological Approaches to Sustainable Soil Systems258
q 2006 by Taylor & Francis Group, LLC
† It is estimated that nearly 500 million ha of land are degraded, approximately 40%
of the total arable area, due principally to the forces of water and wind erosion
(Oldeman, 1994), which have more adverse effects on soils that have diminished

biological integrity.
All these processes have led to declining per capita food production in SSA, which
has resulted in over 3 million tons of food aid yearly (Conway and Toenniessen, 2003).
Inadequate and inappropriate soil systems management has exacerbated these problems
to an alarming extent.
18.2 From an External-Input Paradigm to an Integrated Soil Fertility
Management Paradigm
During the past three decades, the ideas that have shaped soil fertility management
research and development efforts in SSA have undergone substantial change. During the
1960s and 1970s, an external-input paradigm was framing the research and development
agenda. Appropriate use of certain external inputs, whether fertilizers, lime, or irrigation
water, was believed to be able to alleviate any constraints to crop production. Organic
resources were seen as only playing a minor role (Table 18.1). By working within this
paradigm, and benefiting from the development and use of improved cereal germplasm,
bolstered by extensive fertilizer demonstrations and subsidization, what became known as
the Green Revolution boosted agricultural production in Asia and Latin America in ways
not seen before. Seeking similar yield enhancement, subsidies together with government
distribution schemes were introduced in many African countries to promote fertilizer use
by farmers. However, while some of these met with success, overall they did not come close
to overcoming the estimated nutrient depletion rates in SSA or in matching the use rates of
farmers in Asia and Latin America. By the early 1980s, these programs became mostly
financially unsustainable as costs rose and productivity gains were not achieved (Kherallah
et al., 2002).
TABLE 18.1
The Changing Role of Organic Resources in Tropical Soil Fertility Management
Period Soil Fertility Management Paradigm Role of Organic Resources
1960s/1970s External-input paradigm Organic matter plays a
minor role
1980s Biological management of soil
fertility as part of

low-external-input
sustainable agriculture
Organic matter is mainly
a source of nutrients
and especially N
1994 Second paradigm — combined
application of organic
resources and mineral fertilizer
Organic matter fulfils
other important roles
besides supplying nutrients
Today Integrated soil fertility management
(ISFM) as a part of integrated
natural resource
management (INRM)
Organic matter management
has social, economic, and
political dimensions, with
multiple stakeholders’ interests
Integrated Soil Fertility Management in Africa: From Knowledge to Implementation 259
q 2006 by Taylor & Francis Group, LLC
18.2.1 The Search for Less Input-Dependent Agricultural Systems
During the 1980s, exclusive reliance on chemical fertilizers for soil fertility enhancement
was challenged by proponents of low-external-input sustainable agriculture (LEISA) who
correctly argued that organic inputs were viewed as essential to sustainable agriculture
(Okigbo, 1990). Further, it was argued that LEISA was preferable because it was more
accessible to low-income rural households, who could afford little fertilizer and few
agrochemicals. Organic resources were considered to be the major sources of nutrients
(Table 18.1) and substitutes for mineral inputs. Additionally, the logistical problems of
acquiring and transporting fertilizer, the uncertainty and unevenness of its supply in rural

areas, and frequent issues of quality and efficacy reinforced the concern. However, LEISA
approaches had little widespread acceptance, in large part because of technical and socio-
economic constraints, e.g., insufficient training, lack of sufficient organic resources to
apply in the field, and the labor-intensity of these technologies (Vanlauwe et al., 2001a,
2001b).
In this context, Sanchez (1994) proposed an alternative, second paradigm for tropical
soil fertility research and remediation: “Rely more on biological processes by adapting
germplasm to adverse soil conditions, by enhancing soil biological activity and by
optimizing nutrient cycling to minimize external inputs and maximize the efficiency of
their use.” This paradigm, discussed more in Chapter 49, recognized the need for
judiciously combining both mineral and organic inputs to sustain crop production and soil
system fertility. The need for both organic and mineral inputs was advocated because
(i) both resources fulfill different functions related to crop growth, (ii) under most small-
scale farming conditions, neither is available and/or affordable in sufficient quantities
to be applied alone, and (iii) for reasons still not fully researched, there were often
added benefits when applying both inputs in combination, reflecting a degree of
synergy. The alternative paradigm also highlighted the need for improved germplasm
well-adapted to local conditions and able to give the most output from the available land,
labor, water and nutrient inputs
As in the first paradigm, the LEISA approach put more emphasis on the quantity and
quality of nutrient supply than on managing the demand for these nutrients. Obviously,
optimal synchrony or use-efficiency requires that both supply and demand be
coordinated. While organic resources were initially seen as complementary inputs to
mineral fertilizers, over time, as seen in Table 18.1, their role has been seen as more than
a short-term source of N, evolving to emphasize a wide array of benefits that can be
derived from organic inputs to soil systems, both in the short and long term.
18.2.2 The Search for Optimizing Strategies
From the mid-1980s to the mid-1990s, the shift in thinking toward a more combined use of
organic and mineral inputs was accompanied by a movement toward more participatory
involvement of various stakeholders in the research and development process. One of the

important lessons learned was that farmers’ decision-making processes are not driven
primarily by variations in soil and climate but by a whole set of factors encompassing the
biophysical, socio-economic, and political domains (DFID, 2000).
18.2.2.1 Integrated Soil Fertility Management
The ISFM paradigm shown in Figure 18.1 goes beyond Sanchez’s second paradigm to
recognize the important roles that social, cultural, and economic processes play in soil
fertility management strategies and also the many interactions that soil fertility has with
other ecosystem services. ISFM presents a holistic approach to soil fertility research and
Biological Approaches to Sustainable Soil Systems260
q 2006 by Taylor & Francis Group, LLC
practice that embraces the full range of driving factors and consequences related to soil
degradation — biological, physical, chemical, social, economic and political. Organic
resource use has many social, economic, and policy dimensions besides biological and
technical aspects reflected in belowground relationships.
The emergence of the ISFM paradigm parallels the development and spread on a wider
scale of concepts of integrated natural resource management (INRM). It is increasingly
recognized that natural capital (soil, water, atmosphere and biota) not only creates services
that generate goods having market value, e.g., crops and livestock, but also services
that are essential for the maintenance of life, e.g., clean air and water. Organic resource
management is viewed as the link between soil fertility and broader environmental
benefits, particularly ecosystems services such as carbon sequestration and biodiversity
protection (Swift, 1997). Due to the wide array of services accruing from natural capital,
different stakeholders may have conflicting interests in natural capital, and thus thinking
has to extend into social and even political domains. INRM aims to develop policies and
interventions that take both individual well-being and broader social needs into account
(Izac, 2000). Soil system management is one component, but a basic component, of larger
INRM strategies.
18.2.2.2 Tropical Soil Biology and Fertility Research
The Tropical Soil Biology and Fertility (TSBF) Institute, initially a program of UNESCO,
was founded in 1986 to promote and develop capacities for soil biology as a research

discipline benefiting the tropical regions. For over a decade, the program worked closely
with the International Center for Agroforestry Research in Nairobi. However, since 2001 it
has operated as an institute within the International Center for Tropical Agriculture
(CIAT) based in Colombia, while remaining based in Kenya.
The biological management of soil fertility is held to be an essential component of
sustainable agricultural development. The program’s mission is directed toward four goals:
1. Improve understanding of the role of biological and organic resources in tropical
soil fertility and their management by farmers to improve the sustainability of
land-use systems.
Soil
Organic / mineral
inputs
Erosion / deposition
BG biodiversity
Inherent traits (CEC,
SOM, pH, WHC)
Policy
Prices
Markets
Infrastructure
Information
Policy context
Crops/
Livestock
Germplasm
IPM
Livestock
Human
Local knowledge
Land

