Drivers of use and adoption of bundled climate-smart agriculture and climate information services in Northern Ghana

In the past decade, despite advances in breeding and management practices with the potential to improve productivity, the agricultural growth rate in Ghana has faltered (Asravor et al., 2024; Raheem et al., 2021; Bjornlund et al., 2020). Consequently, food and nutrition security problems are persistent, affecting well-being, and this is largely attributed to variation and changes in the climate (Fusco et al., 2020; Teye and Quarshie, 2021). Several promising climate-smart agricultural innovations that support agriculture by boosting productivity, adapting to and mitigating climate change have been scaled out to address the increasing threat posed by climate change and meet sustainable development goals by 2030 (Lipper and Zilberman, 2018). Achieving this goal requires large-scale investment in climate-smart agriculture (CSA) (Davila et al., 2024; Williams et al., 2015). Adoption of climate smart agricultural through a top-down and bottom-up strategies which offers a platform to meaningfully contribute to national policies within the climate change and CSA space (Sam et al., 2021).

Strong government–private sector cooperation is a promising strategy to provide commercial opportunities for private entities to advance CSA practices (Samal and Jena, 2025; Akomea-Frimpong et al., 2024; Casey et al., 2021). In response to threats posed by climate change on the livelihoods of farming population, policymakers have recognized the development of appropriate policy instruments as critical components to boost adaptive capacities (Djido et al., 2021a, 2021b). These steps and initiatives have added to the discussions on adopting the best combination of CSA and climate information service (CIS) technologies with a translated effect on farm productivity. Despite all these attempts to promote CSA practices, adoption rates are low among farmers in Ghana (Akrofi-Atitianti et al., 2018; Issahaku and Abdulai, 2020). In addition, the private sector’s participation and Ghana’s agricultural development in general have been hampered, and the adoption and promotion of CSA practices have also been less impacted (Totin et al., 2018). Several factors affect the adoption of CSA, including subsidies, free inputs and information services, as well as expected benefits (Weniga Anuga et al., 2019). Prioritization of these practices is based on the level of importance the farmer assigns to each climate smart technology. Technology prioritization involves the identification of technologies which are assessed based on factors such as the potential to enhance productivity, build resilience, greenhouse gas emission reduction, as well as its suitability to various socio-economic improvement and suitability to specific agro-ecological zones (Khatri-Chhetri et al., 2017). The outcome of the impact of climate change and the wrongful application of CIS–CSA technologies are the risk associated with these technologies. These risks come alongside low productivity and the associated high cost of production even in the preceding season. Komarek et al. (2020) reviewed literature on types of risk in agriculture and possible solutions to address the various types of risk. They identified five types of risk comprising production risk, market risk, institutional risk, personal risk and financial risk, and findings revealed that about 66% of the 3,283 studies focused on production risk which largely emanates from climate change effects. This further confirms the role of adoption of appropriate combination of CIS–CSA which addresses the multiple types of risks faced by farmers. A shift in focus toward analyzing the multiple contemporaneous types of risk with associated CIS–CSA technologies and approaches may provide a basis that gives farmers greater opportunities and options for coping with climate change impacts and risk management. To enhance the adoption of preferred climate-smart agricultural farm management practices among Ghanaian farmers, the CGIAR Research Program on Climate Change, Agriculture and Food Security (CCAFS) earlier promoted the combined dissemination of CIS and weather that are demand-driven, cost effective and accessible with ease of implementation. This study, therefore, seeks to assess the extent of adoption and use of bundled CSA and CIS and its effect on farm yield in Northern Ghana. Partey et al. (2018) earlier assessed and focused on some of the most preferred climate-smart agricultural options used by smallholder farmers in the Upper West Region of Ghana, specifically on erosion control, integrated pest management, pest-resistant crops, water management and multiple cropping practices combined with weather and CIS (weather forecasts, call center, agro-advisories, input and output market prices). They concluded that the adoption of these CSA practices has proven to be effective in enhancing farm productivity, incomes, food security and conserving natural resources.

