Fertilizer Impacts on Soils and Crops of Sub-
David Weight and Valerie Kelly
MSU International Development Paper
Paper No. 21
Department of Agricultural Economics
Department of Economics
MICHIGAN STATE UNIVERSITY
East Lansing, Michigan 488241999
MSU Agricultural Economics Web Site: http://www.aec.msu.edu/agecon/
MSU Food Security II Web Site: http://www.aec.msu.edu/agecon/fs2/index.htm
MSU is an affirmative-action/equal-opportunity institution.
MSU INTERNATIONAL DEVELOPMENT PAPERS
Carl Liedholm and Michael T. Weber
The MSU International Development Paper series is designed to further the comparative analysis
of international development activities in Africa, Latin America, Asia, and the Near East. The
papers report research findings on historical, as well as contemporary, international development
problems. The series includes papers on a wide range of topics, such as alternative rural
development strategies; nonfarm employment and small scale industry; housing and construction;
farming and marketing systems; food and nutrition policy analysis; economics of rice production
in West Africa; technological change, employment, and income distribution; computer techniques
for farm and marketing surveys; farming systems and food security research.
The papers are aimed at teachers, researchers, policy makers, donor agencies, and international
development practitioners. Selected papers will be translated into French, Spanish, or other
Individuals and institutions in Third World countries may receive single copies of papers published
since 1993 free of charge. Requests for copies and for information on available papers may be
MSU Bulletin Office
10-B Agriculture Hall
Michigan State University
East Lansing, Michigan 48824-1039
Information concerning how to purchase MSU International Development Papers is included in
the back of this publication and requests should be sent to:
MSU Bulletin Office
10-B Agriculture Hall
Michigan State University
East Lansing, Michigan 48824-1039
FERTILIZER IMPACTS ON SOILS AND CROPS
OF SUB-SAHARAN AFRICA
David Weight and Valerie Kelly
This paper is published by the Department of Agricultural Economics and the Department of
Economics, Michigan State University (MSU). Funding for this research was provided by the
Food Security II Cooperative Agreement (PCE-A-00-97-00044-00) between Michigan State
University and the United States Agency for International Development, through the Africa
Bureau’s Office of Sustainable Development, Productive Sector Growth and Environment
Division, Technology Development and Transfer Unit (AFR/SD/PAGE/TAT).
Weight is an independent soil scientist and Kelly a visiting associate professor, Department of
Agricultural Economics, Michigan State University.
© All rights reserved by Michigan State University, 1999.
Michigan State University agrees to and does hereby grant to the United States Government a
royalty-free, non-exclusive and irrevocable license throughout the world to use, duplicate,
disclose, or dispose of this publication in any manner and for any purposes and to permit others to
Published by the Department of Agricultural Economics and the Department of Economics,
Michigan State University, East Lansing, Michigan 48824-1039, U.S.A.
This paper began as an individual effort by the first author to inform social scientists about the
basic debates concerning current soil fertility issues in Sub-Saharan Africa. Over time, Valerie
Kelly’s assistance in revising and expanding the paper led to a jointly authored document. I wish
to acknowledge her efforts in seeing this paper to completion, especially her generosity in availing
the principal author of her wise counsel and editing experience. I would also like to thank the
other members of the Food Security staff who supported and provided assistance to the effort,
including Eric Crawford, Julie Howard, Patricia Johannes, Josie Keel, and Michael Weber.
Special thanks are due to the following individuals who read through various drafts of the paper
and supplied critical and sometimes extensive commentary: Henk Breman (IFDC, Togo); Eric
Crawford (Michigan State University); Bruce James (University of Maryland); Mike McGahuey
(USAID/Africa Bureau); Christian Pieri (The World Bank); and John Sanders (Purdue
University). Thanks also to Richard Harwood and Michel Cavigelli of the Crop and Soils
Department/Kellogg Biological Station, Michigan State University; Joost Brouwer,
University of Wageningen (the Netherlands); and to colleagues, too numerous to list here, who
have provided valuable insights and information. Every effort has been made to assure that they
are cited in the text.
Sincere appreciation is extended to my wife, Linda, and children, Alex and Leah, for their patience
and perseverence during the more intensive periods of writing and revisions and for their love and
The authors would appreciate receiving comments on the paper from readers. They can be sent
via the following e-mail address:
David Weight: email@example.com
Background: Successful agricultural development has resulted in substantial alleviation of
poverty and food security in Asia and Latin America since the 1960s. Much of this success can be
attributed to the introduction of high-yielding varieties of crops, especially wheat and rice, which
have addressed the constraints faced by farmers using traditional varieties. In Sub-Saharan Africa
(SSA), however, productivity levels have remained stagnant despite the introduction of new crop
germplasm. In recent years, scientists have recognized that low soil fertility is the primary
constraint blocking agricultural development in SSA.
Soil fertility problems in SSA can be attributed to soil degradation due to soil mining (associated
with long-term low-input agriculture), tillage, and accelerated erosion. Soil organic matter
(SOM), soil organic carbon (SOC), and nutrients have become depleted in most soils. In the
lower rainfall regions of the continent, the situation is analogous to the "Dust Bowl" era in US
history when SOM levels reached their lowest point after years of agriculture-induced soil
degradation. In these regions of SSA, wind and water erosion are depleting what little remains of
the topsoil, leaving farmers with low-fertility subsoils or desertification, declining or stagnant
yields, and long-term poverty. Fertilizers are considered by many to be critical tools for
increasing crop yields and restoration of soil fertility in SSA. The purpose of this paper is to (1)
evaluate the potential impacts of fertilizer, both positive and negative; (2) suggest ways in which
positive impacts can be maximized and negative impacts minimized; and (3) identify national
strategies that have the greatest potential to achieve positive impacts and address the constraints
Positive Impacts of Fertilizer and Effective Strategies: The primary positive impact of
fertilizers is to increase the biological base of the plant/soil system, measured as net primary
productivity (NPP), resulting in increased crop yields and recapitalization of soils, if appropriate
management systems are introduced. When fertilizers (or organic inputs) are applied, essential
nutrients are supplied for the creation of plant biomass by means of photosynthesis. In the
process, carbon dioxide (CO2) is incorporated or fixed into the biomass from the atmosphere
which is then referred to as organic carbon (C). However, the organic C and nutrients in the plant
biomass can only recapitalize the soil if crop residues are allowed to remain on the soil surface
where they decompose and are transformed into SOM.
In this paper, fertilizers are recommended as the primary nutrient input and organic materials are
recommended as "amendments" to fertilizers. This recommendation is based on the fact that large
quantities of organic material are required to deliver a nutrient load equivalent to fertilizers. Such
large quantities are required due to the low concentrations of nutrients in organic matter. It is
difficult for farmers to obtain such large quantities of organic materials due to competition from
non-agricultural uses (fuel, fodder, construction, etc.). Also, there are declining rates of
biological cover in SSA.
Historical research in the Great Plains (Dust Bowl) region of the US has indicated that the
introduction of fertilizers and return of crop residues to the soil has been a successful strategy for
increasing levels of SOC and SOM, effectively reversing declines. In the 1970s, conservation
tillage (e.g., no-till) as well as use of cover crops, both of which included increased returns of
residues to the soil, were introduced or expanded and also contributed to increases in SOC/SOM.
In South America, no-till systems have been very successful in addressing constraints and
increasing productivity. Besides application of fertilizers, these systems, based on "agro-
ecological" principles, typically include no tillage, green manure cover crops, and rotations.
Primary advantages of such systems are increased yields and profits, reduced costs and labor
requirements, and increased fertilizer and water-use efficiency.
In conventional agricultural systems, especially in the tropics, fertilizer efficiency is typically very
low with the result that the majority of available nutrients are not utilized by the crop (low
"recuperation rates"). This is due primarily to accelerated rates of decomposition and
mineralization which means that outflows of mineralized inorganic nutrients are too great for them
to be utilized efficiently. This leaves them vulnerable to losses, especially leaching of nitrates in
sandy soils. This is the primary reason for the inefficient rates of recuperation of nutrients by
crops in SSA. Roughly twice as many nutrients are lost in SSA compared to other regions.
In integrated "agro-ecological" systems, however, fertilizer-use efficiency is high, primarily due to
better soil structure and aggregation. Improved soil structure and aggregation are associated with
higher levels of SOM in which soil microbes attack particulate organic matter from residues as
sources of C and nutrients. In this process, soil aggregates are formed which have a high capacity
for sequestering C and nutrients. If one observes fields under these systems, there is a much
higher level of biological cover (larger crop canopies, cover crops, and trees) as well as residue
cover compared to conventional fields, which are bare except for the primary monocrop. The
high levels of residues result in high levels of SOM and associated improvements in soil structure
and aggregation. The net result is increased nutrient use efficiency with an estimated potential for
increases in nutrient uptake by the crop of at least two times current rates and parallel decreases in
water pollution from losses to leaching and run off.
One of the most severe constraints in SSA for production as well as for fertilizer use is low
availability of water (relative to other continents). "Agro-ecological" systems are associated with
increased water-use efficiency with estimated increases in crop water uptake of three to five times
current rates. Such efficiency can result in stable or increasing levels of crop yields, even during
periods of drought stress.
Negative Impacts of Fertilizer: If fertilizer use in SSA is increased, the primary negative
impacts that are expected are:
• Acidification of soils by ammonium-N fertilizers which can result in serious declines in
yields and soil quality. This can be addressed by use of non-acidifying nitrate fertilizers
and application of lime or lime plus manure.
• Negative impacts on traditional systems and environments, especially when extensive
management systems are implemented that take over from appropriate traditional soil
management practices. Management systems need to be sensitive to traditional values and
• Non-point source pollution of water resources which is the result of excessive fertilizer
use. This can be addressed by developing more efficient "agro-ecological" systems with
minimal losses to leaching/ runoff and avoiding excessive use of fertilizers beyond crop
• Increased carbon dioxide emissions (greenhouse gasses) associated with fertilizer-based
conventional agricultural systems. More efficient systems sequester increased quantities
of C resulting in lower levels of SOC that are lost to CO2 via decomposition.
Historical Evidence Concerning the Potential of Fertilizer-based Production in SSA: It is
clear from the historical record that, under favorable climatic and soil conditions, farming has been
productive and profitable in SSA, especially on commercial, large-scale farms. The critical factor
for that success has been the implementation of fertilizer-based crop management systems,
especially conventional and/or "green revolution" systems which have focused on improved
cultivars, planting density, and pest/weed control. In many cases, farm management has been
backed up by technological, institutional, and financial support such as research and input
services, credit for fertilizers, and pre-set price levels for farmers.
In regions of lower rainfall, there is very little evidence of successful agriculture on a large scale.
However, on-farm experiments have shown the technical potential for fertilizer-based production
in these zones. Experimental findings suggest that the primary restrictions for use of fertilizers
have been the expense and lack of availability of fertilizers, as well as lack of institutional support
and knowledge about fertilizers and fertilizer-based management systems. Efforts to improve
productivity, especially in the lower rainfall zones, will need to address these constraints.
While it is technically feasible to maintain productive systems, the overwhelming majority of
farmers in SSA are smallholders with severe economic constraints. These farmers do not possess
the financial or technical capacity to implement intensive conventional systems. Rather, strategies
are being sought that take advantage of natural restorative processes and are, therefore, efficient
in terms of fertilizer and water requirements as well as costs and labor.
