integrated soil fertility management - home | food and...

Outlook on AGRICULTURE Vol 39, No 1, 2010, pp 17–24 17

Integrated soil fertilitymanagement

Operational definition andconsequences forimplementation anddisseminationB. Vanlauwe, A. Bationo, J. Chianu, K.E. Giller,R. Merckx, U. Mokwunye, O. Ohiokpehai,P. Pypers, R. Tabo, K.D. Shepherd,E.M.A. Smaling, P.L. Woomer and N. Sanginga

Abstract: Traditional farming systems in Sub-Saharan Africa depend primarily onmining soil nutrients. The African green revolution aims to intensify agriculturethrough the dissemination of integrated soil fertility management (ISFM). Thispaper develops a robust and operational definition of ISFM based on detailedknowledge of African farming systems and their inherent variability and of theoptimal use of nutrients. The authors define ISFM as a set of soil fertilitymanagement practices that necessarily include the use of fertilizer, organic inputsand improved germplasm, combined with the knowledge on how to adapt thesepractices to local conditions, aimed at maximizing agronomic use efficiency of theapplied nutrients and improving crop productivity. All inputs need to be managedin accordance with sound agronomic principles. The integration of ISFM practicesinto farming systems is illustrated with the dual-purpose grain legume–maizerotations in the savannas and fertilizer micro-dosing in the Sahel. Finally, thedissemination of ISFM practices is discussed.

Keywords: agronomic use efficiency; fertilizer; micro-dose; organic inputs; soilorganic matter; grain legume–maize rotation

B. Vanlauwe, A. Bationo, J. Chianu, O. Ohiokpehai, P. Pypers and N. Sanginga are with the TropicalSoil Biology and Fertility Institute, International Centre for Tropical Agriculture (TSBF-CIAT), PO30677, Nairobi, Kenya. E-mail: [email protected]. K.E. Giller is with the Department of PlantSciences, Wageningen University, PO Box 430, 6700 AK Wageningen, The Netherlands. R. Merckxis with the Department of Earth and Environmental Sciences, Faculty of Bioscience Engineering,Kasteelpark Arenberg 20, 3001 Leuven, Belgium. U. Mokwunye is Director, Mokwunye Enterprises,International Consultants, 4th Ayiku Lane, Baatsonaa, Accra, Ghana. R. Tabo is with theInternational Crops Research Institute for the Semi-Arid Tropics (ICRISAT) BP 12404, Niamey,Niger. K.D. Shepherd is with the World Agroforestry Centre (ICRAF), PO 30677, Nairobi, Kenya.E.M.A. Smaling is with the Faculty of Geo-Information Science and Earth Observation (ITC),University of Twente, PO Box 6, 7500 AA Enschede, The Netherlands. P.L. Woomer is withFORMAT, PO Box 79, The Village Market, Nairobi, Kenya.

18 Outlook on AGRICULTURE Vol 39, No 1

Integrated soil fertility management

Earlier versions of a green revolution in South Asia andLatin America boosted crop productivity through thedeployment of improved varieties, water and fertilizer.However, efforts to achieve similar results in Sub-SaharanAfrica (SSA) have largely failed (Okigbo, 1987). Most ofthe recent increases in crop production have resultedfrom expanding crop production areas, at the expenseof traditional fallows. This has caused substantialnutrient mining in areas under continuous cultivationand in conversion of marginal areas to agriculture,thus causing further land degradation (Smaling et al,1997).

The need for sustainable intensification of agriculturein SSA has gained support, in part because of the growingrecognition that farm productivity is a major entry pointto break the vicious circle underlying rural poverty.Recent landmark events include the African Heads ofState Fertilizer Summit held in Abuja, Nigeria (AfricaFertilizer Summit, 2006) and the launching of the Alliancefor a Green Revolution in Africa (AGRA). Kofi Annan, theChairman of the Board of AGRA, has repeatedly stressedthat the African green revolution should be madeuniquely African by recognizing the continent’s greatdiversity of landscapes, soils, climates, cultures andeconomic status, while also learning lessons from earliergreen revolutions in Latin America and Asia (Annan,2008). Besides the aforementioned uniquely Africanfeatures, recent global developments in increased fertilizerand commodity prices, the growing competition for landfor biofuel production, the continuing HIV/AIDSpandemic, climate change and the increasing scarcity ofwater must be incorporated in any strategy to achieve theAfrican green revolution.

Since fertilizer is an expensive commodity, AGRA hasadapted integrated soil fertility management (ISFM) as aframework for boosting crop productivity throughreliance upon soil fertility management technologies, withthe emphasis on increased availability and use of mineralfertilizer. Various definitions of ISFM have been proposed,including: ‘ISFM is a holistic approach to soil fertilityresearch that embraces the full range of driving factorsand consequences of soil degradation – biological,physical, chemical, social, cultural, economic andpolitical’ (TSBF-CIAT, 2005) and ‘The ISFM packageincludes the combined use of soil amendments,organic materials, and mineral fertilizers to replenishthe soil nutrients and improve the efficiency and cost-effectiveness of external inputs’ (IFDC, 2002).

