This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and education use, including for instruction at the authors institution and sharing with colleagues. Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party websites are prohibited. In most cases authors are permitted to post their version of the article (e.g. in Word or Tex form) to their personal website or institutional repository. Authors requiring further information regarding Elsevier’s archiving and manuscript policies are encouraged to visit: http://www.elsevier.com/authorsrights
Transcript
This article appeared in a journal published by Elsevier. The attachedcopy is furnished to the author for internal non-commercial researchand education use, including for instruction at the authors institution
and sharing with colleagues.
Other uses, including reproduction and distribution, or selling orlicensing copies, or posting to personal, institutional or third party
websites are prohibited.
In most cases authors are permitted to post their version of thearticle (e.g. in Word or Tex form) to their personal website orinstitutional repository. Authors requiring further information
regarding Elsevier’s archiving and manuscript policies areencouraged to visit:
Agriculture, Ecosystems and Environment 187 (2014) 1–10
Contents lists available at ScienceDirect
Agriculture, Ecosystems and Environment
j ourna l h om epage: www.elsev ier .com/ locate /agee
Editorial
Evaluating conservation agriculture for small-scale farmers inSub-Saharan Africa and South Asia�
a b s t r a c t
This introductory paper provides an overview of the contributions to this special issue, addressing four key questions related to conservation agriculture(CA) in Sub-Saharan Africa and South Asia: What is the impact of CA on yields? What is the impact of adopting CA on farmers’ profits? What are theenvironmental impacts of adopting CA? How well does CA fit with wider agricultural, social, economic and political contexts for small-scale farmers?Papers in this special issue find that yield increases under CA are possible but uncertain given the low average yields that pertain in these regions, andyield gains are more likely to be observed after several years. CA is not widely adopted in Sub-Saharan Africa and South Asia owing to a lack of economicincentive for smallholder farmers—that the process of conversion to CA is not profitable over planning horizons of most farmers. There is no clear trend forgreater carbon sequestration under CA, so the potential for subsidizing farmers to adopt CA using payments for ecosystem services/carbon credit schemesseems limited in scope. There is early evidence that farmers perceive a benefit from CA adoption in regions that are prone to erratic rainfall, suggestinga potential risk mitigation role. In addition, throughout this overview paper we offer a commentary on some of the scientific issues that constrain ourability to understand the performance of CA in these systems more comprehensively.
Agriculture in Sub-Saharan Africa (SSA) and South Asia (SA)faces significant challenges in coming decades to reconcile the needfor increased food production to feed a growing population with-out significantly increasing the area under agricultural production(Cassman, 1999; The Montpellier Panel, 2013; Stevenson et al.,2013) and while protecting natural resources for future genera-tions (Cassman and Wood, 2005; Hobbs et al., 2008). ConservationAgriculture (hereafter, CA) has been promoted by a number ofscientists and institutions with the expectation that it can helpreconcile these competing objectives and contribute to sustainableintensification (Hobbs et al., 2008; FAO, 2011).
According to the FAO definition, CA is characterized by threelinked principles expected to be adopted together, namely:
• Continuous minimum mechanical soil disturbance (reduced,minimum or zero tillage).
• Permanent soil cover with crop residues or other types of organicmaterials.
• Diversification of crop species grown in sequences (crop rotation)and/or associations.
This definition of CA as a “package” of three components withsynergistic effects among them has been a source of significantdebate and controversy. Each component requires interpretation.What are allowable levels of tillage for “reduced” or “mini-mum” soil disturbance? How much soil cover is required? How
� Guest editorial for special issue of Agriculture, Ecosystems and Environment.
many crop species are required in a rotation to provide sufficientdiversification?1
Controversy surrounds the purported CA benefits (e.g. Sumberget al., 2012) and the underpinning science that supports claimsabout its suitability and effectiveness, especially in smallholderagriculture in Sub-Saharan Africa and South Asia. For example,Giller et al. (2009) argue there are grounds for questioning thepotential for CA – as defined above – to contribute to agriculturaldevelopment in SSA, highlighting a possible mismatch between theconditions required for all CA principles to be adopted by farmersand the circumstances that characterize and constrain smallholderAfrican farming systems.
The mechanisms through which CA has been promoted inSSA and SA have been extensively reviewed elsewhere (Kassamet al., 2009; Andersson and Giller, 2012; Serraj and Siddique,2012). In this Special Issue we focus on critical reviews ofrecent scientific literature on CA for smallholders in develop-ing countries, evaluating the science underpinning controversialclaims and interpretations, as well as offering some new empiricalcontributions to the literature. The papers span a range of disci-plines and expertise—agronomy (Brouder and Gomez-McPherson;Gathala et al.; Kirkegaard et al.; Nyamangara et al.—all this issue),
1 FAO’s AQUASTAT database defines conservation agriculture as follows: ‘Conser-vation Agriculture (CA) is an agricultural practice, whereby the disturbed area is beless than 15 cm wide or 25% of the cropped area (whichever is lower). AQUASTATdistinguishes between 30%-60%, 61-90% and 91% ground cover. Ground cover mustbe measured after planting time. Ground cover less than 30% is not considered CA.Rotation must involve at least 3 different crops. Rotation is not a requirement forCA at this time, but AQUASTAT reports whether rotation is being carried out or not.’AQUASTAT, accessed 22 July 2013.
2 Editorial / Agriculture, Ecosystems and Environment 187 (2014) 1–10
economics (Pannell et al.; Arslan et al.; Dalton et al.—all this issue),soil and environmental sciences (Palm et al., this issue), hydrology(Ahmad et al., this issue), anthropology (Andersson and D’Souza,this issue), and inter-disciplinary perspectives supported by model-ing exercises (Rosenstock et al.; Corbeels et al.; Baudron et al.—allthis issue). Most of these authors, and many other researchersfrom a range of institutions, were participants in a workshopheld in October 2012 at the University of Nebraska, Lincoln, andorganized by the CGIAR Independent Science and PartnershipCouncil (www.sciencecouncil.cgiar.org). The aim of the workshopwas to debate the role CA could play in meeting the strategichigh-level objectives of the CGIAR—a global agricultural researchpartnership that aims to reduce rural poverty, increase food secu-rity, improve human health and nutrition, and manage naturalresources in a sustainable manner in low-income developingcountries (http://www.cgiar.org/our-research/).
Here we provide an overview of the 13 studies included in thespecial issue, pulling out common threads and connections acrossthem (Section 2). A consensus statement on current understandingof potential benefits of CA for smallholder agriculture in SSA andSA and priorities for future research – the “Nebraska Declaration”on conservation agriculture – forms Section 3 of this paper, signedby a group of 44 scientists who participated in the October 2012workshop. Conclusions are summarized in Section 4.
2. Overview of papers in the special issue
The 13 papers in this issue cluster into four groups according tothe central focus of the research questions addressed, with someoverlap because some papers address more than one issue. Table 1provides a summary of the papers, highlighting the topic, produc-tion system, country, methodology and main results reported.
2.1. What is the impact of CA adoption on crop yields?
Average yields of major food crops in SSA and SA are sig-nificantly lower than their water-constrained yield potential(e.g. van Ittersum et al., 2013). Raising agricultural productivitywithout negative environmental impacts (and potentially gen-erating socially valued environmental benefits) is an attractiveproposition, and consequently the concept of sustainable inten-sification has gained attention in agricultural policy circles (TheMontpellier Panel, 2013; Garnett et al., 2013). CA is seen by someas being central to the sustainable intensification concept (FAO,2011).
