Navigating Climate Challenges: Insights from smallholder of Mbangassina’s Local Communities in the Centre Cameroon Region

doi:10.21203/rs.3.rs-4997053/v1

This comprehensive study explored the factors shaping smallholder farmers’ perceptions of climate change in the Mbangassina Subdivision. Data collection methods included structured interviews, focus group discussions, and key informant interviews. The findings revealed that 97.2% of the farmers surveyed noticed changes in temperature and rainfall over the past 30 years. Specifically, 17.8% observed an increase in rainfall, 75.6% noted a decrease, and 6.7% saw no change over the last 10–30 years. Additionally, there has been a significant rise in mean and maximum temperatures. From 1981 to 2022, Mbangassina recorded an average annual rainfall of 1491.26 mm, with yearly fluctuations influenced by seasonal variations, geographical factors, and climatic oscillations. While linear regression offers a simplified model, other factors such as greenhouse gases, natural variability, and regional effects also impact temperature changes. The study identified significant associations between perceived climate change impacts and factors such as age, years of farming experience, annual income, microfinance/banking status, gender, community membership, and socioeconomic level (residential situation) (p < 0.050). Furthermore, adaptive capacity to climate change impacts was influenced by years of farming experience, a usable agricultural area, annual income, microfinance client status, marital status, residential situation, and membership in a farmer organization. The research underscores the importance of understanding local factors that influence climate change perceptions. This knowledge can aid smallholder farmers in better coping with climate challenges and inform the development of effective adaptation strategies. Policymakers can create an enabling environment that empowers smallholder farmers to adapt to climate change and improve their livelihoods.

perception

climate change

factors

adaptation strategies

smallholder

Smallholder farmers play a vital role in rural economies. However, they are the most vulnerable to the effects of climate change (FAO, 2015). There are 500 million smallholder farms around the world that support the livelihoods of approximately 2 billion people [1]. These farmers live in particularly vulnerable areas, including mountain slopes, deserts, and floodplains [2], [3], [4], [5]. Climate change multiplies the threats they face. It endangers the natural resources on which they depend and accelerates environmental degradation [6]. These farmers receive only a tiny fraction of global climate finance, despite their increased vulnerability to climate impacts. According to a report by the UN's International Fund for Agricultural Development (IFAD) and the Climate Policy Initiative (CPI), smallholder farmers in developing countries receive only 1.7% of such finance [7]. These farmers produce 50% of the world's food calories. But they face significant challenges. Rising temperatures, droughts, and floods are destroying their crops and livestock. They are struggling to make a living [8]. Yet they urgently need climate finance to adapt their production and sustain their livelihoods (FAO, 2021). Over the centuries, smallholder farmers have been adaptors to environmental change and climate variability (FAO, 2015). However, the speed and intensity of climate change are now outpacing their ability to adapt. Crop failures and livestock mortality are causing economic losses, food price increases, and threats to food security, particularly in sub-Saharan Africa (FAO, 2021). In combating the effects of climate change, agriculture and forestry play an important role. We can significantly reduce greenhouse gas emissions CIRAD (2023) by adopting better soil management practices, sustainable agriculture, and promoting reforestation. Poor farmers are the precious custodians of our natural resources, often managing large tracts of land and forests. By strengthening their adaptive capacity, we can promote green growth and poverty reduction at the same time [10]. To scale up investments in sustainable agriculture, the International Fund for Agricultural Development (IFAD) is working on several fronts. In partnership with the rural poor, it is focusing on risk prevention, adaptive capacity, governance, and solutions based on local knowledge. Climate change brings with it new challenges, but also new opportunities. In 2020, IFAD adopted a strategy to strengthen the capacity of the rural poor to adapt to climate change by seeking their views throughout the planning process. In short, it is essential to promote innovative approaches to help smallholder farmers build resilience to climate change and benefit from the mitigation measures and financing available to them (FAO, 2021).

Cameroon's revised Nationally Determined Contribution (NDC) for the period 2020–2030 targets a 35% reduction in the carbon footprint of the country's development by 2030 compared to the 2010 baseline. This reduction will be achieved without compromising economic growth. Mitigation options with high co-benefits will be prioritized. Targeted sectors include agriculture, forestry, energy, and transport. Cameroon's agricultural sector is significantly affected by climate change. Cameroon is 80% dependent on agriculture. Climate variability has led to changes in the agricultural calendar, with direct impacts on production and food security in this developing country [11]. Women farmers are disproportionately affected by the effects of climate change. The rains don't come at a fixed time, which makes it impossible for farmers to work properly [12]. They no longer know whether to plant or not because the dry season is either long or absent. This reduces production [13]. Agricultural vulnerability is now a reality, especially in sub-Saharan Africa where agriculture is entirely seasonal [14], [15], [16]. It is important to note that climate change has different impacts in different regions and on different crops [17], [18], [19], [20]. Climate change is also a driver of conflict in Cameroon, forcing thousands of people to flee their homes [21], [22]. Climate change is directly affecting more than 70% of the population whose livelihoods depend on farming (Lambi and Kometa, 2014).