Both these definitions and others are incomplete in thesense that they fall short of defining principles that areunique to ISFM.

The objectives of this paper are: (i) to develop a robustdefinition of ISFM that can be used as a practical meansfor objectively evaluating its implementation; (ii) to applythe definition to relevant technologies with great potentialfor dissemination to smallholder farmers; and (iii) tohighlight factors that will facilitate the adoption ofISFM practices. In the current paper, the term ‘fertilizer’is used for processed agro-minerals and manufacturedfertilizer, while organic inputs include crop residues,manures, composts and other locally available organicresources.

Before proposing a definition for ISFM, it is important

to sketch the context in which the smallholder farmer inSSA operates, thereby recognizing the wide diversity offarming systems and the environments in which theseoccur.

Attributes of smallholder farming systems inSSA

In the 1970s, Sanchez (1976) concluded that African soilswere as variable as (if not more so than) soils in otherregions. Such variability strongly impacts upon soilfertility and its management. At the regional scale, overallagroecological and soil conditions have led to diversepopulation and livestock densities across SSA and to awide range of farming systems (FAO and World Bank,2001). Each of these systems has different crops, croppingpatterns, soil management practices and access to inputsand commodity markets.

At the national level, smallholder agriculture isstrongly influenced by governance, policy, infrastructureand security levels. For instance, fertilizer is subsidized inRwanda, but taxed in the neighbouring DemocraticRepublic of Congo (DRC). Roads also play a major role infostering agricultural intensification through access tofarm inputs and commodity markets. Such issues relatedto infrastructure can drastically change upon crossing anational border, as illustrated above. Some countries seekto control farmer associations and produce markets, whileothers provide incentives for rural collective actions andfree markets. The filière coton in Burkina Faso provides apositive example of services to members throughimproved access to farm technologies and productmarketing (Kherallah et al, 2002).

Within farming communities, a wide diversity offarmer wealth classes, inequality and production activitiesmay be distinguished (Chianu et al, 2008). Traditionally,local indicators of wealth have been identified that canthen be used to classify farming households against a setof thresholds (Defoer, 2002). Tittonell et al (2005a)developed farmer typologies based upon productionobjectives of individual households, as related to theiraccess to production factors. The application of thisknowledge to the process of technology adoption hasbeen demonstrated by Shepherd and Soule (1998), whonoted that farmers with a larger quantity and widerdiversity of resources were able to assume greater risksand venture more readily into new technologies and farmenterprises.

At the individual farm level, it is important to considerthe variability between the soil fertility status ofindividual fields, which may be as large as differencesbetween different agroecological zones (Table 1). Thisvariability has obvious consequences for cropproductivity, resulting in yield ranges between 900 and2,400 kg maize grain ha–1 for different fields within thesame farm, as documented in western Kenya by Tittonellet al (2005b). These within-farm soil fertility gradients(SFGs) most often exist in areas with large populationdensities – resulting in intensive use of land – and whereamounts of farmyard manure are insufficient (Figure 1).SFGs are created by the position of specific fields within asoilscape (Deckers et al, 2002), by the selective allocation

19Outlook on AGRICULTURE Vol 39, No 1


Table 1. Soil fertility status for different agroecological zones (Windmeijer and Andriesse, 1993) and for various fields within a farm inBurkina Faso (Prudencio et al, 1993). Home gardens are near the homestead, bush fields furthest away from the homestead and villagefields are at intermediate distances.

Area Organic C Total N Available P Exchangeable K(g kg–1) (g kg–1) (mg kg–1) (mmol kg–1)

Agroecozones (0–20 cm):Equatorial forest 24.5 1.6 NA NAGuinea savanna 11.7 1.4 NA NASudan savanna 3.3 0.5 NA NA

Fields within a village:Home garden 11–22 0.9–1.8 20–220 4.0–24Village field 5–10 0.5–0.9 13–16 4.1–11Bush field 2–5 0.2–0.5 5–16 0.6–1

NA = not applicable.

Figure 1. A four-week-old maize crop in two different plots on the same farm (about 200 m apart) in westernKenya. Both crops were planted at the same time and managed in exactly the same way. On the left is a responsiveplot near the homestead, while on the right is a less responsive plot with high densities of couch grass (Elymusrepens [L.] Gould subsp. repens), a noxious weed (see insert).

of available nutrient inputs to specific crops and fields,and by improved management (for example, time ofplanting, weeding, etc) of plots with higher fertility(Tittonell et al, 2005b).

The above section sketches a summary of the farmingconditions in SSA and the variability that exists atdifferent scales. Any definition of ISFM must considerthese attributes.