Labor
Finances
ISFM
FIGURE 18.1
The processes and components of integrated soil fertility management (ISFM). BG, belowground; CEC, cation
exchange capacity; SOM, soil organic matter; WHC, water-holding capacity; IPM, integrated pest management.
Integrated Soil Fertility Management in Africa: From Knowledge to Implementation 261
q 2006 by Taylor & Francis Group, LLC
2. Enhance the research and training capacity of national institutions in the tropics in
the fields of soil biology and management of tropical ecosystems.
3. Provide land users in the tropics with methods for soil management that improve
agricultural productivity while conserving soil resources.
4. Increase the carbon storage equilibrium and maintain the biodiversity of tropical
soils in the face of global changes in land-use and climate.
The implementation strategy for achieving these goals has evolved along with the
changes in soil fertility management paradigms described above. In the following section,
this will be seen from two case studies examining the contributions that scientific
investigations have made to better soil system management.
18.3 Translating Science into Practice
Despite the inherent complexity of the problems underlying the widespread decline in soil
fertility in SSA, the good news is that progress is being made. At a 2002 meeting organized
by the Rockefeller Foundation to take stock of progress with soil fertility research for
development, advances were identified in three areas: (i) number and range of stakeholders
influenced, (ii) soil management principles identified or clarified, and (iii) methodological
innovations (TSBF, 2002a). National and international research and development organiz-
ations, networks, NGOs, and extension agencies working in SSA are increasingly using ISFM
approaches (e.g., World Vision, 1999). There has been a rapid increase of membership and
activities of the African Network for Tropical Soil Biology and Fertility (AfNet) coordinated
by TSBF, with growing agreement on how soil systems can be better managed (Bationo, 2004).
International agricultural research has contributed significantly to the development of

sound soil management principles that can help achieve sustainable crop production
without compromising the ecosystem service functions of soil systems. Examples of such
principles are:
† Application of organic resources in optimizing combinations with mineral inputs
so as to maximize input-use efficiencies and farmers’ return to their investment.
† Integration of multiple-purpose woody and herbaceous legumes into existing
cropping systems to increase the supply of organic resources, crop yields, and
farm profits (e.g., Sanginga et al., 2003).
† Enhancement of the soil organic carbon pool as an integrator of various soil-based
functions that are related to production and ecosystem services (Swift, 1997).
† Improved sustainability of nutrient cycles through the integration of livestock
with arable production activities.
† Soil conservation methods to control soil loss and improve water capture and use-
efficiency.
Due to the complex and interactive nature of the major factors that promote poverty and
act at different scales, it has been necessary to develop approaches that deal with such a
complex environment:
† Pro-poor participatory research approaches that increase the appreciation and use
of local knowledge systems in the development of improved soil management
Biological Approaches to Sustainable Soil Systems262
q 2006 by Taylor & Francis Group, LLC
interventions and principles have been developed (e.g., Defoer and Budelman,
2000).
† Tools for scaling-up improved soil management practices, including GIS spatial
analysis to better characterize problems and target interventions and to obtain a
better understanding of information flow pathways, are emerging.
† Rapid assessment techniques using diagnostic indicators of land quality,
e.g., spectrometry techniques such as in Shepherd et al. (2005), are now available.
† Molecular tools are being used to study soil biodiversity and pest population
dynamics.

The following two sections describe areas where scientific principles have been
translated into practice. They also illustrate how the dominant soil fertility management
paradigm has shifted.
18.3.1 The Organic Resource Quality Concept and Organic Matter Management
Although use of organic inputs is hardly new to tropical agriculture, the first seminal
analysis and synthesis on the decomposition and management of organic matter (OM)
was contributed by Swift et al. (1979). Between 1984 and 1986, a set of hypotheses was
formulated in terms of two broad themes for soil system management: synchrony, and soil
organic matter (SOM) (see Swift, 1984, 1985, and 1986). These two focuses built upon the
concepts and principles presented in 1979.
Under the first theme, the organisms-physical environment-quality (OPQ) framework
for understanding OM decomposition and nutrient release, formulated by Swift et al.
(1979), was elaborated and translated into specific hypotheses. These could explain the
efficacy of management options that improved nutrient acquisition and crop growth with
an explicit focus on organic resource quality. Under the second theme, the role of OM in
the formation of functionally-different SOM fractions was stressed. It should be noted,
however, that during this period, organic resources were still mainly regarded as sources
of nutrients, and specifically of N (Table 18.1). Their multiple functions within soil systems
were not much considered.
During the 1990s, the formulation of research hypotheses related to residue quality and N
release led to many research efforts to validate these hypotheses, both within TSBF and
other research groups that dealt with tropical soil fertility. Results from these activities were
entered in the Organic Resource Database (ORD) ( />ORD/) (Palm et al., 2000). This database contains extensive information on organic-
resource quality parameters, including macronutrient, lignin, and polyphenol contents of
fresh leaves, litter, stems, and/or roots from almost 300 species utilized in tropical
agroecosystems. Data on the soil and climate from where the material was collected are also
included, as are decomposition and nutrient-release rates for many of the organic inputs.
Analysis of N-release dynamics revealed four classes of organic resources having
different rates and patterns of N release associated with varying organic resource quality
assessed in terms of their N, lignin, and polyphenol content (Palm et al., 2000). Based on

this analysis and information, a decision support system (DSS) for management of organic
N was formulated (Figure 18.2a). This system distinguishes four types of organic
resources, suggesting how each can be managed optimally for short-term N release to
immediately enhance crop production. Materials with lower N and higher lignin and/or
polyphenol contents are expected to release less N and thus they require supplementary N
in the form of fertilizer or higher-quality organic resources to maintain nutrient supply at
comparable levels.
Integrated Soil Fertility Management in Africa: From Knowledge to Implementation 263
q 2006 by Taylor & Francis Group, LLC
Being based on laboratory incubations, the DSS needed to be tested under field
conditions and was assessed in western, eastern, and southern Africa, using biomass
transfer systems with maize as a test crop. The results clearly indicated that (i) the N content
of the organic resources is an important factor affecting maize production, (ii) organic
resources with a relatively high polyphenol content result in relatively lower maize yields
for the same level of N applied, (iii) manure samples do not observe the general
relationships followed by the fresh organic resources, and (iv) N fertilizer equivalency
values of organic inputs often approach or even exceed 100% of what would be supplied
from inorganic sources.
These results gave strong support for the DSS constructed by Palm et al. (2000), except
for manure samples. Manure behaves differently from plant materials since it has
already gone through a decomposition phase when passing through the digestive
system of cattle, rendering the C less available and thus resulting in relatively less N
immobilization, as discussed in the preceding chapter. The observation that certain
organic resources have fertilizer equivalency exceeding 100% indicates that these organic
materials can alleviate other constraints to maize production besides low soil-available
N. In the short-term, organic resources not only release nutrients; they can enhance
soil moisture conditions or improve the available P in the soil (Nziguheba et al., 2000). In
the long term, continuous inputs of OM influence the levels of incorporated SOM and
N > 2.5 %
yes

yes
yes
no
no
no
Lignin < 15 %
Polyphenols < 4 %
Lignin < 15 %
Incorporate
directly with
annual crops
Mix with fertilizer
or high quality
organic matter
Mix with
fertilizer or
add to compost
Surface apply
for erosion and
water control
Characteristics of Organic Resource
green
no yes
yellow
yes no
Leaves crush to powder when dry
Incorporate
directly with
annual crops
Leaves fibrous (do not crush)

Highly astringent taste (makes
your tongue dry)
Mix with
fertilizer or
add to compost
Surface apply
for erosion and
water control
Mix with fertilizer
or high quality
organic matter
(a)
(b)
Class 1 Class 2 Class 4
Class 1 Class 2
Leaf Color
Class 3
Class 3 Class 4
FIGURE 18.2
A decision tree to assist management of organic resources in agriculture. (a) is based on Palm et al. (2000); (b) is a
“farmer-friendly” version of the same from Giller (2000).
Biological Approaches to Sustainable Soil Systems264
q 2006 by Taylor & Francis Group, LLC
the quality of some or all of its nutrient pools (Vanlauwe et al., 1998; Cadisch and Giller,
2000).
Following field-level testing of the DSS, it has been applied and adapted in a variety of
farmer learning activities. These give farmers the knowledge they need to identify and
evaluate the potential use of organic resources in their environment. Because there is so
much diversity of such resources in any given context, the elements of the DSS provide a
generic, easy-to-use tool for farmers to use when confronted with resources that scientists