Brouder and Gomez-Macpherson (this issue) provide an anal-ysis of evidence on the impact of CA on crop yields, based on reviewof recent literature on CA implementation in major cropping sys-tems of SSA and SA. The authors found it difficult to extract thenecessary data for a systematic review of CA effects because moststudies lacked the data and statistical analyses required to meetcriteria for a credible meta-analysis. Likewise many of these stud-ies lacked descriptions of site characteristics including soil type,weather data, and general agronomic management practices otherthan CA, including specifics about how crop residues were man-aged. Nonetheless, their comprehensive assessment suggests thata key CA component, i.e. zero tillage, may not be initially benefi-cial for crop yields, perhaps resulting from direct short-term effectssuch as increased weed pressure that become less severe as the CAsystem matures. Hence, the short-term negative impact on yieldof maize decreases over time and may eventually lead to a yieldadvantage, particularly in those systems linked with mulching.The paper highlights the need for more accurate and systematicreviews addressing agronomic impacts of CA interventions withappropriate handling of within and between study variance as well
as sensitivity analyses and quantitative assessments of publicationbias. The paper concludes with important recommendations forfuture field studies, including requirements for a minimum dataset,adequate description of management practices, appropriate statis-tical analysis, and systematic approaches in meta-analyses of CAstudies.
Gathala et al. (this issue) tested the hypothesis that adoptingthe three principles of CA together with best crop managementpractices would improve cropping system productivity while low-ering irrigation water use and crop production costs, resulting inhigher profitability. They compare the performance of four cerealcropping system scenarios during 2009–2011 in a field experimentlocated in the north-western Indo-Gangetic Plains, an area withrapid depletion of the groundwater due to excessive irrigation. Thebest performing crop and resource management practices with andwithout CA were compared with current farmers’ practices. Theauthors report that some of the CA management packages resultedin significant benefits in terms of yield, water saving, and profitalthough the magnitude of benefits depended on the cropping sys-tem and component crops. Avoiding the puddling operation onrice and using zero-tillage provided benefits in terms of reducedirrigation requirements, higher yields (via reduced terminal heatstress) or profit for wheat and maize while having no negativeeffects on rice.
While these positive effects of CA management practices onyield and system efficiencies can occur soon after conversion in this3-year study, there is need for measuring the system performanceover a longer time frame to obtain a better estimate of benefits andtheir variability. Likewise, because this study was conducted at aresearch station, the results remain to be validated under actualon-farm conditions. As noted in other studies (e.g. Kirkegaard et al.,this issue), on-farm constraints other than CA technologies per semay limit adoption by farmers.
Rosenstock et al. (this issue) demonstrate the potential for ex-ante analysis of farmer decision-making in the adoption processeven where there may be scant data to help parameterize a model ofhow CA adoption affects yields. Using Monte Carlo simulations – theuse of repeated random sampling from a series of linked probabilitydistributions – Rosenstock et al. forecast the likely yield outcomesfor farmers in two sites in East Africa. These analyses took placein the context of an agricultural development project, prior to theintroduction of activities. Their main conclusion is that predictedyield impacts of adopting CA at these two sites were almost alwaysnegative over the first three years, which helped the project infocusing their efforts. The main issue with this approach is whetherthe model is a reasonable simplification of the real-world situation.
While the underlying linear model that is simulated is too sim-ple to capture interactions among the management practices, thejustification for this approach, and the reason we have includedit in the special issue, is the following. There is a strong consen-sus (reflected in a number of points in the Nebraska Declarationgiven in Section 3) that the outcomes from adopting CA practicesare extremely context-dependent. Results from field studies con-ducted at experiment-stations have very high internal validity—i.e.they can do a good job of isolating the causal effect of a treatment(e.g. switching from conventional production to zero-tillage withfull residue retention) on the outcome of interest (e.g. change inyield) for a specific set of conditions. However, as Giller et al. (2011)discuss, on-station experimental results often have limited rele-vance (i.e. external validity) for forecasting ex-ante the outcomesfrom adopting the same conservation agriculture practices in a par-ticular agro-ecological and socio-economic niche. By formalizingexpert opinion in a Bayesian framework, Rosenstock et al.’s paperis a healthy provocation to agronomists and modelers working onthese systems. Moreover, if the Rosenstock approach is too simple,then the challenge is to develop a modeling approach that can be
Editorial / Agriculture, Ecosystems and Environment 187 (2014) 1–10 3
Table 1Summary of all papers in the special issue.
Authors Topic Production system Methodology Main results
Brouder andGomez-MacPherson
Impact ofzero-tillage onyields
Maize; publishedpapers in SSA andSA
Review • Most papers reporting results from field experiments lack criticaldata or meta-data (e.g. critical supporting or explanatory data onsoil type, prevailing weather, and management practices,including handling of crop residues)• In the short-term, ZT generally results in lower yields• Negative impacts on yield ameliorated with time in some casesaccompanied by greater soil water infiltration and organic matter,particularly when mulch was added
Gathala et al. Experiments ondifferentcomponents of CApackage inzero-tillage
• ZT direct-seeded rice + residue retention give yields similar to, orhigher than, farmer-managed puddled and transplanted rice, withless irrigation water, lower production costs, and higher or similarnet income• ZT maize is as productive and almost as profitable as rice duringthe rainy season while using 90% less irrigation water• ZT wheat combined with improved crop management, enhancedwheat productivity and profitability compared with the farmers’practice but depends on how the previous season’s crop ismanaged, including the residue
Rosenstock et al. Yield effects of CAin two projects
Rainfed maize;Kenya, Tanzania
Decisionmodelling; MonteCarlo
• Predicted yield impacts of adopting CA were negative in nearlyall cases• Decision-modelling can be used by development projects toidentify regions and cropping systems most likely to benefit fromadoption of CA practices
Pannell et al. Economic impactsof adopting CA
Rainfed maize;Model applied toZimbabwe dataset
Ex-antemicro-economicmodel
• CA is more likely to be economically more attractive for:larger, better • resourced farmsfarmers with longer planning horizons and lower discount rates• farmers with less uncertainty on the costs and benefits of CAadoption
Dalton et al. Profits andknowledge asconstraints toadoption
Rainfed maize;Ghana
Farm survey data;farm partialbudgets
• Farmer field schools on CA component technologies increasedparticipant knowledge from a low baseline and increasedknowledge of non-participants through spillovers• Partial budgets of 40 farmers show that adoption is not profitablein short-run• Unfavourable economic outcomes is greater constraint toadoption than lack of knowledge
Arslan et al. Adoption anddis-adoption overtime
Rainfed maize;Zambia
Econometricmodel, nationalpanel data
• Significant dis-adoption of minimum soil disturbance occurredbetween 2004-2008• District-level variation in historical rainfall during growingseason predicts continued adoption of minimum soil disturbance,suggesting farmers view CA as a strategy to reduce risk of rainfalluncertainty.