As a follow up to the CCAFS program and expanding the scope to Northern Ghana, the current paper assessed the level of adoption and use of bundled CIS–CSA and the translated effect on farm yields among farmers.

A multi-stage sampling approach was used for the data collection. In the first stage, we used purposive sampling to select the districts in terms of Tolon, Kasena-Nankana, Bongo, Lawra and Jirapa, which happen to be hotspots of climate change impacts and vulnerability in Ghana (Desjonqueres et al., 2024). In the second stage, we purposely selected eight communities based on past climate change intervention implemented in Dazuuri, Dorggoh, Bompari, Tampala, Yidongo, Yizugu, Nyanpkala and Woribogu. This is shown in Figure 1. These past climate change interventions were implemented under the CCAFS program implemented in the selected regions, districts and communities. These communities were designated as climate smart villages based on participatory approaches, institutional and technological options toward mitigating the effects and impacts of climate change in agriculture (Aggarwal et al., 2018). In the last stage, we randomly selected 30 farmers from each of the selected communities. Data comprised socio-economic factors such as sex, age, marital status of the respondent and household size. Access to credit, education CI, extension service, farmer-based organization (FBO) membership and perception of CI were all captured because they are highly likely to affect the adoption and prioritization of CSA practices. Data was also collected on the various climate smart agricultural technologies and CIS used by farmers. This was characterized by the simultaneous application of more than one CIS–CSA technologies toward productivity, adaptation and mitigation. Prioritization of CSA was considered as an outcome variable and measured based on gender and youth responsiveness, one-health consideration and environmental friendliness. The farmer’s adoption and prioritization of these practices are dependent on several factors, and the ultimate is the perceived benefit that will be gained in the end. This indicates that this research is based on the utility maximization theory. The study outlined 14 CSA practices that were practiced in Northern Ghana. CSA practices under study included: cowpea/maize intercropping; contour stone bunds/contour tillage with tied ridges (minimal/zero rate of water run-off and soil erosion); minimum tillage for cowpea, maize and vegetables production, dual purpose cowpea promotion (fodder and grain); promotion of pest- and disease-resistant maize and cowpea varieties (early maturing, drought, low N and striga); seedbed options-ridging as an alternative for mounding for yam cultivation; promotion of seed yam using mini-sett, aeroponic and hydroponic techniques; staking option trellis/minimum staking for deforestation reduction; promotion of stress-tolerant varieties; enhancing biopesticide usage in cowpea and maize systems (neem, wood ash etc.); using organic amendment for improving soil health; planting leguminous crop as previous crop (farmyard manure, bagged compost); and enhancing access to CI (seasonal calendar). Each of these CSA practices has a perceived utility or gain to the farmer.

The theoretical underpinning of this study is the random utility maximization, where a farmer’s choice of CIS–CSA technology or a combination of technologies is based on the satisfaction derived from usage. Farmers’ adoption and priority of these technologies are influenced by numerous factors, the most important of which is the perceived advantage that will be received in the end.

The utility function measures the anticipated benefit or gain (U) that a farmer will receive from CIS–CSA or a combination of practices (j) provided by:

The farmer selects the practice (j) which gives the greatest benefit. Creating the behavioral model:

The decision to select alternative i over j by a farmer indicates that the farmer maximizes his/her utility by selecting option i.

Adopting different CIS–CSA practices can also provide greater benefits to the farmer because they all have associated benefits, while others are complementary in practice.

A farmer who adopts a CSA practice or a combination of the practices is expected to obtain a greater satisfaction or utility than a respondent who adopts none or one. As a result, each extra CSA practice adopted provides an extra utility. Farmers who adopt more will have more utility than farmers who adopt less.