National Strategies: Currently, there is a need for stronger collaboration between fertilizer-
based "green revolution" programs in SSA, such as Sasakawa-Global 2000 which has been
successful at increasing productivity in Ethiopia and other countries, and "agro-ecological"
programs such as the "Soil Fertility Initiative" (SFI) or various non-governmental organization
(NGO) efforts. This paper argues that the goal should be to combine fertilizer strategies with
"agro-ecological" systems (no-till, cover crops, rotations, agroforestry). For this to happen, the
two "camps" need to cooperate and develop an integrated strategy, especially in light of current
funding constraints in SSA. Such a strategy would have the potential to build on successful
fertilizer-based or ecologically based programs that are already in place, integrating the missing
elements of the alternative approach, rather than trying to develop entirely new and separate
It is important to remember that there are alternative approaches that can be effective in adopting
"agro-ecological" systems, as seen in South America. First, farmers can take the initiative in
developing new strategies, especially through the leadership of active farmer organizations. In
this case, researchers as well as development and extension workers will need to learn from and
assist farmers in their efforts. Secondly, NGO’s can play a critical role in introducing new
technologies or systems on a national scale.
Conclusions: Major findings from this study may be summed up in five key points:
• Declining fertility and SOM in SSA are a result primarily of agriculture-induced
degradative processes (especially soil mining, tillage, and accelerated erosion) that can be
reversed using high levels of nutrient inputs as part of "agro-ecological" farming systems
to recapitalize the soil.
Fertilizer is recommended for recapitalization because nutrients available from organic
sources in low-fertility African ecosystems are not adequate.
The primary positive impact of fertilizers is to increase the biological base of the plant/soil
system resulting in increased crop yields. If the system is properly managed, the outcome
can be a fertile and efficient cycling system for nutrients and water due to improved soil
structure associated with increased levels of SOM. Since there is competition for uses of
crop residues (fuel, construction, animal feed), biomass production needs to increase and
alternatives need to be found to satisfy other demands for crop residues.
• Fertilizers and organic matter are complements rather than substitutes – both are
recommended to recapitalize SSA soils. Fertilizer can increase crop yields and residues,
but maximum levels of residues (or equivalent manure) should be returned to the soil.
Because of the very high quantities of residue or manure required to reverse declines in
SOM and inadequate supplies of these materials, integrated "eco-intensive" systems are
recommended to create an aggrading system, including mulch or conservation tillage and
SSA has an historic opportunity to reverse the current trends of stagnant or declining productivity
and soil fertility. The challenge is to begin the enormous process of moving SSA from the low
point of the soil degradation curve to levels which are close to pre-disturbance (native) fertility.
Effectively, this means that long-term fallows, which accomplished this task in the past, need to be
replaced with (or adapted to) appropriate integrated systems that include fertilizers or other
effective input sources, as well as no-till (or mulch tillage), cover crops, rotations, and/or
agroforestry practices based on sound "agro-ecological" principles. That is, systems that take
advantage of natural restorative processes and are, therefore, efficient in terms of fertilizer and
water requirements as well as costs and labor. This is especially critical for smallholder farmers
who make up the vast majority of agricultural producers in SSA and face severe economic and
technical constraints. Once fertility and SOM levels are restored, ideally to pre-disturbance levels,
the primary objective will be to maintain a "sustainable" balanced system with equivalent
inputs/outputs of nutrients and C, as in a natural, undisturbed system.
ACKNOWLEDGMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iii
EXECUTIVE SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v
LIST OF TABLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiii
LIST OF FIGURES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiii
LIST OF ABBREVIATIONS AND ACRONYMS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xv
1. BACKGROUND, OBJECTIVES AND METHODS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1. Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2. Objectives and Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
2. REVIEW OF ENVIRONMENTAL FACTORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
3. FERTILIZER-BASED STRATEGIES THAT ADDRESS CONSTRAINTS . . . . . . . . . . . . 8
3.1. Agriculture-Induced Constraints . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
3.1.1. Soil Mining . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
3.1.2. Tillage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
3.1.3. Accelerated Erosion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
3.1.4. The Process of Decline in Soil Organic Matter in SSA . . . . . . . . . . . . . . 13
3.2. Geographical Estimates of Constraints and Potentialities of African Soils . . . . . . 14
3.2.1. Distribution of Soil Types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
3.2.2. Addressing the Constraints of Marginally Sustainable (“Marginal") Soils
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
3.3. Fertilizers, Organic Matter and the Carbon Cycle . . . . . . . . . . . . . . . . . . . . . . . . . 18
3.4. Fertilizer-Based Strategies that Address Constraints . . . . . . . . . . . . . . . . . . . . . . 23
3.4.1. Nutrient Management for Soil Fertility: Combining Fertilizers with Organic
Inputs/Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
3.4.2. Conservation or Mulch Tillage as a Complementary System . . . . . . . . . . 34
3.4.3. Cover Crops, Rotations, and Agroforestry as Complementary Systems . 38
4. NEGATIVE IMPACTS OF FERTILIZER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
4.1. Acidification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
4.1.1. Comparison by Fertilizer Type . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
4.1.2. Comparison by Fertilizer Level . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
4.1.3. Comparison by Crop Species . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
4.1.4. Comparison by Soil Type . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
4.1.5. Yield Losses from Acidification Over Time . . . . . . . . . . . . . . . . . . . . . . 56
4.2. Negative Impacts on Traditional Systems and Environments . . . . . . . . . . . . . . . . . 58
4.3. Environmental Quality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
4.3.1. Historical Environmental Degradation . . . . . . . . . . . . . . . . . . . . . . . . . . 61
4.3.2. Non-point Source Pollution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
4.3.3. Carbon Sequestration for Reduction of Greenhouse Gasses . . . . . . . . . . 63
5. HISTORICAL EVIDENCE CONCERNING THE POTENTIAL OF
FERTILIZER-BASED PRODUCTION IN SSA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
5.1. The Sub-Humid Zone . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
5.2. The Semi-Arid to Arid Zone . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
5.3. Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
6. DEVELOPING EFFECTIVE STRATEGIES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
6.1. Using Fertilizers to Increase the Biological Base of the Plant/Soil System while
Avoiding Negative Impacts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
6.2. Developing Effective Fertilizer-Based Programs . . . . . . . . . . . . . . . . . . . . . . . . 75
6.2.1. Addressing Variability of Environmental Conditions . . . . . . . . . . . . . . . . 77
6.2.2. Historical and Regional Examples of Successful Strategies . . . . . . . . . . . 79
6.2.3. Integrating Crop-Based Fertilizer Strategies with Soil-Based Organic
Strategies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
7. CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
LIST OF TABLES
1. Impact of Soil Management Treatments on Crop Yield Trends in Selected Long-term
Experiments from Sub-Saharan Africa . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
LIST OF FIGURES
1. Distribution of Soil Orders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
2. Changes in Soil Organic Carbon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
3. Potential for Sustainable Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
4. Contrasting Profiles of Soil Productivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
5. Millet Grain Yield Response to Fertilizer and Crop Residue Application . . . . . . . . . . . . . . . 29
6. Simulated Total Soil Carbon Levels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
7. A Scenario of Carbon Sequestration over 20 Years Resultant from Nutrient Recapitalization in
the East African Highlands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
8. Yield of Sorghum Grain - 5 year Moving Averages, Saria, Burkina Faso . . . . . . . . . . . . . . . 54
9. Yields of Monocropped Sorghum at Saria, Burkina Faso . . . . . . . . . . . . . . . . . . . . . . . . . . 55
10. Diagrammatic Representation of the Effects of the Introduction of Cotton on Soil
Development in East Senegal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
11. Agroecological Zones in Semiarid and Subhumid Sub-Saharan Africa . . . . . . . . . . . . . . . . 65
LIST OF ABBREVIATIONS/ACRONYMS
kg ha -1
The Center for Agriculture and Rural Development
cation exchange capacity
Conservation Technology Information Center
the national agricultural research organization of Ethiopia
effective cation exchange capacity
Food and Agriculture Organization of the United Nations
Fertilizer Use Recommendation Project
International Centre for Research in Agroforestry
International Crops Research Institute for the Semi-arid Tropics
International Fertilizer Industry Association
International Fertilizer Development Center
International Food Policy Research Institute
International Institute of Tropical Agriculture
kilograms per hectare
net primary productivity
Soil Fertility Initiative
soil organic carbon
soil organic matter
t ha -1
tons per hectare
United States Agency for International Development
West African semi-arid tropics
1 In this document, “fertilizer” means “inorganic fertilizer.”
2 Soil organic matter is defined, in this paper, as all living and dead biotic components of the soil
including plant roots, residues, bacteria, fungi, earthworms, etc.
3 The term recapitalization implies that one is aiming to return to a previous (higher) level of fertility
and SOM. There is also the possibility of building SOM to levels beyond the native (pre-disturbance) level
suggesting perhaps the terms capitalization or intensification. Since the focus of this paper is for SSA as a whole,
1. BACKGROUND, OBJECTIVES AND METHODS
While recognizing the economic obstacles that currently block widespread use, many concerned
with improving agricultural productivity and food security in Sub-Saharan Africa are focusing on
fertilizer1 as a remedy for declining soil quality and stagnant yields. Some have suggested that
SSA needs to increase fertilizer use from 9 to 30 kilograms per hectare (kg ha -1) during the next
decade (Borlaug and Dowswell 1995). Others fear that increased use will have undesirable
environmental impacts (soil acidification, water pollution) that could outweigh the benefits (Pretty
1995). Much of the SSA literature on agricultural productivity and soil quality presents extreme
views for or against fertilizer; supporting arguments are often more ideological than technical.
Both the technical and economic evidence underlying these arguments need to be understood by
those designing policies to promote agricultural productivity and reverse declining trends in SSA
soil quality. This paper focuses primarily on the technical/biophysical evidence, complementing
other Michigan State University research that presents economic evidence (Yanggen et al. 1998;
Weight and Kelly 1998).
1.2. Objectives and Methods
This study reviews agronomic studies from SSA and elsewhere that examine (1) environmental
and agriculture-induced constraints to agricultural production and (2) fertilizer-based strategies
that have the potential for increased productivity and sustainability in SSA that address these
constraints. The following questions will be addressed:
What are the positive impacts of fertilizer?
What are the dangers or negative impacts of fertilizer?
How can the positive impacts be maximized and the negative impacts be reduced?
What national strategies have the greatest potential to achieve positive impacts and
Soil recapitalization, as used in this paper, is the replenishment of SOM2 and associated soil
fertility as C and nutrients are added to the soil (inflows) to replace C and nutrients removed from
the soil (outflows) by (i) decomposition and mineralization and (ii) harvests, erosion, runoff,
leaching, nitrogen (N) volatilization, and denitrification.3 Soil fertility is considered a form of
the term recapitalization is used most extensively because the primary goal is to begin the task of returning to pre-
disturbance levels, a very difficult task on such a large scale. On a local or farm level, other terms may be used.
renewable natural capital with service flows (crop production, food security) that increase with
recapitalization (inflows) and decrease with excessive outflows. The objective of recapitalization
is not to build up maximum stocks of nutrient capital, but “appropriate” stocks of nutrient capital
which can provide sustainable levels of nutrients to crops.
Excessive outflows result in declines in SOM and degradation of soil physical properties,
especially aggregation of soil structure resulting in lower levels of efficiency for delivery of
nutrients and water to crops. The primary vehicle for reversing such degradation of soil quality is
by increasing levels of SOM.