Palm et al. Impacts onecosystem services
Global scope withemphasis onrelevance to SSAand SA
Review • No clear trend for higher soil carbon sequestration under CA.• Process for using equivalent soil mass and measurement to atleast 50 cm depth recommended to avoid errors in past studies• CA provides other benefits, e.g. residue coverthat reduces runoffand surface crusting, increased aggregate stability and waterinfiltration, greater total water supply and water use efficiency
Ahmad et al. Impacts on wateruse of resource-conservingtechnologies
Rice–wheat;Pakistan
Water balancemodel and farmsurvey
• Adoption of resource-conserving technologies was closelyassociated with increased profits despite many farmers (almostone third of zero tillage adopters) reporting yield loss• Reductions in irrigation amount at the field level do not alwaystranslate into “water savings” at larger spatial scales• Medium and large farmers may use field-scale water savingsfrom adoption of CA to increase their irrigated area• Without strong governance at watershed level, widespreadadoption of these technologies could have perverse outcome ofincreasing overall water use for a specific watershed or aquifer
Andersson andD’Souza
Critique of CApromotion inSouthern Africa
General, mostlymaize; Malawi,Zimbabwe, Zambia
Review • CA was initially promoted to prevent soil erosion• In late 1990s, justification shifted and the technology was‘re-framed’ as a productivity increasing and foodsecurity-enhancing technology for resource-poor smallholderfarmers• Potential sources of upward bias in adoption studies includenon-representative sampling and/or financial incentives to use CAtechnologies
Kirkegaard et al. Flexible adoptionof specific practices
Rainfed wheat inmixed systemswith livestock;Australia withinferences for SSAand SA
Review • Australian farmers have commonly followed a pragmatic andstep-wise approach for adapting CA principles tailored to the localbiophysical and socio-economic conditions of their farmingsystems• Further innovations in mixed farming systems are still requiredto minimise inherent biomass trade-offs• The high degree of flexibility in CA principles as practiced insouthern Australian mixed farming systems could provide asuitable lesson for future promotion of CA practices in smallholdersystems of SSA and SA
Author's personal copy
4 Editorial / Agriculture, Ecosystems and Environment 187 (2014) 1–10
Table 1 (Continued)
Authors Topic Production system Methodology Main results
Nyamangara et al. Weed managementand labour demand
Farm survey • Adopters of planting basins sow their plots 12 to 23 days earlierthan conventional practice• Farmers weeded their planting basin plots significantly morefrequently than the conventional plots, yet planting basin plotsstill had greater weed pressure• Labor demand was more than double under planting basins thanin the conventional treatment although returns to investmentwere slightly higher under reduced tillage• The practice of minimum soil disturbance should be targeted tohouseholds with access to draft power for tillage and herbicidesfor weed control to reduce labour demand
Multi-levelmodeling based on10 case-studiesfrom 7 countries
• Important interactions across spatial scales govern systemperformance when CA is studied at field, whole farm, orlandscape/regional levels• CA has potential to increase yields over the long-term, andparticularly at sites with erratic rainfall• CA adoption is constrained by lack of short-term increase inprofits• Trade-offs regarding the use of crop residues for fodder is animportant disincentive to CA adoption
Baudron et al. Trade-offs inresidue use
Mixedcrop–livestockfarming; Kenya,Ethiopia
Yield gap analysis;Trade-off analysis
• Substitution in dairy herd diets, from cereal residue tonutrient-rich feeds, and replacing animal draft power with tractorscan both help to reduce demand for cereal residues (facilitating CAadoption), and help to close large current yield gaps• Benefits from mulching are site-specific; the priority focusshould be on water-use efficiency (water being the yield-limitingfactor on many farms) using technologies most acceptable tofarmers rather than on adoption of a specific package ofconservation technologies
easily parameterized for different contexts but that capture moreof the complexity of farming systems in SSA and SA.2
2.2. What is the impact of CA adoption on farmers’ profits orproduction risks?
Long-term productivity or environmental benefits from CA willnot drive CA adoption in smallholder systems in contexts wherefarmers are poor; shorter-term, tangible benefits to farmers aregenerally needed (Corbeels et al., this issue). Pannell et al. (thisissue) describe a simple but comprehensive model of the netreturns to hypothetical, small resource-poor farms growing cerealcrops (and potentially a legume crop in rotation) from the adoptionof CA practices. The model allows researchers to enter parameters,based on the best available data for a specific cropping system, toevaluate the likelihood that specific practices will be adopted. Suchmodelling of ex-ante adoption outcomes provides a powerful toolfor planning a project to disseminate a package of technologies likeCA.
Pannell et al. present a number of scenarios based on data fromCA research projects in Northern Zimbabwe that have been on-going since 2007. The model is an excellent example of using adegree of economic sophistication sufficient to adequately rep-resent reality, but simple enough to foster collaborations withnon-economists. Consistent with economic intuition, the authorsconclude that CA is more likely to be economically attractive onlarger, better resourced farms; to farmers with longer time horizons(the period over which they plan and make decisions), and lowerdiscount rates (the extent to which costs and benefits felt now havegreater value than future costs and benefits); and to farmers withless uncertainty on the costs and benefits of CA adoption.
2 It would be very interesting to put this to an empirical test – to run the Rosen-stock model using parameters from, say, 1000 villages, with predictions of outcomesgenerated for each one, and then aligned to a project in which CA is promoted inall of them. However, if we think there is some degree of validity in the model,then this approach would be unethical – we would be promoting a technology in asub-sample of villages in which our analysis predicts it won’t work well.
Dalton et al. (this issue) confirm the importance of short-runeconomic constraints in a study of farmer field schools in Ghana.Their paper documents that farmer field schools on a number ofCA practices (such as zero-tillage, residue retention, nutrient man-agement, tied ridges, introduction of legumes in the rotation) wereeffective in increasing participant knowledge from a low baselinelevel, and also increased the knowledge of non-participants in thesame communities through spillover effects. However, detailedpartial budgets constructed based on two years of data from 40farmers, indicated that short-run net returns were higher for con-ventional practice than from use of these CA practices. This suggeststhat educating farmers about these practices is not likely to facili-tate adoption because there is no economic incentive for them toadopt in those specific environments and circumstances. Indeed,their non-adoption is rational behaviour.
Arslan et al. (this issue) provide empirical evidence that pullstogether a number of themes raised in Corbeels et al. (this issue),Pannell et al. (this issue) and Andersson and D’Souza, (this issue).Using a two-period nationally representative panel survey inZambia they document significant dis-adoption of minimum soildisturbance (particularly the use of basins in hand-hoe agriculture)from 2004 to 2008. This is an important finding because Zambia hadpreviously been considered a success story for CA adoption in SSA(FAO, 2009). The authors find extension visits to be strongly corre-lated with continued adoption, but it is not possible to uncoverthe extent to which this reflects conditional provision of subsi-dized inputs, such as fertilizer–something that has been central tothe “donor-dependent” promotion of conservation agriculture inZambia (Umar, 2012; Andersson and D’Souza, this issue).
An encouraging finding from Arslan et al., from the perspectiveof risk mitigation and adaptation to climate change, is that district-level variation in historical rainfall during the growing season wasa good predictor of adoption of minimum soil disturbance. Thus,farmers appear to view CA as a strategy to reduce risk when theyare working in regions of variable rainfall. This finding is reinforcedby the work of Nyamangara et al. (this issue) in Zimbabwe thatreports that adopters of planting basins sow their plots 12–23 daysearlier than conventional practice as they do not have to wait for the
Editorial / Agriculture, Ecosystems and Environment 187 (2014) 1–10 5
first rains to soften the soil before they can plant, giving them flex-ibility to respond immediately irrespective of when the rain doesarrive. Another possible mechanism for a positive yield effect undererratic rainfall is the greater soil organic matter content in some sit-uations under CA (Palm et al., this issue) which can lead to greatersoil water retention and plant available water supply in cases wherethe amount of retained residue on the soil surface is sufficient foraffecting soil water balance by reducing evaporation and increas-ing aggregate stability and water infiltration (Hobbs et al., 2008;Sommer et al., 2012).
