Women’s Fields IV

Lead Organization:

FUMA Gaskiya Federation

Partner Organizations:

National Agricultural Research Institute of Niger (INRAN), IMAAN, IRSAT (Burkina Faso), University of Hohenheim, CATHI-Gao project, NGO RAIL, UDDM through the Sahel-IPM project, UNICEF, NGO Save the Children

Community of Practice:

West Africa






During the last three phases of Women’s Fields, the farmer federation FUMA Gaskiya in Niger’s Maradi region grew from its previous role as a minor project partner to a fully accountable research lead. Using a FRN approach and building on synergies with other projects (Networking4Seed, Cowpea Square, CATHI-Gao, GIMEM, Sahel-IPM), a basket of AEI options was co-created, with options largely based on locally available resources, thereby serving the vulnerable and poor, especially women farmers. Options include new crops (Cassiatora) and robust varieties of pearl millet, sorghum, cowpea, and groundnut; innovative agronomic practices such as partial weeding, localized compost applications, and assisted natural regeneration (ANR) of shrubs and trees; biocontrol of the pearl millet headminer; and sanitized human urine as fertilizer (“OGA”) and seedballs as a climate risk-reducing innovation on poor sandy soils. The latter two options resulted in pearl millet yield increases of up to 30 percent (Oumarou et al., 2022, Nwankwo et al., 2021).

Initially, the options were tested by a few farmers. Using a farmer-support-method (“faire-faire”) built on FUMA’s infrastructure, the project moved to large-N trials (implemented by >1000 farmers) since 2017. Development of an Android app allowed for on-farm digital data collection. Co-creation of a solid database owned by FUMA but accessible to researchers took into account climatic, social, and agronomic variables, and the realization and updating of the farmer typology allowed FUMA and its partners to better understand the variability between farmers and to interpret the heterogeneity of on-farm results. 

While promising results were obtained at plot levels with large numbers of farmers, agricultural problems cannot be dealt with in isolation and at the individual plot-level scale.

Grant Aims:

The overall goal is to change the scale from previous phases and move to farm-/household-level research. This more holistic systems approach takes into account trade-offs at the farm/household level. The team also sees the need to more explicitly consider the challenges of climate variability/change, malnutrition, and inter- and intra-farm/-household disparities, such as management knowledge, access to agricultural inputs, and decision-making. 

Specifically, the project aims to:

  • Build ways and foster learning on moving from AEI of individual plots toward a more comprehensive agroecological transition of whole farms.
  • Generate positive outcomes in terms of climate resilience, reduction of malnutrition, and aspects of equity outcomes.
  • Continue Large-N trials in addition to farm-level studies in the three selected villages along the climate gradient to make the project more inclusive, climate-proof AEI option combinations, and foster FRN learning and exchange, having farmer testers comment on trade-offs they are facing in their context.  

Outputs and Outcomes:


  • Characterization of farms/households along climate gradient, better understanding of trade-offs, and identification of constraints and opportunities/entry points for improved resilience through agroecological transition
  • Co-validation of co-chosen combinations of AE options based on farm/household characterization, taking into account socioeconomic and climatic contexts 
  • Inventory of knowledge on transformation of farm/household food systems around FUMA FRN
  • Data and algorithms defined for responsible AI-based farmer decisions
  • Development of communication tools to support decision-making for different types of farms and co-learning and -sharing information with farmers and other partners toward sustainable agroecological transition


  • Established farm/household typology based on farmer information from survey on households and combined with existing FUMA farmer typology
  • Better understanding of farm/household types existing at FUMA level, resulting in better support to decision-making on AEI options and, ultimately, contextualized AE transitions 
  • Knowledge on millet-based farming system management created to detect potential options and trade-offs for different types of farm households
  • FUMA’S basket of AE options continuously refined and contributions to climate risk management validated
  • Agricultural production and socioeconomic conditions of farms improved 
  • Farm resilience to climate change enhanced
  • Knowledge on traditional and new methods of agricultural product processing created and applied to diversify agricultural production and food sources
  • Diversity of processed foods on farms increased
  • Diversification of income-generating activities at farm/household achieved
  • Defined data and algorithms used to develop essential foundation for responsible AI-based farmer decision support system
  • New communication tools used to disseminate knowledge and stimulate discussion/co-learning 
  • Improved adoption rate of new technology combinations through the FRN 
  • Proof that research at farm/household level is reasonable and feasible within FRN, enabling derived learning models for other farmers, setting basis for future landscape-level studies, and opening up better ways to inform policymakers
  • Improved knowledge toward agroecological transitions at farmer/household levels through academic and non-academic training