This study was conducted in Ahafo Ano North District (06°47′-07°02′N, 02°26′-02°04′W) of Ashanti Region, Ghana, with a total area of 593.7 km² (Fig. 1). The mean annual temperature of Ahafo Ano North District is 28.0°C, the mean annual rainfall is 1750 mm, and vegetation is mostly moist deciduous forest type, which is quite similar to the rain forest in this district. Agriculture employs about 70.0% of the total labor force in this district. Crops cultivated in Ahafo Ano North District include cocoa, plantain, maize, cassava, cocoyam, rice, oil palm, citrus, and vegetables. Ahafo Ano North District was selected because it is one of the major cocoa-producing districts in Ghana (GSS, 2024). Cocoa is very important to the economy of this country and to the livelihoods of many smallholder farmers (Oyekale, 2021). This district is located in the wet semi-equatorial zone, which is vulnerable to the adverse effects of climate change (MoFA, 2024). Furthermore, Ahafo Ano North District has a diverse population of approximately 6.35×10⁴ persons with different ethnics, religious, and socioeconomic backgrounds (GSS, 2024). We selected four intensive-farming communities, namely, Manfo, Odikrom Nkwanta, Tepa, and Subriso, based on the advice of district agricultural extension officers.

Thematic analysis was performed to analyze the qualitative data obtained from FGDs because it provided a thorough understanding of the perceptions of respondents by identifying common themes and patterns (King and Brooks, 2018). Climatic data were not interpreted arbitrarily and were subjected to rigorous statistical tests and uncertainty assessments to determine the significance and confidence of the observed trends.

Therefore, mean annual rainfall and temperature trends were determined using the Mann-Kendall trend test, which is a valuable method for evaluating trends in climatic data because it has several advantages over other methods. Some of its advantages are as follows: (i) it is a nonparametric test, which means it does not assume any specific data distribution and makes it more robust and flexible to deal with different data types, such as skewed, non-normal, or missing values (Wang et al., 2020); (ii) it can detect linear, nonlinear, and monotonic trends such as consistently increasing or decreasing, which is important because climatic data may exhibit complex and nonlinear patterns over time; and (iii) it is less sensitive to outliers and extreme values, which may distort the results of other methods such as linear regression. Here, the magnitudes of mean annual rainfall and temperature trends were estimated using Sen’s slope estimator (Atta-ur-Rahman and Dawood, 2017). The procedure for calculating the Mann-Kendall trend test was as follows:

where S is the statistic value of the Mann-Kendall trend test; m represents the total number of data points during 2002-2022; k is the earlier time point during 2002-2022; j is the later time point at the time series; sign is the mathematical symbolic function; x_j represents the data point at time point j; and x_k represents the data point at earlier time point k. An extremely high positive value of S denotes an upward trend, while a highly negative value of S denotes a downward trend.

Moreover, it was critical to calculate the probability between the statistic value of the Mann-Kendall trend test and the sample size. This study computed probability according to the study of Forthofer and Lehnen (1981) as follows:

where VAR(S) means the variance of the Mann-Kendall trend test; g is the number of tied groups (a tied group is a set of sample data having the same value); and t_q is the number of data points in the q^th group.

A normalized test statistic value Z was computed as follows:

The probability linked with the normalized test statistic value was computed. The probability density function for a normal distribution with a mean of 0 and a standard deviation of 1 was given as follows:

where f(Z) is the probability density function.

The coefficient of variation (CV) was used to estimate rainfall and temperature variability during the study period. The CV was calculated as follows:

where σ is the standard deviation of rainfall or temperature during the study period and μ is the mean value of rainfall or temperature during the study period.

The relative importance index (RII) was also used to rank the climatic and non-climatic factors perceived by respondents. The RII was calculated as follows:

where W is the weight given to each climatic or non-climatic factor perceived by respondents and ranges from 1 to 2 (where “1” indicates “Yes” and “2” means “No”); and A is the highest weight (A=2 in this case). The range of RII is as follows: 0.000<RII≤1.000.