2.3. What are the environmental impacts of CA?
Palm et al. (this issue) confirm that there is no clear trend forhigher soil carbon sequestration under CA—a finding previouslydocumented in a review by Govaerts et al. (2009). The implicationsof this conclusion are many. In the absence of a clear short-termboost in profits through cost-savings for small-scale labor-intensivefarms3 or higher yields from adoption of CA (ambiguous, but likelynegative in the short-term—Brouder and Gomez-McPherson, thisissue), the idea of subsidizing farmers to “break the adoptionbarrier” (World Bank, 2012, p. xxix) has been suggested, on theassumption that it would sequester carbon. Indeed, some farmers inthe United States have been paid subsidies for adoption and main-tenance of zero-tillage based on a carbon price determined in theChicago carbon exchange under the assumption of C sequestrationarising from zero tillage (Baker et al., 2007). This strategy is nowquestioned because of skepticism about sequestration potential ofCA in the USA.
Palm et al. find that long-standing and fundamental scientificerrors are at the heart of the matter regarding soil carbon seques-tration under CA (Baker et al., 2007; Ellert and Bettany, 1995).After much optimism in the early 2000s about CA’s potential tosequester carbon, once the correct method of using equivalent soilmass as the basis for comparison of CA versus conventional practiceis applied, there is no consistent pattern of soil carbon sequestra-tion from CA adoption. However, CA adoption does provide otherecosystem benefits that may be valued by the farmer, particularlythe protection of the surface soil using residues, thereby reduc-ing runoff and surface crusting, increasing aggregate stability andwater infiltration, and greater soil water content and water use effi-ciency (Hobbs et al., 2008; Thierfelder and Wall, 2009; Farooq et al.,2011). These are not typically ecosystem services that would attractpublic subsidies as the benefits are largely captured by the farmer4,although other authors have argued that these do have public goodcharacteristics (e.g. Knowler and Bradshaw, 2007).
Ahmad et al. (this issue) analyzed the constraints and oppor-tunities for water saving and improving productivity throughresource-conserving technologies such as zero tillage, laser land-levelling, and bed and furrow planting in Punjab (Pakistan). Theauthors used both physical measurements and farmer surveys fromthe rice–wheat cropping system to analyse drivers of the adoptionand impacts on productivity and water savings. The survey foundthat adoption was closely associated with increased profits fromthese technologies despite a considerable proportion of farmers(almost one third of zero tillage adopters) reporting yield loss.
Another key finding from Ahmad et al. (this issue) is thatreductions in irrigation application at the field level did notalways translate into “water savings” at higher spatial scales(farm–cropping system–catchment), especially in areas wheredeep percolation from the root zone can be reused as a source
3 Such savings do occur in larger, mechanized farms – Pannell, this issue4 With the possible exception of a reduction in run-off from sloping lands in
important watersheds as discussed below
for groundwater irrigation elsewhere in the watershed. Mediumand large scale farmers tended to use the field scale irrigation sav-ings to increase their irrigated crop production area, suggestingthat without regulations and proper policies to manage water atthe watershed level, widespread adoption of CA could have theperverse outcome of increasing overall water use for a specificcatchment or aquifer.
Given inconsistent effects of CA adoption on emissions of NO2and CH4, and the ambiguous impacts on soil carbon sequestration(Palm et al., this issue), it would seem there is little justifica-tion to build a greenhouse gas mitigation program on subsidiesto farmers who adopt CA. A possible exception to this wouldbe for reduced carbon emissions through the entire life-cycle ofproduction, associated with less ploughing and less pumping inmechanised irrigated systems in SA (Erenstein and Laxmi, 2008).Another would be for reduced soil erosion and reduced run-offfrom sloping lands (Valentin et al., 2008) where there are well-documented positive impacts from adoption of CA. The sloping landprogram in China provides a precedent, in which farmers were paidto discontinue farming marginal lands on steep slopes (Li et al.,2011). However, given the stark differences in political, culturaland agronomic context it is unclear how transferable the idea ofpayments for CA adoption on sloping lands would be in SA andSSA.
2.4. How well does CA fit with wider agricultural, social,economic and political contexts for small-scale farmers?
Despite the benefits that CA can potentially contribute, debatecontinues about the specific conditions for which CA practices arerelevant for smallholder farming systems, especially in SSA (Gilleret al., 2009). Site-specific production constraints can often limitadoption of CA by resource-poor farmers, resulting in lower thanexpected benefits. The low adoption of CA in parts of SSA could bedue to the fact that CA is often promoted as a package, withoutproper fine-tuning of technologies and adaptation to local circum-stances (Tittonell et al., 2012). Factors that reduce the adoption ofCA include weed pressure, livestock demand for feed (Valbuenaet al., 2012), and the potential for increased severity of rootand foliar diseases associated with residue retention. The balancebetween these various enabling and constraining factors in specificsocio-political contexts shape whether CA adoption is appropriateand this is the focus of the final set of five papers in the special issue.Andersson and D’Souza provide a critique of the way in which CAhas been promoted in Southern Africa—focusing on contrasting his-tories in Zambia, Zimbabwe and Malawi. Their detailed analysis ofthe literature, including project documentation from a number ofnational, international and non-governmental organizations in theregion, shows how large-scale promotion of CA in Southern Africacame about. Their qualitative, socio-political analysis of the factorsbehind the “re-framing” of CA, from an erosion control measure toa set of practices with productivity and food security benefits, ishighly informative.
Andersson and D’Souza also attempt to answer the related ques-tions of how many farmers practice CA in Southern Africa andhow do they benefit from doing so? There are a number of poten-tial sources of upward bias in adoption studies, often in relationto sampling (i.e. being systematically non-representative of thepopulation of farmers in a particular country), or where farmersare counted as adopters of technology because they have receivedinputs/subsidies/incentives to do so but would not exercise theirfree choice to adopt in the absence of such inducements. Manynatural resource management technologies have been developedby the international agricultural research centers since the 1970s,but the majority of the ‘promising technologies’ developed havenot been widely adopted (Renkow and Byerlee, 2010), despite
6 Editorial / Agriculture, Ecosystems and Environment 187 (2014) 1–10
the putative promise of delivering environmental benefits rela-tive to current practice. We would expect to have seen take-upof many of these management technologies, at least temporarily,if these had been subsidized. Lack of access to fertilizer in Africa(through an inability to obtain credit and high import prices) is anacute problem, resulting in a continental average application rate ofonly approximately 10 kg/ha (International Fertilizer DevelopmentCenter, 2012). Any project that offers fertilizer to farmers, on thecondition that they practice a particular technology, is likely to bein high demand, but it does not mean that the promoted technologyis appropriate or will remain adopted over the long term.
Grounded in the success of widespread CA adoption in Australia,an analytical review by Kirkegaard et al. (this issue) attempts todraw lessons for the adaptation and adoption of CA principlesby smallholder farmers of SSA and SA. Although socioeconomicfactors of the farming systems considered by Kirkegaard et al.are clearly distinct from SSA and SA, many of the biophysicalparameters and economic drivers are common, e.g., infertile soils,variable and extreme climates, relatively low input levels, inte-grated crop–livestock systems, small profit margins, highly variableincome. Australian farmers have been commonly following a prag-matic and step-wise approach for adapting CA principles tailored tothe local biophysical and socio-economic conditions of their farm-ing systems. Further innovations in mixed farming systems are stillrequired to limit the competition over use of residues between live-stock feeding and retention for soil-cover that is inherent to theintegrated crop–livestock mixed farms. The high degree of flexi-bility in CA principles as practiced in southern Australian mixedfarming systems could provide a suitable lesson for future promo-tion of CA practices in smallholder systems SSA and SA, applyingadaptive research for testing a package of technologies and adapt-ing them to local conditions.