Operational definition of ISFM

We define ISFM as a set of soil fertility management practicesthat necessarily include the use of fertilizer, organic inputs andimproved germplasm, combined with the knowledge of how toadapt these practices to local conditions, aimed at maximizingagronomic use efficiency of the applied nutrients andimproving crop productivity. All inputs need to be managedfollowing sound agronomic principles. The goal of ISFMis optimized crop productivity through maximizing

interactions that occur when fertilizers, organic inputsand improved germplasm, along with the requiredassociated knowledge, are integrated by farmers. Theproposed definition is unique in the sense that it allowsfor objective evaluation of the occurrence of ISFM in aspecific field.

A conceptual presentation of the definition is shown inFigure 2. The definition includes a number of conceptsthat are described below.

Focus on agronomic use efficiencyThe definition focuses on maximizing the use efficiency offertilizer and organic inputs, since these are both scarceresources in the areas where agricultural intensification isneeded. Agronomic efficiency (AE) is defined asincremental return to applied inputs, or:

AE (kg ha–1) = (YF – YC)/(Fappl) (1)



Figure 2. The conceptual relationship between the agronomicefficiency (AE) of fertilizers and organic resource and theimplementation of various components of ISFM, culminating incomplete ISFM towards the right side of the graph. Soils that areresponsive to NPK-based fertilizer and those that are poor andless responsive are distinguished. The ‘current practice’ stepassumes the use of the current average fertilizer application ratein SSA of 8 kg fertilizer nutrients ha–1. The meaning of the varioussteps is explained in detail in the text. At constant fertilizerapplication rates, yield is linearly related to AE.

where YF and YC refer to yields (kg ha–1) in the treatmentwhere nutrients have been applied and in the control plotrespectively, and Fappl is the amount of fertilizer and/ororganic nutrients applied (kg ha–1). Agronomic efficiencyis composed of ‘capture efficiency’ or the proportion ofnutrients taken up and ‘conversion efficiency’ or yieldproduced per amount of nutrients taken up. Captureefficiency is in part regulated by a sufficient supply ofotherwise non-limiting nutrients and plant requirementsfor soil moisture, aeration and physical support.Conversion efficiency is in part regulated by genotypicproperties, including those determining biomassaccumulation and harvest indices, etc. Together, theseresult in favourable AEs (Giller et al, 2006).

Returns are greatest during the first increments offertilizer response where the slope is steepest. As excessnutrients are applied, the response curve attenuates andAE decreases. With the current low overall fertilizerapplication rates in SSA (about 8 kg/ha) and even with the50 kg ha–1 target set by the Africa Fertilizer Summit (2006),aiming at maximizing AE remains a sensible goal. Inmany African farming systems, the production of cerealstraw for livestock feed, building materials, etc, is impor-tant, so that the economic yield is not just the cereal grain,but includes the straw. Note also that maximal AE leads tomaximal economic returns to inputs used, since bothindicators are linearly related to specific input and outputprices.

Fertilizer and improved germplasm

In terms of response to management, two general classesof soils are distinguished: (i) soils that show acceptableresponses to fertilizer (Path A, Figure 2) and (ii) soils thatshow minimal or no response to fertilizer due to otherconstraints besides the nutrients contained in the fertilizer(Path B, Figure 2). We have classified the above soils as‘responsive soils’ and ‘poor, less responsive soils’respectively. For instance, on sandy granitic soils, N useefficiency by maize varied from >50 kg grain kg–1 N on themore fertile fields close to homesteads, to less than 5 kggrain kg–1 N in degraded outfields (Zingore et al, 2007). Insome cases, where land is newly opened, or where fieldsare close to homesteads and receive large amounts oforganic inputs each year, a third class of soil exists wherecrops respond little to fertilizer as the soils are alreadyfertile. These soils need only maintenance fertilization andare termed ‘fertile, less responsive soils’. The ISFMdefinition proposes that application of fertilizer toimproved germplasm on responsive soils will boost cropyield and improve the AE relative to current farmerpractice, characterized by traditional varieties receivingtoo little and suboptimally managed nutrient inputs (PathA, Figure 2). Numerous studies have looked at theresponses of various crops to applied fertilizer in Africa.Results from the FAO Fertilizer Program, for instance,have shown an average increase in yield of 64% afterapplication of NPK as 45–15–10 across SSA (FAO, 1989).Recent experiences with the Millennium Villages projectshowed an average threefold increase in maize yield withfertilizer application (Sanchez et al, 2007).