The effect of socioeconomic characteristics of smallholder farmers on their perceptions of climatic and non-climatic factors was assessed by the binary logistic regression model. The model was shown as follows:

where Y is the dependent variable; p represents the estimated likelihood of the dependent variable; β₀ is the intercept; β₁, β₂, …, β_z are the coefficients of the independent variables; z is the number of independent variables; X₁, X₂, …, X_z are the independent variables; and ε is the error term.

In this study, the dependent variable was smallholder farmers’ perceptions of each climatic or non-climatic factor, while the independent variables were the socioeconomic characteristics of smallholder farmers. A variance inflation factor (VIF) was used to perform multicollinearity tests to check whether the independent variables were highly correlated—a high correlation among independent variables can affect the accuracy and interpretation of the model (Alin, 2010). However, the VIF values of all independent variables were less than 3.0. The VIF value of more than 3.0 for an independent variable in a regression model implies probable multicollinearity (Baffour-Ata et al., 2023a). A 95.0% confidence interval was set for the binary logistic regression model. The fundamental hypothesis was that smallholder farmers’ perceptions of climatic and non-climatic factors would be affected positively or negatively by their socioeconomic characteristics. All statistical analyses were performed using the SPSS version 26 (International Business Machines, Armonk, the United States).

Respondents comprised more males (131 persons) than females (69 persons) (Table 1). This was not surprising as males constitute the highest proportion of the employment sector in Ahafo Ano North District (GSS, 2014). Amengor et al. (2016) and Bolang and Osumanu (2019) reported that males dominate agriculture in Ghana, unlike females who play minor roles in agriculture. This is because farming is regarded as a male-dominated occupation, whereas females are expected to perform domestic and reproductive roles in many rural areas of Ghana (OWP, 2020). Females also face discrimination and stereotypes that limit their access to land, credit, education, agricultural extension services, and markets. Moreover, approximately 58.5% of respondents (117 persons) were between 41 and 60 years old. Szabo et al. (2021) claimed that most smallholder farmers are between 41 and 60 years old in many parts of the world. This finding indicates an aging crisis in agriculture, as young people increasingly choose urban areas over rural areas. This threatens food security and sustainability, as the current generation of experienced smallholder farmers will retire soon, and there will be fewer smallholder farmers to take over the task of growing food for the country.

Specifically, 74.5% of respondents (149 persons) lived in the community above 10 a, and approximately 77.0% of them (154 persons) were natives of the region, implying that they were aware of the past and present climatic conditions in their localities. Most respondents (115 persons; 57.7%) had primary education, indicating that smallholder farmers had some literacy and numeracy skills that could improve their agricultural productivity and income. Most respondents (135 persons; 67.5%) had their farmlands, implying that they had more security, autonomy, and incentives to invest in their land and improve their productivity and income (Tenaw et al., 2009).

Approximately 75.5% of respondents (151 persons) had access to agricultural extension services—this is a good indicator because these services provide smallholder farmers with the necessary goods and services to boost their agricultural output and most importantly increase their awareness of the best adaptation practices to combat climatic and non-climatic factors (Afsar and Idrees, 2019). Less than half of respondents (68 persons; 34.0%) had access to climate and weather information. Such information allows smallholder farmers to make climate-smart decisions that can improve their adaptation and resilience to climate change (Baffour-Ata et al., 2022).

Results showed that mean annual rainfall was variable (CV=0.101) but decreased (Kendall’s tau= -0.110, Sen’s slope= -0.299) from 2002 to 2022 (Fig. 2). The mean annual temperature was also erratic (CV=0.011) but significantly increased (P<0.05) during 2002-2022 in Ahafo Ano North District. Anning et al. (2022) and Baffour-Ata et al. (2023b) suggested that rainfall is erratic and temperature is significantly increasing in most parts of the Ashanti Region in Ghana. The decreasing rainfall and increasing temperature trends significantly impact agricultural production, food security, livelihoods, and income of smallholder farmers in Ahafo Ano NorthDistrict (GSS, 2014). For instance, changes in the timing, amount, and distribution of rainfall affect the availability and quality of water for irrigation, which is essential for many crops, especially during the drought period (Baffour-Ata et al., 2021). Moreover, changes in rainfall patterns affect soil moisture, nutrient cycling, erosion, and salinity, all of which are important factors for crop growth (Rojas Corradi et al., 2019). The changing rainfall patterns also alter the suitability and productivity of different crops in Ahafo Ano North District because different crops have different water requirements and tolerance levels to drought or flood (Kaur and Kaur, 2017).

