4.1. Farmers’ preferences for IMVs attributes
This research provided an overview of farmers' preferences for attributes of IMVs, factors influencing their adoption, and maize production constraints in South-Kivu. The results of the BWS analysis showed that farmers ranked several IMVs attributes. These results are consistent with recent studies on farmers' preferences for IMVs attributes (Dao et al., 2015; Mafouasson et al., 2020; Yin et al., 2020). These studies suggested that yield is not the only criterion for farmers while selecting a variety. Other criteria included early maturity, adaptation to environmental conditions, response to fertilizers, resistance to diseases and pests, availability of improved variety seeds, price of improved variety seeds, and sensorial and nutritional qualities (Sibiya et al., 2013; Dao et al., 2015; Rattunde et al., 2021; Semahegn et al., 2021). The latent class modeling indicated a difference in farmers' preferences and needs in terms of maize variety characteristics, seed technologies, and environmental conditions. Three preference classes, which can also be considered as market segments, were identified. The preference criteria varied from one class to another. These results agreed with other similar studies that revealed differences in farmers' preferences for maize variety attributes (Mafouasson et al., 2020; Yin et al., 2020). The study by Morris and Bellon (2004) reported that crop variety preference varies among farmers and that variety attributes affect adoption of improved varieties. These criteria are closely correlated with adoption factors and can be helpful in identifying economic, technical, and institutional challenges in developing participatory approaches that integrate end-users into plant breeding programs at both local and national levels. This outcome could be related to the production objectives of maize, which vary from farmer to farmer. Some farmers are the main consumers of their production, others may opt for maize processing, and others may sell their production. All farmers are subject to the preferences of final consumers (Rattunde et al., 2021).
High preference for the potential yield was listed as the most important attribute when selecting a maize variety. This attribute would be closely related to the performance of varieties currently used by farmers in South-Kivu and suggests that farmers really want to increase their yields by adopting varieties with high yield potential. Farmers use local varieties, which are often degenerated, exchange seeds among themselves, and do not purchase seeds from accredited seed distributors. Consequently, the average maize yield (~ 800–1200 kg/ha) is very low in the study area (CAID, 2019; Mushagalusa et al., 2019; Kazige et al., 2022). Adoption of high-yielding varieties would allow farmers to increase maize production and achieve their goals (Qaim, 2020). In South-Kivu province and the DRC in general, maize is the main cereal crop. Its demand is higher and often exceeds production due to the use of degenerated varieties and other technical constraints (CAID, 2019).
Early maturity, drought tolerance, and environmental adaptation were also ranked as key attributes looked for in IMVs. The preference for early maturity is related to the climatic disturbances that currently occur in the South-Kivu province and that impact on the duration of cropping seasons, the frequency and distribution of rainfall, the cropping calendar (Balasha et al., 2021). In this region, maize cultivation is dependent on the environmental conditions, so that climate disturbances could negatively impact maize yields, especially for smallholder farmers who cannot afford control and adaptation measures. Several research studies indicated the susceptibility of maize to climate change. Predictions by Kanox et al. (2012) suggested that maize yields will decrease by 5% in Africa by 2050, mainly due to an increase in temperatures and solar radiation, which interact with precipitation to create more stressful conditions for maize plants, and therefore, lead to lower yields. In China, climate change has resulted in a decrease in maize yield of 1.7% due to increased temperatures (WU et al., 2021). The preference for early maturity and adaptation was reported to be related to the environment that farmers face and the resilience of varieties to extreme conditions and disruptions that would arise in the environment (Jin et al., 2021). This could explain the low level of adoption of IMVs in South-Kivu, even though they are supposed to perform better than local varieties. Development and dissemination of varieties adapted to prevailing climatic conditions would be important to ensure the resilience of maize to climatic disturbances and guarantee yield stability in South-Kivu.
Disease resistance and storage pests were ranked 4th and 7th most preferred traits, respectively. The preference for these attributes would be related to the occurrence of several diseases that are prevalent on maize in South-Kivu as well as to crop pests such as the fall armyworm (Spodoptera frugiperda) and other stock pests. An invasion of the S. frugiperda was recently reported in South-Kivu with an incidence of about 57% in maize fields. According to Cokola et al. (2020), the incidence level of S. frugiperda in terms of leaf damage and larval density is about 65% in South-Kivu, causing considerable yield losses because all varieties currently grown are susceptible (CAID, 2019). The lack of disease/pest resistant maize varieties is a major constraint in South-Kivu where more than half of the population depends directly or indirectly on maize for their subsistence.