have not themselves evaluated.
Farm-level adaptation of the DSS began with exercises where researchers and farmers in
selected communities identified all the organic resources available locally as potential soil
inputs. The quality analysis of these materials in one setting (Table 18.2) shows that among
TABLE 18.2
Organic Resources (leaf residues) and Their Chemical Composition, Identified in Farms Around
Emuhaya Division, Vihiga District, Western Kenya
Genus and Species Name
Common Name N P K Lignin PP
a
Class
b
or Local Name % Dry Matter
Markhamia lutea 3.20 0.24 1.77 21.21 3.99 1
Psidium guajava 2.32 0.19 1.50 11.20 14.35 3
Persea americana Avocado 2.07 0.12 0.82 20.25 10.90 4
Not identified Not known 4.98 0.44 6.66 14.93 3.27 1
Bridelia macrantha 2.37 0.17 1.13 18.53 8.31 4
Vernonia spp 4.88 0.42 4.72 11.31 2.44 1
Croton macrostachyus 4.33 0.38 1.75 10.25 8.42 2
Not identified Esikokhakokhe 3.84 0.39 6.59 9.07 1.32 1
Solanum aculeastrium Sodim apple 2.87 0.21 1.25 13.70 2.39 1
Erythrina exselsa 4.99 0.33 2.42 6.63 2.26 1
Buddleja davidi 3.30 0.27 1.46 7.94 6.20 2
Senna didymobotra 5.23 0.39 2.13 4.62 4.08 2
Vernonia auriculifera 3.65 0.35 5.25 14.86 4.93 2
Hurungania madagascariensis 3.21 0.18 1.04 13.31 12.70 2
Spathodea campanulata Nandi flame 3.09 0.21 1.76 17.34 8.58 2
Erythrina abyssinica 2.66 0.20 1.70 11.21 3.36 1
Morus alba Mulberry 2.86 0.43 2.16 4.28 4.62 2

Acanthus pubescens 3.30 0.30 2.11 5.17 7.56 2
Ricinus commus Castor plant 4.21 0.30 2.34 3.39 5.27 2
Maesa lanceolata 2.78 0.22 2.06 10.37 12.04 2
Mangifera indica Mango plant 1.52 0.12 1.00 11.15 12.43 3
Teclea nobilis 3.15 0.22 1.57 9.05 4.83 2
Not identified Libinzu 3.91 0.29 3.28 12.27 5.67 2
Sapium elliptian 3.11 0.18 0.77 6.34 11.73 2
Vangneria apiculata 3.67 0.23 1.76 4.91 4.27 2
Ficus spp 2.55 0.20 2.62 9.55 5.76 2
Ipomoea potatus Sweet potato 5.07 0.34 2.56 4.34 8.81 2
Not identified Omuterema 3.85 0.34 5.27 2.85 1.20 1
Plectranthus barbatus 3.87 0.28 4.01 16.11 4.98 2
Maesa lanceolata 3.80 0.28 3.92 10.70 6.65 2
Vernonia spp 4.26 0.37 3.67 9.80 5.09 2
a
PP, polyphenols.
b
Class refers to classes 1 to 4 indicated in Figure 18.2.
Source: Authors’ data.
Integrated Soil Fertility Management in Africa: From Knowledge to Implementation 265
q 2006 by Taylor & Francis Group, LLC
the plant resources that farmers would consider incorporating into their soils, the large
majority were class 2 resources. Of the 38 organic resources assessed, only eight belonged
to class 1 and could be classified as equivalent to N fertilizer. Tithionia diversifolia had
already been identified as a high-quality organic resource during a previous hedgerow
survey in the same area, also belonging in class 1 (Gachengo et al., 1999).
When these results were presented and discussed with farmers in a second step, the
decision-tree criteria proposed by Palm et al. (2000) were translated into a more farmer-
friendly version, using locally-acceptable criteria that do not require scientific equipment
(Figure 18.2b). This locally-adapted decision tree was then used by local farmer field

schools to design their own experimental trials that tested the validity of the claims
that scientists were making regarding the use and management of organic resources
(TSBF, 2002b).
These trials, conducted at a variety of sites and through several seasons, provided
many opportunities for farmers to compare the effects of these organic inputs under
different conditions. During evaluation activities, farmers ranked the classes of organic
resources in terms of effect on maize yield as: Tithonia (Class 1) . manure . Calliandra
(class 2) .maize stover (Class 3). They confirmed the hypothesis that the differing quality
of organic materials would have a demonstrable impact on crop yields.
Scientists also drew many valuable lessons from this exercise. They found, for example,
that farmers considered the biomass transfer technology being tested to be less practical
and cost-effective than using compost, a common local practice. Their interest in adding
their organic resources to compost heaps before application to the soil has stimulated new
joint research activities between farmers and scientists on how to improve compost quality
(TSBF, 2002b). (Benefits of composting are discussed in Chapter 31.) A second line of
experimentation used the resource-quality concept to assess the use of organic materials,
especially comparatively-scarce, high-quality Tithonia residues, on high-value crops such
as kale rather than on maize (TSBF, 2002b).
18.3.2 Exploring Positive Interactions between Mineral and Organic Inputs
The paucity of class 1 resources at the farm level, and the consequent advice to mix class
2 or 3 resources with minimal amounts of fertilizer N, has led to a diversification of
the research agenda toward the combined application of organic and mineral inputs.
As mentioned above, such a strategy is consistent with the ISFM paradigm and can
potentially lead to added benefits in terms of extra crop yield and/or extra soil fertility
enrichment where there are positive interactions between both inputs, as illustrated in
Figure 18.3.
Although the concept of interaction between two plant growth factors was already
implied in Liebig’s Law of the Minimum, it has recently received new attention in work
dealing with the combined application of fertilizer and organic inputs. Besides adding
nutrients, organic resources also provide C as a substrate for soil organisms and may

interfere with pests and diseases when the plants are grown in situ.
Two sets of hypotheses can be formulated, based on whether the interactions between
fertilizer and organic matter are direct or indirect. Since fertilizer N is susceptible to
substantial losses if not used quickly and efficiently by a crop, direct interactions result
from microbially-mediated changes in the availability of the fertilizer N when there is an
increase in available C. Further, the addition of fertilizer N may also affect the availability
of soil-derived N, although this will be less important whenever the bulk soil is C-limited.
Indirect interactions are the result of a general improvement in plant growth and demand
for nutrients by alleviation, through the addition of organic matter, of another growth-
limiting factor.
Biological Approaches to Sustainable Soil Systems266
q 2006 by Taylor & Francis Group, LLC
The direct hypothesis regarding N fertilizer can be stated as: temporary immobilization
of applied fertilizer N may improve the synchrony between the supply of and demand
for N and also reduce losses to the environment. Observations made under controlled
conditions justify this hypothesis, showing interactions in decomposition or N
mineralization between different organic materials (Vanlauwe et al., 1994) or between
organic matter and fertilizer N (Sakala et al., 2000).
The indirect hypothesis may be formulated for a certain plant nutrient X supplied by
fertilizer amendments as: any organic matter-related improvement in soil conditions
affecting plant growth (except that attributable to nutrient X) may lead to better plant
growth and consequently to enhanced efficiency of the applied nutrient X. The growth-
limiting factor can be located in the domain of plant nutrition, soil physics or chemistry, or
soil (micro)biology.
Most of the mulch effects or benefits of crop rotation could be classified under the indirect
hypothesis. Positive interactions based on the indirect hypothesis may be immediate
through direct alleviation of growth-limiting conditions after applying organic matter,
e.g., improvement of the soil moisture status after surface application of organic matter
as a mulch, or delayed through the improvement of the SOM status after continuous
application of organic matter and an associated better crop growth, e.g., improvement of

the soil’s buffering capacity.
Under on-station conditions, positive interactions can often be observed and measured.
However, explaining the mechanisms underlying these interactions is often more
problematic:
† In a field study in West Africa, Vanlauwe et al. (2002) observed positive
interactions, likely caused by higher soil moisture retention in treatments where
organic resources were applied (Figure 18.4).
† Bationo et al. (1995) observed a doubling of the fertilizer N-use efficiency after
application of crop residues in Sahelian conditions, attributable to much less wind
erosion on treatments when crop residues were applied.
0
500
1000
1500
2000
2500
3000
3500
0 20 40 60 80 100 120 140
Nutrient application (kg ha
−1
)
Maize grain yield (kg ha
−1
)
without OM
with OM
(a)
Fertilizer equivalent of OM
0 20 40 60 80 100 120 140