The quantity and quality of agricultural products and the losses in property and other aspects caused by climate change in Ahafo Ano North District were reported by the District Agricultural Development Unit and the Ghana Living Standards Survey (GLSS). For instance, the average crop yield, biomass, and protein content of grain declined by 15.0%, 12.0%, and 10.0% in the study area, respectively, because of changes in rainfall and temperature (MoFA, 2024). The reports also showed that the average cocoa yield and quality decreased by 20.0% and 25.0%, respectively (GSS, 2024; MoFA, 2024). The GSS (2024) indicated that the average income, asset value, and health status of smallholder farmers declined by 18.0%, 22.0%, and 15.0%, respectively, and the number of floods, droughts, and pest outbreaks increased by 30.0%, 25.0%, and 35.0%, respectively. This information provided evidence for the negative impacts of climate change on the quantity and quality of agricultural products, as well as the losses in terms of property and other aspects (e.g., income reduction, increased costs, reduced agricultural productivity, and increased health risks) in Ahafo Ano North District.

The results indicated that the most ranked climatic factor was extreme heat or increasing temperature (RII=0.498) (Table 2). This was followed closely by erratic rainfall (RII=0.485). Increased windstorms ranked the third (RII=0.475); however, low temperature ranked the lowest (RII=0.035). These findings were validated by FGDs.

Smallholder farmers’ perceptions of erratic rainfall and extreme heat or increasing temperature were consistent with the trend analysis results of mean annual rainfall and temperature in Ahafo Ano North District from 2002 to 2022 (Fig. 2). Similarly, Asamoah and Ansah-Mensah (2020) reported that smallholder farmers in the Bawku area of Ghana perceived increasing temperatures and erratic rainfall as key climatic factors impacting their communities. Moreover, Guodaar et al. (2021) also reported that smallholder farmers in the upper east, upper west, and northern regions of Ghana experienced droughts and intensified heat waves.

Extreme heat or increasing temperature was a critical climatic factor because of its adverse effects on crop production, income, and food security. It can reduce crop yields by affecting the growth, development, and quality of crops and increasing water evaporation from the soil and plants, leading to water stress and drought (Harvey et al., 2018). It can also increase the incidence and severity of pests and diseases that damage crops and reduce their marketability. Furthermore, extreme heat or increasing temperature can affect the health and well-being of smallholder farmers, their families, and livestock by causing heat stress, dehydration, and heat-related illnesses (Harvey et al., 2018). These impacts can reduce the income and food security of smallholder farmers, especially those who depend on rainfed agriculture and have limited access to irrigation, inputs, technology, and insurance. Therefore, extreme heat or increasing temperature was a major challenge for smallholder farmers in Ahafo Ano North District as well as in many other regions worldwide. For example, extreme heat or increasing temperature has reduced the suitability and productivity of some crops, such as cocoa, which is the main cash crop and export commodity of Ahafo Ano North District (Oyekale, 2021). Cocoa requires a humid and cool climate, with optimal temperature in the range of 18.0°C-32.0°C. However, extreme heat or increasing temperature has increased the evapotranspiration and water stress of cocoa trees, leading to lower yield and quality in Ahafo Ano North District. In addition, extreme heat or increasing temperature has increased the vulnerability of crops to pests and diseases, such as the cocoa swollen shoot virus, thereby causing significant losses and income reduction.

