Perceptions and adaptation strategies of smallholder farmers to climate change in Builsa South district of Ghana

Abstract

Climate change (CC) poses a significant threat to small-scale farmers in low-income countries, increasing vulnerability to food insecurity and requiring various methods to mitigate its impact. The current study assessed producers’ perceptions of CC and the adaption measures they adopt to mitigate the effect of CC in Builsa South district of Ghana. A generalized Poisson regression model was used to evaluate the factors affecting adoption of climate change adaptation strategies (CCASs) by the respondents. Farmers’ knowledge of the factors contributing to CC was analyzed by employing a 5-point Likert scale while producers’ perception of the effect of CC on maize cultivation was assessed using Kendell’s coefficient of concordance. The findings indicated that deforestation, bush burning, improper disposal of waste and greenhouse gases are the main activities contributing to CC in the district. The adaptation strategies used by farmers include early planting, adoption of disease resistant and drought-tolerant varieties, crop rotation, mixed cropping, and zero tillage. The study further revealed that years of education, farm size, radio ownership, and crop insurance significantly enhanced adoption of CCASs. The authors recommend more education and training on CC adaptation practices to equip farmers with the skills to alleviate the impact of CC.

Keywords:

• Adaptation strategies
• climate change
• count data model
• Ghana
• perceptions

REVIEWING EDITOR:

• Michael Hardman, University of Salford, Salford, UK

SUBJECTS:

• Agriculture & Environmental Sciences
• Environmental Studies
• Environmental Policy
• Environmental Management
• Environment & Resources

1. Introduction

Agriculture is central to realizing the Sustainable Development Goals (SDGs) relating to zero hunger, reducing inequality and improving incomes and livelihoods in low-income countries (Blesh et al., 2019; Mollier et al., 2017). Agriculture employs most of the economically active citizens in sub-Saharan Africa and remains a major source of livelihood for millions of farm households (Moyo, 2016). The future and role of agriculture in most developing countries is however threatened by climate change and variability (Lemi & Hailu, 2019). Consequently, climate change (CC) has received much attention in recent years as a result of its perceived detrimental effects on a variety of activities, most notably agriculture. Climate change seems to exert more deleterious effects on smallholder farmers who lack the resources and expertise to address the challenges posed by CC to their production activities and the threat to their livelihoods (Acquah & Onumah, 2011).

Climate change denotes long-term variations in mean weather conditions or changes in worldwide climatic patterns (Beaugrand et al., 2019). It refers to statistically significant changes in the weather over a long time, generally a decade or more (IPCC, 2007). CC reveals itself in shifts in the regularity and amplitude of random weather events, for example temperature, rainfall, and wind. CC accounts for the rise in temperatures, as well as the onset, regularity and intensity of rainfall which affects agricultural production.

Globally, climate change is a major environmental and development problem (Skogen et al., 2018). It is a new reality that threatens to drastically alter both human geography and the physical environment, with dire consequences for humanity. Climate change is exerting an increasingly noticeable impact on the agricultural industry. Communities with high levels of poverty have fewer alternatives for adaptation, leaving them in the climate change trap. CC negatively impacts crop yields and thus food security (Gitz et al., 2016). Birthal et al. (2014) indicated that food security is impacted directly by yield losses due to severe environmental circumstances and indirectly by changes in agricultural pest population dynamics and geographic distributions. Ngoma et al. (2021) observed that CC has a negative effect on agriculture and gross domestic product. This is because CC reduces farm output due to high temperatures and drought, resulting in a high poverty rate (Aragón et al., 2021; Gezie, 2019).

As a result of its deleterious effects, it is essential to investigate how CC has impacted Ghanaian smallholder farmers and the various mechanisms by which to reduce its influence. Climate change adaptation is commonly recognized as a critical component of any policy response. CC adaptation is a strategy for lowering vulnerability, improving resilience, and reducing the risk of climatic impacts on people’s livelihoods (GadédjissoTossou et al., 2018).

De Pinto et al. (2012) identified a variety of potential adaptation strategies (ASs) for Ghana’s agricultural sector including (1) measures dealing with risks and uncertainties (crop insurance, weather and climate information, raising awareness, etc.), (2) farming practices and production technologies (resistant varieties, irrigation, extension services and training, crop diversification, etc.), (3) off-farm practices and strategies (improve access to credit, better storage, etc.), and (4) national development policy (agricultural intensification and land use policy, institutional reform, etc.). Indigenous CC adaptation practices among Ghanaian smallholders include application of inorganic and organic fertilizers, adoption of improved crop varieties, legume crop rotations, agroforestry, no-till or reduced tillage practices, use of cover crops, mulching, integrated pest and water management, non-farm and crop diversification, among others (Antwi-Agyei & Nyantakyi-Frimpong, 2021; Asante et al., 2021).

Recent studies indicate that the success of any adaptation efforts will be determined by farmers’ perceptions of CC and variability (Guodaar et al., 2021; Kichamu et al., 2018; Tesfaye & Seifu, 2016). Gaining a better knowledge of individual families’ perceptions and adaption techniques can assist policymakers and interventionists in better addressing the problems of CC. A deeper understanding of producers’ perspectives on climate change (CC), current adaptation techniques, and decision-making processes is required to develop policies that support successful agricultural land-use systems in developing countries.

Maize is a key staple crop in Ghana and cultivated by most farm households. Farmers with larger acreages also produce maize as a commercial activity. Maize represents over half of the nation’s overall cereal output (Ragasa et al., 2014). However, in recent times, maize farmers particularly in Ghana’s semi-arid regions are becoming distressed in their farming activities because of the effect of CC on their output. In the Builsa South district, CC is currently a constraint that accounts for lower crop yield. Without the ability to adapt to CC, the livelihoods of farmers will be in jeopardy. The willingness and urgency to implement adaptive measures on the other hand depend on producers’ knowledge and perceptions of the causes and effects of the change in climatic factors on their production activities and welfare.

The effects of CC and variability, and the ASs that farmers adopt have been well documented in the literature (Adu-Boahen, 2023; Malhi et al., 2021; Yamba et al., 2023). However, information and research on such ASs in the semi-arid savanna area of northern Ghana where the threats of CC are more distinct are scanty. Thus, despite the threat of CC and variability, very little is known about the local adaptation strategies in the context of the Builsa South district and the northern savanna area in general. There is therefore the need to critically examine how farmers perceive CC and their unique adaptation techniques, to help fashion out appropriate measures that are needed to address CC at the local level. We are of the view that assessment of farmer perceptions and adaptation to CC is critical because perceptions and adaptation are key starting points to understand and offset the adverse impact of CC. These concerns therefore informed this study.