Nutrient application (kg ha
−1
)
Inter-
action
(b)
FIGURE 18.3
Theoretical response of maize grain yield to the application of certain levels of nutrients as fertilizer in the
presence or absence of organic matter (a) without interaction, and (b) with positive interaction between the
fertilizer nutrient and organic matter. Source: Vanlauwe et al. (2001a, 2001b).
Integrated Soil Fertility Management in Africa: From Knowledge to Implementation 267
q 2006 by Taylor & Francis Group, LLC
† In Zimbabwe, added benefits ranging between 663 and 1188 kg maize grains ha
21
were observed by Nhamo (2001), possibly because the supply of cations contained
in the manure alleviated constraints to crop growth caused by the low cation
content of the very sandy sites where clay content ranged between 2 and 10% and
CEC varied between 1.2 and 2.5 cmol kg
21
.
Translating these principles into cropping systems that are adaptable by farming
communities has resulted in a series of development innovations, e.g., rotations of maize
with promiscuously-nodulating soybean that combine high N-fixation and the ability to
kill large numbers of Striga hermonthica seeds in the soil; and rotations of millet and dual-
purpose cowpea that greatly enhance the productivity and sustainability of integrated
livestock systems (Sanginga et al., 2003).
These two systems are effectively used for the replenishment of soil nutrients and organic
matter. They contribute positive residual soil N for the following crops while at the same time
providing farmers with seeds for food and fodder for feed, as well as income from marketing
these farm products. Another option offered to any farmers who have manure available is

the opportunity to derive benefits from the combined application of manure and fertilizer to
maize. This practice allows farmers to complement the modest fertilizer quantities that they
can afford with high-quality organic nutrients, thereby benefiting from the synergism that
occurs when combining the two sources of nutrients. Currently, Sasakawa Global 2000 is
testing the above options in Northern Nigeria with promising results.
18.4 Challenges and the Way Forward
Although soil fertility replenishment has had a prominent position on the research and
development agenda in SSA for decades with tangible progress as seen above, widespread
0
200
400
600
800
1000
1200
1400
1600
1800
control 90 urea-N 90 OM-N
(SF)
45 urea-N +
45 OM-N
(SF)
90 OM-N
(INC)
45 urea-N +
45 OM-N
(INC)
Maize grain yield (kg ha
−1

)
AB=479
AB=549
FIGURE 18.4
Maize grain yields in Sekou, southern Benin Republic, as affected by the application of urea, organic materials, or
the combination of both. SF, surface-applied; INC, incorporated; OM, organic matter; AB, added benefits.
Numerical values for treatments are expressed as kg N ha
21
. Adapted from Vanlauwe et al. (2001a, 2001b).
Biological Approaches to Sustainable Soil Systems268
q 2006 by Taylor & Francis Group, LLC
adoption of ISFM strategies is lacking. A full discussion of the reasons for this is beyond
the scope of this chapter, but certain issues that have hampered large-scale adoption of
ISFM options can be singled out.
18.4.1 Adjusting to Variability at the Farm and Community Levels
Farmers’ production objectives are conditioned by a complex set of biophysical as well as
social, cultural, and economic factors. One must also take account of the fertility gradients
existing within farm boundaries. Most soil fertility research has been targeted at the plot
level, but decisions are made at the farm level, considering the production potential of all
plots. In Western Kenya, farmers will preferentially grow sweet potato on their most
degraded fields, while bananas and cocoyam occupy the most fertile fields (Tittonell et al.,
2005). Current recommendations for use of organic resources and mineral inputs do not
take into account these gradients in soil fertility status. On the contrary, recommendations
are often formulated at the national level and disregard the much greater variations that
exist between regions in terms of inherent soil properties and access to input and output
markets (Carsky and Iwuafor, 1999).
18.4.2 Use of Adapted Germplasm to Overcome Abiotic and Biotic Constraints
and Create More Resilient Cropping Systems
Breeding and biotechnology can help small farmers to sustainably increase their
productivity through improved drought-tolerance, soil acidity-tolerance, pest-resistance,

and increased efficiency of N-fixation. ISFM acknowledges the importance of the
interaction between new crop germplasm and more efficient natural resource manage-
ment for intensifying food and forage crop systems. Such a combination would utilize the
best variety for a given environment when grown in an improved soil using appropriate
crop management technologies. Interactions between adapted germplasm and key inputs
such as organic residues, mineral fertilizers, and water can lead to improved use-efficiency
of nutrients and water at a system level. ISFM bridges a commodity focus and an eco-
regional approach, working alongside germplasm development and integrated pest and
disease management.
18.4.3 Market-Led Integrated Soil Fertility Management
ISFM practices require some additional inputs of resources, whether minimal amounts
of mineral fertilizer, more organic matter, improved germplasm, or greater labor. As most
of these inputs require access to financial resources, implementing ISFM strategies will
often require farmers to have access to local or national markets so that they can acquire
more resources to reinvest in improved soil fertility management. It has been hypothe-
sized that improved profitability and access to markets will motivate farmers to invest in
new technology, particularly to integrate use of new varieties with improved soil
management options (John Lynam, 2004, personal communication).
Some current evidence does not show conclusive support for this hypothesis, how-
ever. For instance, the increased movement of bananas to urban markets in Uganda
without replenishment of the soil resource base could lead to a faster degradation of
banana-based systems within the production areas. It is also important to consider
nutritional consequences. Farmers who sell most of their produce could use the money
received for other uses rather than ensuring sufficient and nutritious food for the
household. This could lead to poorer health status with unfavorable consequences
for household labor availability and quality.
Integrated Soil Fertility Management in Africa: From Knowledge to Implementation 269
q 2006 by Taylor & Francis Group, LLC
18.4.4 Scaling Up
The knowledge-intensive nature of ISFM means that the kind of simplistic extension methods

such as “training and visit” promoted by the World Bank in the 1980s and 1990s are
not suitable for disseminating soil management technologies. This lack of suitability
accounted in part for the collapse of training-and-visit extension in the mid-to-late 1990s
(e.g., Gautam, 2000, on Kenya experience). Since then, the move in many countries toward
the decentralization of government services, the improved capacity of NGOs in service
delivery, and the beginnings of farmer groups and collective action have created the precon-
ditions for greater innovation and for the redesign of extension and dissemination systems.
Recognizing the wide diversity in agroecological and socio-economic conditions under
which most farmers work has led to a general realization that research and extension
agencies do not have the capacity to fine-tune their technological recommendations to the
level required by farmers. As extension services have become increasingly marginalized
and nonfunctional, the gaps in knowledge-dissemination and technological improvement
have been largely filled by a variety of NGOs and, in some cases, community-based
organizations. Scaling up information dissemination requires the reinforcement of
communication networks and strengthening of information centers (agricultural input
suppliers, community centers, field schools), as well as supporting farmers in various
ways to transfer knowledge farmer-to-farmer across communities.
18.4.5 Policy Changes
Since the 1980s, most countries in SSA have initiated extensive agricultural market reforms
(Kherallah et al., 2002). The expectation of agricultural market reform is that increasing
crop prices and improving markets will generate a positive supply response, increasing
both agricultural output and income levels. However, the average growth of agricultural
production per capita has been negative in SSA since the 1970s. In many countries, reform
has meant the elimination of government input and credit subsidies. This has kept yields
stagnant or reduced them, or has made input supplies irregular or completely absent,
undermining the stability of local prices. What production growth has occurred has often
been due either to expansion of crop area rather than increases in productivity per unit
area, or to the output of cash-crop farmers still operating within systems who have good
access to credit and inputs.
For ISFM to operate on a broader scale, there is a need for (i) regional policy