Major requirements for achieving production gains on‘responsive fields’ within Path A (Figure 2) include (i) theuse of disease-resistant and improved germplasm, (ii) theuse of the correct fertilizer formulation and rates, and(iii) appropriate fertilizer, crop and water managementpractices. Use of improved germplasm in areas whereconstraints other than the nutrients applied through thefertilizer limit plant growth will certainly not boost the AEof fertilizer. Such constraints include pests, diseases ordrought. When one or more of these constraints isremoved, benefits from fertilizer use are obtained. Forinstance, in areas affected by Striga, a parasitic weedaffecting cereal growth throughout SSA, application offertilizer to herbicide-resistant maize will result in higheryields compared with local maize varieties (Kanampiu etal, 2003). In SSA, most soils are deficient in N and P, withsome areas low in K, S or micronutrients. The relativeimportance of these limitations varies with soil type andpast management. Moreover, different crops havedifferent requirements for specific nutrients. Soil nutrientstatus and crop requirements must be considered whenformulating fertilizers for maximal AE. Lastly, fertilizermanagement practices can substantially affect cropresponses. For instance, incorporating rather than broad-casting urea reduces volatilization losses, banding Pfertilizer on strongly P-adsorbing soils enhances itsuptake, and point-placement of fertilizer with cerealsresults in higher uptake (Aune and Bationo, 2008).Adjusting N applications to seasonal rainfall patterns isone means of reducing nutrient losses and improvingfertilizer use in semi-arid areas (Piha, 1993). Better

Ag

ron

om

ic e

ffic

ien

cy Increase in knowledge

A

B

C

Responsive soils

Poor, less responsive soils

Currentpractice

Germplasm& fertilizer


+ Organicresource mgt


+ Organicresource mgt

+ Localadaptation

‘Complete ISFM’Move towards ‘complete ISFM’



agronomic management may, however, conflict with otherdemands for labour at critical phases of the croppingseason, or may become confounded by lack of experiencein its site-specific application.

Combined application of organic and mineral inputsOrganic inputs contain nutrients that are released at a ratedetermined in part by their biochemical characteristics ororganic resource quality. However, organic inputs appliedat realistic levels seldom release sufficient nutrients foroptimum crop yield. Combining organic and mineralinputs has been advocated as a sound managementprinciple for smallholder farming in the tropics becauseneither of the two inputs is usually available in sufficientquantities, because positive interactions between bothinputs have often been observed (Vanlauwe et al, 2001)and because both inputs are needed in the long term tosustain soil fertility and crop production. Two other issuesarise within the context of ISFM: (i) does fertilizerapplication generate the required crop residues that areneeded to optimize the AE of fertilizer for a specificsituation, and (ii) can organic resources be used torehabilitate ‘less responsive soils’ and make theseresponsive to fertilizer (Path C in Figure 2)? The first issueis supported by data obtained in Niger by Bationo et al(1998). Where fertilizer was applied to millet, sufficientresidue was produced to meet both farm householddemands for feed and food and the management needs ofthe soil in terms of organic inputs and surface protectionof the soil from wind erosion. Evidence also supports thesecond rehabilitation issue. In Zimbabwe, applyingfarmyard manure for three years to sandy soils atrelatively high rates enabled a clear response to fertilizerwhere such a response was not visible beforerehabilitation (Zingore et al, 2007). In south-westernNigeria, integration of Senna siamea residues reducedtopsoil acidification resulting from repeated application ofurea fertilizer (Vanlauwe et al, 2005). Applying the aboveprinciples to maximize AE will require adaptation to theprevailing soil fertility status (SFGs) and other site-specific modifiers of crop growth.

Adaptation to local conditionsAs previously stated, farming systems are highly variableat different scales and one challenge for the African greenrevolution involves adjusting for site-specific soilconditions. First, soil fertility status can vary considerablywithin short distances, resulting in three general soilfertility classes, as explained above and demonstrated inFigure 3. A good proxy for soil fertility status is often thesoil organic matter (SOM) content, provided that thisparameter is not overextrapolated across dissimilar soils.Soil organic matter contributes positively to specific soilproperties or processes fostering crop growth, such ascation exchange capacity, soil moisture and aeration ornutrient stocks. On land where these constraints limitcrop growth, a higher SOM content may enhance thedemand by the crop for N and consequently increase thefertilizer N use efficiency. On the other hand, SOM alsoreleases available N that may be better synchronized withthe demand for N by the plant than fertilizer N.Consequently, a larger SOM pool may result in lower N

Figure 3. Variability in response to applied fertilizer (NPKS) inVihiga and Siaya districts of western Kenya during the long rainyseason of 2004. Three classes of soils can be identified: Class Ifields have relatively low yields without fertilizer and respondlittle to applied NPKS (referred to as ‘poor, less responsivefields’); Class II fields have relatively low yields without fertilizer,but respond well to fertilizer application (referred to as ‘respon-sive fields’); and Class III fields have relatively high yieldswithout fertilizer and respond little to fertilizer applicationbecause of this (referred to as ‘rich, less responsive fields’).

fertilizer AEs. Evidence from western Kenya shows thatfor fertile soils, AE for plant nutrients is less than thatfor less intensively managed outfields (Vanlauwe et al,2006).