Adopting a technique or practice is not a self-contained decision because a farmer may choose from a variety of techniques, the use of one of which may lead to the use of another. This demonstrates the possibility of a link because of the interdependence of the dependent variables. To find the determinants for binary outcome estimations, the Probit model might have been estimated seven times. Because of the likelihood of correlation, the individual Probit model for each practice is prone to bias and inconsistent results. In providing unbiased and consistent determinants of adopting CSA practices, a model that allows for correlation between dependent binary outcomes will be necessary, leading to the selection of the multivariate probit (MVP) model.

The MVP model was used, and the general formulation by Velandia et al. (2009) is presented as follows:

where χ_jm represents the explanatory variables or predictors of every adopted CSA practice; β_m represents the vector of unknown parameters; and ∈_jm m = 1,2…7 represents the disturbance term or error term constant variance (1) and a zero mean.

ϒ_jm represents the CSA practices the farmer selected. χ_jm also represents the farmer’s characteristic vector, which is regressed against the vector of unknown parameters, resulting in the adoption choice.

To experimentally estimate equation (3), many binary probit models for each CSA practice adoption could be used. The maximum likelihood Approach is used to estimate the MVP model. For maximum likelihood estimation, the Geweke-Hajivassilioukeane simulation process is used to compute the likelihood function probabilities and their derivatives. This yields approximations to the m fold multivariate integrals:

where ρ(⋅) is the multivariate normal density. The log probabilities of the observed outcomes are subsequently used to estimate the logarithmic-likelihood model, which has been defined as:

where Z is defined as Z = Smℵ, R is matrix of the correlation, MNV represents the normality density of the Multivariate and Ҭ represents the matrix $t$_mm = 2 y_m − 1.

The Multinomial Logistic model was used to investigate the factors impacting the CIS–CSA technology prioritizing because five of the seven technologies did not follow any natural sequence. Because the dependent variables in this research were mutually exclusive, the MVP model failed in the prioritization of CIS–CSA technologies because of the mutually exclusive nature of the technologies. Farmers selected the CIS–CSA technology options frequently used on their farms.

The multinomial logit was run using the five CIS–CSA technologies, and no practices prioritized at all, yielding six outcomes. Each had its structural function, which included an error term j = 1,2,…,6; the coefficient (b¹, b², b³, b⁴, b⁵, b⁶) was calculated for each outcome; and the anticipated probabilities are given in the resulting equations (Svensson and Wahlström, 2023; Etwire, 2020; Sadiq et al., 2019):

Because there are several solutions for the coefficients (b¹, b², b³, b⁴, b⁵, b⁶) in equation (6), each of the practices will have the same probabilities. Following that, the model was identified by designating the sixth group (with no CSA practice prioritized) as the reference group. The model now looks like this:

After making the sixth group (with no CSA prioritization) ϒ_i = 6 the base outcome, the Relative Risk Ratio is estimated. The Relative Risk Ratio for ϒ_i = 1 is expressed as:

The relative probability for ϒ_i = n is given as:

b(n), (b₁, b₂,…,b_n) being the vector that goes with the explanatory variable (x₁, x₂,…,x_n) when χ_i changes, the impact of prioritizing CSA can be derived as:

Results showed that, the average age of the farmers was 45 years as shown in Table 1, which is younger than the national average of 55 years for Ghanaian (MoFA, 2020). Males also dominated among farmers and confirms males’ participation level in agriculture in Northern Ghana. Years of formal education was averagely three and reveals farmers’ limitation of basic education. In terms of accessing extension services, about 30% of the farmers did not receive any form of extension delivery services. In all, 45% were not members of FBOs and about 89% received CI for the past four years, of which half expressed willingness to pay for CI. About 50% cultivated on their own land, while the other half faced land tenure constraints to use CSA practices with 49% have access to credit for production.