harmonization and policy reform frameworks for improved management within sub-
regional areas, (ii) development of appropriate partnerships to facilitate efficient input-
output markets and strengthen their links to ISFM, (iii) identification of marketing
opportunities through participatory research within a comprehensive, resource-to-
consumption framework, and (iv) development of appropriate seed supply systems and
resilient germplasm. Since not all farmers have the capacity to buy themselves out of
poverty, there is a major need for a series of “stepping stones” that enable poor farmers to
have access to inputs, services, and markets so that they can “climb out of poverty” as their
agricultural productivity increases.
Acknowledgments
The Rockefeller Foundation and the Belgian Directorate-General for Development Co-
operation are gratefully acknowledged for their continued financial support of this work.
Biological Approaches to Sustainable Soil Systems270
q 2006 by Taylor & Francis Group, LLC
References
Bationo, A., Managing Nutrient Cycles to Sustain Soil Fertility in Sub-Saharan Africa, Academy Science
Publishers, Nairobi (2004).
Bationo, A. et al., A critical review of crop-residue use as soil amendment in the West African semi-
arid tropics, In: Livestock and Sustainable Nutrient Cycling in Mixed Farming Systems of Sub-Saharan
Africa, Technical Papers, Vol. 2, Powell, J.M. et al., Eds., International Livestock Centre for Africa,
Addis Ababa, 305–322 (1995).
Cadisch, G. and Giller, K.E., Soil organic matter management: The role of residue quality in carbon
sequestration and nitrogen supply, In: Sustainable Management of Soil Organic Matter, Rees, R.M.
et al., Eds., CAB International, Wallingford, UK, 97–111 (2000).
Carsky, R.J. and Iwuafor, E.N.O., Contribution of soil fertility research and maintenance to improved
maize production and productivity in sub-Saharan Africa, In: Strategy for Sustainable Maize
Production in West and Central Africa, Badu-Apraku, B. et al., Eds., International Institute
for Tropical Agriculture, Ibadan, 3–20 (1999).
Conway, G. and Toenniessen, G., Science for African food security, Science, 299, 1187–1188 (2003).
DFID, Sustainable Livelihoods Guidance Sheets, Department for International Development, London

(2000).
Defoer, T. and Bundelman, A., Managing Soil Fertility in the Tropics: A Resource Guide for Participatory
Learning and Action Research, KIT Publishers, Amsterdam, Netherlands (2000).
Gachengo, C.N. et al., Tithonia and senna green manures and inorganic fertilizers as phosphorus
sources for maize in western Kenya, Agroforest. Syst., 44, 21–36 (1999).
Gautam, M., Agricultural Extension: The Kenya Experience: An Impact Evaluation, World Bank,
Operation Evaluation Department, Washington, DC (2000).
Giller, K.E., Translating science into action for agricultural development in the tropics: An example
from decomposition studies, Appl. Soil Ecol., 14, 1–3 (2000).
Izac, A M., What paradigm for linking poverty alleviation to natural resource management? In:
Proceedings of an International Workshop on Integrated Natural Resource Management in the CGIAR:
Approaches and Lessons, 21–25 August, 2000, Penang, Malaysia, CGIAR Secretariat, Washington,
DC (2000).
Kherallah, M. et al., Reforming Agricultural Markets in Africa, John Hopkins University Press,
Baltimore, MD (2002).
Nhamo, N., An evaluation of the efficacy of organic and inorganic fertilizer combinations
in supplying nitrogen to crops, M Phil thesis, University of Zimbabwe, Zimbabwe (2001).
Nziguheba, G. et al., Organic residues affect phosphorus availability and maize yields in a Nitisol
of western Kenya, Biol. Fert. Soils, 32, 328–339 (2000).
Okigbo, B.N., Sustainable agricultural systems in tropical Africa, In: Sustainable Agricultural Systems,
Edwards, C.A. et al., Eds., Soil and Water Conservation Society, Ankeny, IA, 323–352 (1990).
Oldeman, L.R., The global extent of soil degradation, In: Soil Resilience and Sustainable Land Use,
Greenland, D.J. and Szabolcs, I., Eds., CAB International, Wallingford, UK, 99–118 (1994).
Palm, C.A. et al., Organic inputs for soil fertility management in tropical agroecosystems:
Application of an organic resource database, Agric. Ecosyst. Environ., 83, 27–42 (2000).
Sakala, W.D., Cadisch, G., and Giller, K.E., Interactions between residues of maize and pigeonpea
and mineral N fertilizers during decomposition and N mineralization, Soil Biol. Biochem., 32,
679–688 (2001).
Sanchez, P.A., Tropical Soil Fertility Research: Towards the Second Paradigm, Transactions of the 15th World
Congress of Soil Science, Acapulco, Mexico. Mexican Soil Science Society, Chapingo, Mexico, 65–88

(1994).
Sanginga, N. et al., Sustainable resource management coupled to resilient germplasm to provide new
intensive cereal-grain legume-livestock systems in the dry savanna, Agric. Ecosyst. Environ.,
100,
305–314 (2003).
Shepherd, K.D., et al., Decomposition and mineralization rates of organic residues predicted using
near infrared spectroscopy. Soil Biol. Biochem. in press (2005).
Integrated Soil Fertility Management in Africa: From Knowledge to Implementation 271
q 2006 by Taylor & Francis Group, LLC
Stoorvogel, J.J. and Smaling, E.M.A., Assessment of Soil Nutrient Depletion in sub-Saharan Africa 1983–
2000, Winand Staring Center, Wageningen, Netherlands (1990).
Swift, M.J., Soil Biological Processes and Tropical Soil Fertility: A Proposal for a Collaborative Programme of
Research, Biology International special issue 5. International Union of Biological Sciences, Paris,
France (1984).
Swift, M.J., Tropical Soil Biology and Fertility: Planning for Research, Biology International, Special issue 9.
International Union of Biological Sciences, Paris, France (1985).
Swift, M.J., Tropical Soil Biology and Fertility: Inter-regional Research Planning Workshop, Biology
International Special Issue 13. International Union of Biological Sciences, Paris, France (1986).
Swift, M.J., Special issue: Soil biodiversity, agricultural intensification and agroecosystem function,
Appl. Soil Ecol., 6, 1 (1997).
Swift, M.J., Heal, O.W., and Anderson, J.M., Decomposition in terrestrial ecosystems, Studies in Ecology,
Vol. 5, Blackwell Scientific Publications, Oxford, UK (1979).
TSBF, Soil Fertility Degradation in sub-Saharan Africa: Leveraging Lasting Solutions to a Long-Term
Problem: Conclusions from a Workshop held at the Rockefeller Foundation Bellagio Centre, March 4–8,
2002, Tropical Soil Biology and Fertility Institute, Nairobi (2002a).
TSBF, Demonstration Plot: Harvesting and Evaluations. A Report on August 16 and 22, 2002 Events at the
Farmer–Researcher Demonstration Site in Emuhaya, Tropical Soil Biology and Fertility Institute,
Nairobi (2002b).
Tian, G. et al., Food production in the moist savanna of West and Central Africa, In: Moist Savannas of
Africa Potentials and Constraints for Crop Production, Kang, B.T. et al., Eds., International Institute

of Tropical Agriculture, Ibadan, Nigeria, 107–127 (1995).
Tittonell, P., et al., Exploring diversity in soil fertility management of smallholder farmers in western
Kenya. I. Heterogeneity at region and farm scale. Agric., Ecosyst. Envir. 110, 149-165.
Vanlauwe, B. et al., Maize yield as affected by organic inputs and urea in the West-African moist
savanna, Agron. J., 93, 1191–1199 (2001a).
Vanlauwe, B., Dendooven, L., and Merckx, R., Residue fractionation and decomposition: The
significance of the active fraction, Plant Soil, 158, 263–274 (1994).
Vanlauwe, B. et al., Organic resource management in sub-Saharan Africa: Validation of a residue
quality-driven decision support system, Agronomie, 22, 839–846 (2002).
Vanlauwe, B., Sanginga, N., and Merckx, R., Soil organic matter dynamics after addition of
15
N
labeled Leucaena and Dactyladenia residues in alley cropping systems, Soil Sci. Soc. Am. J., 62,
461–466 (1998).
Vanlauwe, B., Wendt, J., and Diels, J., Combined application of organic matter and fertilizer, In:
Sustaining Soil Fertility in West Africa, Tian, G., Ishida, F., and Keatinge, J.D.H., Eds., SSSA Special
Publication 58, Madison, WI (2001b).
World Vision, Food Security Program Newsletter 4:2, World Vision International, Accra, Ghana (1999).
Biological Approaches to Sustainable Soil Systems272
q 2006 by Taylor & Francis Group, LLC
19
Managing Soil Fertility and Nutrient Cycles through
Fertilizer Trees in Southern Africa
Paramu L. Mafongoya, Elias Kuntashula and Gudeta Sileshi
World Agroforestry Centre (ICRAF), Lusaka, Zambia
CONTENTS
19.1 Fertilizer Trees and a Typology of Fallows 274
19.1.1 Use of Non-Coppicing Fertilizer Trees 274
19.1.2 Use of Coppicing Fertilizer Trees 275
19.1.3 Mixed-Species Fallows 276