It is critical to investigate farmers’ views on CC and their adaptation techniques to provide a well-informed framework for dealing with the issue. Farmers’ ability to comprehend the causes of CC is a necessary precondition for selecting adaptation solutions. Localized forms of climate risk perception are heavily influenced by the cultural, ethnic, and economic contexts in which people are exposed to risk, as perceptions influence behavior (Gebrehiwot & Van Der Veen, 2015). Thus, when deciding on adaptation techniques, policymakers must have a thorough understanding of producers’ perspectives on CC. Understanding the consequences of CC mechanisms will aid effective development programming based on community-driven initiatives to increase resilience.

The study makes a significant contribution to the literature by shedding light on the mix of local climate change adaptation strategies (CCASs) used by farmers in the semi-arid savanna ecological area of the country. In particular, the paper explores the synergies and conflicts between the selected CCASs to guide policy makers on the right promotional measures to address the challenge of CC. This study is relevant because previous studies have only shown the adaptation strategies producers adopt due to CC, and have not highlighted the synergies and conflicts between the strategies.

2. Literature review

2.1. Perception of climate change

Developing appropriate policies for agricultural and food security requires a thorough understanding of farmers’ perspectives on and responses to CC (Fadina & Barjolle, 2018). Whitmarsh and Capstick (2018) claim that the process of how people perceive CC is complex and involves a range of psychological categories, including information, beliefs, attitudes, and concerns about whether and how the climate is changing. Perception is affected by an individual’s characteristics, experiences, information received, and geographic and cultural surroundings (Van der Linden, 2017; Whitmarsh & Capstick, 2018).

People’s views are determined by their experiences, and individuals with personal experience of extreme climatic events are more inclined than others to think that they will probably happen again (de Matos Carlos et al., 2020). Thus, farmers in geographical areas prone to severe weather conditions and climatic variations are more likely to anticipate worsening climatic conditions and hence more likely to implement ASs. Furthermore, the information that someone is exposed to has the potential to influence or modify their perspective on climate change (Weber, 2010). This reinforces the need to strengthen agricultural extension services and farmer groups as channels for transmitting climate information to farmers.

Last but not least, it’s important to keep in mind that perception is in part a subjective process, thus even while people in the same place may encounter similar weather conditions, their perspectives on CC may vary (Simelton et al., 2013). The differences in perspectives on CC is likely to influence the choice of ASs adopted by the individual. Understanding the perceptions of farmers to climate change is therefore critical in finding appropriate mitigation measures at the local level.

2.2. Effects of climate change

The majority of Ghana’s rural areas, including Builsa South, rely heavily on agriculture for their livelihoods. Negative changes in agriculture would have an adverse impact on all other livelihoods that rely on agricultural production. According to Davidson et al. (2003), it is becoming increasingly obvious that CC can substantially hinder the achievement of national development goals. Because it lowers crop output, it has a direct effect on agricultural production and encourages poverty. CC and variability affect agricultural productivity and negatively affect the livelihoods of many farmers (Asante et al., 2021). Birthal et al. (2014) claim that changes in agricultural pest population dynamics and geographic distributions have an indirect effect on food security in addition to having a direct impact on production losses caused by extreme environmental conditions.

Ghana and most poor nations are vulnerable to the effects of CC on the environment, subsistence farming, and food security, with peasant farmers being especially at risk because their livelihoods depend on biodiversity resources (Acquah & Onumah, 2011). Due to inadequate CC feedback and adaptation strategies, the effects of CC are far worse in African countries compared to other parts of the world (Urama & Ozor, 2011). Chilunjika and Gumede (2021) claim that the African continent runs the possibility of becoming the focus of a significant global food crisis if climate change issues are not addressed locally.

The agro-ecology in Ghana’s northern regions is thought to be the poorest in the nation, and this region is more vulnerable to the effects of CC and variability than any other (Antwi-Agyei & Nyantakyi-Frimpong, 2021). For the past two to three decades, crop failure due to rainfall has been a frequent occurrence in this area (Antwi-Agyei & Nyantakyi-Frimpong, 2021). Despite this, research and policy in this area receive little attention.

2.3. Adaptation strategies to climate change

The agricultural sector must be able to adapt to the adverse effects of CC so as to safeguard the livelihoods of the population that directly depend on agriculture (Asfaw et al., 2016). Adaptation to CC refers to measures to mitigate the impact of CC (IPCC, 2007). It relates to a range of behaviors, strategies, processes, and policies that adjust to real or anticipated changes in the climate in order to minimize the effects on humans, communities, and the economy. It connotes the ability to respond and adjust to real or projected impacts of CC conditions in ways that inflict low harm and are beneficial to the environment.

In order to adapt, people first need to appreciate that the climate is changing or has the potential to change and must be prepared to act on their conviction about the changes in the climate conditions (Eakin et al., 2014). Consequently, one could argue that public knowledge of CC is key to the uptake of CCASs (Simelton et al., 2013; Makuvaro et al., 2018).

In order to develop adaptation techniques that can be used to lessen vulnerability to CC, it is vital to comprehend how producers perceive climate risk (Sarr et al., 2015). Even if urgent mitigation actions are implemented, it is crucial to take adaptation measures because the weather patterns have already changed and are predicted to do so again (Katsaruware-Chapoto et al., 2017). CC is particularly unfavorable for the agricultural sector without adaptation techniques; however, vulnerability can be significantly decreased with adaptation (Jiri et al., 2015). It is necessary to discover adaptation techniques in the management of crops by the smallholder farmers in a changing environment in order to increase food security because adaptation is essential for doing so (Msongaleli et al., 2015). Including farmer perspectives in research for the creation of food security programmes is crucial in discovering sustainable solutions to agricultural production constraints. Farmers’ perceptions of CC, its effects, and local alternatives to address climate change conditions are expected to be highlighted by this current study to enable farmers to better deal with the challenges of CC and variability.

3. Methodology

3.1. Study area

The Builsa South district in Ghana’s Upper East Region served as the location of the study. The district was selected for the study because it lies in the semi-arid savanna ecological zone of the country where the effects of CC and variability are more pronounced due to the harsh weather conditions. Farmers in the area encounter several challenges due to climate variability and change such as low soil fertility, new pests and diseases, unpredictable rainfall pattern, high daily average temperatures, torrential rains, among others. These challenges contribute to low agricultural production and productivity, and food insecurity. The Builsa South district is also noted for the production of charcoal which is a major cause of deforestation, and a factor influencing low agricultural productivity among maize producers in the area. There are however, no empirical studies into the causes and effects of CC in the study area, leading to the choice of Builsa South district as the study site.