In this paper, a combination of fertilizers, organic inputs, and beneficial agricultural practices are
recommended which have the potential to recapitalize soils and provide an efficient soil physical
environment for delivery of nutrients and water to crops. The ultimate goal is to return to a level
of soil fertility across SSA that approximates pre-disturbance levels of fertility and results in
increased yields, efficiency, and profitability.
4 USDA soil classification system will be used in the paper.
5 African Alfisols typically have considerable constraints including weak soil structure, vulnerability to
erosion and lack of infiltration due to compaction (Lal 1997).
6 Unless they are cultivating specific crops that prefer acid environments
2. REVIEW OF ENVIRONMENTAL FACTORS
The most critical environmental factors that determine a sustainable system are climate and soils.
Systematic analysis of these factors can serve as a basis for the determination of guidelines and
policies (Stewart et al. 1991). Most African soils have inherent difficulties for agriculture in terms
of fertility, acidity, or drainage which, in many cases, can be overcome with proper management.
A recent general classification of African soils provides data for distribution of soil types: “acid
infertile soils” (Oxisols and Ultisols)4 21%, “very infertile sandy soils” (Psamments) 13% and
“poorly drained soils” (Aquepts) 2%. On the other hand, “Moderately fertile, well-drained soils”
(Alfisols,5 Vertisols, Mollisols, Andepts, Tropepts and Fluvents), which account for 33% of Asian
soils, represent only 19% of African soils (Brady 1990; Eswaran et al. 1997) (Figure 1).
Low soil nutrient reserves are common in tropical ecosystems: “About 36% of the tropics (1.7
billion ha) are dominated by soils with low nutrient reserves, defined as having <10% weatherable
minerals in the sand-and-silt fraction. This constraint identifies highly weathered soils with limited
capacity to supply C, phosphorus (P), potassium (K), magnesium (Mg), and sulfur (S). Soils with
low nutrient reserves are more extensive in the humid tropics (66%) and in the acid savannas
(55%) but are locally important in the Sahel. It is relevant to note that about two-thirds of the
soils in the tropics (64%) do not suffer from low nutrient reserves.” (Sanchez and Logan 1992,
Because farmers universally seek out soils that are high-base-status, non-acidic soils,6 the
percentage of African cultivated soils that are moderately fertile is expected to be considerably
higher than the above data suggest. Precise information for cultivated soils is not available due to
the lack of a land use database classifying cultivated lands (Russell Almarez, NRCS World Soil
Resources, personal communication).
It should be pointed out that certain tropical soils which are commonly categorized as “infertile”
such as the Oxisols actually possess great potential for agricultural production if managed
The management of Oxisols presents both problems and opportunities. Most of them
have not been cleared of their native forest vegetation or have been tilled using only
primitive methods. The few instances where modern farming techniques have been used
have met with mixed success. Heavy fertilization, especially with phosphorus-rich
Figure 1. Distribution of Soil Orders
Source: Reproduced with permission from USDA/Natural Resources Conservation Service
(Figure 1. Removed to reduce download time. Please download separately.)
7 The rate at which organic matter is decomposed.
8 Atmospheric CO2, water and inorganic ions.
9 Litterfall is the annual transfer of living plant material to non-living forms of organic matter, from both
above-ground (leaves, grasses, etc.) and belowground (roots) sources.
10 The contribution of perennials to vegetative biomass production increases with rainfall and with the
transition from one to two rainy seasons (Breman and Kessler 1995).
materials, is required. Deficiencies of micro nutrients are common. In some areas,
torrential rainfall makes cultural practices that leave the soil bare extremely harmful....
Research on crop production on Oxisols suggests that the potential of some of these soils
for food and fiber production is far in excess of that currently being realized. In Brazil and
central Africa, selected areas of these soils have been demonstrated to be high in
productivity when they are properly managed. (Brady 1990, 73)
Organic matter levels, however, are not inherently lower in the tropics and Africa than in the
temperate zones, despite early literature to the contrary. The “myth” of low organic matter levels
in the tropics was initially based on misunderstandings of soils such as Oxisols that had red
coloring rather than black/brown coloring associated with organic matter content. Later scientific
studies in North America led to an apparent inverse relationship between annual temperature and
organic matter content. Subsequent studies revealed that such findings could not be extrapolated
to the tropics. When comparisons were made between soils of the same order in temperate and
tropical ecosystems, it was found that organic matter levels were comparable (Greenland, Wild,
and Adams 1992). It is the turnover rate of organic matter7 in the tropics that is different.
Research has shown that, in the humid tropics, decomposition losses8 in natural systems are
balanced by high biomass input, all of which are caused by high temperatures, high levels of
available moisture, and a 12-month period for decomposition (vs. 8-9 months in temperate
regions). For example, litter fall9 and decomposition/turnover rates are approximately five times
higher in tropical forest soils than in temperate forests (Parton et al. 1989). The critical problem
is that when agriculture is introduced, there are increased losses of SOM due to accelerated
In the Sahel, vegetative cover is very limited due to the extreme aridity of the dry season which
severely limits perennial plant life compared with areas with the same total annual rainfall. Thus,
levels of biomass input and SOM are lower under these conditions (Breman and Kessler 1995).10
In this case, agricultural disturbance exacerbates an even more difficult fertility situation, depleting
already low native levels of SOM. Also, the low levels of clay in Sahelian soils mean that SOM is
minimally (physically) protected from accelerated decomposition.
A critical factor concerning tropical and African soils is their level of diversity or variability which
is based on the high level of environmental diversity in tropical ecosystems. “In view of the
immense environmental diversity encountered in the tropics, often over short distances, the
complexity and variability of the resultant soils patterns should come as no surprise. The small
island of Puerto Rico may serve as an example: in an area of <9000 km2, soils representing 10 of
the 11 orders currently recognized in Soil Taxonomy have been identified.” (Eswaran et al. 1992,
Large-area reconnaissance maps of African soils were made after World War II and were based
primarily on surveys by a few individuals. Understanding and appreciation of the diversity of
these soils has only come about in recent years with the advent of more detailed soil surveys using
modern scientific methods including the Food and Agriculture Organization of the United Nations
(FAO) Soil Map of the World using a scale of 1:5,000,000 which was published in the 1970s
(Eswaran et al. 1992).
Research indicates that water is less available in Africa than elsewhere. African levels of available
water from rainfall, measured as precipitation minus evaporation (cm year -1 ) are low relative to
other continents. One survey of data (Brady 1990 citing Mather 1984) found African levels at
12.7 cm year -1 vs. North America at 25.8 and South America at 64.8 with a world average of
24.9. Water is absolutely critical for fertilizers to be effective. Besides the fact that water is a
requirement for plant growth, fertilizers are distributed to and within the rooting zones of crops in
To compound matters, low soil quality results in low water use efficiency in many regions of SSA.
For example, in the Sahel, it is common for plants to utilize only 10 to 15% of rain water (Penning
de Vries and Djiteye 1991). This is a serious constraint considering that rainfall in the region is so
limited in the first place. Also, variability of rainfall is a critical factor affecting fertilizer efficiency
and in determining risk-aversion strategies for farmers in SSA (Bationo 1998; Brouwer and
The high intensity of storms is the primary cause of high levels of soil erosion in the tropics when
compared with erosion levels of temperate regions. "Rains in the tropics, particularly those
caused by thunderstorms, have sharp, high-intensity peaks. Because tropical rains are caused by
convection, they are generally accompanied by lightning and thunder, are localized, and are
intense.... The result is intense downpours, high rates of rainfall per unit time, and relatively high
drop size.... Both the amount and the rate of rainfall, or its intensity, affect soil erosion. The same
amount of rain falling over a short time causes more erosion than when it is distributed over a
relatively long time and falls as a gentle rain of low intensity." (Lal 1990, 29-32) Likewise, the
high intensity of wind in tropical storms is responsible for severe wind erosion in the semi-arid
regions of SSA, especially in the Sahel. Vulnerability of the soil to wind and water erosion in the
region is exacerbated by the extremely limited vegetative cover.
Soil compaction is another serious constraint of some African soils. Coarse-textured soils with
low activity clays, such as West African Alfisols, are especially prone to compaction. This results
in soils with high bulk density and low total porosity with impaired seedling establishment,
inhibited root development, and low fertilizer and water-use efficiency since it is difficult for water
to infiltrate into the soil. In contrast, soils with a higher sand/clay ratio are more likely to be
limited by low nutrient content (Lal 1987).
Finally, drought stress is a major constraint in many regions of SSA. "The term ’drought stress’
implies crop response to the integrated effects of low available water-holding capacity, high
evaporative demand, and high soil and ambient temperatures. Compaction and high water run off
cause severe and frequent drought stress even in regions of high annual rainfall.... Crops
susceptible to drought do not respond to fertilizers and other chemical amendments." (Lal 1987,
3. FERTILIZER-BASED STRATEGIES THAT ADDRESS CONSTRAINTS
3.1. Agriculture-Induced Constraints
Scientists agree that the introduction of agriculture has caused significant declines of SOM and
associated soil quality relative to the undisturbed system (Woomer et al. 1994). Effects include
decreased aggregation, water-holding capacity, nutrient-holding capacity, soil macro-structure
and infiltration. Studies have been carried out that trace the long-term evolution of SOM under
various cultivation scenarios. Woomer et al. (1994) have reviewed a study by Resck et al. (1991)
showing trends in loss of SOM in disturbed systems in South America. Annual decline in SOM is
generally in the 1-2% range. One example is from the Cerrados region of Brazil. Under 2 years
of upland rice cultivation, SOM increased. The authors surmise that this is due to the
decomposition of root residue from the native vegetation. Rice was followed by soybean
cultivation. After 11 years of total cultivation, SOM and soil quality, especially aggregation,
declined at about 2% per year, consistent with declines in other studies.
Cultivation with low-input methods (no fertilizer) in the humid savanna zones of SSA can induce
a 30% loss of SOM after 12 years and 66% after 46 years, with rainfed rice yields declining from
1 ton/ha to only 300 kg ha -1 at the end of the period (Pieri 1992, citing Siband 1974).
Beyond natural constraints, the following factors are considered to be responsible for such
3.1.1. Soil Mining
Soil mining is the process by which farming removes more nutrients (and C) from the system than
are replaced. It is associated with low input agriculture in which low-nutrient value organic
amendments are used which are insufficient to replace nutrients extracted by crop harvests,
resulting in a negative balance of nutrient elements. Estimation of nutrient balances (inflows
minus outflows) is a common method used to evaluate soil mining. Estimates for 38 countries in
SSA suggest that annual loss of nutrients per hectare during the 1980s was 22 kg of N, 2.5 kg of
P, and 15 kg of K (Smaling 1993b).
A historical U.S. study shows that rapid loss of SOC – approximately 70% over a period of 35
years or 2% annually – occurs on virgin Kansas prairie when native vegetation is replaced with
annual wheat crops (Figure 2). Crop residues returned to the soil are not sufficient to offset the
annual removal of nutrients and C by crops and increasing erosion. Thus, inflows are not
sufficient to offset outflows and there is a negative balance in the system (“mining”) (Brady 1990;
Balesdent, Wagner, and Mariotti 1988). In a comparison with timothy grass plots, it is shown
that decline is greater under wheat because of the negative effects of wheat tillage on SOC and
that clay is the most important soil fraction for protection of C. “... the loss of soil C was clearly
greater under cultivation of wheat than under timothy grass, supporting the idea that the annual
tillage exposes physically protected organic materials to degradation by soil biota. Under both
11 Clay is known to physically protect soil C from decomposition loss to CO2.
12 As the primary building block of SOM, SOC is the most commonly used indicator for SOM. SOM
values (%) can be estimated by multiplying SOC values (%) by a factor of 1.7.