Farmers’ extent of CIS–CSA technology adoption and prioritization are presented in Figure 2. The results showed that leguminous crop rotation was adopted by 39.81% of the farmers, making it the most adopted CSA technology. This is not surprising given that farmers prefer solutions based on local ecological knowledge, as reported in a similar study by Mensah et al. (2021) in Ghana. This practice has traditionally been associated with farmers’ practices, as it is an easier way to fix nitrogen to the soils added to the dual-purpose nature of most leguminous crops. Previous studies such as Abdul-Rahaman et al. (2021) and Fosu-Mensah and Mensah (2016) have confirmed the role of leguminous cropping as a CSA adaptation measure to boosting yields and farmers income in Ghana and Northern Ghana specifically. However, leguminous crop rotation and organic amendment of the soil received the same level of prioritization by about 13% (n = 28) of the total respondents. This can be attributed to the ability of leguminous crop residue with its ability to fixing nitrogen to the soil as an alternative to organic amendment of soils in the study area. The existence of related practices like intercropping as revealed by Ouédraogo et al. (2018) and Peterson (2014) further influenced the prioritization of the use of leguminous crop as a previous crop to a main crop. CIS was the second most adopted practice by about 38% (n = 83) farmers and was the most prioritized practice among the farmers, representing 55 (25.46%) of the farmers. CI is a basic requirement to making informed decision on the choice and bundling of CSAs among farmers and hence given much attention.

Drought- and stress-tolerant varieties were the third (24.54%) adopted but were the least prioritized CSA at 4.63% and indicates that, among farmers, this technology is successfully used when the right CI is provided. This is confirmed by Antwi-Agyei et al. (2021) that access to appropriate CIS influences and addresses the importance of drought- and stress-tolerant varieties in North-Eastern Ghana.

Organic amendment of farm soil was adopted by approximately 24% of the farmers and was the second most prioritized technology with 13% of the farmers. Gamage et al. (2023), Omari et al. (2018) and Martey (2018) revealed that organic fertilizer is the least expensive and sustainable, the most efficient approach to enhancing soil fertility and soil health among smallholder farmers. Findings revealed that the respondents did not massively prioritize conservation tillage (minimum tillage) although about 22.22% of the farmers adopted the same technology. This implies that, most of the farmers still consider tractor ploughing as a major form of farming in most parts of Northern Ghana. Although an old practice among farmers in the Northern, the practice of using tractor ploughing is cheaper compared to use ridgers to minimally till the land and also difficult to access ridgers. Pest- and disease-tolerant varieties was adopted by only 37 (17.13%) of the farmers and was the fourth prioritized practice recording about 10.19% of the farmers. Improved seed varieties are common and available to farmers in Northern Ghana, although cost per kil of the seed is generally beyond the income level of farmers. Farmers would, therefore, prioritize other practices more than other practices, as seed varieties can be sourced from neighboring farmers, open market, family and friends and agro-input dealers. This is confirmed by Natogmah et al. (2023), who found that majority of farmers obtain seed from the informal seed system in Northern Ghana, and this largely attributed to relatively high cost of certified seed which is above the income level of smallholder farmers, especially in Northern Ghana. Finally, water management, specifically mulching, was the least adopted CSA practice with 9.26% farmers, with no farmer giving priority to this technology. This is confirmed by the view that most of the CSA practices almost address the role of mulching on a farm in terms of organic amendment which improves the soil health with less evapotranspiration. Of the seven CIS–CSA practices, only five were prioritized, implying that some of the practices are adopted based on some external conditions but not wholly the derived utility. This is shown in Figure 2.

The MVP model revealed a differential adoption of CSA practices, and this difference is driven by the age of the farmer. The outcome in Table 2 indicates that older farmers are less likely to adopt organic amendment of soils, and this is attributed to the labor-intensive nature of the practice. Similarly, older farmers were more inclined to adopt pest- and disease-tolerant varieties, as this practice requires less labor days and less cost compared to other pest and disease control measures. This age differential findings are consistent with the findings of Akrofi-Atitianti et al. (2018) and Djido et al. (2021a, 2021b). Numerous studies have shown that education is a positive and important determinant of innovation adoption and prioritization (Dagunga et al., 2020; Bilaliib Udimal et al., 2017), as the findings also indicated that as the years of a farmer attaining formal education increase, so does their probability to adopt CI, drought- and stress-tolerant varieties, pest- and disease-tolerant crop varieties and mulching. The study revealed that increasing years of attaining formal education reduces the possibility of a farmer adopting organic soil improvement, and this is attributed to the role of agriculture and farming practices among persons of higher education as alternative livelihood and food security measure. This allows for less time allotment to agriculture, and hence, people of higher education in the study area spend less time in practicing composting as organic amendment of the soils.