The district is situated between latitudes 100 20’ North and 100 50’ North, and longitudes 10 05’ West and 10 35’ West. The area experiences monthly average temperatures ranging from 21.9 °C to 34.1 °C, with a peak of 45 °C in March. The district experiences a single rainy season, beginning in April and reaching a peak in August/September, and ending in mid-October. Rainfall ranges from 85 to 1150 mm per year, with dry spells in June and July. These dry spells in-between the onset and the peak of the rainy season make the decision on when to plant quite challenging to farmers. Every year, some farmers lose their crops during the dry spells that characterize the rainy season in northern Ghana.

3.2. Sampling and data collection

Green’s (1991) sample size formula was used to determine the study’s total respondents, based on the model’s explanatory variables. From the formula $50+8(m)$, where m is the number of explanatory variables, we obtain $50+8\left(18\right)=194$ respondents, which was adjusted to 250.

A multistage sampling approach was adopted to select the respondents for the study. The Builsa South district was purposively selected because of the effect of CC on agriculture in the area. Charcoal burning is a major economic activity in the area and this competes with agriculture in terms of removal of forest trees for charcoal production. The study intended to investigate the perceptions of the inhabitants on the effects of CC as a way to increase awareness of CC. Five communities in the district were selected at random and from each community, 50 farmers were randomly selected to give a sample size of 250 respondents. The study communities included Uwasi, Naadema, Yerinsa, Logmisa and Kasiesa. Respondents were engaged face-to-face in the interviewing that made use of pretested questionnaires. The interviews were conducted by the first author who was assisted by a team of trained field assistants.

All subjects gave their informed consent for inclusion before they participated in the study. The study protocol was approved by the Research Seminar and Supervisory Committee of the Department of Agricultural and Food Economics, University for Development Studies. The informed consent was verbal because majority of smallholder farmers in the study area cannot read or write.

3.3. Method of data analysis

The objectives of the study were analyzed using both descriptive statistics and econometric modelling. The first objective, that is, assessment of farmers’ knowledge of the activities that contribute to CC was analyzed descriptively based on the 5-point Likert scale used in the data collection with 1 indicating ‘very low’, 2 indicating ‘low’, 3 indicating ‘moderate’, 4 suggesting ‘high’, and 5 indicating ‘very high’. On an ordinal scale, the Likert scale effectively captures answers to qualitative knowledge claims. For example, perception questions such as ‘How does deforestation contribute to CC?’, ‘to what extent does bush burning contribute to CC?’, etc., could be effectively analyzed using a Likert scale. Specifically, frequencies and percentages were generated for the responses provided by the respondents. Also, means and standard deviations were calculated and presented in a tabular form.

The second objective, that is, assessment of farmers’ perception of the effect of CC on maize production, was evaluated using Kendall’s coefficient of concordance (W), which is a rank-based statistic that establishes agreement amongst respondents. It is a metric for how well the respondents agree on a collection of elements they rate. The identified challenges are ranked from the least to the most influential. The Kendall’s Coefficient of Concordance (W) assesses the degree of agreement between the farmers in the ranking of the constraints.

Kendall’s W is represented by the equation: (1) $$W=\frac{12\left[\sum T^2\hspace{0.17em}-\hspace{0.17em}\frac{{(\sum T)}^2}{n}\right]}{\left(n^3\hspace{0.17em}-\hspace{0.17em}n\right)m^2}$$(1) where n = total number of objects ranked, m = total number of farmers, and T = sum of factors being ranked. After W has been estimated, it is tested for significance using the F-test.

The third objective, that is, evaluation of the factors affecting adoption of CCASs by the respondents, was analyzed using a count data model, specifically, the generalized Poisson (GP) regression model. The rationale behind using a generalized Poisson regression model to evaluate the factors affecting the adoption of CCASs by farmers is that the combination of CCASs adopted by a farmer is a count, hence the need to use a count data model. The reason behind the use of a count data model (i.e. the generalized Poisson regression) is that a farmer can choose to use a single strategy or multiple of strategies as CC adaptation strategy. For adaptation strategies with synergies between them, the more practices that are adopted, the better the expected outcome (e.g. reduction of the effects of CC, increase in productivity, etc.). The model therefore measures the intensity of adoption as against the adoption of each individual strategy. The generalized Poisson model has been used to analyze the intensity of adoption in other studies such as Anang and Asante (2020), and Awuni et al. (2018).

When the data is count, as with the case of adoption of multiple adaptation strategies, some methodological questions arise which must be addressed to ensure that valid results are obtained. The key question relates to whether the data is underdispersed, overdispersed or equidispersed. Negative binomial (NB) regression, zero-inflated Poisson (ZIP) and zero-inflated negative binomial (ZINB) models are suitable for analyzing overdispersed count data, that is when the mean of the count data is less than the variance. When the mean of the count data is greater than the variance, it represents an underdispersed data, and the GP model is the appropriate analytical tool to use. The standard Poisson model is unsuitable for underdispersed data, but appropriate for data exhibiting equidispersion, that is, when the variance and the mean are equal. A careful examination of the data used in this study reveals undersdispersion (the mean is greater than the variance as shown in Table 8), which justifies the use of the GP regression model. Furthermore, the dispersion parameter was found to be less than zero suggesting that the standard Poisson model is inappropriate for the analysis. Hence, the GP model was used to estimate objective three of the study, that is, the determinants of adoption intensity of CCASs, following after other previous studies (Awuni et al., 2018; Harris et al., 2012; Aker, 2011).

Table 8. Adoption intensity of climate change adaptation strategies.