Adaptation to local conditions also includesaccompanying measures that are needed to addressconstraints that are unlikely to be resolved by fertilizerand organic inputs. These adjustments include applicationof lime on acid soils, water harvesting techniques on soilssusceptible to crusting under semi-arid conditions, or soilerosion control on hillsides. For ISFM to remain apractical approach to better soil management for farmers,it must become readily integrated into field practices thatare dictated by these local conditions.

A move towards ‘complete ISFM’Several intermediary phases are identified that assist thepractitioner’s move towards complete ISFM from thecurrent 8 kg ha–1 fertilizer nutrient application with localvarieties. Each step is expected to provide themanagement skills that result in yield and improvementsin AE (Figure 2). Complete ISFM comprises the use ofimproved germplasm, fertilizer, appropriate organicresource management and local adaptation. Figure 2 is notnecessarily intended to prioritize interventions, but rathersuggests a need for sequencing towards complete ISFM. Itdoes, however, depict key components that lead to bettersoil fertility management. For instance, in areas wherefarmyard manure is targeted towards specific fieldswithin a farm, local adaptation is already taking place,even if no fertilizer is used, as is the case in much ofCentral Africa. For less responsive soils, investment in soilfertility rehabilitation will be required before fertilizer AEwill be enhanced. It is important to note that the differentsteps are part of ISFM, but only when all steps are takencan one expect maximal AE or ‘complete’ ISFM. For

8,000

7,000

6,000

5,000

4,000

3,000

2,000

1,000

0

Maiz

e g

rain

yie

ld(w

ith N

PK

S a

dded)

(kg h

a–

1)

Maize grain yield (control) (kg ha–1)

0 1,000 2,000 3,000 4,000 5,000 6,000 7,000 8,000

Vihiga district

Siaya district

1:1 line

III

II

I



instance, a farmer adopting good agronomic practices forapplied fertilizer is going to improve the AE of thoseinputs and is thus implementing one component of ISFM.However, land managers can only be considered completeISFM practitioners when they also recycle organic inputs,plant improved germplasm and use the requiredaccompanying measures. Lastly, the above evidence forthe different ‘steps’ is fragmented and derived fromdifferent cropping systems and agroecozones, and aconcerted effort with a standardized multilocationaldesign, encompassing the existing variability in soilfertility, is needed to determine which circumstances aremost crucial in the stepwise improvement of fertilizer andorganic input AE.

ISFM and environmental issuesISFM interacts with environmental quality by minimizingnutrient losses to the environment and maximizing cropproductivity per unit of nutrient applied. In this way,agricultural production is intensified and pressure toconvert additional lands to agriculture is reduced. Whilesoil fertility is most often defined as the capacity of a soilto grow crops, soil health is usually defined in broaderterms, and encompasses the potential of soils to provide afull range of ecosystems services. The MillenniumEcosystem Assessment (2005) defined healthy soils asthose soils that were capable of delivering essentialprovisioning, regulating and supporting ecosystemservices on a sustained basis. Examples of the latter arenutrient cycling, water flow regulation or maintenance ofsoil biological diversity. Many of these services aredirectly or indirectly related to the soil organic matterpool, although knowledge of how much SOM and ofwhich quality is required to retain specific servicefunctions is currently limited. ISFM is a viable entry pointfor improved soil health, and more so when ISFMpractices increase the soil organic matter pool over time.Recycling of organic inputs is likely to result in increasedSOM and restoration of degraded soils. Whether the ‘seedand fertilizer’ strategy is able to produce sufficient cropresidues to impact positively upon the SOM pool isuncertain and ultimately regulated by soil microclimate,texture and structure, organic input strategies and soiltillage regime.

Principles embedded within the definition of ISFMneed to be applied within existing farming systems(Figure 4). Two examples clearly illustrated theintegration of ISFM principles in existing croppingsystems: (i) dual-purpose grain legume–maize rotationswith P fertilizer targeted at the legume phase and Nfertilizer at rates below the recommended rates targeted atthe cereal phase in the moist savanna agroecozone(Sanginga et al, 2003) and (ii) micro-dose fertilizerapplications in legume–sorghum or legume–milletrotations with retention of crop residues and combinedwith water harvesting techniques in the semi-aridagroecozone (Bationo et al, 1998; Tabo et al, 2007).

As for the grain legume–maize rotations, application ofappropriate amounts of mainly P to the legume phaseensures good grain and biomass production, the latter inturn benefiting a subsequent maize crop and thusreducing the need for external N fertilizer (Sanginga et al,2003). Choosing an appropriate legume germplasm with a

Figure 4. Components of a typical farming system, indicatingflows of nutrients and cash (adapted from Tittonell et al,2005a). The boxes with grey tints give examples of specific ISFMinterventions.

low harvest index will favour accumulation of organicmatter and N in the non-harvested plant parts, andchoosing adapted maize germplasm will favour amatching demand for nutrients by the maize. Applicationof a sufficient amount of legume crop residues can alsoimprove other soil conditions, thus leading to improveduse efficiency of the applied N fertilizer (Sanginga et al,2003). Selection of fertilizer application rates based onlocal knowledge of the initial soil fertility status withinthese systems would qualify the soil managementpractices as complete ISFM.