In Cameroon, the climate plays a critical role in agricultural production. Low rainfall can hurt crop development and productivity [23], [24]. The main impact of climate change in the country is a reduction in agricultural production, which requires adaptation measures. Agriculture is one of the key sectors of the economy of Cameroon. According to the Ministry of Agriculture and Rural Development, its contribution to GDP is 22.9% and its share in total exports is about 23% (Lambi and Kometa, 2014; Molua, 2007). However, according to the World Bank, by the year 2022, agriculture, forestry, and fisheries will account for 16.98% of the GDP of Cameroon. Agriculture is directly responsible for the livelihoods of about 36% of adult Cameroonians or 2.75 million households. Agriculture is also the main source of employment in Cameroon. It employs about 60% of the working population, mainly on family farms.

Climate change poses a serious threat to the livelihoods and food security of smallholder farmers in Cameroon, especially in areas that are already vulnerable to climate variability and socio-economic challenges [7], [15], [25], [26]. There is a lack of research on the same topic in Cameroon, especially in the Mbangassina subdivision, although there are several studies on factors influencing smallholder farmers' perceptions of climate change in sub-Saharan Africa [15], [27], [28], [29], [30]. Smallholder farmers in the Mbangassina subdivision depend on food crops and cocoa production for their survival. However, this subdivision is also affected by landlocked conditions and poverty. This exposes farmers to complex problems and multiple stressors and uncertainties that influence their perceptions and responses to climate change. Understanding how these smallholders perceive and cope with climate change is important. What local factors influence how smallholders in Mbangassina perceive climate change? How do these perceptions affect how they choose to adapt? This question explores local farmers' specific perceptions and knowledge of climate change, and how they influence how they choose to adapt. This study aims to provide an in-depth analysis of the factors influencing how smallholders in the Mbangassina Subdivision perceive climate change to be changing.

First, we will assess smallholders' current knowledge of climate change. We will measure their understanding of its causes and effects. Next, we will identify vulnerability factors. We will look at the socio-economic, geographic, and environmental aspects that make smallholders more vulnerable to climate change. This may include their access to resources and their dependence on certain types of crops. Finally, we will look at adaptation practices in place: what strategies are farmers currently using to cope with climate change? Through the identification of these factors, we will formulate practical recommendations on how to strengthen their capacity to adapt, while maintaining their food security and livelihoods. These recommendations could include training, infrastructure, and local policies, contributing to concrete solutions to support smallholder farmers in the face of climate challenges.

Study area

Mbangassina, nestled in the heart of the Centre Region of Cameroon, is a place where community resilience meets the challenges posed by climate change. Geography and Location: Coordinates: Situated at 4°37’47" N latitude and 11°42’25" E longitude (Fig. 1). Encompassing approximately 638 square kilometers of land. Riverside Charm: Mbangassina straddles both banks of the Mbam River, adding to its picturesque landscape. The origin of the name “Mbangassina” carries intriguing tales: Local Mourning: Some believe it results from the fusion of two words: “Mbanga” (meaning “I weep” in the local language) and “Messina” (the name of a young wife who tragically passed away). Mispronunciation: Another theory suggests that it emerged from a mispronunciation of the word “magasin” (store), as a large rice-processing and storage facility once stood in the area (PCD, 2013). It has 20 villages: Badissa, Biakoa, Bialanguena, Biapongo, Biassamba, Biatangana, Biatombo, Bilomo, Bindamongo, Boukê, Boura I, Boura II, Enangana, Goura, Nyamanga II, Nyamballa, Talba, Tchamongo II, Yanga et Yebekolo (PCD, 2013).

Mbangassina has an equatorial climate. It is characterized by four distinct seasons: Short and long dry, short and long rainy. The average temperature is 23–25°C. At the time of the 2005 census, Mbangassina had a population of 41,180. 4,306 people lived in the town itself. In developing adaptation strategies, the community's knowledge and perceptions of climate change play a key role. The main language spoken in Mbangassina is Ossananga, a variant of Tuki. It is widely spoken outside the region [18], [31]. Through dictionaries and grammars, efforts are underway to document and preserve this ancient language. The Mbangassina are largely subsistence-based. Crops grown include plantains, cassava, peanuts, and maize. Agriculture provides a livelihood. There is also untapped tourism potential. Educational facilities include public and private nursery schools, primary and secondary schools, and vocational training institutions. The community has a strong commitment to the improvement of education opportunities for its young people. With its rich history, its resilient people, and its environmental challenges, Mbangassina is a microcosm of the global struggle to combat climate change.

Methodology

Methodological framework

Drawing on various publications on climate change technologies in different contexts and periods, the methodological framework developed in this sub-section. It focuses on the capacity of rural households to adapt to change. Institutional, economic, political, environmental, and socio-demographic factors are the main aspects considered. In shaping adaptation strategies, specific socio-economic elements play a crucial role. These include land tenure, the availability of labor to adapt to climate change, the cost of agricultural inputs, the number of livestock owned, and the social activities and capital of farmers. In addition, factors such as farming experience and involvement in off-farm activities influence household food security decision-making. From a climate change perspective, this framework aims to influence multiple dimensions of the household environment. These include biophysical, agricultural, socio-cultural, and socio-economic aspects. Resources, assets, and risk factors are considered. The framework includes research design, literature review, field data collection, and data analysis methods (Fig. 2).