19.1.4 Biomass Transfer Using Fertilizer-Tree Biomass 276
19.2 Mechanisms for Improved Soil Fertility and Health 279
19.2.1 Biomass Quantity and Quality 279
19.2.2 Biological Nitrogen Fixation and N Cycles 279
19.2.3 Deep Capture of Soil Nutrients 280
19.2.4 Soil Acidity and Phosphorus 280
19.2.5 Soil Physical Properties 281
19.3 Effects on Soil Biota 282
19.4 Sustainability of Fertilizer Tree-Based Land Use Systems 285
19.5 Discussion 286
Acknowledgments 287
References 287
Low soil fertility is increasingly recognized as a fundamental biophysical cause for
declining food security among small-farm households in sub-Saharan Africa (SSA)
(Sanchez et al., 1997). Because maize is the staple food crop in most of southern Africa, it
will be our focus in this chapter. In 1993, SSA produced 26 million metric tons of maize on
approximately 20 m ha; approximately 54 million metric tons is expected to be needed by
2020. Meeting this maize production goal will depend on sustaining and improving soil
fertility levels that have been declining in recent years.
Soil fertility is not the only significant constraint; lack of appropriate, high-quality
germplasm, unsupportive policies, and inadequate rural infrastructure also limit maize
production. However, protecting and enhancing soil fertility is the most basic requirement
for achieving production goals. As discussed in Chapters 40 and 41, even controlling the
parasitic weed Striga hinges on this fundamental factor.
In most cases, nitrogen is the main nutrient that limits maize productivity, with
phosphorus and potassium being occasional constraints. Although inorganic fertilizers
273
q 2006 by Taylor & Francis Group, LLC
are used throughout the region, the amounts applied are seldom sufficient to meet crop
demands due to their high costs and uncertain availability. Most countries in southern

Africa have formulated fertilizer recommendations for all their major crops, sometimes
with regionally specific adaptations. However, the amount of fertilizer used in southern
Africa is very small in comparison to other parts of the world. For most smallholders,
fertilizer use averages as low as 5 kg ha
21
year
21
(Gerner and Harris, 1993).
While the need for increasing the availability of soil nutrients in southern Africa is quite
apparent, increasing their supply is very challenging. A high-external-input strategy
cannot rely on standard fertilizer-seeds-credit packages without addressing other
requirements for successful uptake of Green Revolution technologies, including reliable
irrigation, credit systems, infrastructure, fertilizer manufacture and supply, and access to
markets. Most African conditions differ starkly from those in the prime agricultural
regions of Asia. Approaches that produced successes in Asia are not readily transferable to
the African continent. Considering the acute poverty and the limited access to mineral
fertilizers in SSA, therefore, an ecologically robust approach of promoting “fertilizer trees”
is discussed here. This is a product of many years of agroforestry research and develop-
ment by the International Center for Research on Agroforestry (ICRAF), now called the
World Agroforestry Center, working with various partners in eastern and southern Africa.
19.1 Fertilizer Trees and a Typology of Fallows
Improved fallows involve the deliberate planting of fast-growing species, usually woody
tree legumes, referred to here as fertilizer trees, for the rapid replenishment of soil
fertility. Improved fallows were not a major area for research during the Green Revolu-
tion due to its focus on eliminating soil constraints by use of mineral fertilizers. Biological
approaches to soil fertility improvement began to receive attention in connection with the
articulation of a second soil-fertility paradigm based on adaptability and sustainability
considerations (Sanchez, 1994). Research on fertilizer trees had begun increasing from
the mid-1980s, so by the mid-1990s they had growing justification in research
(e.g., Kwesiga and Coe, 1994; Drechsel et al., 1996; Rao et al., 1998; Snapp et al., 1998).

Large-scale adoption of fertilizer trees by farmers is now taking place across southern and
eastern Africa. A more general consideration of fallows is presented in Chapter 29.
19.1.1 Use of Non-Coppicing Fertilizer Trees
Non-coppicing species do not resprout and regrow when cut at the end of the fallow
period, typically after 2 years of growth. Non-coppicing species include Sesbania sesban,
Tephrosia vogelii, Tephrosia candida, Cajanus cajan, and Crotalaria spp. Since the work of
Kwesiga and Coe (1994) on Sesbania fallows, much has been learned about the
performance of improved fallows using tree species that do not coppice. There has been
extensive testing of various species and fallow length on-farm to determine their impact
on maize productivity and to assess the processes that influence fallow performance. The
performance of Sesbania and Tephrosia under a wide range of biophysical conditions is
shown in Table 19.1.
Trials at Msekera Research Station, Zambia, have shown that natural regeneration of
Sesbania fallows is possible through self-reseeding, but it is highly erratic. Improved
fallows of 2-year duration using either Tephrosia or Sesbania significantly increased maize
yields well above those of unfertilized maize, the most common farmer practice in the
region. While it was true that fertilized maize usually performed better than improved
Biological Approaches to Sustainable Soil Systems274
q 2006 by Taylor & Francis Group, LLC
fallows in most cases, this required a greater cash outlay, so improved fallows could be
more profitable. The problem demonstrated in these trials was that the residual effects
of these improved fallows on maize yield declined after the second year of cropping
(Table 19.1). In a third year of cropping, maize yields following fallow were similar to
those of unfertilized maize. The marked decline of maize yields two or three seasons
after a non-coppicing fallow is probably related to depletion of soil nutrients and/or to
deterioration in soil chemical and physical properties.
19.1.2 Use of Coppicing Fertilizer Trees
Coppicing species include Gliricidia sepium, Leucaena leucocephala, Calliandra calothyrsus,
Senna siamea, and Flemingia macrophylla. Fallowing with a coppicing species, in contrast to
a non-coppicing species, shows increases in residual soil fertility beyond 2–3 years

because of the additional organic inputs that are derived each year from coppice regrowth
that is cut and applied to the soil. An experiment was established in the early 1990s at
Msekera Research Station to examine these relationships. These plots have now been
cropped for 9 years during which time both maize yields and coppice growth were
monitored.
The species evaluated showed significant differences in their coppicing ability and
biomass production, with Leucaena, Gliricidia, and Senna siamea having the greatest
coppicing ability and biomass production, while Calliandra and Flemingia performed
poorly. The trends in maize yields have been tracked carefully. In the plots with Sesbania
fallow, while maize yields were high for the first three seasons, they then declined to the
same level as control plots. Flemingia and Calliandra showed low maize yields over all
years. There were no significant differences in maize grain between Gliricidia and
Leucaena fallows over the seasons.
The effects of different fallow species on maize yield can be explained partly by the
different amounts of biomass added and the quality of the biomass and coppice regrowth.
Species such as Leucaena and Gliricidia, which have good coppicing ability, produce large
amounts of high-quality biomass with high nitrogen content and low contents of lignin
and polyphenols, thereby contributing to higher maize yields (Mafongoya and Nair, 1997;
Mafongoya et al., 1998). While Sesbania produces high quality biomass, its lack of coppice
regrowth means that it cannot supply nutrients for an extended period of cropping.
Species such as Flemingia, Calliandra, and Senna siamea, on the other hand, produce low-
quality biomass, high in lignin and polyphenols and low in nitrogen. Their use as fallow
species leads to N immobilization and reduced maize yields.
TABLE 19.1
Effect of Fallows on Maize Grain Yield Across 18 Locations in
Zambia
Land Use
Maize Grain Yield (t ha
21
)