Figure 2. Changes in Soil Organic Carbon
Changes in amount and origin of soil organic C accompanying
the long-term cultivation of wheat on formerly virgin prairie
soil. Open circles denote total C, and solid circles represent C
of prairie origin with upper and lower points at different dates
for the 0- to 10- and 10- to 20-cm depth samples, respectively.
The straight, solid line shows the level of stable C.
Source: Reproduced with permission from Balesdent, Wagner, and
Mariotti (1988, 121).
cultures, C associated with clay was the most persistent.”11 (Balesdent, Wagner, and Mariotti
While most texts attribute declines in SOM and SOC12 primarily to tillage, some authors consider
mining, especially of C, to be more important. "The loss of SOC upon conversion to arable
agriculture is traditionally attributed to the physical effect of tillage, which can disrupt soil
aggregates and expose previously inaccessible materials to rapid decomposition. Although this
process is undoubtedly a factor, the SOC loss upon cultivation reflects multiple effects. Indeed,
the primary reason for C loss may be the enhanced removal of C from agroecosystems; since the
intent of agriculture is to trap C in a marketable form, C inputs in agricultural systems are usually
lower than those in native systems (e.g., Voroney et al. 1981)." (Janzen et al. 1997, 61-2)
The Role of Soil Nutrient Budgets in Assessing Nutrient Balances: If sufficient nutrients and
carbon are supplied by fertilizer and residue, the potential is created to recapitalize SOM so that it
has the physical capacity to maximize retention and minimize losses of nutrients. Smaling
(1993a) has written extensively on the subject of soil nutrient budgets and developed the
following system of inflows and outflows for N, P and K. Inflows are fertilizers, organic matter,
wet and dry deposition (from the atmosphere). Outflows are harvested product, crop residue
(accounting for the fraction removed), leaching, gaseous losses (e.g. denitrification) and erosion.
In commercial agriculture, the point is to maximize the first outflow, harvested product.
Minimizing all other outflows has the effect of channeling NPK to the harvested product.
In a study of one district in Kenya, the two strongest outflows were harvested product (a positive
outflow-from the point of view of the farmer) and erosion (a negative outflow). Kissii is a
highland district with high rainfall resulting in runoff on sloping clay soils. A management system
that is responsive to these factors would be comprised of first, inflow management with sufficient
fertilizer to support crop growth and second, outflow management that promotes high yield
(harvested product), high levels of residue return and minimal erosion. The optimal combination
of inflow/outflow management would result in high profits that were sustained over time.
In an undisturbed system, a balanced equilibrium exits where inflows and outflows are balanced.
In a degrading system, as seen in the early years of the Kansas example, there are high outflows
relative to inflows. Under recapitalization, we seek ways to have high inflows relative to outflows
with the goal of increasing SOM and, thus, reaching a “fertile” level of nutrient capital in which
sustainable levels of nutrients are supplied to crops. When this occurs, higher levels of nutrients
can be stored in organic forms and released over time. Traditionally, when fertilizers are applied,
high levels of N can be lost by leaching, particularly in sandy soils. High levels of P can be
chemically bonded so that the P is unavailable to crops grown in clay soils. Storage of nutrients in
organic forms (in SOM) can minimize these problems, increasing fertilizer efficiency and
providing sustainable nutrient capital. If nutrient release is synchronized with the crop growth
cycle, nutrients become available at the time of greatest need.
Nutrient budgets can provide a valuable tool for measurement of nutrient flows. There are
serious flaws, however, in the current methodologies that are being used as a basis for budget
estimates, especially for soil erosion. Typically, controlled soil erosion experiments are carried
out on bounded small plots which produce highly inflated erosion rates, not taking into account
deposition of soil that occurs on an actual slope. These exaggerated rates have been used to
support the idea that soil erosion is severe throughout SSA (Stocking 1996; Sanchez 1998).
More recently, scientists H. Breman and N. de Ridder, working in West Africa, have developed a
method to translate data from small plots into an estimate of run-off on large plots (Hank Breman,
Director, International Fertilizer Development Center-Africa, (IFDC) personal communication).
More precise analyses suggest that while erosion is often severe in SSA, there is a great deal of
variability in the level of vulnerability of soils to erosion.
13 Biomass inputs to tropical systems are higher compared to temperate systems, again due to
temperature. This balances the high rates of loss due to decomposition of SOM in the natural system. Agriculture
disrupts this equilibrium by increasing decomposition rates beyond equilibrium levels.
Soil tillage has been used since the dawn of agriculture to alter the physical condition of the soil
and to prepare the seedbed for cultivation. Often its benefits, such as breaking down of clods and
increasing infiltration, have been found to be short-lived with the result that it is necessary to
repeat the operations regularly. Conventional tillage has been found to have significant
destructive influences on soils, especially in the case of intensive and mechanical tillage. In some
cases, the constraints for which tillage is used (e.g. bulk density, infiltration) are actually made
worse by tillage (see northern Nigeria experiment below). Cavigelli (1998, 26) lists the following
negative impacts associated with conventional tillage:
erosion (both wind and water)
burying residues so that they are not able to protect the surface from erosion and
are exposed to greater microbial activity
exposing SOM to oxygen (aeration) and increasing soil temperature both resulting
in increased decomposition
physically breaking up soil aggregates and exposing the internal SOM to microbial
Reviewing the above points, erosion can result in the loss of the fine particles of the surface soil
which are associated with organic and nutrient content with the result that soils can become
completely denuded of organic material and vegetation. Secondly, residues are a critical factor in
protecting the surface from erosion and will decompose more rapidly when buried and exposed to
soil microbes. Thirdly, research in SSA has indicated that clay fractions in soils physically protect
SOM and organic N from aeration and high temperatures associated with increased
decomposition. If the soil is torn open by tillage with exposure of SOM, increased decomposition
will result. Finally, aggregation is perhaps the most important factor in building a well-structured
soil and may be described as the “glue” that holds the soil matrix together. Continuous
conventional tillage results in the destruction of this fragile component of the soil matrix. The
primary cumulative impact of tillage, based on these factors, is increased decomposition of soil
organic matter in which organic C is lost to the atmosphere as CO2 and SOM declines. Since
rates of decomposition are higher in the tropics due to increased temperatures, the impacts of
tillage are more severe with greater declines in SOM over time.13
Tillage is also associated with decreases in soil quality, due to SOM losses. Four-year
experiments at Zaria in semi-arid northern Nigeria concluded "that soil bulk density measured 7
weeks after planting increased with increasing intensity of mechanical tillage. Accordingly, the
percentage of water stable aggregates was more in untilled and less intensively tilled than in tilled
14 Lower erosion rates on no-till plots supports the relationship, described previously, that tillage is
associated with accelerated erosion.
treatments." (Lal 1987, 578, citing Dunham and Aremu 1979, Aremu 1980, and Dunham 1982)
In another Nigerian experiment at Ife, scientists also observed higher bulk density as well as lower
porosity in tilled plots. Compaction and crusting were most severe on plots which had received
the most intensive tillage treatments (Lal 1987, citing Aina 1979). While these are the most
typical results, there are also contradictory findings at a number of sites.
3.1.3. Accelerated Erosion
"Soil erosion began with the dawn of agriculture, when people began using the land for settled
and intensive agriculture. In fact, soil erosion has been a quiet crisis and has plagued the land
since people began practicing agriculture by removing the protective vegetation cover and
growing food crops on disturbed soil surface.... Soil erosion is severe in all regions, temperate and
tropical, wherever the land is used beyond its capability by crop and soil management systems that
are ecologically incompatible." (Lal 1990, 10-12)
Long-term studies on agricultural plots in North America have quantified the impacts of
cultivation on erosion. Specific practices are associated with greater levels of erosion, even on
slight slopes (Paustian et al. 1997, 27) provide the following examples: "For the Sanborn plots
[Missouri], Gantzer et al. calculated that topsoil thickness had been reduced by 56% under
continuous corn and by 30% under corn rotated with oats, wheat, and perennial crops, compared
with permanent timothy grass. In long-term plots at Wooster, Ohio, Dick et al. estimated that
conventionally tilled plots lost about 3.7 cm more soil than no-tilled plots over 18 years, an
amount equivalent to about 500 g C m-2...."14
Lal has summarized regional erosion trends in SSA as follows: "The Sahel suffers from severe
wind erosion during the dry season and accelerated gully erosion during the much-awaited rains.
(FAO 1979) reported that in Africa, north of the equator, 11.6 % of the total land area is affected
by water erosion. High erosion rates are especially prevalent in the coastal regions of northwest
Africa. Soil erosion is equally serious in eastern Africa and is particularly menacing in the
Ethiopian highlands. The Ethiopian highlands are believed to lose over 1 billion t/yr of topsoil
(Brown 1981). Gully erosion is catastrophic in some parts of southeastern Nigeria. Soil erosion
is also severe in southern Africa whenever large-scale farming is practiced without appropriate
conservation measures." (Lal 1990, 15)
3.1.4. The Process of Decline in Soil Organic Matter in SSA
Woomer et al. (1997, 154) have described the process of decline in SOM in smallholder farms of
SSA as follows:
15 In humid savanna zones of SSA there is no soil quality improvement for a fallow of less than 10 years;
in many cases fallows of 15 to 30 years are required to recapitalize soils adequately (Pieri 1992, citing Charreau
and Nicou 1971).
16 Based on data from the Kenya Fertilizer Use Recommendation Project (FURP 1994) in Woomer et al.
System carbon dynamics within small hold agriculture may be viewed as a three-step
process. The first two steps are essentially the same as the well-documented case of
shifting agriculture (Nye and Greenland 1960; Sanchez and van Houten 1994). Initially,
carbon and nutrient stocks which have accumulated within natural vegetation are
mobilized through land conversion. Felling large volumes of vegetation often necessitates
burning simply to obtain access to the land for cultivation and as a result much of the
above-ground biomass carbon is lost to the atmosphere.... Next, the soil resource base is
exploited through productive cropping for several years while nutrient-rich, mineralizable
organic matter and root residues decompose. Reduced yields are often associated with
declining soil organic matter levels. At this point traditional shifting agriculture abandons
land to fallow allowing for re-accumulation of carbon and nutrients in vegetation and soils.
This is also the point of departure of smallholder systems in Africa, where population
pressure has decreased farm size to the point where natural fallows are precluded.15 Thus
the third step in carbon dynamics in small hold farms, where the most labile soil organic
matter fractions have become mineralized at a lower-level equilibrium of soil organic
matter is approached. Continued productivity becomes dependent upon the application of
external inputs or the development of indigenous solutions which make better use of
locally available and under-utilized organic resources (Swift et al. 1994). Typical
responses by African farmers to reduced fallow have been crop rotations, manuring, and
composting (Binns 1992).
While the three-step process for carbon dynamics in SSA is different from processes in temperate
ecosystems, the outcome is similar; a significantly lower level of SOM (lower level of equilibrium
vs original pre-disturbance levels). In the earlier Kansas example, only 30% of original SOM
remained at the lower equilibrium level (see Section 3.1.1.). In a national survey in Kenya, only
28 to 33% of SOM remained at the lower level, depending on soil type. While the levels are
similar, the amount of time required to reach these levels was very different. In the Kansas
example, the degraded level was arrived at after 35 years. In the Kenya example, the equivalent
degraded level was achieved in only 24 years.16 This is not surprising considering the increased
rates of decomposition of tropical SOM.