Strategies	Frequency	Percentage
0	17	6.80
1	60	24.00
2	70	28.00
3	57	22.80
4	25	10.00
5	14	5.60
6	7	2.80
Total	250	100.00
Mean	2.332
Variance	2.188

Suppose $Y_i$ symbolizes the response variable, the probability mass function can be represented by (2) $$f\left(y_i,{\theta }_i,\delta \right)=\frac{{\theta }_i{({\theta }_i+\delta y_i)}^{y_i-1}{e}^{-{\theta }_i-\delta y_i}}{y_i!};(y_i=0,1,2,\dots .)$$(2) where ${\theta }_i>0$, and $\mathrm{\text{max}}(-1,-{\theta }_i/4)<\delta <1.$ The GP model has mean $({\mu }_i)$ and variance ($\mathit{\text{var}}(y_i)$) specified according to eq. (3)(3) $${\mu }_i=E\left(Y_i\right)=\frac{{\theta }_i}{1-\delta },\mathit{\text{var}}\left(Y_i\right)=\frac{{\theta }_i}{1-{\delta }^3}=\frac{{\theta }_i}{1-\delta }E\left(Y_i\right)=\varphi E\left(Y_i\right)$$(3) . (3) $${\mu }_i=E\left(Y_i\right)=\frac{{\theta }_i}{1-\delta },\mathit{\text{var}}\left(Y_i\right)=\frac{{\theta }_i}{1-{\delta }^3}=\frac{{\theta }_i}{1-\delta }E\left(Y_i\right)=\varphi E\left(Y_i\right)$$(3) where $\varphi E(Y_i)$ is a dispersion factor. When $\delta =0$, the data shows equidispersion and the GP model reduces to the standard Poisson model; when $\delta >0$ we have overdispersion, and when $\delta <0$, there is underdispersion.

The log-likelihood function of the GP model is specified as follows $$L=\sum _{i=1}^{n}L({\theta }_i,\delta ,y_i)=\mathrm{ }\sum _{i=1}^n\mathit{\text{lnL}}({\theta }_i,\delta ,y_i)$$ (4) $$L=\sum _{i=1}^n\{\mathit{\text{ln}}{\theta }_i+(y_i-1\left)\mathit{\text{ln}}\right({\theta }_i+\delta y_i)-({\theta }_i+\delta y_i)-\mathit{\text{ln}}{y}_i!\}$$(4)

The number of adaptation strategies adopted by the respondents (the response variable) is non-negative, ranging from 0 to 6. The response variable is regressed on a number of explanatory variables to estimate the GP regression model. According to Consul and Famoye (1992) and Consul (1989), the regressors are introduced into the GP regression model as shown in eq. (5)(5) $$\mathrm{\text{log}}\frac{{\theta }_i}{1-\delta }=\sum _{r=1}^p{x}_{\mathit{\text{ir}}}{\gamma }_r$$(5) . (5) $$\mathrm{\text{log}}\frac{{\theta }_i}{1-\delta }=\sum _{r=1}^p{x}_{\mathit{\text{ir}}}{\gamma }_r$$(5) where $x_{\mathit{\text{ir}}}$ designates the ith observation of the rth regressor, p specifies the number of regressors, and ${\gamma }_r$ denotes the rth coefficient.

The empirical GP regression model takes the following form (6) $$Y_i={\gamma }_0+\sum _{j=1}^{11}{\gamma }_j{x}_{\mathit{\text{ji}}}+v_i$$(6) where $x_{\mathit{\text{ji}}}$ are covariates assumed to affect uptake of the adaptation strategies. Inclusion of the covariates was informed by the extant literature and the study’s a priori expectations.

4. Results and discussions

4.1. Description of the study’s variables

Table 1 presents the variables used in the study and their summary statistics. The data shows that farmers adopted an average of two (2) adaptation strategies, with a minimum of zero (0) and a maximum of six (6). Thus, farmers typically adopt the adaptation strategies in combination due to the complementarity in their usage. Also, majority of the sampled farmers are males (90.8%), which agrees with the results of Mudege et al. (2015) indicating that more men are involved in farming compared to women. Men are more likely to grow maize while women typically grow other crops such as vegetables, soybean, groundnuts, etc. because maize is the household’s main crop for subsistence and men usually take charge of its cultivation. This could explain the high proportion of men in the sample.

Table 1. Description of the study’s variables.

Variable	Mean	Std. Dev	Min	Max
Number of strategies (number)	2.332	1.411	0	6
Sex (dummy: male = 1; 0 otherwise)	0.908	0.290	0	1
Age (years)	41.94	7.966	24	60
Years of formal education (years)	6.788	4.710	0	18
Farming experience (years)	12.97	7.765	3	30
Farm size (acres)	10.43	3.534	5	20
Cattle ownership (dummy: 1 for cattle ownership)	0.420	0.495	0	1
Radio ownership (dummy: 1 for cattle ownership)	0.868	0.339	0	1
Number of extension visits (number)	1.208	0.938	0	3
Access to subsidy (dummy: 1 for access to subsidy)	0.792	0.407	0	1
Access to credit (dummy: 1 for credit access)	0.320	0.467	0	1
Awareness of crop insurance (dummy: 1 if aware)	0.344	0.476	0	1
Experience of drought (dummy: 1 for drought incidence)	0.520	0.501	0	1
Distance from home to market (kilometers)	9.968	2.599	3	13
Distance from home to farm (kilometers)	6.744	2.035	3	10

The average age of a farmers in the sample is 42 years. This shows that the study area is endowed with agricultural labor force for farming activities. In terms of farming experience, the sampled farmers have been in farming for 13 years. The respondents also had 10 acres of farmland, and had only 1 extension visit during the cropping season. Access to credit and subsidy stood at 79% and 32% respectively, while 42% own cattle and 87% own radio. Radio is a key source of information on agriculture for many rural farmers, hence ownership of radio is anticipated to improve uptake of CCASs. Close to 34% of the sampled farmers were aware of crop insurance as a risk-mitigating tool, while 52% experienced drought in the course of their production.

The respondents have attained an average of 7 years of formal education. Farmers’ adoption of technology and development of new skills have been observed to be impacted by level of education (Dokyi et al., 2021). There is the perception that most Ghanaian farmers are not formally educated, and this can lead to low economic return from their production since modern agriculture is beyond tilling of the soil. The result agrees with the finding of Bryan et al. (2009) which shows a positive correlation between level of education and access to knowledge on adaptation strategies in Ethiopia and South Africa. On the average, farmers travel almost 10 km from their homes to the market, and 7 km from their homes to the farm. These distances are relatively long and could have negative effect on adoption of CCASs.

4.2. Ranking of farmers’ perceptions of the activities that contribute to climate change

Table 2 shows the results of the ranking of farmers’ perceptions of the activities that contribute to CC. A five-point Likert scale was used to solicit these responses. Regarding the extent to which deforestation contributes to CC, the respondents indicated a mean response of ‘high’. Similarly, the extent to which bush fires contribute to CC was rated as ‘high’. Deforestation and bush fires are major challenges confronting farmers in the study area emanating from charcoal burning, indiscriminate bush burning and other unsustainable farming practices. These practices have negative consequences on agricultural productivity in the area. The result is insightful because both deforestation and bush burning are direct consequences of agriculture. The fact that farmers are aware of these as factors causing CC should provide an entry point for educating farmers on how to be more environmentally-friendly in farming to fight against CC.