Erratic rainfall severely affected crop production, income, and food security of smallholder farmers. Erratic rainfall leads to some adverse consequences such as reduced crop yields, increased incidence and severity of pests and diseases, reduced farm income because of lower yields, higher production costs and lower market prices, and reduced food security (Boillat et al., 2019; Mujeyi et al., 2021). In Ahafo Ano North District, erratic rainfall has affected pest and disease dynamics because some pests and pathogens have benefited from the changing moisture and temperature conditions. For example, erratic rainfall increased the incidence and severity of fungal and bacterial diseases, such as cocoa black pod, reducing cocoa yield and quality (Amfo and Ali, 2020; Oyekale, 2021). Erratic rainfall has also favored the emergence and spread of insect pests such as stem borers, which cause significant damage to maize and other crops in Ahafo Ano North District (Baffour-Ata et al., 2023b).

In addition to these factors, increased windstorms were perceived as a critical climatic factor by smallholder farmers because they can cause severe damage to their crops, livestock, infrastructure, and livelihoods. For instance, increased windstorms destroy crops by uprooting, breaking, or flattening crops (FAO, 2020). Furthermore, increased windstorms can harm livestock by injuring, killing, or displacing them, reducing their productivity and health. They damage infrastructure by blowing away roofs, fences, irrigation systems, and storage facilities, increasing production costs and postharvest losses (FAO, 2020). Increased windstorms enhance vulnerability by exposing smallholder farmers to heightened risks of hunger, poverty, debt, and displacement (FAO, 2020).

In contrast, low temperature was ranked as the least climatic factor in Ahafo Ano North District. Compared with other climatic hazards such as drought, flooding, heat stress, pests, and diseases, low temperature events were less likely to occur and had less impact on crops and livestock. A survey of 860 smallholder farmers of coffee and primary grain in Central America revealed that only 9.0% of smallholder farmers reported negative impacts of low temperature on crop production, 87.0% of smallholder farmers claimed that drought has negative impacts on crop production, 66.0% of them thought that heat stress had negative impacts on crop production, 54.0% of smallholder farmers found negative impacts of pests and diseases on crop production, and 40.0% of smallholder farmers discovered negative impacts of flooding on crop production (Harvey et al., 2018). Similarly, a study of 600 smallholder farmers in Ethiopia showed that low temperature was the least frequently observed climatic factor, and it affected only 8.0% of smallholder farmers, whereas drought, erratic rainfall, and high temperature affected 98.0%, 96.0%, and 88.0% of smallholder farmers, respectively (Frost et al., 2023). One possible reason for the low perceptions of low temperature as a climatic factor is that most smallholder farmers live in low-latitude regions such as Africa, South and Southeast Asia, and Latin America, where mean temperature is relatively high and temperature variability is low. Therefore, low temperature is not a common or severe threat to agricultural systems.

Another possible reason is that some crops and livestock can tolerate or even benefit from low temperatures as long as the temperature does not reach freezing levels. For example, some coffee varieties can produce higher-quality beans in cooler areas (Woetzel et al., 2020), and some animals can cope with cold weather by growing thicker fur or feathers. However, this does not mean that low temperatures do not pose a problem for smallholder farmers. In some regions, such as the highlands of East Africa and South America, low temperatures caused frost damage to crops, thereby reducing crop yields (Adhikari et al., 2015). Low temperatures can also affect the health and productivity of livestock by increasing their energy requirements, reducing their feed intake, and lowering their immunity (Savsani et al., 2015). Moreover, climate change may increase the frequency and intensity of cold spells in some areas, posing new challenges for smallholder farmers who are unprepared or have not adapted to cope with them. Therefore, although low temperature may not be the most pressing climatic factor for smallholder farmers in general, it should be considered and addressed in some situations.

The results indicated that the highest ranked non-climatic factor perceived by respondents was high cost of farm inputs (RII=0.485) (Table 3). This was followed by high cost of healthcare (RII=0.435). Poor condition of roads to farms was ranked the third perceived by respondents (RII=0.415). The lowest ranked non-climatic factor was lack of electricity (RII=0.095). These findings were validated using FGDs.