Weed pressure and high labour demands are among the mostimportant factors that reduce the benefits and limit adoption of CAby smallholder farmers in SSA (Giller et al., 2011; Erenstein et al.,2012). Although the reduction in tillage generally implies the needfor alternative weed control measures, including herbicides, a num-ber of alternative management approaches (e,g, stale-seed-bedtechnique) could be integrated to develop sustainable weed man-agement strategies under CA (Farooq et al., 2011). In the absenceof chemical weed control, manual weeding is often associatedwith disadoption, whereas manual hoe weeding also implies con-siderable soil movement and thereby reduces the benefits of CA(Erenstein et al., 2012). Nyamangara et al. (this issue) report oneffects of reduced tillage based on hand-hoe dug planting basinson weed pressure, labour demand, and returns to investment com-pared with an animal-drawn mouldboard plough (conventionaltillage), in contrasting agro-ecological zones in Zimbabwe. The sur-vey found that farmers weeded their reduced tillage basin plotssignificantly more frequently compared to the conventional plots,yet the reduced tillage basin plots still had more weeds com-pared to conventional plots. Labour demand was more than doubleunder reduced tillage than in the conventional treatment althoughreturns to investment were slightly higher under reduced com-pared to conventional tillage. The authors conclude that reducedtillage should be targeted to households with access to draft powerfor tillage and herbicides for weed control to reduce labour demandand consequently increase its adoption both in terms of number offarmers and cultivated area in southern Africa.
Corbeels et al. (this issue) bring a wealth of local-level data tobear on the question of interactions across spatial scales when CAis studied at field, whole farm or landscape/region levels. Basedon field and farm-level data from seven countries in Sub-SaharanAfrica they find that CA has potential to increase yields over thelong-term, and particularly at sites with erratic rainfall. Corbeelset al. also show that CA adoption is constrained by lack of short-term
increase in profits (supporting the modelling work in Pannell et al.,this volume), and that the trade-offs regarding the use of residuesfor fodder is an important disincentive to CA adoption.
Along similar lines, Baudron et al. (this issue) model how spe-cific technical innovations may potentially ease trade-offs betweenalternative demand for crop residues for soil-cover mulch or as live-stock fodder on mixed crop–livestock farming systems in SSA. Inparticular, they use data from livestock feeding trials to examinehow substitution in the diets of dairy herds, from cereal residueto nutrient-rich feeds, affects milk production and the unit cost ofproduction. They also examine how replacing animal draft powerthrough small-scale mechanisation can reduce demand for cerealresidues, and how closing the wide yield gap (to reach 90% of water-limited yield potential) can increase residue production. However,the idea that a wide range of technologies exist for increasingyield in the systems they study (Western Kenya and Ethiopia) isoptimistic—what is needed are technologies that increase farmprofits given uncertainty aversion (unwillingness to adopt owingto lack of information about the distribution of potential outcomes,e.g. Dalton et al., this issue), risk aversion (unwillingness to adoptbecause of the possibility of having a negative outcome; e.g. Pannell,this issue) and the widespread market failures that limit the adop-tion of agricultural technologies (Jack, 2013). Baudron et al. proposesteps towards a process of greater residue retention for the long-term health of African soils. They argue, however, that benefitsof mulching are site-specific, and that the priority is a focus onwater-use efficiency rather than on adoption of a specific set oftechnologies.
3. Discussion
Based on the papers in this special issue and the discussionat a workshop on CA held at the University of Nebraska, Lincolnon 15 and 16 October 2012, a consensus emerged around anumber of key issues. In the weeks following the workshop, wedrafted and agreed the “Nebraska Declaration” on conservationagriculture. The signatories to the Declaration (listed in AppendixA) represent the major academic disciplines relevant to thestudy of CA for small-scale resource-poor farmers in developingcountries, and we believe it summarizes the main findings andimplications from the papers in this special issue. The Declaration(http://www.sciencecouncil.cgiar.org/fileadmin/templates/ispc/documents/Meetings and events/Workshops/NE declaration onCA - FINAL.pdf) comprises 13 points as listed in Box 1.
Given low yields, low income, and soil degradation of small-holder farming systems in SSA, and to some extent in SA, thereis urgent need to increase productivity, profit, and soil quality asa means to achieve sustainable intensification (The MontpellierPanel, 2013). There are several potential pathways to achieving thisgoal. For instance it is well known that diversified crop rotationsand introduction of legumes in cereal-based production systemsare important elements of good agricultural practices, not only forCA but also in the context of a range of practices referred to as Inte-grated Soil Fertility Management (ISFM; Vanlauwe et al., 2010).ISFM aims to sequentially improve farming system performanceby maximizing the efficiency of fertilizer use, and increasing cropproductivity and biomass, which could provide incentive for theadoption of CA by smallholder farmers in SSA (Lahmar et al., 2012).Efforts for development and promotion of CA technologies havebeen in progress for several decades in some international organi-zations. While adoption would appear to be increasing worldwide(Friedrich et al., 2012; the FAO data are based on “expert opinion” atnational level so are inevitably imprecise), CA codified as a packageof three practices has not succeeded in the rainfed systems of SSAand SA (Fig. 1).
Editorial / Agriculture, Ecosystems and Environment 187 (2014) 1–10 7
Box 1: The Nebraska Declaration on Conservation Agriculture
1 We, a group of 44 scientists from a wide range of disciplines and international agricultural research and development organizations,have discussed and debated the scientific evidence regarding conservation agriculture for small-scale, resource-poor farmers inSub-Saharan Africa (SSA) and South Asia (SA), and we have reached consensus on the following points.
2 The goals of conservation agriculture are to concomitantly:• Improve household food security and/or increase profit• Achieve substantial increases in crop yields and greater yield stability on existing farmland with climate and soils suitable for
sustainable intensification.• Reverse trends of natural resource degradation associated with crop production, with particular emphasis on increasing water
capture and retention in soils, avoiding erosion, and improving or maintaining soil quality.• Contribute to mitigation of emissions by reducing greenhouse gas emissions per unit of production.• Help farmers adapt to climate variability and change.
3 Most efforts to date in developing countries have promoted conservation agriculture as a package of three practices: minimumdisturbance of soil (zero/minimum/reduced tillage); retention of sufficient crop residue to provide surface coverage; and diversifiedcropping patterns that include at least three plant species including one legume.
4 There is little evidence of widespread adoption of conservation agriculture in SSA or SA when strictly defined as this three-componentpackage. In contrast, there is some evidence of adoption of one or two of these components in some parts of SSA and SA.
5 Like most farming practices, the main driver of adoption for conservation agriculture is a positive impact on profit and/or householdfood security (including reduced risk of crop failure, particularly important for resource-poor farmers). For most resource-poorfarmers, the positive impacts on soil properties or ecosystem services are important determinants of adoption only through theirpotential short-term effect on profits or reduced risk. Technologies that simultaneously meet farmers’ short-term objectives whileimproving provision of ecosystem services must be identified by research and extension in order to achieve widespread adoptionof practices with environmental benefits.
6 There are sound agronomic, economic, and/or social reasons why farmers have not adopted the three-component conservationagriculture package in SSA and SA. Typically this is because one or more of the components is not consistent with the objectives ofsmall-scale, resource-poor farmers in these regions, or cannot be implemented given the constraints they face. For example, suchfarmers may choose not to adopt conservation agriculture due to an inability to access or purchase machinery, equipment, or inputs(e.g., small-scale seeders, or herbicides) necessary for conservation agriculture to perform effectively.