Findings further revealed that, female farmers were more likely to use CI, minimum tillage and water management, as this offered lower cost compared to the other CSA practices, although water management in the form of stone bundling required more labor. Specific to women, increased tendency of using CI in the study area was also attributed to the planning required to address time constraints, as women farmers have a responsibility to attend to farm activities, attend social occasions and undertake off-farm livelihood activities. These non-farm activities require critical decision-making after receiving information on the weather and climate. This is confirmed by similar findings by Baffour-Ata et al. (2022), who found that daily and weekly forecast of CI provides useful information to assist in critical decisions including drying of crops.

To boost soil fertility, male farmers used pest- and disease-tolerant varieties, and they also planted leguminous crop as a preceding crop to a primary crop. Antwi-Agyei and Amanor (2023) and Takahashi et al. (2020) reiterated the effort of research and development of improved varieties suitable for the various agroecological zones of Ghana, and diffusion of these varieties is critical to boosting crop yield and income of farmers. This practice mainly focused on cowpea because of its nitrogen fixing ability, dual purpose nature for livestock fodder, beans and leaves as food for the households. In rural settings in Northern Ghana, specific female farmers have little control over land and capital. The practice, which requires a large amount of capital and land area, can only be adopted by insignificant female farmers, thereby warranting the outcome. This can also be reflected in the studies Mensah et al. (2021) and Sam et al. (2020).

Access to CI is a complementary service to any of the CSA practices and technologies used by farmers, as farmers required CI at various scales and sources for planning and decision-making. Farmers who were unwilling to pay for CI were also more inclined to use pest- and disease-resistant varieties as a CSA adaptation strategy. This is because most farmers rely on recalls on past weather events and other sources of CI such as on radio and television, family and friends, free SMS alerts and indigenous knowledge from traditional rulers in the farming communities.

Drought- and stress-tolerant varieties were also adopted by farmers cultivating on leased or borrowed lands which is a common land tenure arrangement in Ghana. Small-scale farmers were the worst affected by the negative effects of the current land tenure system, where land owners can request for the land for purposes other than farming. The land tenure systems do not allow for continuous practice of appropriate CSA technologies, given that these lands could be taken away at any time. As a result, farmers in the study area are hesitant to invest in enhancing their current fields, which they may lose at any time.

Farmers that have access to extension services are more likely to use CI, as well as pest- and disease-resistant varieties, at a significant level of 1% and 10%, respectively. Asante et al. (2024) and Abegunde et al. (2019) also reported that, extension services positively influence CSA technology adoption but Ayisi Nyarko and Kozári (2021) reported the minimal contact hours between agricultural extension officers and farmers and, therefore, recommended the training of local extension agents from among the farmers and to equip them with the ability to interpret CI to other farmers in the communities. Farmers who could not access the extension services were unlikely to adopt organically improved soil based on indigenous ecological knowledge, like neem extract, for pest and disease control to limit the effects of climate change.

Findings also demonstrated that membership in a FBO influences the adoption of CSA technologies, as examined by Diallo et al. (2019) and Israel et al. (2020). Members of FBOs are more inclined to use pest- and disease-resistant crops and water management techniques, and this is attributed to the numerous capacity building trainings received by FBOs in the study area. FBOs serves as technology scaling points including easy access to inputs (Zakaria et al., 2020). Farmers who are not members of FBOs had higher adoption rates of CI, as these farmers rely on radio and television and friends and relatives for CI services. The years a farmer received CI was found to positively influence farmer’s decision to adopt CIS–CSA technology. As farmers continue to receive information on climate, adoption rates of CI and drought- and stress-tolerant varieties significantly increases by 5% and 10%.