As for the micro-dose technology, spot application ofappropriate amounts of fertilizer to widely spaced cropssuch as sorghum or millet substantially enhances its useefficiency, with further enhancements obtained whencombined with physical soil management practices aimedat water harvesting. Recycling crop residues can reducewind and water erosion (Bationo et al, 1998) and thusfurther benefit growth and nutrient demand of afollowing cereal. Rotating a legume, such as cowpea, withsorghum or millet has been shown to increase cerealyields further (Aune and Bationo, 2008). Although bothsystems are good examples of complete ISFM, theselection of management priorities is in line with theoverall economic conditions (for example, the presence ofa market for the grain legumes produced) and theresource endowment of the farm family (for example,sufficient financial resources to purchase the requiredinputs). Certain conditions enable the dissemination andretention of ISFM practices.

Dissemination of ISFMThe gradual increase in complexity of knowledge as onemoves towards complete ISFM (Figure 2) has implicationsfor the strategies in adapting for widespread

HOME

LIVESTOCK

COMPOST

OTHER

ENTERPRISES

FARM

MARKET

INPUT OUTPUT LABOUR

CashNutrients

REMITTANCES

Recycling of

crop residues

Soil

Soil

Soil

Crop

Crop

Crop

FOOD CROPCONSUMPTION

FOOD CROPMARKET

CASH CROPMARKET

Use of manure

or compost

and fertilizer

Proper manure

and compost

management

Choice of

tolerant variety

Use of appropriate

fertilizer types

Varying fertilizer

rates for

different fields

Appropriate timing

and placement

of fertilizer

Appropriate crop

rotation with targeted fertilizer

Targeted use of

manure/compost for

less-responsive fields



dissemination of ISFM. Furthermore, a set of enablingconditions can favour the uptake of ISFM. The operationsof every farm are strongly influenced by the larger ruralcommunity, policies, plus supporting institutions andmarkets. Not only are farms closely linked to the off-farmeconomy through commodity and labour markets, but therural and urban economies are also stronglyinterdependent. For example, as noted above, it is quitecommon for small farm households to derive a significantpart of their income – often 40% or more – from off-farmactivities. Farming households are also linked to ruralcommunities and social and information networks, andthese factors provide feedback that influences farmerdecision making; for example, the negative impact of thestructural adjustment programmes of the 1980s and early1990s on the formal agricultural extensioninstitutions. Because ISFM is a set of principles andpractices to intensify land use in a sustainable way, uptakeof ISFM is facilitated in areas with greater pressure onland resources. In areas with substantial fallow land, suchfallow periods can continue to regenerate the soil fertilitystatus, as was the case in most of SSA until recently, andstill is the case in many humid forest environments, withincentives for adoption of ISFM being reduced.

The first step towards ISFM acknowledges the need forfertilizer and improved varieties. An essential conditionfor its early adoption is access to farm inputs, producemarkets and financial resources. To a large extent,adoption is market-driven as commodity sales provideincentives and cash to invest in soil fertility managementtechnologies, providing opportunities for community-based savings and credit schemes. Policies towardssustainable land use intensification and the necessaryinstitutions and mechanisms to implement and evaluatethese are also what facilitates the uptake of ISFM. Policiesfavouring the importation of fertilizer, its blending andpackaging, or smart subsidies, are needed to stimulate thesupply of fertilizer as well. Specific policies addressingthe rehabilitation of degraded, non-responsive soils mayalso be required since investments to achieve this may betoo large to be supported by farm families alone. Anotherfactor that may facilitate the dissemination of ISFMinvolves the promotion of improved nutrition, forexample, through incorporation of legume-based productsin local diets.

While dissemination and adoption of complete ISFM isthe ultimate goal, substantial improvements in productioncan be made by promoting greater use of farm inputs andgermplasm within market-oriented farm enterprises. Suchdissemination strategies should include ways to facilitateaccess to the required inputs, simple information fliersspread through extension networks, and knowledge onhow to avoid less responsive soils. The latter can be basedon earlier experiences of the farmer, since most localindicators of soil fertility status are linked to theproduction history of specific fields (Mairura et al, 2008).A good example of where the ‘seeds and fertilizer’strategy has made substantial impact is the Malawifertilizer subsidy programme. Malawi became a net foodexporter through the widespread deployment of seedsand fertilizer, although the aggregated AE was only 14 kgof grain per kg of nutrient applied (Chinsinga, 2008).Such AE is low and ISFM could increase this to at least

double its value, with all the consequent economicbenefits to farmers.