The study adopts a mixed-methods approach to explore the factors influencing smallholder farmers' local perceptions of climate challenges in the Mbangassina subdivision. The study area is located in the north of Cameroon's central region, where smallholder farmers depend on food crop production for their livelihoods. The study comprises three main stages: documentary analysis, quantitative survey, and qualitative interviews.

Literature review

A comprehensive review of relevant literature on climate change, small-scale farming, and local perceptions was carried out. This review aimed to identify the key concepts, theories, and frameworks underpinning the study. It also helped formulate the research questions and hypotheses.

Quantitative survey

A structured questionnaire was designed and administered to a sample of 120 small-scale farmers in the Mbangassina subdivision. The questionnaire collects data on respondents' socio-demographic characteristics, farming practices, perceptions of climate change, and coping strategies. Data are analyzed using descriptive, multivariate, and inferential statistics, such as frequency, mean, standard deviation, correlation, and regression.

Qualitative interviews

A semi-structured interview guide is developed and used to conduct in-depth interviews with 120 respondents selected for the survey. The purpose of the interviews is to gather respondents' views, experiences, and stories about climate change and its impact on their food crop production and livelihoods. The interviews are recorded, transcribed, and analyzed using thematic analysis. The study adopts a convergent parallel design, in which quantitative and qualitative data are collected and analyzed separately, then integrated and compared in the discussion and conclusion sections. In addition, the WorldClim website (https://power.larc.nasa.gov/data-access-viewer/) provides online bioclimatic data coordinates from 1981 to 2022. Agroclimatic data on precipitation and temperature were combined with altitude and latitude to describe the environmental characteristics of each site's location.

Data Analysis

The data analysis consists of two parts: quantitative and qualitative. The quantitative data analysis is performed using Jamovi software, while the qualitative data analysis is performed using SPSS software.

Quantitative Data Analysis

The quantitative data analysis aims to test the hypotheses and answer the research questions related to the factors influencing the local perceptions of smallholder farmers in the Mbangassina subdivision regarding climate challenges. The analysis involves the following steps:

Data cleaning and screening: The data collected from the questionnaire is checked for missing values, outliers, and errors. The data is also screened for normality, linearity, and multicollinearity assumptions.
Descriptive statistics: The data is summarized using descriptive statistics, such as frequency, mean, standard deviation, and percentage. The descriptive statistics provide an overview of the socio-demographic characteristics, farming practices, climate change perceptions, and adaptation strategies of the respondents.
Inferential statistics: The data is analyzed using inferential statistics, such as correlation, chi-square, t-test, and regression. The inferential statistics test the relationships and differences among the variables of interest. The regression analysis is used to identify the factors that influence the local perceptions of climate change and the adoption of adaptation practices by smallholder farmers. The regression models are evaluated based on the R-squared, F-test, and significance of the coefficients.
Results presentation: The results of the quantitative data analysis are presented using tables, graphs, and charts. The results are also interpreted and discussed concerning the research questions and hypotheses.

Qualitative Data Analysis

The qualitative data analysis aims to explore the views, experiences, and stories of smallholder farmers related to climate change and its impacts on their food crop production and livelihoods. The analysis involves the following steps:

Data transcription and translation: The data collected from the interviews is transcribed verbatim and translated from the local language to English. The transcripts are checked for accuracy and completeness.
Data coding and categorization: The data is coded and categorized using SPSS software. The coding process involves assigning labels or tags to the segments of the data that reflect the meaning or theme of the text. The codes are then grouped into categories based on their similarity or relevance.
Data interpretation and synthesis: The data is interpreted and synthesized using thematic analysis. The thematic analysis involves identifying, analyzing, and reporting the patterns or themes that emerge from the data. The themes are then linked to the research questions and the literature review.
Results presentation: The results of the qualitative data analysis are presented using quotes, narratives, and diagrams. The results are also interpreted and discussed concerning the research questions and the literature review.

To analyze perceptions in studies, the multinomial logistic model is often used. However, in the case of farmers who simultaneously selected multiple climate change perceptions and adaptation strategies, this model may not be appropriate to avoid confounding the interpretation. Therefore, to analyze climate change perceptions and adaptation strategies among smallholder farmers in Mbangassina, we chose a binary logistic regression method. Several papers, including [32], [33], [34], [35], and [36], have also used this approach to study adaptation strategies in farming systems. Through this analysis, we will have a better understanding of smallholder farmers' perceptions of climate change and the benefits of adaptation in these contexts.