Varietal response to fertilizer was reported as the sixth most important attribute in farmers' varietal choice. This trend is linked to the fact that in South-Kivu, soil fertilization is essentially based on the use of various forms of organic matter, since mineral fertilizers are not widely used in the study area (Heri-Kazi & Bielders, 2020). Only a limited number of farms use mineral fertilizers. The scarcity of organic fertilizers in this region and the high costs of mineral fertilizers limit their use by local smallholder farmers (Ndeko et al., 2019a; Mushagalusa et al., 2020). Farmers prefer to grow varieties with high nutrient-use efficiency to maximize maize productivity at low production costs. It would be very important to focus breeding programs on selecting for nutrient-use efficiency, since currently available improved varieties require more nutrients. It has been reported that the use of high nutrient-use efficient varieties can reduce mineral fertilizer inputs and ensure efficient and sustainable management of natural resources (Ali et al., 2018). In addition to the varietal selection approach, the exploitation of other strategies based on the use of beneficial microorganisms found in the rhizosphere of plants such as arbuscular mycorrhizal fungi and plant growth promoting bacteria (PGPB) could also increase the nutrient-use efficiency of crops (Adesemoye et al., 2008 ; Meena et al., 2017 ; Ndeko et al., 2019b, 2020).
The preference for high seed availability and reasonable seed price of improved varieties is related to the lack of breeding programs and functional seed delivery systems in DRC in general and in South-Kivu in particular. The market for improved variety seeds is not well coordinated and their price is very high and fluctuates widely (Seed systems Group, 2019). Jandrain (2020) reported that in DRC and South-Kivu in particular, farmers are suffering from seed insecurity because they have lost their seed autonomy and are now struggling for access to high-quality seeds. Furthermore, the Congolese seed policy would be the result of instructions from funding organizations and not the result of the needs of the population. This situation would be the main reason for the low rate of adoption of improved varieties and the persistence of local varieties that are often degenerated in the region. However, the availability and affordability of improved seeds are key factors in influencing the adoption of improved varieties (Mafouasson et al., 2020). On the other hand, the preference for seed longevity would be explained by farmers' seed management practices. Farmers save part of their harvest that will be used as seed for the next season. Farmers may exchange seeds among themselves, leading to a mixture of varieties. For an improved variety, high yields and a longer storage period are the qualities most desired by farmers (Mushagalusa et al, 2020; Manandhar et al., 2018). The study by Jin et al. (2020) showed that this seed management method reduces production costs by increasing seed availability. Actions should, therefore, be taken to improve farmers’ practices of seed storage to ensure they can effectively save quality seed from their harvests to increase availability of seed in the absence of a functional seed delivery system. Studies in Ethiopia, Cameroon (Mafouasson et al., 2020), Burkina Faso (Dao et al., 2015), Ghana (Danso-Abbeam et al., 2017) and Uganda (Mastenbroek et al., 2021) have also demonstrated the importance of maize variety attributes such as yield, disease resistance, early maturity, response to fertilizer, improved seed availability and price of improved seed, in farmers' varietal preference and choice.
Preferences for cooking quality and taste were important for farmers in classes 2 and 3. In both classes, these attributes were related to other criteria such as potential yield, early maturity, environmental adaptation, seed longevity, and pest/disease resistance. Other studies have reported the preference for these attributes and have emphasized the relevance of considering yield and crop consumption criteria in breeding programs (Teeken et al., 2018; Jin et al., 2020). In South-Kivu, the preference for cooking qualities and taste would be explained by the fact that subsistence farming is the most predominant production model. Most farmers are consumers of their own products and only a limited quantity participate to market (Ansoms et al., 2011; Cirimwami et al., 2019). Therefore, the introduction and dissemination of biofortified varieties in the province should be accompanied by the investigation of the sensory qualities. Thus, participatory selection is encouraged to ensure farmers' preferences are met, and consequently, increase the probability of IMVs adoption by farmers.