3.2. Geographical Estimates of Constraints and Potentialities of African Soils
3.2.1. Distribution of Soil Types
17 The buffering capacity of a soil is a measure of its ability to resist changes in pH; highly buffered soils
are more resistant to acidification, which inhibits crop growth, than poorly buffered soils.
The African landscape has been mapped by Eswaran et al. (1997) based on estimates of its
potential for sustainable development (Figure 3). Such maps provide an estimate of the types of
soil constraints that need to be addressed in intensification efforts. Broad categories of lands are
Prime land contains highly buffered soils with high levels of SOM and good water retention.17
Soils are deep, with excellent tilth, and have few impermeable layers. They comprise
approximately 10% of the African land surface and exhibit little to no decline in SOM or fertility
under various soil management systems. As a result, they have the greatest potential for
High potential land is similar to prime land with some limitations such as “extended period of
moisture stress, sandy or gravelly materials, or with root restricting layers in the soil.” (Eswaran
et al. 1997, 16) These soils (7% of land) are vulnerable to declines in SOM and fertility under
low-input agriculture but they have good potential for recapitalization. However, if they are
mismanaged with continuous mining, they may become degraded with low soil fertility and
Medium to low potential land. These soils (28% of land) have significant constraints and are very
vulnerable to declines in SOM when cultivated with low-input techniques. Risks of crop failure
can be very high unless proper management techniques are applied. “The constraints include
adverse soil physical properties including surface soil crusting, impermeable layers, soil acidity and
specifically subsoil acidity, salinity and alkalinity, and high risks of wind and water erosion.”
(Eswaran et al. 1997, 16) Many areas of the southern Sahel region are classified under this
category as low potential lands.
Humid tropical forest soils, especially Oxisols, found in Central and West Africa, are also
considered to be medium potential soils. Constraints are primarily acidity and significant fixation
of phosphorus. As described earlier, farmers may only achieve good yields if proper management
practices are followed. Otherwise, crops will fail and soils will be degraded.
Marginally sustainable and unsustainable land (57% of African total) has poorly buffered soils
with very low SOM and very poor water retention. A large share is not arable (e.g., the Sahara
Desert). The arable portions are on the fringes of deserts where both water availability and
nutrients are limiting. These lands “are considered to be fragile, easily degraded through
management, and in general are not productive or do not respond well to management. They are
generally highly erodible and generally require very high investments for any kind of
Figure 3. Potential for Sustainable Development
Source: Reproduced with permission from USDA/Natural Resources Conservation Service
(Figure 3. Removed to reduce download time. Please download separately.)
Source: Reprinted with permission from Anderson, J., and J. Thampapillai. 1990. Soil
Conservation in Developing Countries: Project and Policy Intervention, p. 9.
Washington, DC.: The World Bank.
agriculture.” (Eswaran et al. 1997, 15) The process of intensification for soils in this classification
is not fully understood, but there is some evidence that it is possible.
3.2.2. Addressing the Constraints of Marginally Sustainable (“Marginal") Soils
There is a concern by some scientists that “marginal” (or “fragile”) soils, with low native levels of
SOM and cation exchange capacity (CEC) which have been further degraded by agricultural
activities may not have the potential to regain their fertility or productivity.
Figure 4 represents two hypothetical recapitalization scenarios – one for “highly-buffered soils,”
as found in the Kansas prairie, and another for “fragile soils”commonly found in semi-arid regions
of SSA. The lower “fragile” curve suggest that there is a point-of-no-return at which SOM is so
depleted that soil productivity may not recover after the introduction of soil conservation
practices (Anderson and Thampapillai 1990, citing Ragland and Boonpuckdee, forthcoming).
This concept of “fragile” soils or lands is based on the idea that marginal poorly-buffered soils
with low CEC and SOM can reach a point where they are permanently “damaged” and may not
recover their fertility or productivity. According to Pieri (1992, 180): “... experiments on less
fertile soils which occur more frequently, showed that there is a critical level for soil organic
Figure 4. Contrasting Profiles of Soil Productivity
18 This critical level of SOM formulated by Pieri is based on particle size distribution. It is a function of
the level of clay and silt particles: (SOMcrit in g kg-1) = 0.05 x (%clay + %silt). Groot et al. (1998), citing Pieri
19 Similar increases in SOM were also found with addition of manure, also in Niger (Bationo and
20 One important finding from the study was the strong decrease in percent Al + H saturation with
addition of residues. The decrease was from 48% to 20% saturation with residue alone; it changed to 16% with
residue and fertilizer. This suggests that a key factor linking residues with increases in crop yield is suppression of
Al toxicity in soils.
matter at or above which yields are maintained.18 While there is no need to keep organic matter
levels greatly above that level, and it would not be practically possible to do so, to maintain the
critical level is not impossible and it is essential to do so. Once organic matter impoverishment
passes below the barrier, yields decrease catastrophically.”
Some soil scientists have estimated that soils below a critical level of 0.6% organic matter suffer
damage to soil structure resulting in irreversible erosion that precludes the possibility of
recapitalization (van der Pol 1992). “... There is no smooth relationship between the decline in
soil properties, organic matter, and nutrients and especially structural organization of the soil
profile and yield. Yields decline seriously only when soil properties fall below a critical level.”
(Pieri 1992, 113)
In this paper, the term “marginal” is used rather than “fragile” because the term “fragile” implies
permanent damage and the evidence for such a permanent effect is inconsistent. First, there are
research results on marginal soils which show that yields can be increased significantly using
fertilizers with crop residues or other organic amendments. For example, one three-year study on
sols fatigues (“tired soils”) at Sadore, Niger, showed a fifteen-fold increase in millet yield when
fertilizer and residues were added (Bationo and Mokwunye 1991).19 Soil organic matter levels
increased from a low 0.24% (well below the 0.6% “critical” level) to 0.33% in these treatments,
showing the potential for some level of recapitalization over time (Bationo, Christianson, and
Bationo and Mokwunye (1991, 221) also discuss another study from Niger by Geiger, Manu, and
Bationo (1988) in which SOM does not increase after long-term additions of residues: “after 5
years of the addition of crop residue, the levels of calcium (Ca), Mg, and K had increased
significantly in the top soil (0-20 cm).... It was also found that the organic matter levels in the soil
did not increase significantly after 5 years with the addition of crop residue.”
Based on SOM studies including simulation and validation models, Breman and Sissoko (1998)
conclude that it is technically feasible to restore soils that are below the "critical" level of SOM to
higher levels of fertility but such restoration will require intensive management (see below).
Organic matter is vulnerable to leaching in “marginal” sandy soils. While the rate of leaching of
organic matter is less than that of inorganic nitrate, it can still be significant. In one experiment on
sandy soil in Niger: “Calculations showed that within 1 year after the application of 10 tons per
21 For example, millet is used in Northern Australia in rotation with sorghum to bring "lost" nutrients
back to the surface (Breman, personal communication).
hectare (t ha -1) of cattle manure (a rate that does occur in farmers’ fields), soil nutrient store ...
had increased by 91 kg ha -1 of N, 19 kg ha -1 of P, and 1070 kg ha -1 of organic C.” (Brouwer and
Bouma 1997, 23, citing a study by Brouwer and Powell 1995) Unfortunately, these levels were
stored at a depth of 1.5 to 2.0 meters! The authors suggest that similar levels of nutrients and
organic C were also stored below the 2.0 m depth. Such severe leaching, in this region, is
associated with torrential rainstorms and specific micro-topography variables. Thus, it is possible
that leaching could be responsible for the lack of success in recapitalization of these soils; the
build-up could be occurring in the subsoil rather than the surface soil where it is needed. There
are plant root systems in semi-arid soils of the Sahel which extend to a 1.5 to 2 m depth which
have the capacity to access subsoil nutrients (Penning de Vries and Djiteye 1991; Breman and
Numerous studies have been conducted which show the potential afforded by significant organic
inputs for the restoration of degraded soils including the improvement of soil structure and
fertility. Among these, there are several which address the potential to restore soils to a higher
level of fertility/SOM than native levels. For example, Padwardhan et al. (1997), citing Johnson
(1995) have developed a hypothetical model in which SOM reaccumulates under "new
management" (after significant losses with agricultural disturbance) at a higher (new) steady-state
than the native, pre-disturbance steady state. Breman (1998) actively supports efforts of "eco-
intensification" to achieve such new/higher steady-state levels of soil fertility for soils in SSA.
However, it is difficult to increase SOM levels in some regions, especially in the Sahel, due to the
length and severity of the dry period which is long enough to decompose most biomass. It is
estimated that it takes between 10-30 years of intensification to build up SOM in this region,
depending on soil type, quantity, and quality of inputs. Breman (personal communication)
suggests that soil fertility initiatives with fertilizer use should begin in regions with higher potential
than marginal regions (both agroecological and socio-economic) to create fertilizer demand and
input market development, both of which are required for addressing the constraints of marginal
3.3. Fertilizers, Organic Matter and the Carbon Cycle
When combined with recycling of organic materials, the primary positive impact of fertilizers is to
increase the biological base of the plant/soil system resulting in increased crop yields and
recapitalization of soils. When fertilizers or organic inputs are applied, essential nutrients are
supplied for the creation of plant biomass by means of photosynthesis. In the process, CO2 is
incorporated or fixed into the biomass from the atmosphere which is then referred to as organic
C. “Ecologists call the production of plant biomass from sunlight, water, atmospheric CO2 and
nutrients primary production. Primary production is based on photosynthesis and is the basis for
the global food chain. During photosynthesis, energy from sunlight is stored in the chemical
bonds holding carbon atoms together.” (Robertson 1998, 6) When fertilizers are applied, the
22 Mineralization is the conversion of an element from an organic to an inorganic form by microbial
decomposition. Plants require mineralized, inorganic nutrient elements for growth.
increased availability of nutrients to the plant creates an increased capacity for absorption of the
above “ingredients,” thereby increasing the biology and productivity of the soil system.
The biological health and sustainability of an ecosystem is typically assessed by its NPP which is
“the amount of plant biomass produced during a given time period within a particular ecosystem.
Ecosystem NPP depends on the plants’ photosynthetic efficiency, leaf area, leaf duration and on
water and nutrient availability.” (Robertson 1998, 7) Again, the Sahel provides an example of
environmental limitations of agroecosystems in SSA. The region has severely limited NPP due to
"sub-optimal" conditions which result in limited biomass, SOM, nutrients and water levels. Rates
of growth of plants and crops are three to five times less than the maximum "production
potential" which could be achieved if nutrients and water were not limited. NPP is also limited
due to the short season of rapid growth (Penning de Vries and Djiteye 1991).
Plants are typically 40-45 % C on a dry-weight basis (Cavigelli 1998). When the plant is
harvested, approximately 50 % of the above-ground biomass is removed as grain in North
American ecosystems (Robertson 1998). However, in SSA, this "harvest index" is much lower,
averaging 15-30 % (Breman, personal communication). For example, a local millet variety in
Niger has a harvest index of approximately 20%. That which remains, including below-ground
root biomass, is considered plant residue, a primary precursor to SOM.
Residue organic matter (OM) serves as a critical nutrient source for crops. Over 95% of N and S
of surface soils are found in SOM as well as 20-75% of P (Duxbury et al. 1989). Residue micro
nutrients that the crop has taken up from the soil are also recycled back into the soil as SOM.