Analysis of climatic trends

Excel's pivot table function was used to aggregate daily temperature and rainfall data into monthly averages. The monthly data were then imported into Jamovi version 2.4.8 and analyzed. Jamovi's graph generator was used to describe rainfall and temperature trends [37]. Linear regression was used to determine temporal trends in annual rainfall over the short rainy seasons (March-June) and long rainy seasons (August-November), and the R² value was used to determine whether or not the trends were significant. The data analysis results allow us to present the findings in the context of perception and practices for climate change adaptation in the Mbangassina forest-savanna transition zone. Le test de Mann-Kendell (MK), un test non paramétrique de rang, a été utilisé pour tester la signification des tendances de la température et des précipitations sur la base des données météorologiques obtenues. Le test de Mann-Kendell a été développé par Mann et Kendell et est fondamental pour détecter la linéarité des tendances. Dans ce test, l'existence (H₁) et la non-existence (H₀) des tendances ont été testées. Le test de Mann-Kendell a été exécuté sur XLSTAT.

Socio-demographic characteristics

The analysis results indicate that 26.7% of respondents were women and 73.3% were men, with no significant differences observed between communities (p = 0.705). Regarding education levels, 8.9% had received Quranic education, 8.9% had non-formal education, 35.6% had primary education, and 46.7% had secondary education, again showing no significant differences between municipalities (p = 0.651). Marital status among respondents revealed that 11.1% were single, 11.1% were widowed, 64% were married, and 13.3% were divorced, with no significant differences between communities (p = 0.1961). Additionally, 97.8% of respondents were permanent residents, while 2.2% were temporary residents, with no significant difference between communities (p = 0.367). Community membership showed that 82.2% were members and 17.8% were not, with no significant difference between communities (p = 0.587). In terms of access to agricultural credit, 53.3% of respondents had access, while 46.7% did not, with no significant difference between communities (p = 0.236). Finally, 40.0% of respondents had received agricultural training, while 60% had not, with a significant difference found between communities (p = 0.004).

The results indicate that the average age of respondents ranged from 32.7 to 47 years, with no significant differences between villages (F = 0.47, p = 0.711). Household sizes varied from 2.9 to 6.9 members, and the statistical test (F = 1.28, p = 0.291) confirmed no significant differences between villages. Farming experience among respondents ranged from 10 to 20.6 years, with the statistical test (F = 0.61, p = 0.611) indicating no significant differences between villages. The average farm size was between 2.4 and 4.5 hectares, and the statistical test (F = 0.76, p = 0.521) showed no significant differences in useful agricultural areas between villages. Annual incomes ranged from 2,834,000 to 9,089,533.3 FCFA, with the statistical test (F = 0.22, p = 0.881) confirming no significant differences between regions (Fig. 3).

Variations in weather indicators

Temperature trends from 1981 to 2022

The results indicate a significant increase in mean and maximum temperatures. Consequently, hypothesis H₁, which suggests the existence of a trend, was accepted for precipitation, maximum, and mean temperatures. Conversely, hypothesis H₀, which suggests the absence of a trend, was accepted for minimum temperature (see Table 1). It is important to note that although the Mann-Kendall test accepts H₀ for minimum temperature, Sen’s slope generally shows an increasing trend for minimum, maximum, and mean temperatures (Table 1).

Table 1

The Mann-Kendall trend test for rainfall and temperature at Mbangassina.
Variables	Rainfall	Temperature
Variables	Rainfall	Mean	Maximum	Minimum
Kendall’s tau	-0.385	0.663	0.446	0.131
S	-331.00	568.00	384.00	107.00
Var (S)	8507.67	8504.67	8513.34	7925.67
p-value (Two-tailed)	0.00034	7.832e^− 10	3.311e^− 05	0.23
Alpha	0.05	0.05	0.05	0.05

The bold values indicate statistically significant values at 5% or below.

Three types of temperature, average, maximum, and minimum, are shown on the graph. All three temperature categories show a positive slope over the years. This indicates an overall increase in temperatures. The linear regression lines for each type of temperature are as follows: (i) maximum temperature: (y = 0.0984x + 31.226), with an (R²) value of 0.3124; (ii) average temperature: (y = 0.0369x + 23.079), with an (R²) value of 0.714; and (ii) minimum temperature: (y = 0.0206x + 14.825), with an (R²) value of 0.0332 (Fig. 4). Positive slopes indicate a warming trend over the period studied. R² indicates how well linear regression lines fit data. Higher R²'s indicate a better-fitted model. Linear regression provides a simplified model. Other factors (such as greenhouse gases, natural variability, and regional effects) contribute to temperature changes.

Precipitation trends from 1981 to 2022

The data show that between 1981 and 2022, Mbangassina received an average of 1491.26 mm of rain per year. During this period, there have been fluctuations in annual rainfall (see Fig. 5). The graph shows the amount of rainfall in millimeters (mm). The blue line fluctuates between about 1,000 mm and a little more than 2,000 mm. A slight decrease in precipitation over the years is shown by the linear trend line (dotted line). The linear trendline is equated as follows (y = -10.933x + 1751.8). The value of the coefficient of determination (R²) (0.2378) indicates that this linear model is not a perfect fit for the data. This decreasing trend indicates that annual precipitation continues to decrease. Several factors may be contributing to this trend. These include climate variability, land use change, and regional influences. Precipitation patterns are complex and influenced by many interactions, including seasonal variations, geographic features, and climate variability.