Table 2. Ranking of farmers’ perceptions of the activities that contribute to climate change.

Knowledge of activities that contribute to climate change	Frequencies and Percentages (%)					Mean	Std. Dev
	Very Low (1)	Low (2)	Moderate (3)	High (4)	Very High (5)
To what extend does deforestation contribute to CC?	1 (0.40)	11 (4.40)	19 (7.60)	169 (67.60)	50 (20.00)	4.02	0.69
To what extend does bush burning contribute to CC?	13 (5.20)	16 (6.40)	49 (19.60)	146 (58.40)	26 (10.40)	3.62	0.94
To what extent can improper disposal of waste contribute to CC?	93 (37.20)	41 (16.40)	43 (17.20)	67 (26.80)	6 (2.40)	2.41	1.29
How does burning of fossil fuel contribute to CC?	138 (55.20)	54 (21.60)	42 (16.80)	11 (4.40)	5 (2.00)	1.76	1.01
How does greenhouse gases contribute to CC?	147 (58.80)	62 (24.80)	33 (13.20)	2 (0.80)	6 (2.40)	1.63	0.91
To what extent does urbanization contribute to CC?	163 (65.20)	62 (24.80)	15 (6.00)	3 (1.20)	7 (2.80)	1.52	0.88
To what extend does power plants contribute to CC?	161 (64.40)	63 (25.20)	17 (6.80)	3 (1.20)	6 (2.40)	1.52	0.87

Note. Figures in brackets are percentages.

The respondents rated the extent to which improper disposal of waste contributes to CC as ‘low’. Similar response (‘low’) was adduced for the extent to which fossil fuel, greenhouse gases, urbanization, and power plants contribute to CC. Hence, farmers do not perceive these to contribute significantly to CC. The result is plausible because for most farmers in rural areas, these are not critical issues they face on a daily basis.

4.3. Ranking of farmers’ perception of the effect of climate change on maize production

Kendall’s Coefficient of Concordance was used to rank the effects of CC on a scale of 1–7 (1 = least important and 7 = most important). The results are presented in Table 3.

Table 3. Ranking of the effects of climate change on maize production.

Effects of climate change	Mean	Rank
Incidence of pests and diseases	6.01	1^st
Loss of soil fertility	5.41	2^nd
High temperature	4.15	3^rd
Change in onset of the rainy season	4.12	4^th
Strong winds	3.28	5^th
Drought	2.85	6^th
Excessive rainfall	2.19	7^th
Number of observations	250
Kendall’s W	0.40
Chi-Square	600.4
Degree of freedom	6
Asymp. Sig.	0.000

The results reveal that the foremost effect of CC on farmers’ production is the incidence of pests and diseases. According to the respondents, new pests and diseases attack their crops due to CC. Producers find it challenging to prevent these pests and diseases from destroying their crops because they are difficult to control. The findings of Ward and Masters (2007) reveal that CC is associated with pest invasion, which supports the finding of this study. As indicated by Aljaryian and Kumar (2016), CC causes global warming which increases the risk of pest infestation. One of such pests is the fall-army worm that emerged in 2016 and became a main challenge to maize production (Bariw et al., 2020). Fall-army worm infestation is reported to have caused a decrease in farmers’ income in northern Ghana by raising the cost of production and worsening household food security (Bannor et al., 2022). The results are supported by the findings of a study in northern Ghana by Anang (2021), which indicated that close to half of the respondents stated that they experienced pest and disease infestation which had a negative effect on their production.

The respondents ranked loss of soil fertility as the second most important effect of climate change on maize production. Due to climate change, the soil in the area is gradually losing its fertility and thereby reducing output levels (Bakari, 2015). Soil infertility is a critical challenge in the semi-arid regions of Ghana and the result from this study is not surprising. The savanna ecology is very susceptible to bushfires and run-off during the rainy season, which is exacerbated by the long dry season leading to loss of soil fertility. The finding of this study resonates with other studies. For example, Yahaya (2015) observed that declining soil fertility accounts for the low agricultural productivity in West Africa, while Wood (2013) observed that farmers in northern Ghana face the dual challenge of managing depleted farmlands and controlling pests and diseases, which rank among the greatest challenges they face.

The third challenge posed by CC to farmers in Builsa South district is rise in average temperature which according to the farmers account for the decline in farm output in the district. An increase in average daily temperature affects the rate of plant development (Hatfield & Dold, 2018). High average day and night temperatures have detrimental effects on transpiration, plant growth and yield. High daily temperatures also affect soil nutrient quality and lead to loss of soil fertility.

In addition, change in the onset of the rainy season ranked as the 4^th most important effect of CC on farmers’ production. According to the farmers, some twenty years ago, they used to start sowing in April. However, in contemporary era, the onset of the rainy season has shifted to May/June, which does not favour the cultivation of long-maturing crop varieties. Changes in the onset of the rainy season therefore affect farmers’ decision making on time to plant and choice of crop to suit the duration of the rainy season, which may have implications on crop yield.

The incidence of strong winds is another effect of CC identified by the respondents. Ranked as the 5^th perceived challenge, strong winds can destroy crops on the farm and reduce yield levels by causing damage to plants, flower abortion, among others. Drought is another effect of CC and it ranked 6^th in the list. Drought leads to a reduction in the output of farmers by limiting the availability of water for crop use. Drought causes severe reduction in yield of farmers as indicated by Ray et al. (2018).

Furthermore, excessive rainfall was ranked as the least effect of CC in the study area. Tropical rainfall is usually intense and can cause severe damage, especially in the savanna ecological zone which lacks dense vegetative cover. Excessive rainfall can destroy farm produce as a result of flooding or washing away of crops on the farm. Excessive rainfall also increases the occurrence of soil erosion and loss of soil fertility through leaching, thereby reducing crop yield.

4.4. Farmers’ adaptation strategies to climate change

Table 4 shows the ASs used by farmers to mitigate the effect of CC. The most commonly adopted strategy is early planting, followed by adoption of mixed cropping, and application of zero tillage. The least adopted strategy is crop rotation.

Table 4. Adaptation strategies adopted by farmers.