These micro and macro-nutrients in SOM may be considered as a source of nutrient capital which
is mineralized and becomes available to crops over the long term.22
Second, carbon in the residue OM becomes a food/energy source for soil microorganisms
including fungi, bacteria and nematodes. Five to fifteen percent of residue C is incorporated into
microbial biomass in this way (Cavigelli 1998).
Although the microbial biomass carbon pool generally represents less than five percent of
the total soil organic carbon pool, it is fundamental to the functioning of any ecosystem
and is crucial in developing SOM. As a result of microbial activity, carbon undergoes
many complex chemical transformations that are collectively known as decomposition.
Decomposition rates are influenced by factors that influence microbial activity:
temperature, moisture, aeration, pH, amount and quality of residue, residue particle size
and degree of burial in soil....
A certain portion of the carbon in residues and manure is readily decomposed and is thus
called short-term SOM. Short-term SOM provides some benefits to soil physical
condition, but it is mostly important as a short duration (one to three years) source of
plant nutrients (primarily nitrogen, phosphorus, and sulphur). Manipulating this portion in
23 This does not imply that these factors are not important in other regions as well.
seasonal patterns is absolutely essential to nutrient use efficiency and preventing nutrient
loss to the environment. (Cavigelli 1998, 21)
Remaining portions of decomposed C from crop residues (10-35%) are incorporated into long-
term SOM or humus. This more recalcitrant matter includes lignin and hemicellulose and has a
turnover rate in the 100s to 1000s of years. Although scientists have traditionally associated
aggregation and its structural benefits with humus, more recent research points to intensive
decomposition of residues in short-term SOM for this benefit . "Recently incorporated particulate
organic matter was shown to initiate aggregation by acting as a substrate for the fungi and
bacteria which aggregate soil particles through their associated mucilages or physical
enmeshment.... Golchin et al. (1994) proposed a model of micro aggregate (20-250 um)
formation around plant residues. They suggested that particulate SOM entering the soil is rapidly
colonized by a microbial population. The micro flora and its by-products have strong adhesive
properties, and mineral particles adhere to them.... The plant fragments are thereby rapidly
encrusted by mineral particles and become the center of water-stable aggregates." (Angers and
Chenu 1997, 201)
This explains the direct relationship that has been observed between crop residues and
aggregation and associated physical benefits including moisture infiltration and retention (water-
holding capacity), reduced erosion, and nutrient sorption (retention) or base saturation. Thus,
residues and other particulate organic matter increase SOM levels and aggregation/improvement
of soil physical structure simultaneously. Manure also has high potential for build-up of SOM and
aggregation. In this case, the processes involved may be more related to long-term humus due to
higher levels of recalcitrant products in manure, especially lignin.
Research has indicated that organic materials have the ability to reduce the P sorption (retention)
capacity of soil and thus increase P availability to crops. Addition of organic inputs can be
especially useful on certain soils, such as Oxisols, that are known to have low availability of P due
to P sorption despite the presence of medium to high levels of P in the soil. A variety of complex
organic reactions are responsible for this effect (Palm, Myers, and Nandwa 1997). Also, organic
inputs have been associated with increased root-length density resulting in increased uptake of P.
This is an important benefit since P is immobile and the plant depends on the root system to scarf
P (as opposed to N which is mobile in soil solution). Bationo et al. (1993, 318) write: "Hafner et
al. (1993) ... reported an increase of root-length density with [crop residue] CR application which
led to an increase in total P uptake from 3.4 to 10.6 kg P/ha."
Bationo and Mokwunye (1991, 218), citing work by Charreau and Nicou (1971) and Poulain
(1980), listed the primary benefits of OM in the West African semi-arid tropics (WASAT)23 as:
• Improvement of soil macro-structure.
• Increased water-holding capacity of the soil.
24 It should be pointed out that there are management systems, especially in the West, which do not
depend to a great degree on the soil for macronutrients or water but rather on reliable, very high inputs of
fertilizers with irrigation. In SSA, where such resources are not commonly available, it is necessary to rely on the
soil for nutrient and water storage and delivery for sustainability, especially for those years when fertilizers are not
readily available and rainfall is limited.
• Improved infiltration and erosion control.
• Prevention of soil hardening.
• Improved soil cation exchange capacity. This is of particular importance for the
sandy soils of the WASAT. For example, for the millet-producing soils of West
Africa effective cation exchange capacity (ECEC) is more correlated with the
organic matter content of the soil than with the clay content.
• Increased supply of slowly released inorganic nutrients ... ensures a steady release
of nitrogen at a time when the established crop can use it. This minimizes losses of
readily available nitrate-nitrogen through leaching.
• Development of a favorable environment for microbial activity in the soil.
• Prevention of phosphate fixation by iron and aluminum oxides.
• Certain substances like quinones and benzo-quinones which appear in the course
of transformation of organic matter may play a specific physiological role and
might increase the absorption capacity and length of roots.
• Increase in the resistance of roots to some diseases.
In the WASAT, residues on the surface of the soil are also critical for protection of the soil from
the desiccation and high temperatures of the dry season and the potentially severe wind and water
erosion of the rainy season (Mokwunye, Uzo and Hammond 1992).
Large quantities of crop residues are required to be effective in recapitalization of soils and
promoting the above factors. This is due to the fact that 60-75% of original residue C is respired
by soil organisms back into the atmosphere as CO2 (Cavigelli 1998). “Because a large proportion
of added residues and a portion of already existing SOM is converted to CO2 during microbial
decomposition, large amounts of residue are required to maintain or increase SOM levels.”
(Cavigelli 1998, 23) Even higher levels of residue are required in SSA due to more rapid
decomposition (turnover) rates that are associated with higher temperatures and 12 month periods
To build a sustainable soil base, it is critical that appropriate contributions are made to SOM.24
Certain crops are especially efficient at building up long-term SOM by providing high levels of
carbon from photosynthetic C. “Some plants (notably corn), warm-season grasses and common
25 Different portions of the plant residue have different C/N ratios and , thus, different “quality.”
Therefore, certain portions of the plant residue are recommended as forage for livestock. In this paper, the
“quality” of specific crop residues is based on the average C/N ratio of the crop residue and not that of a particular
26 um denotes a micrometer or micron which equals 1 millionth of a meter.
27 Immobilization is the microbial conversion of an element from the inorganic to the organic form.
weeds have a photosynthetic pathway dominated by four-carbon (C4) molecules. At high
temperatures these C4 plants can photosynthesize at much higher rates than their three-carbon
molecules (C3) counterparts such as wheat, soybeans and cool-season grasses.” (Robertson 1998,
7) C4 maximum rates of photosynthesis are estimated to be about 50% higher than C3 rates
(Penning de Vries and Djiteye 1991).
This higher level of C fixation into the plant biomass in C4 plants results in higher levels of NPP.
Not surprisingly, the quantity of residues is also higher. Thus, higher levels of associated energy
and nutrients are incorporated into the soil system. As a C4 plant, maize/corn produces one of the
best crop residues (of primary crops) for build-up of SOM, exhibiting a high carbon content.
Contrary to traditional research findings, rotation with annual leguminous crops, such as
soybeans, results in lower levels of SOM, organic C and N (Omay et al. 1997). However,
rotations remain important for achieving agronomic goals, especially control of plant disease and
As in the case of C, researchers have “concluded that the major contribution of N in crop
residues, particularly low-quality [i.e., high C/N ratio] gramineous residues [such as maize], is
through the soil organic matter [SOM].”25 (Myers et al. 1994, 92) Nitrogen in maize residue is
sequestered primarily in long-term SOM. Feller and colleagues found that 25% of N15 added as
maize stover residue to a sandy soil in Senegal was found in the new plant biomass with the
remaining 75% in the soil with no losses to leaching or runoff. Most of the remaining N was
located in the larger, more recalcitrant >50 um particle size fraction of SOM associated with
long-term SOM (Myers et al. 1994, citing Feller, Chopart, and Dancette 1987).26 The remaining
N is cycled relatively rapidly through the microbial biomass of the short term SOM. First, it is
tied up or immobilized into the microbial biomass. After a typical period of 1-3 years, it is
released or mineralized as inorganic N in soil solution as a crop nutrient.27
There is an inverse relationship between biomass production and "quality" (nutrient content and
digestibility by animals and microorganisms). For example, C4 plants such as maize have high
levels of production (with high carbon levels) but very low quality (low N levels) (Penning de
Vries and Djiteye 1991). Maize has a high C/N ration of 60:1. As a result, maize residues often
result in significant N limitations during the immobilization phase because soil microbes need all of
the available N in order to utilize C. “Decomposition of materials [such as maize] with N
concentrations of less than 2% (or C/N >25) lead initially to immobilization of mineral N, whereas
materials with higher than 2% N (or C/N <25) release mineral N.” (Myers et al. 1994, 91). Thus,
N becomes less available to the crop and crop growth is limited during the length of the
28 This immobilization effect will vary according to plant species.
immobilization phase. Soybean is an example of a low productivity/high quality plant with a C/N
ration of 30:1.
The negative impact of cereal residues, such as maize, on crop growth has been shown to be
consistent in agronomic trials. In one study, addition of 2.5 to 5.0 t ha-1 of maize stover resulted
in a 30 to 60% decline in available N (Palm, Myers, and Nandwa 1997, citing Ishuza 1987).
Losses of nitrate are most common and the length of the “nitrate depression period” can range
from several weeks to the entire length of the growing season depending on the quality of the
residue (Brady 1990). Since the addition of low-quality residues is common in SSA without
complementary N inputs, it is likely that long nitrate depression periods are a significant
contributor to low crop yields on the continent. Research has shown that the immobilization
effect can be offset by the addition of N fertilizer and/or high-quality organic inputs (Palm, Myers,
and Nandwa 1997).28
The IFDC-Africa Division, based in Togo, has been working with other international research
centers to learn which combinations of fertilizer with various organic inputs (residues, manure,
cover crops, green manure) from a variety of quality classes result in a beneficial "nutrient
equivalency value." This value is based on those soil fertility factors described earlier which
provide water and nutrient use efficiency and other benefits. It is considered to be more critical
than nutrient content alone (Breman, personal communication).
3.4. Fertilizer-Based Strategies that Address Constraints
Management systems or strategies for SSA will be most effective when they respond successfully
to site-specific environmental and agriculture-induced constraints outlined above and restore
depleted soil fertility and SOM. This approach is suggested because soil and water constraints are
frequently severe and have resulted in declines in productivity and sustainability in SSA. Lal et
al. (1997) list the following four "Site Specific Soil Management Options for C Sequestration:"
1. Soil Fertility and Nutrient Management (Macro nutrient [N, P, K], Micro nutrient
[S, zinc (Zn), copper (Cu)], Strengthening nutrient cycling mechanisms to
2. Tillage Methods and Residue Management (conservation tillage, cover crops,
3. Water Management (supplementary irrigation, surface and subsoil drainage, soil-
water management, water harvesting)
4. Erosion Control (runoff management, vegetative barriers, soil surface management
and mulch farming).
In SSA, it is not unusual that intensified management systems have focused primarily on nutrient
management and improved cultivars, similar to "green revolution" strategies in Asia. However,
because they have not addressed the other constraints via appropriate management systems, they
have proved to be unsustainable. Stewart et al. (1991, 142), citing El-Swaify et al. (1985)
provide the following example pertaining to Alfisols of West Africa:
The most abundant soils in the semiarid tropics are Alfisols, and these soils are extremely
vulnerable to erosion, crusting, compaction, drought, and limited rooting depth. Alfisols
contain predominantly low-activity clays and have low plant-available water reserves.