Farmers' perceptions of climate-related changes and factors influencing perceptions

According to the statistical analysis, 97.2% of the farmers interviewed said that they had noticed a change in temperature and rainfall in the last 30 years. The results of analyzing perceived climate change and its effects in different villages: Biakoa, Bilomo, Mbangassina, and Talba in the Mbangassina subdivision. For the late arrival of the rains, 42.2% of the respondent farmers noted this phenomenon, and for the rain spots in the middle of the dry season, 13.3% of the respondent farmers noted this phenomenon. For the irregularity of the rainy season, only 2.2% of the farmers interviewed noted this phenomenon. Finally, 42.2% of the respondents noted the phenomenon of a long and intense drought. This variable showed a significant difference between villages (p = 0.035). About the perception of rainfall in the last 10–30 years, 17.8% of the farmers interviewed noted an increase in rainfall, 75.6% of the farmers interviewed noted a decrease in rainfall and 6.7% of the farmers interviewed noted no change. The p-value is 0.001. There is a significant difference between the different villages. Concerning temperature over the past 10–30 years, 20.0% of respondents reported no change, 77.8% of respondents reported an increase in temperature, and only 2.2% of respondents reported a decrease in temperature, with p = 0.025. Regarding drought over the past 10–30 years, 25.0% of respondents reported no change, 72.8% of respondents reported an increase in drought, and only 2.2% of respondents reported a decrease in drought, with p = 0.023. About the onset of the dry season over the past 10–30 years, 48.9% of respondents reported a later onset of the dry season, 8.9% of respondents reported no change, and 42.2% of respondents reported an earlier onset of the dry season, p = 0.424 (Table 2).

Regarding the end of the dry season in the last 10–30 years, 4.4% of respondents found no change, 2.2% of respondents found a later end of the dry season, and 93.3% of respondents found an earlier end of the dry season with p = 0.222. Regarding the opinion on these changes, 42.2% of the respondents found these changes acceptable, only 2.2% of the respondents found these changes good, and 55.6% of the respondents found these changes bad. The p = 0.010 indicates that there is a significant difference between the villages. Finally, regarding the manifestation of climate change, 11.1% of the respondents did not observe any manifestation of climate change and 88.9% of the respondents observed a manifestation of climate change, with p = 0.020.

Table 2

Farmers' perception of climate change.
Perception	Modality	Biakoa	Bilomo	Mbangassina	Talba	Total	p value
Seasonality	Late arrival of rains	36.4%	40.0%	36.4%	53.8%	42.2%	0.035²
	Spots of rain in the middle of the dry season	0.0%	0.0%	9.1%	38.5%	13.3%
	Irregular rainy season	9.1%	0.0%	0.0%	0.0%	2.2%
	Long, intense drought	54.5%	60.0%	54.5%	7.7%	42.2%
Rainfall over the last 10–30 years	Increased	45.5%	20.0%	9.1%	0.0%	17.8%	0.001²
	Decreased	27.3%	80.0%	90.9%	100%	75.6%
	Unchanged	27.3%	0.0%	0.0%	0.0%	6.7%
Temperatures over the past 10–30 years	No change	27.3%	0.0%	0.0%	46.2%	20.0%	0.025²
	Increased	72.7%	100%	100%	46.2%	77.8%
	Decreased	0.0%	0.0%	0.0%	7.7%	2.2%
Drought over the past 10–30 years	No change	27.3%	0.0%	0.0%	46.2%	25.0%	0.023²
	Increased	72.7%	100%	100%	46.2%	72.8%
	Decreased	0.0%	0.0%	0.0%	7.7%	2.2%
Beginning of dry season during the last 10–30 years	Late	45.5%	60.0%	54.5%	38.5%	48.9%	0.424²
	No change	0.0%	0.0%	9.1%	23.1%	8.9%
	Early	54.5%	40.0%	36.4%	38.5%	42.2%
Late dry season over the last 10–30 years	No change	0.0%	0.0%	0.0%	15.4%	4.4%	0.222²
	Late	9.1%	0.0%	0.0%	0.0%	2.2%
	Early	90.9%	100%	100%	84.6%	93.3%
Avis sur ces changements	Acceptable	0.0%	40.0%	72.7%	53.8%	42.2%	0.010²
	Good	0.0%	10.0%	0.0%	0.0%	2.2%
	Poor	100%	50.0%	27.3%	46.2%	55.6%
The manifestation of climate change	No	36.4%	0.0%	0.0%	7.7%	11.1%	0.020²
The manifestation of climate change	Yes	63.6%	100%	100%	92.3%	88.9%	0.020²

1. Chi-squared test for given probabilities; 2. Pearson's Chi-squared test.

The Multiple Correspondence Analysis (MCA) revealed that 59.58% of the information is distributed across axes 1 and 2, with respective contributions of 30.82% and 28.76% to the total variance (Fig. 5). The parameters “community member” (β = 0.637; p = 0.049) and “housing situation” (β = 3.137; p = 0.043) have positive coefficients, indicating that as these variables increase, the probability of answering “yes” also increases (2.604).

For drought conditions over the past 10–30 years, an increase is associated with a coefficient of -6.177, while a decrease is associated with 2.213. The perception of climate change manifestations in the last 10 years shows a negative coefficient for “no” (-2.086) and a positive coefficient for “yes” (2.086). Similarly, being a microfinance/bank client shows a negative coefficient for “no” (-1.995) and a positive coefficient for “yes” (1.995).