Variable	Mean	Std. Dev
Adoption of disease-resistant varieties	0.288	0.454
Adoption of drought-tolerant varieties	0.296	0.457
Application of zero tillage	0.372	0.484
Early planting	0.608	0.489
Adoption of crop rotation	0.180	0.385
Adoption of mixed cropping	0.588	0.493

Shifting planting dates, particularly early planting has been identified as a climate change adaptation strategy (Ahmad et al., 2020; Getachew et al., 2021). Early planting as a CCAS is important in the context of northern Ghana which is savanna ecology and experiences a single rainfall regime. Early planting is connected to the nature of the wet season, which is a key factor determining crop production in the study area. The duration and regularity of the rainy season are important factors affecting crop production in the study area. With limited irrigation facilities, farmers are compelled to plan their cropping activities within the short rainy season between May/June and September/October. A change in the onset of the rainy season presents a serious threat to farmers ability to accurately predict the planting season. Also, many farmers lack access to accurate weather information, and thus tend to misjudge the right time to plant. A common occurrence is when farmers cultivate crops at the onset of the rainy season (May to June) but encounter intermittent dry spells between June and early July, resulting in some farmers experiencing crop loss. The adoption of early planting by the respondents in this study therefore aligns with the study’s a priori expectation and agrees with previous studies. For example, in a study by Acharjee et al. (2019) in Bangladesh, early planting was recommended as a CCAS for high-yielding rice varieties as a way to boost yields.

Adoption of mixed cropping is a diversification strategy which is also an adaptation strategy to CC and variability (Himanen et al., 2016; Sawe, 2022). Typically, smallholder farmers prefer mixed cropping because it helps to reduce the risk of complete crop failure thereby safeguarding the farm household’s livelihood. Also, farmers are conscious of synergies between certain crop combinations, and the fact that mixed cropping reduces pest and disease build-up. Monocultures may be easier to manage and yield more if proper management practices are put in place, but they are more prone to pest and disease build up that can lead to massive crop loss for resource-poor smallholder farmers.

The adoption of zero tillage as a CC adaptation strategy is another important development in smallholder agriculture. Below et al. (2010) identified agricultural conservation as one of the risk management instruments used by farmers to alleviate climatic risks. Zero tillage is a conservation strategy adopted by farmers to mitigate climate risks and conserve soil fertility. With zero tillage, farmers avoid the practice of employing heavy tillage equipment on the farm, and resort to the use of herbicides to control weeds. With zero tillage, the crop residues from the previous year’s farming are left on farm to decompose and fertilize the soil. A challenge that farmers face in adoption of zero tillage is that there are frequent bush fires during the dry harmattan season, while cattle grazing may also lead to removal of the crop residues intended to fertilize the soil.

The need to develop and adopt improved varieties that have superior qualities such as drought-tolerance and disease-resistance have gained acceptance among both researchers and farmers in recent times (Prasanna et al., 2020; Yamano et al., 2018). The savanna ecology is prone to drought conditions hence the preference for drought-resistant varieties. Also, the savanna ecology is prone to pest and diseases that flourish in hot climates such as prevails in northern Ghana. Thus, farmers spend so much resources to control crop pests and diseases and therefore prefer to adopt disease-resistant crop varieties to alleviate the impact of CC. Below et al. (2010) identified crop and variety diversification as risk management instruments used by farmers to mitigate climatic risks. Farmers in the study area also employ variety diversification such as adoption of disease-resistant and drought-tolerant varieties to manage climatic risks. It is therefore essential for crop breeders to factor this behavior of farmers into improved variety development to enhance acceptability and adoption.

Another adaptation strategy to CC identified is the adoption of crop rotation. The practice of crop rotation helps to improve soil fertility and prevent build-up of insect pests. The savanna ecology is prone to pests and diseases as well as declining soil fertility due to high daily temperatures and soil erosion that depletes soil nutrients in the area. Cropping rotation is a rational decision for farmers to take when the fertility of the soil is low such that it cannot support the productivity of a crop. Cropping rotation, particularly cereal-legume crop rotation, is regarded as an adaptation mechanism to regenerate depleted soil nutrients (Debie, 2021; Miheretu & Yimer, 2017). Also, there is the assertion that the fertility status of the soil is an important factor that drives adoption of legume-cereal crop rotation as reported by Debie (2021). In other words, when farmers perceive their soils to be low in fertility, they rationally opt for crop rotation to rejuvenate the nutrient level of the soil. This practice is common among Ghanaian farmers, especially in areas where agricultural land is not scarce.

Table 5 depicts the differences in adoption of CCASs among maize farmers in Builsa South district. Generally, women had higher adoption of early planting and mixed cropping as adaptation strategies relative to men. Women farmers typically produce crops that are part of the household diet and contribute to household food and nutrition security (such as cereals, legumes and vegetables), hence are more disposed to engage in mixed cropping. Also, because women play multiple roles in the household (keeping of the home, farming, business, among others), they prefer to plant early.

Table 5. Differences in adoption of CCASs based on sex of the farmer.

Adaptation strategy	Male farmers (N = 227)		Female farmers (N = 23)
	Mean	Std. Dev	Mean	Std. Dev
Disease-resistant varieties	0.286	0.453	0.304	0.470
Drought-tolerant varieties	0.295	0.457	0.304	0.470
Zero tillage	0.370	0.484	0.391	0.499
Early planting	0.604	0.490	0.652	0.487
Crop rotation	0.185	0.389	0.130	0.344
Mixed cropping	0.568	0.496	0.783	0.422

In Table 6, we explore the differences in adoption of CCASs based on geographic location within the district. Adoption of the CCASs does not follow any predetermined pattern as shown by the results. Adoption of Disease-resistant varieties was highest in Uwasi and least in Kasiesa, while adoption of drought-tolerant varieties was highest in Yerinsa and least in Kasiesa. Some communities (Yerinsa and Naadema) had higher adoption of early planting compared to the remaining communities. Also, adoption of crop rotation was low in Yerinsa, Uwasi and Logmiisa compared to Naadema and Kasiesa, while Naadema recorded the highest adoption of mixed cropping compared to the remaining communities.

Table 6. Differences in adoption of CCASs based on geographic location within the district.