Improved management systems for conventional cropping of Alfisols have succeeded in
increasing yields of conventional crops, largely due to improved cultivars and use of
fertilizers. Effective practices for improving soil and water conservation, however, have
not been developed. This is primarily because of the extreme structural instability of these
soils.... Alfisols are inherently low in soil organic matter, even under native vegetation, and
once they are tilled, the organic matter becomes critically low.
The authors continue (p. 136) citing Hartmans (1983):
’Why do most tropical soils become unproductive and useless after only a few years, and
what can be done to arrest this deterioration?’ He stated that results of 15 years of
research at the International Institute of Tropical Agriculture (IITA) at Ibadan, Nigeria, in
pursuit of these questions are quite clear. Chemically, the land becomes more acid very
rapidly. Physically, the soil seems to collapse on itself. It becomes more dense, and
erosive forces often cause the finer particles to disappear, leaving a sandy or gravelly
material. The soil loses its capacity to form stable aggregates because the binding
material, the soil organic matter, is gone. The result is a rapid downward spiral of soil
This illustrates the direct relationship between loss of SOM (the primary determinant of fertility)
and soil structure with resulting declines in productivity.
3.4.1. Nutrient Management for Soil Fertility: Combining Fertilizers with Organic
Since this is a paper about fertilizer use and impacts, the primary management system being
addressed is nutrient management and soil fertility. However, this system cannot function
sustainably, especially in SSA, without close complementarity with the other three systems:
tillage methods and residue management, water management, and erosion control. These systems
and associated benefits will also be addressed.
Recent research and writings support the use of fertilizers in combination with organic inputs as
part of intensification strategies to drive sustainable growth in agricultural production in SSA and
end the long cycle of agricultural and economic stagnation ( Bationo and Mokwunye 1991;
Bekunda, Bationo, and Ssali 1997; Breman and Sissoko 1998; Pieri 1992; Quinones, Borlaug, and
Dowswell 1997; Reardon 1997; Swift 1996; Wallace 1997; Yanggen et al.1998). There is a
29 When possible, efforts should be made by researchers using labeled carbon to determine if apparent
increases in fertility are simply internal SOM transfers from long-term SOM to shorter-term biomass SOM or are
from sequestered C inputs. Brady (1990) notes that 60 years after the initiation of wheat farming, most of the
organic matter that remains (after soil mining) is the original native prairie organic matter which is a very “stable”
long-term form of SOM. After the initiation of recapitalization in 1950, an increasing portion of SOM can be
estimated to be derived from wheat residue rather than from long-term SOM since long-term levels remain
30 U.S. Alfisols do not have the severe structural instability of many African Alfisols.
consistent perspective, in these works, that neither input strategy, on its own, is capable of
achieving production goals. Quinones, Borlaug, and Dowswell (1997, 83) point out that
Increased fertilizer use in Africa can create a win-win situation, by promoting more
efficient crop production and reducing soil degradation. Mineral fertilizers should
be at the core of strategies to restore soil fertility and raise crop productivity,
although their use should be part of integrated systems of nutrient management in
which organic fertilizer sources are included. Organic sources of nutrients,
however, will be complementary to the use of mineral fertilizers, and not the other
way around. Exclusive use of organic fertilizers will increase food production at
best by 2% yr -1 (Hiyami and Ruttan 1985), well below the population growth rate,
and not even close to the 5 to 6% required to reduce poverty and assure food
The Kansas prairie study, (comparable to “high potential SSA land” described previously)
provides an example of how increased nutrient input via a combination of fertilizers and crop
residues can result in increased yields and C sequestration to SOC/SOM over time. In the study,
as in most of North America, fertilizer use increased dramatically in the 1950's, increasing crop
yields and the amount of unharvested crop residue that was returned to the soil (Figure 2) (Brady
1990, citing Balesdent, Wagner, and Mariotti 1988). As a result, SOC/SOM increased at an
annual rate similar to the decline noted in section 3.1.1. (1-2% per yr).29 The prairie soil is an
Alfisol which, by definition, includes a base saturation > 35% (in non-frigid climates) which
indicates that it is a well-buffered soil. Such U.S. prairie soils have been used as models to show
how soils can be improved with both fertilizer and residue input.30
Janzen et al. (1997, 71-2) have reviewed crop nutrient/ litter input/SOC/SOM relationships for
The rate of plant litter input in agroecosystems is closely related to crop yield. Numerous
studies have shown strong correlations between crop residue inputs and SOC contents
(e.g., Campbell and Zentner 1993; Biederbeck et al. 1994; Nyborg et al. 1995; Gregorich
et al. 1996). Many of the SOC gains in response to improved management practices can
be directly linked to higher yields arising from better crop nutrition, more efficient nutrient
and water utilization, and higher yielding crops. In part, the variable response of SOC to a
given management change depends on whether the new practice elicits a yield response.
For example, under semiarid conditions of western Canada, adoption of no-tillage [with
31 Success of no-till in humid regions of the U.S. suggests that its effectiveness is not limited to low
32 The percent of the total amount of an applied nutrient which is recuperated or extracted by the crop
from the soil.
33 It should be noted that not all nutrients are cycled to crops but are stored in SOM to improve the C/N,
C/P, and C/S ratio.
high inputs of crop residues] can maintain or enhance crop yields (Lafond et al. 1992)
because of greater moisture retention, thereby favoring higher SOC (Campbell et al.
1995). Under humid conditions like those in eastern Canada, however, reduced tillage
may have little yield advantage and therefore elicit only limited gains in SOC (Angers et al.
1995; Angers and Carter 1996).31
Research trials by the IFDC have shown the impact of increased nutrient use efficiency (from soil
improvement) to increased yields. “IFDC showed on its research fields in Togo that the efficiency
of the “national recommended package of fertiliser” increased 2 to 3 times through soil
improvement. At the start, on relatively good soils (> 10 year fallow), the nutrient recovery was
only about 30%, leading to an increase of maize yield with 900 kg. After 4-7 years of soil
improvement, the nutrient use efficiency increased 2-3 times, and maximum yield increases of
2000 to 3000 kg/ha have been measured with the same fertiliser package.” (Hank Breman,
personal communication) He suggests that the principal cause for the increase is not simply
supply of nutrients but, more importantly, the improvement of soil structure with increased SOM
levels resulting in increased efficiency of nutrient and water supplies to crops, increased
infiltration rates, decreased erosion and improved plant rooting. Without such improvement,
crops in the Sahel region, for example, have "recuperation rates"32 on average of only 35% for N
and 15% for P which are approximately half of typical rates elsewhere. Due to low efficiency and
high losses, the amounts of fertilizers required are too great to interest farmers in most cases
(Breman 1998, 6). With improved fertility and efficiency, there is a realistic potential to increase
these rates to 50% and 30% respectively (Groot et al. 1998).33
Lessons from Long-Term Experiments in SSA: Bekunda, Bationo, and Ssali (1997) reviewed
findings from long-term experiments (7-27 years) in SSA which compared fertilizer with organic
inputs and liming – both alone and in various combinations (Table 1). Countries represented were
Kenya, Nigeria, Uganda, Zambia, Tanzania, Chad, Burkina Faso, Senegal, and Cote d’Ivoire.
Use of fertilizer alone resulted in “measurable yield declines” in 9 of 13 cases. According to the
authors (p. 71), “Such declines [for fertilizer alone] might result from (i) soil acidification by the
fertilizers, (ii) mining of nutrients as higher grain and straw yields remove more nutrients than
were added (Scaife 1971), (iii) increased loss of nutrients through leaching as a result of the
downward flux of nitrate when fertilizer N is added, and (iv) decline of SOM.” According to the
extensive research of Pieri (1992), the first negative impact (soil acidification), especially from N
fertilizers, is the primary impact of fertilizers used alone (see below).
Table 1. Impact of Soil Management Treatments on Crop Yield Trends in Selected Long-term Experiments from Sub-Saharan Africa
descriptionduration Test crops
residues (C)Number Site
A + B or
A + C
A + Liming
1969 to 1990
1969 to 1990
1957 to 1974
1957 to 1974
3 Burkina Faso
1962 to 1984
1960 to 1983
Groundnut, millet +D
1981 to 1988
1981 to 1988
1966 to 1981
1966 to 1981
1966 to 1981
SerereFerralsol1937 to 1964Cotton, millet,
SamaruFerruginous1964 to 1975 Cotton, millet,
KabeteNitisol1976 to 1996Maize, bean +D+S+D++S
† +, yield higher than control; ++, yield relatively higher than + within the row; S, stable yield trend; D, measurable yield decline; DD, sharp yield decline.
‡ Sources: 1, Traore & Harris (1995); 2, 3, 4, Pieri (1995); Laryea et al. (1995); 5, 6, Singh & Goma (1995); 7, Research Reports, Serere Experiment Station; J.B. Byalebeka (1996, personal
communication); McWalter and Wimble, (1976); 8, Singh and Balasubramanian (1997); 9 = Swift et al. (1994) and S. Nandwa (1996, personal communication).
§ Approximate USDA Soil Taxonomy equivalents: Ferralitic, Oxisol; Ferruginous, Alfisol; and Luvisol, Alfisol.
Source: Reproduced from Bekunda, Bationo, and Ssali (1997, 72) with permission from The American Society of Agronomy and Soil Science Society of America.
34 The experiments were considered “successful” from the perspective of stable yield trends. They do not
include soil recapitalization parameters, however, for measurement of success. Thus, it is not possible to know if
the use of residues and manure are having the effect of building up SOM and nutrient storage.
35 As Bekunda et al. is a broad review of experiments across SSA, experimental details indicating types,
levels and timing of organic inputs are not included.
36 It is also possible that residues in the experiments were not protected from (uncontrolled) grazing or
other exports. Such lack of protection of residues in SSA experiments is not unusual.
One of the four experiments where there was no significant yield loss from fertilizer use alone was
Bebedjia, Chad. Crop yields were stable over time. According to Pieri (1992), the soils at this
site are unusually fertile and well-buffered with no erosion or deep leaching which may explain
their lack of vulnerability to these inputs. Presumably, these soils may be categorized as “prime
land” since they are not vulnerable to degradation.
Experiments which were “successful” in all cases were fertilizer combined with manure or
residue.34 As described previously, fertilizer, particularly when used on fertilizer-responsive
crops with high biomass production, “primes the photo synthetic pump,” helping the plant use
more of the available CO2 and water; resulting in more biomass production. When crop residue is
recycled, increased biomass nutrients and C (from the plant CO2) are captured into SOM, creating
a sustainable system for delivery and storage of plant nutrients and water.
Liming or fertilizer with liming was successful in maintaining stable yields in 4 of 5 experiments
which indicates the importance of pH in maintaining soil fertility for sustainable crop management.
Results for residue alone and manure alone were mixed. Manure alone resulted in stable yields
in 3 of 4 trials. Residue alone, however, led to a decline in yield in 3 of the 4 trials.35 The authors
suggest that these declines may have been caused by residues with high C/N rations (typically
from the primary crop, e.g. maize) leading to short-term N deficiencies due to N immobilization.36
There are experiments in SSA indicating that residue alone can be moderately successful in
increasing yields and, again, that fertilizer plus residue is a superior combination, even on
“marginal” lands. “In 1983, at the [International Crops Research Institute for the Semi-Arid
Tropics] ICRISAT Sahelian Center at Sadore, Niger, a trial was set up to study the effect of crop
residue and fertilizer on pearl millet production. Crop residue (4 tons ha -1 pearl millet stover)
was added to the soil surface in the first year to prescribed plots. In subsequent years the residues
produced were simply placed on the plot surface. After 3 years, addition of crop residue alone
had resulted in statistically the same amount of millet grains as plots to which fertilizers had been
applied.” (Bationo and Mokwunye 1991, 221) Specifically, residue alone raised yields from about
200 to 750 kg ha -1; fertilizer alone raised yields to about 900 kg ha -1 and; crop residue plus
fertilizer increased yields to about 1700 kg ha -1 (Figure 5) (Bationo and Mokwunye 1991).