4.2. Factors influencing the adoption of improved maize varieties
Several groups of factors influence the adoption of improved varieties in rural areas, including demographic, socio-economic, and variety-specific factors. These are often then farmer age, level of education, household size, farmer experience, membership of a farmer's organization, access to credit, availability of labor, availability of improved seeds and prices of improved seeds (Danso-Abbeam et al., 2017 ; Jin et al., 2020 ; Mafouasson et al., 2020 ; Mugumaarhahama et al., 2021). In addition to these commonly studied factors, other research has reported that adoption of agricultural innovations could be explained by institutional factors such as the existence of effective extension services and regular contacts of farmers with these services (Sisay et al., 2015).
The results of this study showed a positive relationship between the age and the adoption of IMVs. This result confirmed our expected positive effect of age on the adoption of improved maize varieties in South-Kivu, but strongly disagreed with previous results by Lunduka et al. (2012) and Danso-Abbeam et al. (2017) that showed that older farmers generally prefer to grow local varieties and that age negatively influences the IMVs adoption. In contrast, Islam et al. (2012) showed that age was significantly and positively correlated with the adoption of high-yielding rice varieties. The positive relationship between age and adoption of improved maize varieties in the study area could be attributed to the fact that older farmers have some experience in maize production. This category of farmers prefer either the improved varieties or to combine the two types of varieties in order to avoid the risk of yield loss by using the local varieties. In addition, older farmers have access to production resources and information and are inclined to easily adopt the IMVs (Asante et al., 2017).
Farmers' experience in maize production showed a significant influence on the adoption of IMVs. Consistent with this study, Danso-Abbeam et al. (2017) found a significant positive relationship between farmers' experience and IMVs adoption. The positive relationship between famers’ experience and IMVs adoption would be explained by the fact that the more experienced farmers become, the more they would have acquired greater knowledge about profitable farming activities and maize production practices. Therefore, they easily understand the challenges related specifically to the use of local varieties as opposed to improved varieties (Rattunde et al., 2021; Mastenbroek et al., 2020). This category of farmers are more interested in boosting their yields through the use of improved varieties than less experienced farmers (Byerlee et al., 2008; Abebe et al., 2013). It also explains the negative relationship between farmers' experience and the adoption of the combination of local and improved varieties, because farmers tend to discard local varieties for improved ones.
Institutional factors such as membership in the farmers' organization and contact with extension services were significantly and positively correlated with the probability of adopting IMVs. The first reason may be that there are several agricultural development projects in the region collaborating with FBOs to disseminate new technologies. The members of the FBOs receive training and seminars on best agricultural practices and new technologies. Secondly, FBOs also increase farmers' contact with extension services and consequently increase farmers' exposure to agricultural innovations and their uptake of improved varieties (Danso-Abbeam et al., 2017; Mugumaarhahama et al., 2021). Rattunde et al. (2021) reported that membership in farmers’ organizations and contacts with extension services are important sources of useful information that can stimulate adoption of IMVs.
We also observed a positive and significant relationship between land ownership and adoption of IMVs. This result is consistent with findings by a study by Ogada et al. (2014) in Kenya that showed that land ownership security is a very important parameter in the adoption of IMVs, which can increase the probability of adopting the varieties by more than 4% compared to households without land ownership security. The relationship between land ownership and adoption could be explained by the land conflicts that prevail in South-Kivu and the limited availability of farmlands (Mushagalusa et al., 2019; Angélique et al., 2022). Because of this situation, sharecropping is the most common mode of land acquisition in the study area. In addition, security of land tenure could allow households to rent part of their land in exchange for some money to purchase improved technologies. Land ownership also stimulate the producer to invest in more demanding or long-term agricultural technologies than if not owned. However, these results disagreed with Mansaray et al. (2019) and Mugumaarhahama et al. (2021) who showed that farmers who cultivated their own land had a higher tendency of remaining non-adopters of improved varieties, but those who rented farmland or were engaged in sharecropping had higher adoption rates of improved rice and potato varieties to make the investment profitable. Access to agricultural credit positively and significantly influenced the adoption of IMVs. This result aligned with the study by Danso-Abbeam et al (2017) who found that farmers with access to agricultural credit were 7.7 times more likely to adopt IMVs either alone or in combination with local varieties compared to those without access to agricultural credit. Access to agricultural credit allows farmers to cover additional production costs and to invest more in research and purchase of improved maize seed (Lunduka et al., 2012).