As efforts to promote the ‘seed and fertilizer’ strategyare under way, activities such as farmer field schools ordevelopment of site-specific decision guides that enabletackling more complex issues can be initiated to guidefarming communities towards complete ISFM, includingaspects of appropriate organic matter management orlocal adaptation of technologies. The latter will obviouslyrequire more intense interactions between farmers andextension services and will take longer to achieve itsgoals. Farmer adoption of ISFM may further beaccelerated through the implementation of balanced ruralcampaigns that combine all of these considerations byoffering farmers information, technology demonstrations,product samples, financial incentives and opportunities todevelop their skills within their own farms.

Conclusions

The African green revolution aims at intensifyingagriculture through large-scale dissemination of ISFMpractices. An operational definition of ISFM is proposedas a set of soil fertility management practices that necessarilyinclude the use of fertilizer, organic inputs and improvedgermplasm, combined with the knowledge on how to adapt thesepractices to local conditions, aimed at maximizing agronomicuse efficiency of the applied nutrients and improving cropproductivity. All inputs need to be managed following soundagronomic principles. This definition acknowledges theneed for integration of improved germplasm, organicinputs, plus variability in production factors formaximum fertilizer AE. ISFM practices are also bestintegrated in farming systems shown to have greatpotential for dissemination. Although complete ISFM isknowledge-intensive, much can be achieved through thepromotion of the ‘seed and fertilizer’ strategy whilesimultaneously setting up structures and institutions toaddress the complexity of complete ISFM. Variousaccompanying measures can facilitate the adoption ofISFM practices in priority farming systems. Widespreadadoption of ISFM has the potential not only to improvefarm productivity and farmers’ welfare but also to bringabout environmental benefits.

Acknowledgments

The authors gratefully acknowledge the Bill & MelindaGates Foundation for supporting various workshops anddiscussion sessions during which the main ideaspresented in this paper were developed.

ReferencesAfrica Fertilizer Summit (2006), Abuja Declaration on Fertilizer for

African Green Revolution, African Union Special Summit of theHeads of State and Government, Abuja, Nigeria.

Annan, K. A. (2008), ‘Forging a uniquely African greenrevolution’, address by Mr Kofi A. Annan, Chairman of AGRA,30 April, Salzburg Global Seminars, Austria.

Aune, J. B., and Bationo, A. (2008), ‘Agricultural intensification inthe Sahel – the ladder approach’, Agricultural Systems, Vol 98,pp 119–125.



Bationo, A., Lompo, F., and Koala, S. (1998), ‘Research on nutrientflows and balances in west Africa: state-of-the-art’, Agriculture,Ecosystems and Environment, Vol 71, pp 19–35.

Breman, H., Fofana, B., and Mando, A. (2005), ‘The lesson ofDrente’s “Essen”: soil nutrient depletion in sub-SaharanAfrica and management strategies for soil replenishment’,paper presented at the 2005 IHDP Open Meeting, Sessionon ‘Impact of Land Use Change on Soil Resources’, 11 October,Bonn.

Chianu, J. N., Ajani, O. I. Y., and Chianu, J. N. (2008), ‘Livelihoodsand rural wealth distribution among farm households inwestern Kenya: implications for rural development, povertyalleviation interventions and peace’, African Journal ofAgricultural Research, Vol 7, pp 455–464.

Chinsinga, B. (2008), Reclaiming Policy Space: Lessons from Malawi’s2005/2006 Fertilizer Subsidy Programme Future Agricultures,Institute of Development Studies, Brighton.

Deckers, J. (2002), ‘A systems approach to target balancednutrient management in soilscapes of sub-Saharan Africa’, inVanlauwe, B., Diels, J., Sanginga, N., and Merckx, R., eds,Integrated Plant Nutrient Management in sub-Saharan Africa: fromConcept to Practice, CABI, Wallingford, pp 47–62.

Defoer, T. (2002), ‘Learning about methodology development forintegrated soil fertility management’, Agricultural Systems, Vol73, pp 57–81.

FAO – Food and Agriculture Organization (1989), Fertilizers andFood Production. The FAO Fertilizer Programme 1961–1986, FAO,Rome.

FAO and World Bank (2001), Farming Systems and Poverty.Improving Farmers’ Livelihoods in a Changing World, FAO, Rome,and World Bank, Washington, DC.

Giller, K. E., et al (2006), ‘Resource use dynamics and interactionsin the tropics: scaling up in space and time’, AgriculturalSystems, Vol 88, pp 8–27.

IFDC (2002), Collaborative Research Programme for Soil FertilityRestoration and Management in Resource-Poor Areas of Sub-SaharanAfrica, Technical Bulletin IFDC – T-67, IFDC, Muscle Shoals, AL.