Adaptation strategy	Yerinsa		Logmiisa		Naadema		Kasiesa		Uwasi
	Mean	S.D	Mean	S.D	Mean	S.D	Mean	S.D	Mean	S.D
Disease-resistant varieties	0.22	0.42	0.26	0.44	0.36	0.49	0.08	0.27	0.52	0.51
Drought-tolerant varieties	0.44	0.50	0.26	0.44	0.34	0.48	0.12	0.33	0.32	0.47
Zero tillage	0.36	0.49	0.22	0.42	0.56	0.50	0.50	0.51	0.22	0.42
Early planting	0.78	0.42	0.42	0.50	0.82	0.39	0.52	0.51	0.50	0.51
Crop rotation	0.02	0.14	0.16	0.37	0.36	0.49	0.24	0.43	0.12	0.33
Mixed cropping	0.44	0.50	0.68	0.47	0.90	0.30	0.62	0.49	0.30	0.46

4.5. Correlations (synergies or conflicts) between different CCASs

In Table 7, we present the correlation between the individual adaptation strategies to assess the synergies (complementarities) and conflicts (substitutability) between them. A positive correlation indicates a synergy or complementarity between the strategies while a negative correlation indicates substitutability or conflict between the strategies. The results indicate that adoption of zero tillage has a positive correlation with adoption of early planting, crop rotation and mixed cropping. Thus, zero tillage has complementary relationships with these adaptation strategies.

Table 7. Correlations (synergies or conflicts) between different CCASs.

Climate change adaption strategies	Drought-resistant variety	Disease-tolerant variety	Zero tillage	Early planting	Crop rotation	Mixed cropping
Drought-resistant variety	1.000
Disease-tolerant variety	0.575***	1.000
Zero tillage	−0.014	0.117	1.000
Early planting	−0.014	0.144**	0.177***	1.000
Crop rotation	0.070	0.0383	0.178***	0.035	1.000
Mixed cropping	−0.042	−0.009	0.157**	0.027	0.265***	1.000

The result further indicate that adoption of drought-resistant and disease-tolerant varieties exhibit a positive correlation, suggesting that they complement each them. Also, the correlation between adoption of disease-tolerant varieties and early planting portrays that the two are complementary adaptation strategies. The adoption on one strategy leads to the adoption of the other. Similarly, adoption of crop rotation and mixed cropping were found to be complementary strategies. Where the strategies show synergy between them, they are anticipated to positively influence agricultural production and productivity as they are expected to have a greater influence on mitigating the impact of CC. Thus, adoption of synergistic or complementary adaptation practices are to be encouraged to enhance farm performance. Even though some of the strategies reported negative correlation coefficients indicating substitutability between them, the correlation coefficients were not statistically significant.

4.6. Adoption intensity of climate change adaptation strategies

Table 8 shows the adoption intensity of the measures employed by farmers to lessen the effects of CC on their production. Farmers usually adopt technologies in combination, and this is reflected in the adoption of CCASs by the respondents. While 6.8% of the respondents did not adopt any of the strategies, 24% adopted a single strategy, while 28% and 22% adopted 2 and 3 strategies, respectively. In addition, 10% adopted 4 strategies, 5.6% adopted 5 strategies while 2.8% adopted a total of 6 adaptation strategies.

4.7. Factors affecting adoption intensity of climate change adaptation strategies

The study’s overarching objective is to assess the factors that determine adoption of multiple CCASs to mitigate the impact of CC on farmers’ production. Due to the nature of the dependent variable, the study adopted the Generalized Poisson regression model to analyze the data. Adoption intensity was measured over a range of 0 and 6, and represents a count data for the dependent variable, which make the use of a count data model appropriate for the analysis. The appropriateness of the estimation was tested using the Chi-square test which is significant at 1% level, indicating that the model adequately represents the data and the explanatory variables jointly explain the adoption of CCASs in the Builsa South district. The results are shown in Table 9.

Table 9. Factors affecting adoption intensity of farmers’ adaptation strategies to climate change.

Variables	Coefficient	Standard error	P > | z |
Sex	−0.117	0.135	0.384
Age	−0.007	0.006	0.282
Years in education	0.014*	0.009	0.093
Years in farming	−0.006	0.007	0.377
Farm size	0.029***	0.010	0.005
Cattle	0.132*	0.075	0.079
Radio	0.196*	0.108	0.070
Extension visits	−0.020	0.042	0.633
Subsidy	0.072	0.088	0.418
Credit for farming	0.053	0.072	0.462
Crop insurance	0.179**	0.083	0.032
Drought	0.140**	0.070	0.044
Home to market	−0.118***	0.022	0.000
Home to farm	−0.032**	0.016	0.049
Naadema	−0.298**	0.118	0.012
Yerinsa	−0.449**	0.190	0.018
Logmisa	−0.102	0.120	0.392
Kasiesa	−0.558***	0.150	0.000
Constant	2.131***	0.379	0.000
Chi^2(1)	33.90
Prob > Chi^2	0.000
Pseudo R^2	0.117
Dispersion	−0.321
Number of observations	250

*** Significant at 1%,.

** Significant at 5%,.

* Significant at 10%. Note. Uwasi is the reference community.

The result indicates a positive association between farmers’ years of education and the adoption of ASs at 10% level of significance. Education enables farmers to acquire more knowledge about CCASs, increasing their desire to adopt adaptation practices to reduce the effects of CC on their production. The result is in line with that of Ali and Erenstein’s (2017) work, which shows that the intensity of adoption is positively correlated with respondents’ years of education. Kassem et al. (2019) also showed that an education increases the willingness of farmers to adopt CCASs.

Farm size correlated positively with uptake of multiple CCASs and is highly significant at 1%. Thus, farmers with larger acreages tend to adopt more strategies to mitigate the impact of CC. The result could mean that the larger the farm, the more susceptible it is to the negative impact of CC, which therefore requires effective ASs to mitigate these effects. The finding agrees with Fosu-Mensah et al. (2012) who observed that farm size influences adoption of CCASs.

Cattle ownership positively influenced uptake of multiple CCASs with a significance level of 10%. Thus, farmers who own cattle tend to adopt the adaptation strategies more than those who do not own cattle. Cattle ownership is used as a wealth indicator in this study, and the result supports the assumption that wealthier households are more likely to adopt CCASs. Wealthier farm households have the ability to adoption improved technologies compared to less wealthy farm households.

The findings indicate a positive association between radio ownership and uptake of multiple ASs, which aligns with a priori expectation. Radio, television, mobile phones, and similar media devices are sources of information to farmers which promote adoption of adaptation strategies. All things being equal, farmers who own radios tend to receive more information on adoption and are educated on how to practice the adaptation strategies. Farmers who own radios are anticipated to have greater access to information as compared to those without radios, all things being equal. The finding resonates with that of Thinda et al. (2020) which indicates that access to information through media such as radio, television, and mobile phones promotes adoption of CCASs.