Year 1 Year 2 Year 3
Sesbania sesban fallow 3.9 1.7 1.1
Tephrosia vogelii fallow 2.4 0.8 0.9
Traditional grass fallow 1.1 0.7 0.7
Unfertilized maize 1.0 0.7 0.6
LSD 0.8 0.6 0.6
Source: Authors’ data.
Managing Soil Fertility and Nutrient Cycles through Fertilizer Trees in Southern Africa 275
q 2006 by Taylor & Francis Group, LLC
Both Gliricidia and Leucaena have shown good potential as coppicing fallows. Over
9 years of cropping, cumulative maize yield of these fallows is greater than unfertilized
maize, maize grown after Sesbania, and traditional grass fallow. Continuous nutrient
replenishment is achieved by applying the coppice regrowth as mulch to the soil. This trial
will be continued for another three seasons to test the sustainability of coppicing fallows in
terms of nutrient budgets such as for NPK. On-farm trials have already been established to
evaluate responses more widely and to screen more coppicing fallow species.
19.1.3 Mixed-Species Fallows
Improved fallow practices using shrub legume species such as Sesbania have become
popular agroforestry systems for soil fertility management in southern Africa and western
Kenya. Large increases in maize yields have been reported following short-duration
fallows of 9–24 months with single species. Sesbania has been the main focus for these
improved fallows given its ability to provide large amounts of high-quality biomass and
fuel wood. Dependence upon a few successful fallow species has revealed some
drawbacks, however. Sesbania is susceptible to root nematodes and the Mesoplatys
beetle. The introduction of any new species can lead to an outbreak of new pests and
diseases, as was observed with Crotalaria grahamiana in western Kenya (Cadisch et al.,
2002). Thus, there is an urgent need to diversify the fallow species and types offered to
farmers. Mixing species with compatible and complementary rooting or shoot-growth
patterns in fallow systems should lead to more diverse systems and maximize growth and
resource utilization above- and belowground. Sowing herbaceous legumes under open-

canopy tree species can increase the use of photosynthesis radiation by the whole canopy
and thus enhance the system’s primary production.
Mixing shallow-rooted species with deep-rooted species can enhance the soil-water and
nutrient-uptake zone within the soil profile. More important, it enhances the utilization of
subsoil nutrients such as the nitrate that is otherwise lost through leaching. Mixing species
in fallows may also reduce the risks with fallow establishment, e.g., if one species is
susceptible to water stress, diseases or pests, another can survive and even prosper.
Obtaining multiple products from mixed fallows as well as increasing the biodiversity of
the system makes the whole system more robust. We have assessed a variety of mixed
fallows of tree legumes or tree legumes with herbaceous legumes to test these hypotheses.
Mixing a coppicing fallow species such as Gliricidia sepium with a non-coppicing species
like Sesbania (Chirwa et al., 2003) significantly increased maize yields compared to single-
species fallows (Table 19.2). However, mixtures of non-coppicing species did not increase
maize yield compared to sole species (Table 19.3). Mixing coppicing and non-coppicing
species reduces the level of subsoil nitrate, and we found that it reduces Mesoplatys
beetles (Sileshi and Mafongoya, 2002). We have found also that mixing Gliricidia,
Tephrosia, or Sesbania with herbaceous legumes such as Mucuna or Dolichos reduces tree
growth, and hence maize yield. Such mixtures also lead to a build-up of the Mesoplatys
beetle, which can cause more damage (Sileshi and Mafongoya, 2002).
19.1.4 Biomass Transfer Using Fertilizer-Tree Biomass
Traditionally, resource-poor farmers in parts of Southern Africa have collected leaf litter
from secondary forest, called miombo, as a source of nutrients for their crops. In the long
term, this practice is not sustainable because it mines nutrients from the forest ecosystems
in order to enhance soil fertility in croplands. Also, the miombo litter is of low quality
and may immobilize N instead of supplying N immediately to the crop (Mafongoya and
Nair, 1997). An alternative means of producing high-quality biomass is through the
Biological Approaches to Sustainable Soil Systems276
q 2006 by Taylor & Francis Group, LLC
establishment of on-farm “biomass banks” from which the biomass is cut and transferred
to crop fields in different parts of the farm. In western Kenya, for example, the use of

Tithonia diversifolia, Senna spectabilis, S. sesban, and Calliandra calothyrsus planted as farm
boundaries, woodlots, and fodder banks has proven to be beneficial as a source of
nutrients for improving maize production (Palm, 1995; Palm et al., 2001). A study by
Gachengo (1996) found that Tithonia green biomass grown outside a field and transferred
into a field was quite effective in supplying N, P, and K to maize, equivalent to the amount
of commercial NPK fertilizer recommended. In some cases, maize yields were higher with
Tithonia biomass than with commercial mineral fertilizer.
Biomass transfer using fertilizer-tree species is a more sustainable means for main-
taining nutrient balances in maize and vegetable-based production systems, as the tree
leafy materials are able to supply to the soil N (Kuntashula et al., 2004). Synchrony
between nutrient release from tree litter and crop uptake can be achieved with well-timed
TABLE 19.2
Maize Grain Yield (t ha
21
) from 3-Year Coppicing Mixed-Fallow Species
Treatments at Msekera, Eastern Zambia
Species 2003 2004
Fertilized maize 5.9 3.4
Acacia angustissma (34/88) 3.7 1.3
Acacia angustissma þ Sesbania sesban 4.6 2.2
Gliricidia sepium (Retalhuleu) 4.1 2.9
Gliricidia sepium þ Sesbania sesban 4.6 2.7
Gliricidia sepium þ Tephrosia vogelii 3.3 2.1
Leucaena diversfolia 3.6 1.5
Leucaena diversfolia þ Sesbania sesban 4.3 2.0
Sesbania sesban 3.9 1.9
Tephrosia vogelii 4.3 2.6
Tephrosia vogelii þ Sesbania sesban 4.3 2.0
Traditional grass fallow 2.5 1.3
Unfertilized maize 1.7 1.4

SED: 0.5 0.8
F probability , 0.001 , 0.05
TABLE 19.3
Maize Grain Yield (t ha
21
) from 2-Year noncoppicing Mixed-
Fallow Species Treatments at Msekera, Eastern Zambia
Species 2002 2003
Maize with fertilizer 4.7 4.3
Tephrosia vogelii þ Cajanus cajan 4.7 2.0
Sesbania sesban þ Tephrosia 4.4 1.3
Sesbania sesban þ Cajanus cajan 4.0 1.8
Tephrosia vogelii alone 3.9 1.6
Sesbania sesban alone 3.4 1.0
Cajanus cajan alone 2.7 0.9
Maize without fertilizer 1.3 0.4
SED 0.9 0.4
F probability , 0.001 , 0.001
Managing Soil Fertility and Nutrient Cycles through Fertilizer Trees in Southern Africa 277
q 2006 by Taylor & Francis Group, LLC
biomass transfer. The management factors that can be manipulated to achieve this are litter
quality, rate of litter application, and method and time of litter application (Mafongoya
et al., 1998).
Biomass transfer technologies require more labor for managing and incorporating
the leafy biomass, however. If used for the production of low-value crops such as maize,
the higher maize yield from biomass-transfer technologies may not be enough to
compensate for the higher labor cost. Most economic analyses have concluded that it is
unprofitable to invest in biomass transfer when labor is scarce and its cost is thus high.
However, when prunings are applied to high-value crops like vegetables, the technology
becomes profitable (ICRAF, 1997). This practice has been found quite suitable for vegetable

production in dambo areas of southern Africa (Kuntashula et al., 2004).
Dambos are shallow, seasonally or permanently waterlogged depressions at or near the
head of a natural drainage network, or alternatively they can occur independently of a
drainage system. All together, dambos serve approximately 240 million ha in all of sub-
Saharan Africa (Andriesse, 1986), of which 16 million ha are in southern Africa. Though
dambos are extremely vulnerable to poor agricultural practices, rising population pressure
has caused their agricultural use to become increasingly important (Kundhlande et al.,
1995). Without applying fertilizers or cattle manure, smallholder farmers cannot produce
vegetables successfully in dambos that are degraded due to their continuous cultivation
for over 25 years (Raussen et al., 1995). Inorganic fertilizer is not always available to
smallholder farmers, and cattle manure is accessible only to those with animals. This
calls for alternatives such as biomass transfers for fertilizing vegetables in dambos of
southern Africa. Additional results of such evaluations are given in Section 29.3.2.1.
Farmer participatory experiments conducted in 2000–2004 by Kuntashula et al. (2004)
have shown that biomass transfer using Leuceana leucocephala and Gliricidia sepium is
tenable for sustaining vegetable production in dambos. In addition to increasing yields of
vegetables such as cabbage, rape, onion, tomato, and maize grown after vegetable
harvests, biomass transfer has shown potential to increase yields of other high-value
crops such as garlic (Table 19.4). Our studies suggest that biomass transfer has greatest
potential when (a) the biomass is of high quality and it rapidly releases nutrients, (b)
when the opportunity costs of labor are low, (c) when the value of the crop is high, and
(d) when the biomass does not have other, valued uses apart from being a reliable source
of nutrients.
TABLE 19.4
Selected Vegetable Yields (t ha
21
) in Dambos Using Inorganic Fertilizers or Organic Inputs from
Manure or Tree Leaf Biomass in Chipata District, Zambia
Treatments
Cabbage Yield