In this research, we showed evidence that access and availability of improved seeds influence the adoption of IMVs among farmers in South-Kivu. This result could be linked to low availability of improved seeds, high prices of improved seeds and the absence of a functional seed delivery system (Semahegn et al., 2021; Rattunde et al., 2021). Therefore, the majority of farmers recycle their own seed. This result is in accordance with Ogada et al (2014) and Lunduka et al (2012) who showed that seed availability is an important factor in the adoption of improved varieties. Based on Lunduka et al. (2012), the adoption rate was higher among farmers who received the maize seed vouchers, which give a short-term effect. Labor availability positively and significantly affected the adoption of IMVs alone or in combination with local varieties. It is closely related to household size. According to Abebe et al. (2013), large family size means high labor force, which could positively influence the adoption decision of agricultural innovations.
4.3. Farmers’ perceptions of the maize production constraints
In South-Kivu province, farmers face several production constraints that affect maize production. Results of this study showed that crop diseases and pests, lack of cultivable lands, lack of labor, lack of seeds of improved varieties, lack of farm inputs, and low soil fertility are the most recognized constraints by farmers. However, among all these constraints, crop diseases, lack of land and lack of labor are the three most important constraints for small-scale maize producers in South-Kivu. Previous studies have mentioned crop diseases and pests as one of the most production-limiting constraints (Mafouasson et al., 2020, Semahegn et al., 2021). Disease and pest pressure on maize, especially the invasion by the S. frugiperda, would explain the particular importance that farmers accorded to this constraint in the study area. The lack of land is also mentioned as a major constraint because land is the most important factor in agricultural production but which is increasingly scarce because of the high population density that characterizes the eastern DRC. Walungu and Kabare territories covered by this study are the highest densely populated areas of DRC (~ 300 habitants per km2) (Mugumaarhahama et al., 2021), with farm size being on average less than 0.5 ha per household. Low land ownership could impact both the agricultural production of households and their food security status (Ogada et al., 2014). The other constraint reported by farmers is the availability of labor, especially in Ruzizi plain where youths are not attracted by agriculture, considering it as less productive and burdensome (Furaha et al., 2016; Mondo et al., 2020). Consequently, farmers are importing labor from neighboring countries Rwanda and Burundi to ease the labor shortage in the area (Mondo et al., 2020). The lack of farm labor for the most important work may limit the technical efficiency of the farm, and therefore, decrease crop yields or increase significantly the production cost as in the Ruzizi plain (Dahlin & Rusinamhodzi, 2019; Mondo et al., 2020).
4.4. Implications for maize breeding programs and seed policies
In this study, we found poor adoption of IMVs compared to local varieties, which are still widely used among farmers. Despite the dissemination efforts of IMVs during the last five years, farmers prefer to use local varieties that are lower yielding and degenerated compared to improved varieties. This would be related to several institutional factors listed in the previous sections that constitute barriers to the adoption of IMVs in South-Kivu. Due to the absence of a reliable seed delivery system and absence of effective seed policies, farmers consider local varieties to be easily available and well adapted to their environmental conditions (Jandrain, 2020). In addition, breeding programs do not consider farmers' local/regional preferences, which would be the main determinants of the adoption of improved maize varieties in South-Kivu, as Mondo et al. (2019) noted for cassava varieties in the same study area.
According to the findings of this study, the low adoption rate of improved maize varieties in the study area is explained by a lack of collaboration between farmers, extension services, and variety improvement programs. Table 8 shows the different implications of the study and the potential interventions to be considered according to the model of Kuivanen et al. (2016), in order to improve the level of adoption of IMVs in South-Kivu. Results showed that farmers are classified into three classes (market segments) according to their preferences. This implies that interventions planned for new variety development should be adapted to this classification. Therefore, a particular effort should be made to develop a seed policy, a functional seed delivery system and an effective extension service by decision-makers at all levels. This will boost the dissemination and use of improved seeds in the study area. At the same time, special emphasis should be placed on research in order to develop varieties that meet farmers' expectations.