7 Benefits from retention of crop residues in the soil are small at the low average yields typical of many parts of SSA and SA. Whilethere is evidence of reduced runoff and erosion, and greater infiltration in some cases, with relatively small amounts of residuesleft on the soil surface (e.g. 1.5 t/ha), crop residues are of high value as fodder or fuel in many agricultural systems of SSA and SA,and can account for a large portion of total crop value. Understanding and quantifying tradeoffs between the benefits of residueretention to future crop productivity and soil quality, versus its value for other uses, is an important research priority for majoragricultural systems in SSA and SA.
8 To play a significant role in low-productivity, resource-poor agricultural systems in SSA and SA, efforts to reach the goals outlinedin point 2 should be broadened beyond a focus on the package of three main CA agronomic practices. Additional emphasis shouldbe placed on diagnostic agronomy and participatory on-farm research to identify biophysical and socioeconomic constraints toincreased crop production, and to guide farmers in finding solutions from among a broader range of sound agronomic practices thatachieve the underlying or fundamental goals of conservation agriculture. An expanded list of practices would include: crop rotation,relay-cropping and inter-cropping; retention of crop residues; green manure, cover or fodder crops; zero, reduced, or minimumtillage; appropriate use of organic and inorganic nutrients; improved weed and disease management techniques, equipment andherbicides; use of physical soil or water conservation structures (bunds, drainage, grass contours and waterways, planting basins).There is no fixed recipe for how these practices should be optimally combined for a given agricultural system.
9 There is a critical need to better understand how these component practices affect yield, farmers’ profits, soil quality, water con-sumption, ecosystem services, and the trade-offs among these factors at multiple scales of analysis (field, farm, watershed, region)in major agricultural systems of SSA and SA.
10 Systematic efforts are needed to assess the suitability and viability of management options and practices, given farmers’ objectivesand constraints, to better target the extrapolation domain of conservation agriculture practices and technologies within existingagricultural systems of SSA and SA. Long-term investment is required in collaborations that bring together researchers withfarmers, farmers’ associations, extension agents, non-governmental organizations, community groups, public administrators, andprivate sector providers of CA-related products and services. Diagnostic agronomy and participatory on-farm research are usefulapproaches, alongside model simulations, to evaluate ex-ante the potential impacts of adopting CA practices and technologies onfarm-level objectives and on ecosystem services. Outputs and outcomes from these coordinated research efforts will enhance effec-tiveness of extension approaches to support adoption of a broader range of conservation agriculture practices and technologies(outlined in point 8) at scale.
11 Rigorous studies are needed to assess and better understand the process of adoption of conservation agriculture in SSA andSA. Such studies will be instrumental in explaining who adopts conservation agriculture practices and why. Equally important insuch studies is to understand the reasons for non-adoption and dis-adoption. These studies should lead to better targeting andrecommendation domains for policies that address socioeconomic and biophysical constraints to adoption.
12 Also needed is a better understanding of the role that financial or in-kind incentives have played in promoting conditional acceptanceof conservation agriculture practices used in some development projects in SSA and SA. At issue are the most effective kinds ofeconomic instruments that can incentivise long-term, sustained adoption. Randomized control trials of extension programs orincentive mechanisms designed to reduce the costs or risks associated with adoption offer a rich set of targets for future studies.
13 Based on critical review of the literature on impacts of conservation agriculture on soil carbon sequestration and GHG emissions,payment for carbon credits does not appear to be a viable driver for promoting widespread adoption of conservation agriculturetechnologies by smallholders. However, there may be cases where economically efficient payment schemes can be established forecosystem services seen as public goods such as reduced erosion and nutrient loss (thus preventing water pollution and siltationof waterways and reservoirs), and building or maintaining soil productivity for future generations.
Author's personal copy
8 Editorial / Agriculture, Ecosystems and Environment 187 (2014) 1–10
Fig. 1. Most recent estimate (year of estimate ranges from 2006 to 2011) for countries in FAO’s dataset on conservation agriculture adoption.
A general blueprint for a CA approach based strictly on thethree principles is highly unlikely to work in SSA and SA becauseof diverse political, socio-economic and agroecological contexts(Kienzler et al., 2012). Cost-effective technologies and agriculturalimplements specifically suited for small-scale farm enterprises andresource conserving practices can be made available (Serraj andSiddique, 2012) but as many of the papers in this special issue show,their use may often not be profitable in the 3-year planning horizonthat typifies many resource-poor farmers (Pannell et al., this issue).As we see in Brazil, adoption among wealthier, market-integratedfarmers can spread rapidly (Ekboir, 2003). With the emergenceof infrastructure in “development corridors” in a number of Sub-Saharan African countries (Weng et al., 2013), there is considerablepotential for well-capitalized commercial farmers to adopt con-servation agriculture over the coming decades. For resource-poorsmallholders, some form of contract farming arrangement seemsto be the most likely mechanism through which we might expectwidespread adoption of conservation agriculture in the absenceof conditional subsidies. Both contract farming and conservationagriculture feature strongly in the World Bank report “AwakeningAfrica’s Sleeping Giant” on the prospects for commercial agricul-tural development in West Africa’s Guinea savannah zone (WorldBank, 2009).
Many of the papers in this special issue can be classified asex-ante impact assessment of CA technologies. There are someimportant methodological challenges in producing evidence aboutCA potential benefits and actual realized impacts. In essence thereis a tension between rigor and relevance to the context of small-holding farmers in SSA and SA (Andersson and D’Souza; Palm et al.;Kirkegaard et al., all this issue). Studies carried out on experimentalstations have a high internal validity, owing to their highly con-trolled conditions, allowing for a very rigorous assessment of theimpact of using technology (e.g. Gathala et al., this issue). How-ever, the Achilles heel of experimental station studies is preciselythat their highly controlled conditions do not allow for an assess-ment of how the technology fits in the local socioeconomic context(Giller et al., 2011). Survey-based studies that compare outcomes(e.g. yield) for adopters with outcomes for non-adopters in the samecontext must contend with the difficult challenge of controlling forselection bias (De Janvry et al., 2011) and heterogeneity in returnsto adoption (Suri, 2011)—the fact that those who adopt technolo-gies early tend to be better farmers (in ways that it is difficult tocollect data on to control for it) means that internal validity of thestudy is in jeopardy.
A number of large-scale agricultural research and developmentprograms have been established in recent years to offer a moreintegrated approach to researching the constraints to sustained
improvements in agricultural productivity. Examples include theCGIAR research program on Climate Change Agriculture and FoodSecurity (CCAFS); Economics and Policy Innovations for ClimateSmart Agriculture (EPIC); the Sustainable Agriculture and NaturalResources Management (SANREM) Collaborative Research SupportProgram (CRSP); as well projects specifically focused on CA at CIRAD(Centre de Coopération Internationale en Recherche Agronomiquepour le Développement) and at CIMMYT (International Maize andWheat Improvement Center) and other CGIAR centers. Researchersfrom all of these programs are represented in this special issue.The collaborations among agronomists, livestock scientists, socialscientists and agricultural economists that are fostered by theseprograms offer the promise of both improving our understandingof whether new technologies fit into farming systems, and howprojects can best be designed to help overcome the constraintsfaced by smallholder farmers in Sub-Saharan Africa and SouthAsia. Mechanisms for achieving progress might include throughimproved extension messages, research-based policy advice, or tar-geted project interventions.