(n 5 31) (2000)
Green Maize
Yield After
Onion
(t ha
21
)
Onion Yield
(n 5 12) (2001)
Green Maize
Yield After
Cabbage
(t ha
21
)
Garlic Yield
(n 5 6) (2004)
Manure 10 t þ 1/2 rec.
fertilizer
66.8 11.6 96.0 11.7 9.1
Recommended fertilizer 57.6 8.4 57.1 10.4 7.2
Gliricidia sepium (12 t) 53.6 12.4 79.8 17.3 - -
Gliricidia sepium (8 t) 43.1 10.9 68.3 14.9 10.3
Leucaena leucocephala 2 12 t 32.6 - - - - 13.0 - -
Nonfertilized 17.0 6.4 28.1 7.8 4.2
SED 5.3 2.06 11.2 3.04 1.2
F probability , 0.001 , 0.001 , 0.05 , 0.05 , 0.05
- -, treatment not evaluated.
Biological Approaches to Sustainable Soil Systems278
q 2006 by Taylor & Francis Group, LLC

19.2 Mechanisms for Improved Soil Fertility and Health
19.2.1 Biomass Quantity and Quality
The success of maize crop rotations with fertilizer trees depends very much on processes
for pruning biomass and on their nutrient yields. Analysis of maize yields across several
sites with different fertilizer trees shows that maize yield is most closely correlated with
the N content of prunings, with rainfall, and with the quantity of biomass applied. Low
and insufficient biomass yields, combined with low quality of prunings in most instances,
have contributed to frequent low performance of the technology. The low production of
biomass for pruning may result from the use of unsuitable species, poor tree growth
due to low soil fertility, soil acidity, moisture stress, or poor management of the species.
Work carried out for many years has shown how organic decomposition and nutrient
release are affected by the levels of polyphenol, lignin, and nitrogen content of the organic
inputs (Mafongoya et al., 1998). Recently, we have also found that maize yields after
fallows with various tree legumes were negatively correlated with the (L þ P) to N ratio
and positively correlated with recycled biomass. Fallow species with high N, low lignin,
and low polyphenols such as Gliricidia and Sesbania gave higher maize yields compared
to species such as Flemingia, Calliandra, and Senna. This work has shown that it is not
the quantity of polyphenols that is critically important, but rather their quality as measured
by their protein-binding capacity (Mafongoya et al., 2000). Legume species for improved
fallows can be screened for their suitability based on the above characteristics.
19.2.2 Biological Nitrogen Fixation and N Cycles
The contribution of leguminous trees to crop yield through N
2
fixation is well
recognized, although not all legumes fix N
2
. Numerous nonleguminous species have
N fixed in their roots and root zones through associations with N-fixing bacteria
(Chapter 12). Nitrogen fixation in alley cropping systems in the humid and subhumid
zones of Africa has been reviewed by Sanginga et al. (1995). There has been little work

carried out quantifying N
2
fixation by trees in southern Africa, however. Such analysis
has been difficult due to constraints in the methodologies for measuring the N
2
fixed. A
series of multi-location trials has been set up to measure the amount of N
2
fixed by
different tree genera and provenances using the
15
N natural abundance method. The data
on percent N derived from atmospheric N
2
fixation (Ndfa) shows high variability among
species and provenances of the same species. Greater variation was also recorded for the
same species across different locations. So the measurement task is a challenging one.
Sanginga et al. (1990) found that the Ndfa ranged from 37 to 74% for different
provenances of Leucaena leucocephala. The initial data show a huge potential of trees to fix
N
2
and increase N inputs in N-deficient soils. In future analysis we will focus on factors
responsible for the variability in N
2
-fixation across sites and on how to optimize N
2
fixation under field conditions.
An estimated value of the level of inorganic N in soil before a cropping season begins
is an accepted test for assessing prospective soil productivity. Results of studies in
Southern Africa show that preseason inorganic N can also be an effective indicator of the

N that is plant-available after fallow with different species (Barrios et al., 1997). Studies
we conducted at 18 locations in eastern Zambia have indicated that in a tropical soil with
a pronounced dry season, total preseason inorganic N (i.e., NO
3
þ NH
4
) is more closely
related to maize yield (R
2
¼ 0.62; b ¼ 0.27, se ¼ 0.03) than to preseason NO
3
alone. While
large amounts of NH
4
can accumulate during a dry season, it may not be nitrified when
Managing Soil Fertility and Nutrient Cycles through Fertilizer Trees in Southern Africa 279
q 2006 by Taylor & Francis Group, LLC
the soil is sampled at the beginning of the rainy season. We have concluded that
preseason inorganic N is a relatively rapid and simple index that is related fairly well to
maize yield on N-deficient soils, and hence it can be used to screen fallow species and
management practices.
19.2.3 Deep Capture of Soil Nutrients
The retrieval and cycling of nutrients from soil below the zone exploited by crop roots is
referred to as nutrient pumping (Van Noordwijk et al., 1996; also Chapters 20 and 21). Soil
nutrients not accessible to annual crops such as maize can be extracted by perennial trees
through deep capture. The distributions and density of roots, the demand of plant for
nutrients, and the distribution and concentration of plant-extractable nutrients and water
will influence deep capture of nutrients by fertilizer trees (Buresh et al., 2004). Deep capture
is favored when perennials have a deep rooting system and a high demand for nutrients,
when water or nutrient stress occurs in the surface soils, and/or when considerable

extractable nutrients or weatherable minerals occur in the subsoil (Buresh and Tian, 1997).
These conditions were observed in eastern Zambia where nitrate accumulated in the
subsoil during periods of maize growth, and fertilizer trees grown in rotation with maize
could then effectively retrieve the nitrate in the subsoil that had been “lost” to maize.
Intercropping rather than rotating fertilizer trees with crops appears to improve the
long-term efficiency of nutrient use in deep soils. When perennials such as G. sepium are
intercropped with maize, they remain always present in the agroecosystem compared
with non-coppicing trees such as S. sesban. In a mixed fallow, Gliricidia provides a safety-
net function to reduce nitrate leaching. In the Sesbania-maize rotation, there is no active
perennial legume. Therefore, nitrate leaches into deep soil below the effective rooting
depth of maize. Intercropping with fertilizer trees such as Gliricidia may thus be more
effective for pumping of soil nutrients than a Sesbania-maize rotation. In base-rich deep
soils of Msekera, eastern Zambia, there is potential for subsoil accumulation of highly
mobile cations such Ca, Mg, and K, due to the weathering of minerals and leaching of
cations that accompany NO
3
leaching in fully fertilized maize crops without any trees
present. The introduction of Gliricidia with maize rotation has a great potential for deep
capture of Ca and Mg compared to continuously fertilized monoculture maize.
19.2.4 Soil Acidity and Phosphorus
Acidic soils cover approximately 27% of the land in tropical Africa. Acidic soils are
characterized by low pH, deficiencies of phosphorus, calcium, and magnesium, and
toxic levels of aluminum. This is why finding strategies that offset soil acidity and low P
availability is so important. Here, we discuss how agroforestry systems can address these
two related constraints. In Chapter 37, there is a more detailed consideration of such
a strategy, focused in Western Kenya.
Lime application is the most widely used remedy for high acidity in countries such as in
Brazil and U.S.A., but it is financially prohibitive for resource-poor farmers in southern
Africa and cannot be considered a viable solution to the problem. Numerous laboratory
experiments have recorded increased soil pH, decreased Al saturation, and improved

conditions for plant growth as a result of the addition of plant materials to acid soils
such as tree prunings, which also supply base cations such as Ca, Mg, and K. The value
of tree prunings as a “liming” material for acid soils is related in their cation content
(Wong et al., 2000). There is evidence from field experiments (see Wong et al., 1995) that
the lateral transfer of alkalinity can be achieved by pruning pure stands of agroforestry
trees and applying their pruned biomass to a maize crop.
Biological Approaches to Sustainable Soil Systems280
q 2006 by Taylor & Francis Group, LLC