Table 8
Main implications of preference criteria analyses of maize variety attributes and production constraints for appropriate varietal improvement strategies and interventions in South-Kivu and DRC
Preferred traits of IMVs | Implications in targeting breeding programs and policies | Market segment |
C1 | C2 | C3 |
High yield potential | Develop high-yielding maize varieties, specific to different environments, adopt participatory varietal selection, consider conducting trials with farmers for quick IMVs uptake | x | x | x |
Taste, high flour density | Targeting improvement objectives for the sensory characteristics of the final products with regard to the grain quality | x | x | x |
Resistance to storage pests | Improving varietal resistance to storage pests | | | x |
Seed price | Subsidize farm inputs and ensure the availability of high-quality seeds to farmers, support research programs | | | x |
Early maturity | Develop new short-cycle varieties which adapt to changing environmental conditions | x | x | x |
Drought tolerance | Develop drought tolerant varieties that adapt to changing environmental conditions | x | | x |
Pest/disease resistance | Develop new varieties with multiple resistance to major maize diseases and pests, emphasizing the fall armyworm | x | | x |
Seed longevity | Improvement of seed storage aptitudes, establishment of conservation infrastructures for plant genetic resources in the short- and long-time frames in an effort to better preserve agricultural biodiversity | | | x |
Fertilizer responsiveness / Environment adaptation | Develop new varieties tolerant and adapted to changing environmental conditions | x | | x |
Seed availability | Ensure the seed's availability to farmers by subsidizing them and developing an effective pro-poor seed delivery system across the territory | | | x |
Production constraints and adoption factors of IMVs |
Lack of farmland | Increasing the access to land, promoting agricultural intensification, promoting high yielding varieties for food crops, and reducing inequalities in land ownership | x | x | x |
Low availability of labor | Develop family labor experience and enhance indigenous resources or mechanize some labor-intensive operations. | x | x | x |
Crop disease | Develop varieties combining high yield potential and resistance to major diseases | x | x | x |
Unavailability of IMVs seed | Consolidate and transform the traditional seed system, improve collaboration between farmers and breeding programs, improve the public extension service for effective dissemination of agricultural innovations, especially IMVs registered in the national catalog, create sale ponys in different locations, support acquisition of new varieties’ seeds | x | x | x |
Unavailability of farm inputs | Promote the use of local resources in the control of crop diseases and pests, soil fertility, etc. | x | x | x |
Lack of market demand | Encourage the value addition of harvests by developing processing facilities in the production areas | x | x | x |
Low soil fertility | Promote the use of leguminous crops for biological nitrogen fixation, the use of green manure, burying crop residues and promoting the integration of agriculture and livestock | x | x | x |
Post-harvest losses | Improve post-harvest storage to reduce yield losses and maximize farmers' profits. | x | x | x |
Access to credit | Facilitating access to agricultural credits, creating funds to subsidize agricultural inputs through refundable or non-refundable loans | x | x | x |
Age | Encourage more youths in agriculture and train them in agricultural activities through practical sessions and accompaniment, developing income-generating activities and processing of agricultural products | x | x | x |
Membership of FBO | Encouraging farmers to join farmers' organizations, strengthening the capacity of farmers' organizations through regular training | x | x | x |
Extension service | Restructure extension services, tighten collaboration with farmers and crop improvement programs | x | x | x |
It was reported that farmers should have access to information about new technologies before they adopt them (Lunduka et al., 2012). This increases the level of exposure and awareness of farmers to agricultural innovations. Breeding programs should, therefore, adopt the participatory plant breeding approaches. As defined by Morris & Bellon (2004), this approach has several advantages over traditional methods because it is well adapted to developing varieties that are more adaptable not only to physical environment but also to human preferences and needs, increases breeding efficiency and empowers rural communities. However, the application of this method requires overcoming several technical, economic and institutional challenges in order to achieve the best results.
Improved high-yielding maize varieties that meet farmers' needs and expectations should, therefore, be developed for South-Kivu. The use of these varieties by farmers will increase maize production in the area. Breeders must work with extension specialists to ensure that developed varieties are grown with the appropriate technical packages for successful results and a satisfactory adoption rate. This research will provide guidance for future breeding programs in order to increase the use of IMVs and hence, maize productivity in eastern DRC.
Table 8 Here