Axis 2 perceptions include unchanged temperature over the past 10–30 years (-2.907), decreased temperature (6.140), unchanged drought conditions (-2.907), and decreased drought conditions (6.140). Significant associations were found between Koranic and non-formal education, temporary residence, and microfinance/banking membership with increased and decreased precipitation over the past 10–30 years, unchanged temperature, unchanged drought conditions, and perceived climate change manifestations.

Conversely, increased drought, increased temperature, increased rainfall, and the perception of no climate change manifestations are significantly associated with primary education level, permanent housing status, not being a microfinance/bank client, and secondary education level. Finally, decreased temperature and decreased drought conditions over the past 10–30 years are not associated with any socio-demographic variables (Fig. 6).

Farmers' adaptation to climate change

The results of the survey on the adaptation measures adopted by the different villages show that 15.6% of the respondents have adopted the Short Cycle Variety, with a significant difference between the villages (p = 0.028). Late planting was adopted by 13.3% of the respondents. There was a highly significant difference between the communities (p = 0.001). Early planting was adopted by 13.3% of the respondents. There was a significant difference between the communities (p = 0.028). Another practice mentioned was crop rotation. Only 6.7% of respondents used this measure, with no significant difference between communities (p = 0.815). Reducing the area sown shows that 8.9% of respondents have adopted this measure. However, the difference between villages is not statistically significant (p = 0.083). 13.3% of the respondents have adopted Billonage. However, the difference between the municipalities is not statistically significant (p = 0.232). Crop rotation shows that only 4.4% of the respondents have adopted this measure. There is no significant difference between the communities (p = 0.4931). As for crop association, 6.7% of the respondents adopted this measure, with a significant difference between the communities (p = 0.019). Finally, abandoning/adopting new crops or varieties, according to the analyses, 7% of respondents adopted this measure, with a significant difference between communities (p = 0.019) (Table 3).

Table 3

Adaptation measures of different Mbangassina villages.
Adaptation measures	Modality	Biakoa	Bilomo	Mbangassina	Talba	Total	p-value
Short-cycle variety	No	100%	100%	81.8%	61.5%	84.4%	0.028¹
Short-cycle variety	Yes	0.0%	0.0%	18.2%	38.5%	15.6%	0.028¹
Late sowing	No	90.9%	50%	100.0%	100%	86.7%	0.001¹
Late sowing	Yes	9.1%	50%	0.0%	0.0%	13.3%	0.001¹
Early sowing	No	100%	80%	63.6%	100.0%	86.7%	0.028¹
Early sowing	Yes	0.0%	20%	36.4%	0.0%	13.3%	0.028¹
Crop rotation	No	90.9%	100%	90.9%	92.3%	93.3%	0.815¹
Crop rotation	Yes	9.1%	0.0%	9.1%	7.7%	6.7%	0.815¹
Decrease in planted area	No	100%	100%	72.7%	92.3%	91.1%	0.083¹
Decrease in planted area	Yes	0.0%	0.0%	27.3%	7.7%	8.9%	0.083¹
Billoning	No	72.7%	80.0%	100.0%	92.3%	86.7%	0.232¹
Billoning	Yes	27.3%	20.0%	0.0%	7.7%	13.3%	0.232¹
Crop rotation on the same surface	No	100%	90.0%	90.9%	100.0%	95.6%	0.493¹
Crop rotation on the same surface	Yes	0.0%	10.0%	9.1%	0.0%	4.4%	0.493¹
Association of crops	No	72.7%	100%	100.0%	100.0%	93.3%	0.019¹
Association of crops	Yes	27.3%	0.0%	0.0%	0.0%	6.7%	0.019¹
Abandon/adopt new crops or varieties	No	72.7%	100%	100.0%	100.0%	93.3%	0.019¹
Abandon/adopt new crops or varieties	Yes	27.3%	0.0%	0.0%	0.0%	6.7%	0.019¹

^{1 Pearson's} Chi-squared test.

The survey results indicate that 17.8% of respondents do not plan to adapt their farming practices to climate change in the next 10 years. In contrast, 33.3% intend to adapt, and 48.9% are uncertain but might adapt their practices. The p-value of 0.002 suggests that these responses are statistically significant. Regarding climate change awareness, 26.7% of respondents had no awareness at all, 44.4% sometimes perceived climate change, and 28.9% perceived it very often. The p-value of less than 0.001 indicates that these results are statistically significant, reflecting a real difference in climate change perceptions among respondents. Additionally, 51.1% of respondents had no accommodation in place, while 48.9% did.

The results show that 48.9% of respondents had accommodations in place, while 48.9% did not. The p-value of 0.004 indicates that these results are statistically significant, reflecting a true difference in the level of accommodation among respondents. Regarding the impact of climate change information on decision-making, 73.3% of respondents indicated that the information had no impact, while 26.7% reported that it did. The p-value of 0.002 confirms that these results are statistically significant, indicating a real difference in the impact of climate change information on decision-making. Additionally, 73.3% of respondents were not users of climate change information, whereas 26.7% were. The p-value of 0.002 suggests that this difference is statistically significant, reflecting a genuine difference in the use of climate change information among respondents.