Source: Reprinted from Bationo, A., and A.U. Mokwunye. 1991. Role of
Manures and Crop Residues in Alleviating Soil Fertility Constraints to Crop
Production: With Special Reference to the Sahelian and Sudanian Zones of West
Africa. In Alleviating Soil Fertility Constraints to Increased Crop Production in
West Africa, ed. A.U. Mokwunye, Fig. 5, p. 222. With kind permission from
Kluwer Academic Publishers..
It should be pointed out that research station results may more accurately reflect the reality of
production on medium to large scale western farms with access to investment capital, advanced
education and sources of information rather than actual conditions in SSA.
Despite the low average crop yields obtained under smallholder farm conditions, large
yield levels are achieved under research station conditions. What is common in ... all Sub-
Saharan African countries is a huge yield gap between on-station results compared with
those obtained under smallholder farmers’ fields. This huge yield gap can mainly be
attributed to the problems of managing soil fertility faced by smallholder farmers who are
constrained by cash to purchase farm inputs. Poor crop husbandry practices, which also
directly impinge on soil fertility, contribute greatly to the observed low crop yields. It is
also being recognized that technological innovations currently recommended to
smallholder farmers have not been fully adopted by this target group. The recommended
packages have failed to take into account the resource constraints and limitations of the
smallholder farmers, hence the low rate of technology adoption....
... smallholder farmers [in Malawi] who cultivate small land holdings of between one and
two ha per farm family of five people, are constrained by lack of cash to purchase mineral
fertilizers, and have limited access to credit facilities (Saka, Green, and Ng’ong’ola 1995,
Figure 5. Millet Grain Yield Response to Fertilizer and Crop Residue Application
Besides these factors, adoption of recapitalization is also dependent on farmers’ perceptions and
understanding of the technical potential of residue recycling. Farmers may not accurately value
the nutrient content of crop residues. For example, the straw produced on one hectare of millet
yielding 1.2 tons of grain contains 74 kg of K that could be recycled as residue. The harvested
(exported) grain actually contains only 6 kg of K (Bationo and Mokwunye 1991). As yield and
total biomass increase, crop residue availability also increases. In much of SSA, farmers use crop
residues as animal feed, fuel, or construction materials; this severely diminishes the role that
increased biomass associated with fertilizer use can play in recapitalizing SOM. A partial
solution to this problem is to target fertilizer to cropping situations where the increased
production of biomass will be the greatest, thereby providing farmers with biomass production to
meet both traditional and recapitalization needs. A more sustainable system is to develop
alternative sources for fuel, construction and forage so that all crop residues are available to
soils and crops.
Comparisons of Organic Inputs and Fertilizers: As illustrated in the previous section, there is a
growing consensus regarding the complementarity of fertilizer and organic amendments.
Fertilizer and organic matter each contain nutrients required by plants to create biomass via
photosynthesis, but sustainable intensification of SSA soils can be achieved by a combination of
both types of inputs. Fertilizer makes very little (if any) direct contribution to soil macro-
structure, increased water-holding capacity, improved infiltration and erosion control, prevention
of soil hardening, or improved nutrient holding capacity. But organic matter and fertilizer
combined have the capacity to make positive contributions to all of these factors.
There are limitations on the amounts of organic matter that are available as agricultural inputs in
SSA. Giller et al. (1997, 170), citing work by several other researchers, provide the following
example of the large area of grazing land that is required in West Africa to provide sufficient
manure to produce a significant maize crop. “Sandford (1989) estimated that 16 to 47 ha of
grazing land were required to produce sufficient manure for sustained maize production of 1 to 3
t ha -1 in a semiarid environment in West Africa. It is clear that there is insufficient manure to
sustain even such moderate yields in many parts of West Africa (Fernandez-Rivera et al. 1995;
Williams et al. 1995). There is also a danger of long-term degradation of grazing lands, as there
is substantial nutrient removal over prolonged periods.”
In many regions of SSA, significant quantities of manure go to non-agricultural uses which limits
their availability for agriculture. In Ethiopia, where livestock numbers are high, manure is used
primarily as a cooking fuel and rarely to improve soil fertility (Quinones, Borlaug, and Dowswell
1997, citing Giller et al. 1997). As described in a previous section, crop residues are often
exported for non-agricultural purposes as well. Breman (personal communication) estimates that
the organic matter that is available in the high potential cotton zone of Mali is only one-third of
the amount required to maintain crop production. He concludes that organic matter should not
be considered primarily as a nutrient source (since it is so limited in availability) but rather as a
complement to fertilizers (i.e., "organic amendment" rather than "organic fertilizer") which can
improve nutrient use efficiency and other beneficial properties of the soil.
Aside from high decomposition rates of residues, the reason that such large quantities of organic
materials are required for crops is the low concentration of these materials, especially when
compared with fertilizers. “Animal manures and plant material contain from 1 to 4% N (10-40 g
N kg -1 ) on a dry weight basis, while inorganic fertilizers contain from 20 to 46% N (200-460 g
N kg -1 ) and are already dry. To haul the 100 kg N generally needed for a 4 t ha -1 maize crop, it
would take 217 kg of urea or 20 t of leaf biomass with 80% moisture and a 2.5% (25 g N kg -1 )
N concentration on a dry weight basis. Furthermore, organic inputs are very low suppliers of P
because of their low concentrations.” (Sanchez et al. 1997, 8-9, citing Palm 1995 and Palm et al.
It is obvious from the figures in the previous paragraph that it is technically possible for a farmer
to use only organic inputs and have sufficient nutrients for yield goals and there are sufficient
examples of such practices in SSA in the literature. However, it would be imprudent for a
national government to recommend such a strategy when there are insufficient organic resources
available on a national or regional scale. Secondly, it is more difficult for African organic
farmers to achieve satisfactory yields because they are usually dealing with depleted soils and can
not count on the soil for significant nutrient inputs. Finally, most African farmers need to
recapitalize their soils for long-term sustainability. If one were to calculate the amount of
organic inputs (manure or plant material) that are required to satisfy both crop demands and
recapitalization requirements, the total amount would be enormous. Fertilizers provide a
concentrated response to this demand.
There are two principal reasons for the emphasis in this paper on crop residues as organic inputs.
First, residues are typically provided internally by the crop providing a direct and, therefore,
efficient source of organic inputs. That is, they are not imported from elsewhere thus requiring
an investment by the farmer as well as depleting the soil of OM in another location. Ideally, all
residues should be left on the ground because the quantity required for recapitalization is so high.
To the extent that livestock are raised and residues used as forage, all subsequent manure should
be returned to the soil. While there is some efficiency lost in the process, the quality of manure
as an amendment makes up for it due its high lignin content. Second, manure does not provide
vegetative cover for the soil (vs. residues which do), especially for erosion control. Therefore,
in a livestock/crop system using manure inputs, other management practices need to be
implemented such as low-lying cover crops to provide this function.
Fertilizer and Organic Matter Practices in SSA: Residue use as a soil amendment is common in
SSA compared to fertilizer use which is encouraging for the development of residue-based
strategies outlined in previous sections. Bationo and Mokwunye (1991, 217, citing Poulain
1980) state that “the amounts of the nutrients in crop residue of developing countries are seven
to eight times higher than the quantities of these nutrients applied as fertilizers in these
countries.” Segda (1991) cited in Bationo et al. (1993, 307-8) reviewed crop residue availability
in the Sudanian zone of central Burkina Faso concluding "that the production of cereal straw can
meet the currently recommended optimum level of 5 t/ha every two years. However, the
competition with other uses was not accounted for in this study."
Bationo et al. (1993, 307-8) write: "At the onset of the rains the residual stover on-farm was
only between 21 and 39% of the mean stover production at harvest time ... cattle grazing is likely
to be responsible for most of the disappearance of the crop residues. Similar losses were
reported by Powell (1985) who found that up to 49% of sorghum and 57% of millet stover
37 Other than a few comments, such as the preceding, this paper does not attempt to address the complex Download full-text
issues of integrated livestock/crop systems. Sources/researchers that do address this issue are: McIntire, J., D.
Bourzat, and P. Pingali 1992. Crop-livestock Interactions in Sub-Saharan Africa. Washington, D.C.: World Bank;
as well as the works of Salvador Fernandez-Rivera, Pierre Hiernaux, J. Mark Powell, Matthew Turner, and
disappearance in the subhumid zone of Nigeria was due to livestock grazing." Typically, in the
WASAT "grazing animals remove more biomass and nutrients from cropland than they return in
the form of manure, an exception being reported from Burkina Faso."
“Traditionally, many farmers burn whatever is left of their CR once their needs for fuel, animal
feed, or housing and fencing material have been fulfilled. Economic data collected recently show
the rationality of this strategy as mulched millet stalks increase weed growth and subsequently
labor requirements at weeding (Lamers, unpublished data). However, the same farmers may
conscientiously apply CR at a rate of up to 6 t/ha to counteract erosion and build up soil fertility
on selected spots of poor millet growth (Lamers and Feil 1993).”
In regions where livestock are an important component of the agro-ecological system, innovative
approaches need to be found by which crops, soils and livestock can be sustained. For example,
“intensive rotational grazing” incorporates perennial crops, such as alfalfa, into rotation with a
primary crop, such as maize (Cavigelli 1998) in areas of higher rainfall. Some fields may be
under maize while others are being grazed. Livestock are carefully controlled by portable fences.
The system has the advantage of building up SOM via perennial residues without the usual
export losses of residue for livestock fodder.37 In areas of lower rainfall, the system could be
adapted to include local grass species that are appropriate to the ecosystem.
There are regions in SSA, however, where use of fertilizers as well as residues and manure are
high. For example, in Central Kenya, 83% of farmers use fertilizers, 98% use crop residues and
manure; In Mutoko district of Zimbabwe, 98% use fertilizer, 77% use crop residues and 86% use
manure (Palm, Myers, and Nandwa 1997). Despite these pockets of high fertilizer use, in most
of SSA it is the adoption of fertilizers that is the greater obstacle to recapitalization efforts.
The Role of Fertilizers in Maximizing Water Use Efficiency: As stated in the first section, SSA
has less available water than most other continents. For this reason, there are those who state
that there can never be a truly productive agricultural sector in SSA. On the other hand, there is
convincing historical evidence that successful agricultural production is possible on the continent,
despite these limitations (see Section 5). There will need to be an in-depth analysis of water use
issues and the development of appropriate practices that can maximize water use efficiency if
SSA is to be successful in developing a successful and sustainable agricultural base.
The traditional point-of-view on water use is that fertilizers simply require more water than
organic sources to be effective and create a water deficit relative to organic inputs. Research
results, however, suggest that the reverse is true: fertilizer actually promotes water use efficiency
and conservation. By increasing plant biomass, canopy, leaf area, and root development, the
plant develops increased capacity for capturing and retaining water from rainfall (a process that
may be observed more dramatically in tropical rain forests). For example, field experiments were
conducted in Pakistan comparing two levels of NPK fertilizer treatment over three rainfall