4. Conclusion
The papers in this special issue give insight into the state of thescience on CA across a range of disciplines—agronomy, economics,environmental science and social sciences. The conclusions fromthese papers suggest that: (1) CA is not widely adopted in Sub-Saharan Africa and South Asia owing to a lack of economic incentivefor smallholder farmers—that without a short-term cost-savingof the kind achieved on mechanised farms in other regions of theworld (i.e. through reduced fuel costs associated with ploughing),the process of conversion to CA is not profitable over planninghorizons of most resource-poor farmers; (2) Yield increases underCA are possible but uncertain given the low average yields, andlarge yield gaps, that pertain in these regions, and yield gains aremore likely to be observed after several years; (3) Given that thereis also no clear trend for greater carbon sequestration under CA, thepotential for subsidizing farmers to adopt CA using payments forecosystem services/carbon credit schemes seems limited in scope;(4) There is early evidence that farmers perceive a benefit from CAadoption in regions that are prone to erratic rainfall, suggestinga potential risk mitigation role but this trend still needs to bevalidated through further research; (5) An emerging consensus(reflected in a number of points in the Nebraska Declaration)suggests that sustainable intensification in smallholder systems inSub-Saharan Africa and South Asia requires incremental positive
Editorial / Agriculture, Ecosystems and Environment 187 (2014) 1–10 9
change brought about by focused and long-term research anddevelopment programs.
Appendix A. Signatories to the Nebraska Declaration
1. Mobin-ud-Din Ahmad CSIRO Land and Water, Canberra, Australia2. Juliana Albertengo Asociación Argentina de Productores en Siembra
Directa (AAPRESID)3. André Bationo Alliance for a Green Revolution in Africa (AGRA)4. Humberto
Blanco-CanquiUniversity of Nebraska-Lincoln
5. Ademola Braimoh World Bank6. Sylvie Brouder Department of Agronomy, Purdue University7. Kenneth Cassman CGIAR Independent Science and Partnership
Council8. Lieven Claessens International Crop Research Institute for the
Semi-Arid Tropics (ICRISAT)9. Marc Corbeels Centre de Coopération Internationale en Recherche
Agronomique pour le Développement (CIRAD)10. Tim Dalton Kansas State University11. Domingos Dias Instituto de Investigac ao Agrária de Mozambique
(IIAM)12. David Feindel International Center for Agricultural Research in
the Dry Areas (ICARDA)13. Lydiah Gatere Agriculture and Food Security Center, The Earth
Institute/Columbia University14. Bruno Gerard International Maize and Wheat Improvement
Center (CIMMYT)15. Helena
Gomez-MacphersonInstitute of Sustainable Agriculture (IAS-CSIC),Cordoba, Spain
16. Bram Govaerts International Maize and Wheat ImprovementCenter (CIMMYT)
17. Patricio Grassini University of Nebraska-Lincoln18. Peter Hobbs Cornell University19. Liz Humphreys International Rice Research Institute (IRRI)20. Solomon Jemal Melkassa Agricultural Research Center, Ethiopia21. Alpha Kamara International Institute for Tropical Agricutlure
(IITA)22. Francis Kihanda Kenyan Agricultural Research Institute (KARI)23. John Kirkegaard CSIRO Sustainable Agriculture Flagship, Canberra,
Australia24. JK Ladha International Rice Research Institute (IRRI)25. Bruce Linquist University of California, Davis26. Martha Mamo University of Nebraska-Lincoln27. Paswel Marenya International Maize and Wheat Improvement
Center (CIMMYT)28. Stephen Mason University of Nebraska-Lincoln29. Nancy McCarthy LEAD Analytics30. Justice Nyamangara International Crop Research Institute for the
Semi-Arid Tropics (ICRISAT)32. Cheryl Palm Agriculture and Food Security Center, The Earth
Institute/Columbia University32. David Pannell School of Agricultural and Resource Economics,
University of Western Australia33. John Passioura CSIRO, Canberra, Australia34. John Reganold Department of Crop & Soil Sciences, Washington
State University35. Rachid Serraj ICARDA, presently CGIAR Independent Science and
Partnership Council Secretariat36. Bharat Sharma International Water Management Institute (IWMI)37. David Spielman International Food Policy Research Institute (IFPRI)38. James Stevenson CGIAR Independent Science and Partnership
Council Secretariat39. Nils Teufel International Livestock Research Institute (ILRI)40. Christian Thierfelder International Maize and Wheat Improvement
Center (CIMMYT)41. Jessica Torrion University of Nebraska-Lincoln42. Leigh Winoweicki International Center for Tropical Agriculture (CIAT)43. Christian Witt Bill and Melinda Gates Foundation (BMGF)44. Charles Wortmann University of Nebraska-Lincoln
References
Andersson, J., Giller, K., 2012. On heretics and God’s blanket salesm Contested claimsfor Conservation Agriculture the politics of its promotion in African smallholderfarming. In: Sumberg, J., Thompson, J. (Eds.), Contested Agronomy: AgriculturalResearch in a Changing World. Routledge, London, pp. 22–46.
Baker, J.M., Ochsner, T.E., Venterea, R.T., Griffis, T.J., 2007. Tillage and soil carbonsequestration—What do we really know? Agric. Ecosyst. Environ. 118, 1–5.
Cassman, K.G., 1999. Ecological intensification of cereal production systems: Yieldpotential, soil quality, and precision agriculture. Proc. Natl. Acad. Sci. U. S. A. 96,5952–5959.
Cassman, K.G., Wood, S., 2005. Cultivated systems. In: Hassan, R., Scholes, R., Ash,N. (Eds.), Ecosystems and Human Well-Being: Current State and Trends: Find-ings of the Conditions and Trends Working Group of the Millennium EcosystemAssessment. Island Press, Washington, DC, pp. 747–794.
De Janvry, A., Dustan, A., Sadoulet, E., 2011. Recent Advances in Impact AnalysisMethods for Ex-post Impact Assessments of Agricultural Technology: Optionsfor the CGIAR. Independent Science and Partnership Council, Rome.
Ekboir, J.M., 2003. Research and technology policies in innovation systems: zerotillage in Brazil. Res. Policy 32, 573–586.
Ellert, B.H., Bettany, B.H., 1995. Calculation of organic matter and nutrients storedin soils under contrasting management regimes. Can. J. Soil Sci. 75, 529–538.
Erenstein, O., Laxmi, V., 2008. Zero tillage impacts in India’s rice–wheat systems: Areview. Soil Tillage Res. 100, 1–14.
Erenstein, O., Sayre, K., Wall, P., Hellin, J., Dixon, J., 2012. Conservation agriculturein maize- and wheat-based systems in the (sub)tropics: lessons from adapta-tion initiatives in South Asia, Mexico, and Southern Africa. J. Sustain. Agric. 36,180–206.
FAO, 2009. Scaling-up Conservation Agriculture in Africa: Strategy and approaches.FAO, Addis Ababa.
FAO, 2011. Save and Grow: A Policymaker’s Guide to the Sustainable Intensificationof Smallholder Crop Production. Food and Agriculture Organization of the UnitedNations, Rome http://www.fao.org/ag/save-and-grow/
Farooq, M., Flower, K.C., Jabran, K., Wahid, A., Siddique, K.H.M., 2011. Crop yield andweed management in rainfed conservation agriculture. Soil Tillage Res. 117,172–183.
Friedrich, T., Derpsch, R., Kassam, A., 2012. Overview of the global spread of conser-vation agriculture. Field Actions Sci. Rep. 6, 1–7.