These findings are in line with those of Antwi-Agyei et al. (2017), who reported that smallholder farmers in Ghana perceived high cost of farm inputs, lack of money, and limited access to the market as the key non-climatic factors. Smallholder farmers perceived high cost of farm inputs as a critical non-climatic factor because it affects their profitability and productivity. Farm inputs such as seeds, fertilizers, pesticides, and irrigation systems are essential for improving crop yields and quality and reducing crop losses and environmental impacts (Fan and Rue, 2020). However, many smallholder farmers face barriers to accessing these inputs, because of a lack of credit, insurance, information, infrastructure, and market linkages (Langyintuo, 2020). Therefore, they often have to buy inputs at relatively high prices from informal or unreliable sources or use low-quality or inappropriate inputs that may harm their crops or the environment (Langyintuo, 2020). These factors can reduce the returns on investment and income of smallholder farmers and limit their ability to adopt more sustainable and resilient farming practices.

High cost of healthcare was also perceived as a critical non-climatic factor in Ahafo Ano North District because it can affect the health, income, and food security of smallholder farmers and their families. According to the World Business Council for Sustainable Development (2023), smallholder farmers are highly vulnerable to health shocks and diseases, such as malaria, tuberculosis, etc., because of their limited access to resources, credit, and basic healthcare facilities. These health problems can reduce their productivity and income, and increase their medical expenses (World Economic Forum, 2022). Moreover, high cost of healthcare can force smallholder farmers to sell their assets (e.g., land, livestock, or crops) to pay for their treatment or cope with income losses. This can reduce their food security and resilience.

Poor condition of roads to farms negatively affected the production, marketing, and income of smallholder farmers in Ahafo Ano North District. It reduces crop quality and quantity because of increased vibrations and damage during transportation (Steyn, 2016). Additionally, Lokesha and Mahesha (2016) claimed that poor condition of roads to farms increases transport costs because of higher fuel consumption, vehicle maintenance, and repair expenses. Such roads also reduce the market value of crops and affect access through delays, spoilage, and loss of crops; and they also reduce farm income because of lower returns on investment, higher production costs, and lower market prices (Sieber and Allen, 2016). Therefore, poor condition of roads to farms poses a substantial challenge for smallholder farmers, especially in Ghana.

The lowest ranked non-climatic factor was lack of electricity, possibly because smallholder farmers in Ghana did not rely heavily on electricity for their farming activities. Most smallholder farmers use traditional or manual methods of cultivation, harvesting, processing, and storage that do not require electricity. Moreover, some smallholder farmers in rural areas of Ghana may have access to off- or mini-grids that provide reliable and affordable electricity for remote areas. Therefore, lack of electricity may not affect the productivity, income, or food security of smallholder farmers to the same extent as other non-climatic factors. However, this does not mean that electricity is not essential or beneficial for smallholder farmers. Electricity can help them improve the efficiency, quality, and added value of their products and increase access to information, markets, and services that can enhance their livelihoods and resilience. Therefore, it is important to improve the access, reliability, affordability, and sustainability of electricity for smallholder farmers.

Results showed that socioeconomic characteristics affected smallholder farmers’ perceptions of climatic factors (Table 4). For example, gender significantly affected (β= -1.348, P=0.001) smallholder farmers’ perceptions of increasing rainfall in Ahafo Ano North District. Males and females may have different tasks and expectations related to farming, such as crop selection, land preparation, irrigation, harvesting, and marketing. These tasks and expectations may influence how they observe and interpret the patterns and impacts of rainfall. For example, females may be more concerned about water availability for domestic use and food security, whereas males may focus more on income generation and market opportunities (Piya et al., 2019). However, mean annual rainfall indicated a decreasing trend in Ahafo Ano North District (Fig. 2a). This trend agrees with the fact that males and females perceived rainfall to be decreasing in the study area (Table 2).