Results of the adoption of bundled CIS–CSA technologies among farmers in the study area showed that the correlation coefficients were both positive (complementary) and negative (substitutes) among the paired CIS–CSA technologies (Table 3). Given the demographic and institutional characteristics of the respondents, it was found that drought- and stress-tolerant varieties; organic amendment of soils; mulching and minimum tillage; and pest- and disease-tolerant varieties were complementary practices. These technologies serve as soil water management strategies which is relevant in water-stress communities, especially in Northern Ghana. As an adaptation strategy, farmers use complementary technologies such as cover cropping, ridging to conserve water between ridges, mulching and early maturing crop varieties to meet the crop water requirement early enough to support crop growth. A similar study by Kpadonou et al. (2017) and Birhanu et al. (2019) emphasized on soil moisture, water stress and soil infertility as important factors hindering higher agricultural productivity in semi-arid and drylands, hence the need for current soil and water conservation practices in water-constrained regions.

Pest- and disease-tolerant varieties; organic amendments of soils; water management and organic amendments of soils; access to CI and pest- and disease-tolerant varieties; water management and access to CI; minimum tillage; drought- and stress-tolerant varieties/crops; and leguminous crop as previous crop to a main crop all had negative correlation coefficients indicating that, each CSA technology or CIS substitutivity is attributed to farmers’ resource-constraints, less education on CSA technologies and largely engaged in labor-intensive production systems.

The likelihood ratio test further reported in Table 3 shows a chi² (21) of 73.825 and is highly significant at 1%, suggesting that the independence of the error components in individual equations is empirically rejected. This suggests that the residual of a farmer implementing multiple CIS–CSA technologies is possibly dependent on other suitable CSA and CIS technologies. Also, the error term of a farmer implementing multiple CSA technologies is not independent of the others and confirms the role of bundling as an adaptation measure to address climate change impacts. The presence of a correlation in the residuals justifies using this MVP model for analyzing the data rather than estimating a binary model for each outcome. The models statistics of wild chi² = 282.672 with a statistical significance of 1% further indicates that the explanatory variables highly explain farmers’ CIS–CSA technology adoption decision.

The marginal effects of the Multinomial Logit estimation as shown in Table 4 presents the extent of the effect of the variables influencing CIS–CSA technology prioritization among farmers in the study area. Results revealed that farmers’ years of schooling had a good and significant influence on prioritizing CI access, organic soil amendment and pest- and disease-tolerant varieties. This implies that, farmers who attain more years of formal education will have about 8.48%, 10% and 1.51% likelihood of prioritizing CI access, organic amendment of soil and pest- and disease-tolerant varieties, respectively. Curtis (2022) and Sam et al. (2019) both confirmed that education has a positive relationship with technology adoption among farmers. Farmers are learning more about these CIS–CSA technologies and their contributions to production, adaptation and mitigation, and it has been found that increasing education and knowledge capacitation among farmers takes them away from the field, giving them less time to operate, in addition to prioritizing particular strategies. Ironically, an increase in the age of a farmer led to a 35% likelihood of not giving attention to stress-tolerant varieties. This is attributed to the fact that experienced farmers’ adoption of climate in formation (CI), organic soil amendments and pest- or disease-tolerant varieties is enough to address stress tolerance.

Being a female farmer increased the tendency of accessing CI prioritization by about 2.1%, and this indicates that male farmers were less likely to give more attention to accessing CI by about 97.9%. Female farmers spend relatively less time on the farm because of additional household responsibilities, and this got them closer to a CI medium (radio, television, time with groups) that enhanced their knowledge of CI for effective planning. Observations from the study showed that produce drying, seed planting and sowing are largely female tasks, and CI is an essential aspect of carrying out this work (Gyan et al., 2020; Shee et al., 2019; Sugri et al., 2021). The accessibility and readily available nature of CI have been proven to be crucial for the progress of climate-smart technologies (Tahidu, 2024; Khatri-Chhetri et al., 2017) and, hence, high prioritization of CI.