The impact of cooperation among small-scale farmers on their resilience to climate change

Collaborative knowledge sharing enhances climate resilience among smallholder farmers in Mbangassina. The correlation analysis positions socio-economic variables and adaptation-related variables on two axes of the Multiple Correspondence Analysis (Fig. 7). The first axis accounts for 50.45% of the total variance, while the second axis explains 16.67%.

Axis 1 highlights the contrast between different types of survey sharing or learning about climate change to better plan adaptation measures. Group 1 includes individuals with secondary education, non-customers of microfinance/banks, Koranic education, members of farmer organizations, and temporary residents. This group shows limited engagement in farmer-to-farmer exchanges on climate change, with responses ranging from “not at all” to “occasional.” They also exhibit low access to climate change information and minimal decision-making on adaptation measures.

In contrast, Group 2 is characterized by active decision-making on adaptation, frequent exchanges among farmers on climate change, and better access to climate change information. This group includes individuals with non-formal education, non-members of farmer organizations, permanent residents, microfinance/bank clients, and those with primary education. These variables are associated with a higher likelihood of adaptation in the next 10 years and the influence of climate information on decision-making (Fig. 7).

Binary logistic regression

Binary logistic regression on climate change perception

Quantitative data

The p-value of less than 0.05 (Pr > Chi² = 0.046) indicates that this value is statistically significant. We therefore reject the null hypothesis and conclude that the variable affects the model. The association between − 2 Log (Likelihood) and Pr > Chi² is equal to 0.046. Since this value is less than 5%, the model is considered significant. The independent variables or factors studied in the analysis are age, household size, number of years of farming experience, useful agricultural area, and annual income. The table shows that age (β = 0.021; p = 0.043) and annual income (β = 0.079; p = 0.046) are significantly positively correlated with the dependent variable (Perception of climate change). On the other hand, the number of years of experience factor (β = -0.044; p = 0.045) is significantly negatively correlated (Table 4).

Table 4

Standardized coefficients.
Source	Value (β)	Standard error	Wald Chi-Square	Pr > Chi²	Wald Lower bound (95%)	Wald Upper bound (95%)
Age	0.021	0.252	0.005	0.043	-0.473	0.516
Household size	-0.063	0.178	0.103	0.526	-0.412	0.287
Number of years of farming experience	-0.044	0.251	0.011	0.045	-0.536	0.448
Useful agricultural area	0.290	0.292	0.785	0.121	-0.283	0.863
Annual income	0.079	0.265	0.068	0.046	-0.441	0.598
-2 Log (Likelihood)	/	/	2.634	0.046	/	/
Score	/	/	2.220	0.081	/	/
Wald			2.034	0.084	/	/

Qualitative data

The results are statistically significant, as the value of Pr > Chi² is less than 0.05 (0.022). In addition, the goodness-of-fit measure of the statistical model, associated with the − 2 Log (Likelihood), indicates a better model fit. Variables such as microfinance/bank client status, gender, community membership, and climate change perception are all statistically significant at different confidence levels (Pr > Wald and/or Pr > LR values below 0.05). This means that these variables have a significant effect on the dependent variable in the logistic regression model. Examining the coefficients of the model, we find that microfinance/bank customer status (β = -0.471; p = 0.046) and gender (β = -0.627; p = 0.049) have negative coefficients. This means that as these variables increase, the probability of the "yes" response decreases according to the binary logistic regression results. In contrast, parameters such as community membership (β = 0.637; p = 0.049) and residential status (β = 3.137; p = 0.043) have positive coefficients, indicating that when these variables increase, the probability of the "yes" response increases (Table 5).

Table 5

Standardized coefficients.
Source	Value (β)	Standard error	Wald Chi-Square	Pr > Chi²	Wald Lower bound (95%)	Wald Upper bound (95%)
CMB-No	-0.471	0.256	3.371	0.046	-0.973	0.032
SEX-Fe	-0.627	0.344	3.315	0.049	-1.301	0.048
NSC-Co	-0.364	0.262	1.931	0.165	-0.878	0.150
NSC-Nf	2.614	297.973	0.000	0.993	-581.403	586.630
NSC-Pr	0.053	0.264	0.041	0.840	-0.464	0.570
MCO-No	0.637	0.363	3.084	0.049	-0.074	1.348
SRE-Pe	3.137	352.939	0.000	0.043	-688.610	694.885
MOP-Nx	-0.342	0.257	1.766	0.184	-0.845	0.162
RFA-No	-0.107	0.223	0.230	0.631	-0.545	0.331
-2 Log (Likelihood)	/	/	19.405	0.022	/	/
Score	/	/	15.780	0.072	/	/
Wald	/	/	8.499	0.485	/	/

Legend: CMB: Microfinance/banking customer, SEX: sexe, NSC: Level of education, MCO: Member of the community, SRE: Residential situation, MOP: Member of a farmers' organization.