From Table 4 we can see that the age of smallholder farmers had a significant effect on their perceptions of extreme heat or increasing temperature (β=7.773, P=0.032), erratic rainfall (β=2.931, P=0.028), increased incidence of floods (β=0.823, P=0.050), and increasing rainfall (β= -1.186, P=0.004). For example, older smallholder farmers have different levels of exposure and adaptation to extreme heat or increasing temperature compared with younger smallholder farmers. These differences may affect smallholder farmers’ perceptions and interpretations of temperature changes. For instance, older smallholder farmers may not have difficulty remembering past weather patterns, paying attention to current weather forecasts, processing information quickly, and making appropriate decisions based on temperature changes (Eggenberger et al., 2021). It is important to highlight that most smallholder farmers surveyed for this study were older smallholder farmers (Table 1). Additionally, age may affect the perceptions of erratic rainfall for several reasons. For example, older smallholder farmers may have more experience and memory of past rainfall patterns and how they have changed over time. They may be able to compare the current rainfall situation with their historical observations and notice deviations and irregularities (Habtemariam et al., 2016). Furthermore, older smallholder farmers may be more exposed and vulnerable to the effects of erratic rainfall, such as droughts, floods, crop failures, and food insecurity. These factors may make them more aware of and concerned about rainfall variability and its impacts on their livelihoods (Huda, 2013).

Furthermore, smallholder farmers’ perceptions of non-climatic factors were affected by their socioeconomic characteristics. From Table 5 we can see that access to agricultural extension services significantly affected their perceptions of lack of money (β=1.290, P=0.026), high population (β= -1.618, P=0.006), and damage to crops by livestock (β=1.427, P=0.013). Access to agricultural extension services affected their perceptions of lack of money in different ways such as the quality, relevance, and availability of services. According to some studies (Danso-Abbeam et al., 2018; Maake and Antwi, 2022), access to agricultural extension services had positive effects on farm productivity and income, in turn reducing poverty and food insecurity among smallholder farmers. However, these effects were not uniform and depended on various factors such as the education level and age of smallholder farmers, farm size, and type of extension methods used (Maake and Antwi, 2022). Some smallholder farmers perceive access to agricultural extension services as ineffective if they do not meet their needs, preferences, and expectations. For example, some smallholder farmers prefer extension services that are demand-driven and equitable and that prioritize their problems and interests (Maake and Antwi, 2022). Others may value extension services that provide information on improving agricultural production, such as cultivation practices, fertilization, plant protection, and marketing. Moreover, some smallholder farmers may need access to the technologies required for implementing the extension recommendations, such as improved seeds, fertilizers, pesticides, and irrigation systems. Thus, if access to agricultural extension services is effective and responsive to the needs and conditions of smallholder farmers, they can help them increase their farm output and income, thereby improving their livelihoods. However, if access to agricultural extension services is ineffective and irrelevant to the situations and goals of smallholder farmers, it can generate frustration and dissatisfaction among smallholder farmers, thus worsening their sense of poverty.

In addition, estimated farm income per season of smallholder farmers significantly affected their perceptions of high cost of farm inputs (β= -5.399, P=0.034), high cost of education (β=0.742, P=0.046), lack of drinking water (β=0.984, P=0.006), and damage to crops by livestock (β=0.989, P=0.013). For instance, estimated farm income per season affected their perceptions of high cost of farm inputs in different ways. According to Myers (2022), farming input costs are rising faster than commodity prices, making it harder to break even. Some of the main inputs that have become more expensive are fertilizers, pesticides, seeds, and land. Some smallholder farmers may perceive high cost of farm inputs as a major challenge threatening their livelihoods and food security. They will struggle to afford the necessary inputs to maintain or increase production and quality. They may face difficulty in accessing credit or insurance to cope with price volatility and weather risks. They have to reduce their input use, diversify their crops, adopt low-cost technologies, or seek off-farm income sources to survive (FAO, 2016). Other smallholder farmers perceive high cost of farm inputs as an opportunity to improve their efficiency and competitiveness. They invest in inputs with high returns and benefits, such as improved seeds, fertilizers, andirrigation systems, and adopt best management practices, such as integrated pest management, conservation agriculture, and precision farming to optimize their input use and reduce their environmental impact, thereby boosting their farm income, and also seek market information, value addition, certification, or collective action to enhance their bargaining power and profitability (FAO, 2016).