Garnett, T., Appleby, M.C., Balmford, A., Bateman, I.J., Benton, T.G., Bloomer, P.,Burlingame, B., Dawkins, M., Dolan, L., Fraser, D., Herrero, M., Hoffmann, I.,Smith, P., Thornton, P.K., Toulmin, C., Vermeulen, S.J., Godfray, H.C.J., 2013.Sustainable intensification in agriculture: premises and policies. Science 341,33–34.
Giller, K.E., Corbeels, M., Nyamangara, J., Triomphe, B., Affholder, F., Scopel, E., Tit-tonell, P., 2011. A research agenda to explore the role of conservation agriculturein African smallholder farming systems. Field Crop Res. 124, 468–472.
Giller, K.E., Witter, E., Corbeels, M., Tittonell, P., 2009. Conservation agriculture andsmallholder farming in Africa: The heretics’ view. Field Crop Res. 114, 23–34.
Govaerts, B., Verhulst, N., Castellanos-Navarrete, A., Sayre, K.D., Dixon, J., Dendooven,L., 2009. Conservation agriculture and soil carbon sequestration: between mythand farmer reality. CRC Crit. Rev. Plant Sci. 28, 97–122.
Hobbs, P.R., Sayre, K., Gupta, R., 2008. The role of conservation agriculture in sus-tainable agriculture. Philos. Trans. R. Soc. Lond. B Biol. Sci. 363, 543–555.
Jack, B.K., 2013. Constraints on the adoption of agricultural technologies in devel-oping countries. Literature review, Agricultural Technology Adoption Initiative,J-PAL (MIT) and CEGA (UC Berkeley).
Kassam, A., Friedrich, T., Shaxson, F., Pretty, J., 2009. The spread of conservation agri-culture: justification, sustainability and uptake. Int. J. Agric. Sustain. 7, 292–320.
Kienzler, K.M., Lamers, J.P., McDonald, A., Mirzabaev, A., Ibragimov, N., Egamberdiev,O., Ruzibaev, E., Akramkhanov, A., 2012. Conservation agriculture in CentralAsia—What do we know and where do we go from here? Field Crop Res. 132,95–105.
Knowler, D., Bradshaw, B., 2007. Farmers’ adoption of conservation agriculture: areview and synthesis of recent research. Food Policy 32, 25–48.
Lahmar, R., Bationo, B.A., Dan Lamso, N., Guéro, Y., Tittonell, P., 2012. Tailoring con-servation agriculture technologies to West Africa semi-arid zones: Building ontraditional local practices for soil restoration. Field Crop Res. 132, 158–167.
Li, J., Feldman, M.W., Li, S., Daily, G.C., 2011. Rural household income and inequalityunder the Sloping Land Conversion Program in western China. Proc. Natl. Acad.Sci. U. S. A. 108, 6–7721.
Renkow, M., Byerlee, D., 2010. The impacts of CGIAR research: A review of recentevidence. Food Policy 35, 391–402.
Serraj, R., Siddique, K.H.M., 2012. Conservation agriculture in dry areas. Field CropRes. 132, 1–6.
Sommer, R., Piggin, C., Haddad, A., Hajdibo, A., Hayek, P., Khalil, Y., 2012. Simulatingthe effects of zero tillage and crop residue retention on water relations and yieldof wheat under rainfed semiarid Mediterranean conditions. Field Crop Res. 132,40–52.
Stevenson, J.R., Villoria, N., Byerlee, D., Kelley, T., Maredia, M., 2013. Green Revolutionresearch saved an estimated 18 to 27 million hectares from being brought intoagricultural production. Proc. Natl. Acad. Sci. U. S. A. 110, 8–8363.
Sumberg, J., Thompson, J., Woodhouse, P., 2012. Why agronomy in the developingworld has become contentious. Agric. Human Values 30, 71–83.
Suri, T., 2011. Selection and comparative advantage in technology adoption. Econo-metrica 79, 159–209.
International Fertilizer Development Center, 2012. Improving Fertilizer Supplies inSub-Saharan Africa. IFDC Report 37:2, 10–11.
The Montpellier Panel, 2013. Sustainable intensification: A new paradigm for Africanagriculture, London.
Thierfelder, C., Wall, P.C., 2009. Effects of conservation agriculture techniques oninfiltration and soil water content in Zambia and Zimbabwe. Soil Till. Res. 105,217–227.
Author's personal copy
10 Editorial / Agriculture, Ecosystems and Environment 187 (2014) 1–10
Tittonell, P., Scopel, E., Andrieu, N., Posthumus, H., Mapfumo, P., Corbeels, M., vanHalsema, G.E., Lahmar, R., Lugandu, S., Rakotoarisoa, J., Mtambanengwe, F.,Pound, B., Chikowo, R., Naudin, K., Triomphe, B., Mkomwa, S., 2012. Agroecology-based aggradation-conservation agriculture (ABACO): Targeting innovations tocombat soil degradation and food insecurity in semi-arid Africa. Field Crop Res.132, 168–174.
Umar, B.B., 2012. Reversing agro-based land degradation through conservationagriculture: emerging experiences from Zambia’s smallholder farming sector.Sustain. Agric. Res. 1, 72–87.
Valbuena, D., Erenstein, O., Homann-Kee Tui, S., Abdoulaye, T., Claessens, L., Duncan,A.J., Gérard, B., Rufino, M.C., Teufel, N., van Rooyen, A., van Wijk, M.T., 2012.Conservation Agriculture in mixed crop–livestock systems: Scoping crop residuetrade-offs in Sub-Saharan Africa and South Asia. Field Crop. Res. 132, 175–184.
Valentin, C., Agus, F., Alamban, R., Boosaner, A., Bricquet, J.P., Chaplot, V., de Guz-man, T., de Rouw, A., Janeau, J.L., Orange, D., Phachomphonh, K., Podwojewski,P., Ribolzi, O., Silvera, N., Subagyono, K., Thiébaux, J.P., Vadari, T., 2008. Runoffand sediment losses from 27 upland catchments in Southeast Asia: Impact ofrapid land use changes and conservation practices. Agric. Ecosyst. Environ. 128,225–238.
van Ittersum, M.K., Cassman, K.G., Grassini, P., Wolf, J., Tittonell, P., Hochman, Z.,2013. Yield gap analysis with local to global relevance—A review. Field CropRes. 143, 4–17.
Vanlauwe, B., Bationo, A., Chianu, J., Giller, K.E., Merckx, R., Mokwunye, U., Ohiokpe-hai, O., Pypers, P., Tabo, R., Shepherd, K.D., Smaling, E.M., Woomer, a., Sanginga,P.L.N., 2010. Integrated soil fertility management: Operational definition andconsequences for implementation and dissemination. Outlook Agric. 39, 17–24.
Weng, L., Boedhihartono, A.K., Dirks, P.H.G.M., Dixon, J., Lubis, M.I., Sayer, J.A., 2013.Mineral industries, growth corridors and agricultural development in Africa.Global Food Security 2, 195–202.
World Bank, 2009. Awakening Africa’s Sleeping Giant: Prospects for CommercialAgriculture in the Guinea Savannah Zone and BeyondPN World Bank. WorldBank, Washington, DC.
World Bank, 2012. Carbon Sequestration in Agricultural Soils. In: Agriculture andRural Development Report no. 67395-GLB. World Bank, Washington, DC.
James R. Stevenson a,∗
Rachid Serraj a
Kenneth G. Cassman b
a CGIAR Independent Science and Partnership CouncilSecretariat, FAO, Rome, Italy
b Department of Agronomy and Horticulture,University of Nebraska, Lincoln, NE 68583,