Kanampiu, F. K., et al (2003), ‘Multi-site, multi-season field testsdemonstrate that herbicide seed-coating herbicide-resistancemaize controls Striga spp. and increases yields in severalAfrican countries’, Crop Protection, Vol 22, pp 697–706.

Kherallah, M., et al (2002), Reforming Agricultural Markets in Africa,International Food Policy Research Institute, Washington, DC.

Mairura, F. S., et al (2008), ‘Scientific evaluation of smallholderland use knowledge in central Kenya’, Land Degradation &Development, Vol 19, pp 77–90.

Millennium Ecosystem Assessment (2005), Ecosystems and HumanWell-Being: Synthesis, Island Press, Washington, DC.

Okigbo, B. N. (1987), Towards a New Green Revolution: FromChemicals to New Biological Techniques in the Improvement ofTropical African Agriculture, Elsevier, Amsterdam.

Piha, M. I. (1993), ‘Optimizing fertilizer use and practical rainfallcapture in a semi-arid environment with variable rainfall’,Experimental Agriculture, Vol 29, pp 405–415.

Prudencio, C. Y. (1993), ‘Ring management of soils and crops inthe West African semiarid tropics: the case of the mossi farmingsystem in Burkina Faso’, Agriculture, Ecosystems and Environ-ment, Vol 47, pp 237–264.

Sanchez, P. A. (1976), Properties and Management of Soils in theTropics, Wiley, New York.

Sanchez, P., et al (2007), ‘The African millennium villages’,Proceedings of the National Academy of Sciences, Vol 104, pp16775–16780.

Sanginga, N., et al (2003), ‘Sustainable resource managementcoupled to resilient germplasm to provide new intensivecereal–grain legume–livestock systems in the dry savanna’,Agriculture, Ecosystems and Environment, Vol 100, pp 305–314.

Shepherd, K., and Soule, M. J. (1998), ‘Soil fertility managementin west Kenya – dynamic simulation of productivity, profitabil-ity and sustainability at different resource endowment levels’,Agriculture, Ecosystems and Environment, Vol 71, pp 133–147.

Smaling, E. M. A., Nandwa, S. M., and Janssen, B. H. (1997), ‘Soilfertility is at stake!’ in Buresh, R. J., Sanchez, P. A., andCalhoun, F., eds, Replenishing Soil Fertility in Africa, SSSASpecial Publication Number 51, Soil Science Society of America,American Society of Agronomy, Madison, WI, pp 47–61.

Tabo, R., et al (2007), ‘Improving cereal productivity and farmers’income using a strategic application of fertilizers in WestAfrica’, in Bationo, A., et al, eds, Advances in Integrated SoilFertility Management in Sub-Saharan Africa: Challenges andOpportunities, Kluwer Publishers, Dordrecht, pp 201–208.

Tittonell, P., et al (2005a), ‘Exploring diversity in soil fertilitymanagement of smallholder farms in western Kenya. I. Hetero-geneity at region and farm scale’, Agriculture, Ecosystems andEnvironment, Vol 110, pp 149–165.

Tittonell, P., et al (2005b), ‘Exploring diversity in soil fertilitymanagement of smallholder farms in western Kenya. II. Withinfarm variability in resource allocation, nutrient flows and soilfertility status’, Agriculture, Ecosystems and Environment, Vol 110,pp 166–184.

Tropical Soil Biology and Fertility Institute of the InternationalCentre for Tropical Agriculture – TSBF-CIAT (2005), IntegratedSoil Fertility Management in the Tropics: From Knowledge toImplementation: TSBF-CIAT’s Strategy and Work Plan, 2005–2010,Centro Internacional de Agricultura Tropical (CIAT), Cali,Columbia.

Vanlauwe, B., et al (2005), ‘Long-term integrated soil fertilitymanagement in South-western Nigeria: crop performance andimpact on the soil fertility status’, Plant and Soil, Vol 273, pp337–354.

Vanlauwe, B., Tittonell, P., and Mukalama, J. (2006), ‘Within-farmsoil fertility gradients affect response of maize to fertilizerapplication in western Kenya’, Nutrient Cycling inAgroecosystems, Vol 76, pp 171–182.

Vanlauwe, B., Wendt, J., and Diels, J. (2001), ‘Combined applica-tion of organic matter and fertilizer’, in Tian, G., Ishida, F., andKeatinge, J. D. H., eds, Sustaining Soil Fertility in West-Africa,SSSA Special Publication Number 58, Madison, WI, pp 247–280.

Windmeijer, P. N., and Andriesse, W. (1993), Inland Valleys in WestAfrica: An AgroEcological Characterization of Rice-GrowingEnvironments, International Institute for Land Reclamation andImprovement, Wageningen.

Zingore, S., et al (2007), ‘Soil type, management history andcurrent resource allocation: three dimensions regulatingvariability in crop productivity on African smallholder farms’,Field Crops Research, Vol 101, pp 296–305.

integrated soil fertility management - home | food and...

Documents