Farmers who cultivate on their own farmland are about 13.8% more likely than farmers who produce on borrowed or rented land to prioritize access to CI. Farmers who do not own land are also substantially likely by about 8.94% of prioritizing organically improving soil health compared to farmers who own lands, and this is because of the limited land for cultivation and, hence, low soil fertility regenerative strategies which requires permanent lands and time to yield results. Studies have also proven that improving soil health organically is the best sustainable and cheapest practice (Hammad et al., 2020; Haque et al., 2021).

The farmers who accessed agricultural extension services would likely prioritize leguminous crop rotation compared to their counterparts by about 12.45%. The access had a 12.45% and 6.99% negative influence on pest- and disease-tolerant varieties and drought- and stress-tolerant varieties, respectively, at a significant level of 5%, but the positive a priori expectation was not met. But according to Donkoh et al. (2019), some extension strategies discourage sharing experience, which may be the cause of the negative effect. Farmers who considered CI to be beneficial were 13.35% more inclined to prioritize access to CI than those who did not. The farmers who claimed that CI is helpful may have had success by using the information at a point or still using. CI on temperature, intensity of rainfall and pattern helps farmers to manage soil nutrients loss, moisture, temperature and time of planting and, hence, an increase productivity (Igberi et al., 2022; Mensah et al., 2021). Results further showed that, the number of people in a household is positively related to the likelihood of prioritizing organic amendment of soils, as household members would usually participate in farm operations. Compared to the application of fertilizer, organic amendment of soils requires a lot of labor. The household size also showed an inverse relationship with the likelihood of the farmer prioritizing pest- and disease-tolerant varieties. Farmers in most developing nations prefer methods that are inexpensive to execute (Williams et al., 2019). The cost of household feeding itself will be high, and farmers savings are not adequate to procure improved or certified varieties for planting. This explains why pest- and disease-tolerant cultivars had a 0.008% lower chance of being prioritized.

Farmers’ access to credit also influences prioritization pattern of CIS–CSA technologies, but pest- and disease-tolerant varieties will be prioritized compared to credit access.

Findings suggest that farmers annually increase access to CI influenced choice of CSA technology to prioritize. It demonstrates that if a farmer receives CI for an additional year, then the probability of prioritizing climate-information improvement increases by 1.63%. Cultivating leguminous crop after cereals will increase the prioritization probability by 1.42%. Probability for prioritizing the promotion of stress-tolerant varieties will increase by 0.72%.

The key CIS–CSA focus of the farmers is improving access to CI, leguminous crops as previous crops, an organic amendment to improve soil health, promotion of disease and pest-tolerant maize and cowpea varieties and promotion of stress-tolerant improved maize and cowpea, which were both adopted and prioritized. Resource limitations and externalities lead to water management and minimum tillage not being a farmer’s priority. Education and training (extension services) are the main drivers for effectively using CSA practices that help stakeholders achieve their CSA goals. These two key capacity building approaches are enablers to an efficient utilization and, hence, adoption of the prioritized CSAs among farmers. It is also evident that farmers are concern about the cost effectiveness of technologies while mindful of the yields, mitigation and adaptation abilities of these technologies.

It was unveiled that promoting complementary CIS–CSA is ideal for increasing technology adoption among farmers, as combined technologies jointly yield great impact such as organic amendment to improve soil health and promote stress-tolerant improved maize and cowpea varieties. It is also clear that nature-based solutions area preferred for sustainability and soil health improvement. It is further evident that the role of key stakeholders is required to provide CIS–CSA technology products which addresses the needs of farmers to enhance adoption and farmers should also make conscious effort to take advantage of existing government program to address climate change effects.

This study was conducted with data culled from the Accelerating Impacts of CGIAR climate Research for Africa (AICCRA) project under the Ghana cluster. This work was also supported by the CGIAR Initiative on Mixed Farming Systems which is now under the CGIAR Sustainable Farming Program.