Introduction

Targeting Conservation Agriculture (CA) remains a major challenge in the context of Africa’s heterogeneous agro-ecologies, farmers and farming systems, socio-economic and institutional environments. Although CA can stabilize and increase yields, conserve and improve soil quality, its outcomes are site-specific (Rosenstock et al, 2014; Rusinamhodzi et al, 2011). In sites where results of using CA are positive, not all farmers achieve the required level of management, notably early planting, timely weeding and adequate use of chemical fertilizer (Gatere et al, 2014; Aune et al, 2012; Baudron et al, 2014; Grabowski & Kerr, 2013; Vanlauwe et al., 2014).

CA adoption studies potentially provide ex-post insight into the suitability of particular CA practices and packages for particular categories of farmers in specific agro-ecological, socioeconomic and institutional contexts. Yet, methodological weaknesses of these studies limit their value for CA targeting. First, data collection in CA adoption studies is often biased as it generally takes place in the context of on-going development projects that incentivize CA adoption through input support (fertilizers, seeds, CA implements). Second, these studies generally use unclear and/or reductionist definitions of what is an adopter or what (set of) practices constitutes ‘CA adoption’ – often using the implicit definition ‘practicing minimum tillage on some part of the farm in year x’. Third, CA adoption studies tend to be limited to analyses of farm-level ‘adoption determinants’, largely disregarding higher scale influences affecting CA uptake, such as rural livelihoods, agro-ecological, socio-economic and institutional factors (Andersson & D’Souza, 2014).

Although on-station and on-farm agronomic experiments (e.g. Thierfelder & Wall, 2010 et al, 2013), farm-level economic studies (Pannell et al, 2014), and (market) institutional analyses (Mazvimavi & Twomlow, 2009) have contributed significantly to our understanding of the adoption potential of particular CA practices in African smallholder agriculture, ex-ante targeting of diverse CA packages remains a high priority for donors, policy makers and other rural interventionists. This paper aims to contribute to the development of a CA suitability or ‘adoption potential’ tool that can quickly identify suitable CA options in area, and the categories of farmers that are likely to adopt these practices. Building on the expanding literature on CA use in African smallholder agriculture and experience with the use of qualitative assessment tools – notably the QAToCA15 tool (Ndah et al, 2010, 2014) the authors explore the key elements of a quick identification tool for targeting CA interventions in Africa.

Material and Methods

QAToCA is a simple tool that was built with the aim to enable regional experts, research teams and/or managers of development projects with a focus on CA to assess the ‘relative likelihood of CA adoption’ (Ndah et al 2010: 2) or ‘adoption potential of site-specific CA practices’ in Africa (Ndah et al, 2014: 2). The tool consists of a list of questions with answer statements and scores that together determine the potential for CA adoption in a given project region (Ndah et al, 2014). Questions deal with characteristics of CA as an object of adoption, the capacity of the promoting organization(s), attributes of the dissemination strategy; institutional frame conditions at village and regional level, market conditions at the village and regional levels, and the community’s perception towards CA. In this way, the tool allows for diagnosing the supporting and hindering factors of CA adoption in a given area. QAToCA has been used as a quick assessment guide in a range of CA research and development projects across Africa (Ndah et al, 2014).

Results and Discussion

From the authors’ experience with the use of QAToCA, they identified a number of limitations for its use as an ex-ante identification tool for CA suitability or CA adoption potential. These shortcomings will be addressed as the authors develop a new tool. First, since it was developed as a self-assessment guide for CA proponents in on-going CA projects, it strongly focuses on the promoting organizations and their dissemination strategy (extension); the underlying assumption being that knowledge is limiting the adoption of CA, rather than the farming context (which receives less attention in the assessment). Second, in its current state, QAToCA treats all assessment criteria with the same level of importance and does e.g. not deal with factors that may fully impede adoption. This unweighed aggregation of criteria makes it difficult to sort priorities in CA targeting. Third, there is a strong possibility of bias in assessments done by stakeholders (interested parties), and as a result of the construction of knowledge in public social events such as focus group discussions (cf. Mosse, 1994). The answers people give are likely to be strongly influenced by their own expectations of, or interests in the project (Andersson and D’Souza, 2013). Lastly, the higher level analysis (at village/region scale) is unable to understand farmers’ CA adoption decision-making in the context of diverse CA packages and the heterogeneity of farmers, their production objectives and constraints.

Critical components for an identification tool of CA adoption potential Identifying socio-ecological niches for CA practices/packages – The authors suggest that a quick identification tool for targeting new CA interventions in Africa uses the concept of the socioecological niche (Ojiem et al. 2006, Giller et al. 2009), which provides a practical framework for ‘ideotyping’ the contexts within which CA has most to offer. As series of bio-physical, socio-economical, and institutional factors and their interactions delineate the socioecological niche for a type of CA practice or package. For each factor, several criteria boundaries have to be established and are used to set the limits for the niche.

Need for multi-scale analysis – Following Sumberg (2005), with the tool the authors will explicitly distinguish adoption constraints (farm-scale) from prerequisite conditions (higher scales); the former referring to the ‘goodness-of-fit’ between the CA practice or package and the farmer (type), while the latter focuses on contextual factors, and that cannot be influenced by the CA development and dissemination process. Thus, in comparison with the QAToCA tool, the tool seeks to more explicitly identify the suitability of different CA practices/technologies at farm-level.

Heterogeneity of farmers and different technologies – CA comprises of different practices, each with their specific requirements for labour, equipment, fertilizer, etc., that are suitable to different types of farmers (and farming environments). For example, direct seeders have a high equipment cost, although there are cheaper alternatives such as rippers. In general, CA will be most rapidly adopted by smallholder farmers with adequate resources of land, cash and labour, and not by the most resource-constrained groups. Functional farm typologies based on farmers’ production objectives and resource endowments (including the importance of farm size) will help in better targeting CA packages/technologies. To start with, a clear

Conference paper presented at the First Africa Congress on Conservation Agriculture- Lusaka, Zambia.

Authors: Andersson, Jens A.1, Corbeels, Marc

Corresponding author: j.andersson@cgiar.org