Awareness of crop insurance positively influenced adoption of multiple CCASs at 5% significant level. Crop insurance is a risk-reduction mechanism even though the practice is less pronounced among smallholder farmers. The result is plausible because farmers who are aware of crop insurance are anticipated to have more knowledge about the impact of CC and therefore willing to adopt the ASs.

The results further indicate that farmers who experienced drought on their farms were more likely to adopt multiple CCASs. It may be argued that farmers who have experienced drought problems in previous seasons have no choice but to implement CCASs. Farmer who experienced drought-spells are expected to be more conscious of the impact of CC and therefore more likely to adopt adaptation strategies. Most smallholder farmers in Ghana operate under rainfed conditions because they do not have access to irrigation facilities, and thus unable to overcome the challenge posed by prolonged drought.

The distance from farmer’s home to the local market negatively correlated with uptake of CCASs at 1% significant level. The result lines up with the study’s a priori expectation because the longer the distance, the higher the transaction costs which may be a disincentive to adoption. Ojo and Baiyegunhi (2020) observed that the propensity to adopt CCASs is inversely related to the distance to the source of credit. The result however deviates from that of Bedeke et al. (2019) which indicated that distance to market positively correlated with adoption of conservation tillage as a CCAS in Ethiopia. The results further revealed that the distance between respondents’ home and farm had an inverse relationship with uptake of multiple CCASs at 5% significant level. Hence, adoption decreases when the farm is further away from the home.

Moreover, it was identified that the community-level dummy variables influenced adoption of CCASs. Using Uwasi as the reference community, Naadema and Yerinsa communities had a negative correlation with the adoption of the CCASs at a significance level of 5%, while Kasiesa community had an inverse relationship with the adoption of the strategies at 1% significance level. This means that farmers in Uwasi have higher adoption intensity than their peers in these three communities which shows that location factors play a role in adoption of CCASs.

5. Conclusion and recommendations

The study assessed farmers’ perceptions and determinants of adoption of CCASs in Builsa South district of Ghana, using cross-sectional data. The data was analyzed using descriptive statistics and Poisson regression model. Deforestation and bush burning were identified to contribute highly to CC, while the contribution of improper waste disposal, fossil fuel, greenhouse gases, urbanization, and power plants to CC was rated as low. Farmers rated the incidence of new pests and diseases as the most serious effect of CC on their production followed by loss of soil fertility, high temperatures, and change in the onset of the rainy season. In all, the study identified six major CCASs adopted by farmers. Most farmers adopted these strategies in combination. Early planting, adoption of mixed cropping and application of zero tillage were the topmost strategies adopted by farmers. The factors determining adoption intensity of CCASs include years of education, farm size, ownership of radio and cattle, awareness of crop insurance, market distance, distance from home to farm, experience with drought, and location effects.

On the basis of the study’s findings, the following recommendations are put forth.

First, farmers should be provided with education and training on how to prevent deforestation and bush burning to enable them to carry out environmentally-friendly agricultural practices. Providing farmers in rural areas with formal and informal education will go a long way to improve their human capital and knowledge of CC change and the adaptation strategies. Some effective methods or approaches that can be used to deliver education and training to farmers in rural areas include the use of farmer organizations. To be effective, these groups must be well managed and incentivized. Also, the Ministry of Food and Agriculture should collaborate with the Ministry of Communication and agencies such as the National Commission for Civic Education (NCCE) to carry out farmer sensitization workshops and community forums to educate farmers on climate change and its deleterious effects. Agricultural extension agents should also be adequately equipped to train farmers on CCASs that are effective in the study area. Climate change adaptation strategies should be made a central subject in agricultural extension education provided to rural farmers.

With regards to curbing deforestation and bush burning, there is the need to create alternative livelihoods for the inhabitants to reduce the rate of deforestation and cutting of trees for fuelwood and charcoal production. Also, the district assembly must come out with by-laws prohibiting bush burning especially during the dry harmattan season. Traditional rulers also have a role to play by outlawing practices such as felling of economic trees for firewood and charcoal production. In addition, formation of community watchdog groups to safeguard the forest and prevent deforestation and bush burning is a promising mitigation measure by the community to curb deforestation and bush burning.

Also, the use of mass media to educate farming communities on the causes and impacts of CC is critical. Radio ownership contributed positively to adoption of CCASs and should therefore be used as a medium to educate farmers on CC. In recent times, farmers are becoming increasingly technologically savvy. The use of mobile phones, radio and television for agricultural purposes is on the ascendancy. An innovative way to effectively disseminate information on climate change and its impacts to farmers is through the use of radio, where programs on climate change are aired by extension agents and climate change experts to farming communities. Additionally, such programs should be telecast on television on a regular basis. There is also the need to promote e-agriculture in Ghana with a mobile phone component so that farmers can receive broadcast messages on climate change adaptation strategies, weather information, date of planting, among others. Mobile video vans could also be deployed to show extension programs on CCASs to promote adoption of the adaptation strategies.

Also, smallholders need information on the time to plant and how to effectively adopt mixed cropping and zero tillage to enhance crop yield. This information can be provided to farmers via extension workers and the use of mass media such as radio, farmers fora, and community durbars.

Finally, even though crop insurance is not common among smallholder farmers, efforts should be made to promote its use as it enhances adoption of CCASs. The promotion and uptake of crop insurance can be encouraged among smallholder farmers in low-income countries to enhance their resilience to climate change through the formation of innovation platforms that allow for farmer experiential learning. Innovation platforms promote technology adoption through farmers learning from each other. Crop insurance providers must work with these innovation platforms and use other channels such as the radio, television and mobile phones to educate farmers on the benefits of crop insurance. Agricultural extension education should also emphasize the need for farmers to insure their crop farms. Designing appropriate ways of paying the insurance premium is another important factor to consider in promoting crop insurance uptake among smallholder farmers in low-income countries.

Author contributions

John Adeboa conceived and designed the study, carried out the data analysis and drafted the paper. Benjamin Tetteh Anang interpreted the data and revised the manuscript critically for intellectual content. Both authors approved the final version of the manuscript and are accountable for all aspects of the work.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

Data supporting this work is available upon reasonable request from the corresponding author.

Additional information

Notes on contributors

John Adeboa

John Adeboa holds a BSc degree in Agriculture Technology with specialization in Agricultural Economics and Extension. His research interests include climate change economics and agricultural finance.

Benjamin Tetteh Anang

Benjamin Tetteh Anang holds a PhD in Agricultural Economics and has published extensively in peer-reviewed academic journals. His research interests include agricultural production economics, impact evaluation of agricultural programs and projects, and improvement of smallholder production systems.