Binary logistic regression on climate change adaptation measures

Quantitative data

The table of fit coefficients provides several indicators of the quality of the model (or fit). According to the results, the variables contribute significantly to explaining the variability of the response variable. Indeed, the probability associated with the Chi² test is less than 0.0001, indicating that the − 2 Log (Likelihood) provides a significant amount of information (p < 0.0001). Analysis of the results reveals that certain variables, such as age, household size, number of years of farming experience, useful agricultural area, and annual income, help to explain the response variable. The "age" variable had a negative coefficient (-0.160), suggesting a negative relationship with the dependent variable. However, the p-value (0.110) is greater than 0.05, meaning that this relationship is not statistically significant. For "household size", the coefficient is positive (0.066), indicating a positive relationship. Nevertheless, the p-value (0.104) exceeds the 0.05 threshold, meaning that this relationship is not statistically significant. The coefficient for "years of farming experience" is negative (-0.076), also suggesting a negative relationship. The p-value (0.045) is less than 0.05, confirming that this relationship is statistically significant. For "useful agricultural area", the coefficient is negative (-0.123), indicating a negative relationship. The p-value (0.038) is less than 0.05, validating the statistical significance of this relationship. Finally, for "annual income", the coefficient is positive (0.202), indicating a positive relationship. The p-value (0.037) is less than 0.05, confirming that this relationship is statistically significant (Table 6).

Table 6

Standardized coefficients.
Source	Value	Standard error	Wald Chi-Square	Pr > Chi²	Wald Lower bound (95%)	Wald Upper bound (95%)
Age	-0.160	0.242	0.135	0.110	-0.634	0.315
Household size	0.066	0.174	0.144	0.104	-0.275	0.407
Number of years of farming experience	-0.076	0.242	0.067	0.045	-0.550	0.399
Useful agricultural area	-0.123	0.215	0.046	0.038	-0.544	0.298
Annual income	0.202	0.227	0.045	0.037	-0.242	0.646
-2 Log (Likelihood)	/	/	62.361	< 0.0001	/	/
Score	/	/	38.233	0.058	/	/
Wald	/	/	0.000	1.000	/	/

Qualitative data

The results indicate that the variables provide a significant amount of information to explain the variability of the response variable. Indeed, with a probability of less than 0.049, the values of Pr > Chi² and − 2 Log (Likelihood) provide a significant amount of information (p = 0.049). This means that there is a significant association between the two variables tested at the 5% significance level. The coefficient value is -0.675 for the "microfinance customer/non-bank" variable, with a Wald Chi-Square statistic of 4.527 and a p-value of 0.033. This variable is therefore significant at the 5% level. The coefficient value is -0.231 for the "gender-female" variable, with a Wald Chi-Square statistic of 0.520 and a p-value of 0.471. This variable is not significant at the 5% level. The coefficient value is -2.655 for the "school level-Quran" variable, with a Wald Chi-Square statistic of 0.000 and a p-value of 0.995. This variable is not significant at the 5% level. The results show that certain variables provide a significant amount of information to explain the variability of the response variable. The "marital status-celibate" variable has a value of 0.157 with a Wald Chi-square of 3.147. The probability (Pr > Chi²) is 0.041, which is less than 0.05, indicating that this variable is significant. The "community member-no" variable has a value of 0.498 with a Wald Chi-square of 1.774. The probability (Pr > Chi²) is 0.183, which is greater than 0.05, indicating that this variable is not significant. The "residential situation-pe" variable has a value of -0.957 with a Wald Chi-square of 3.023. The probability (Pr > Chi²) is 0.048, which is less than 0.05, indicating that this variable is significant. The "member of a peasant organization-no" variable has a value of -0.040 with a Wald Chi-square of 4.022. The probability (Pr > Chi²) is 0.028, which is less than 0.05, indicating that this variable is significant. The "Training in agriculture-No" variable has a value of 0.400 with a Wald Chi-square of 2.577. The probability (Pr > Chi²) is 0.108, which is greater than 0.05, indicating that this variable is not significant (Table 7).

Table 7

Standardized coefficients.
Source	Value	Standard error	Wald Chi-Square	Pr > Chi²	Wald Lower bound (95%)	Wald Upper bound (95%)
CMB-No	-0.675	0.317	4.527	0.033	-1.297	-0.053
SEX-Fe	-0.231	0.320	0.520	0.471	-0.858	0.397
NSC-Co	-2.655	391.112	0.000	0.995	-769.221	763.911
SIM-Ce	0.157	0.408	3.147	0.041	-0.644	0.957
MCO-No	0.498	0.374	1.774	0.183	-0.235	1.231
SRE-Pe	-0.957	472.071	3.023	0.048	-926.198	924.285
MOP-No	-0.040	0.270	4.022	0.028	-0.568	0.489
RFA-No	0.400	0.249	2.577	0.108	-0.088	0.889
-2 Log (Likelihood)	/	/	23.791	0.049	/	/
Score	/	/	17.953	0.209	/	/
Wald	/	/	9.273	0.813	/	/

Legend: CMB: Microfinance/banking customer, SEX: sexe, NSC: Level of education, SIM: Marital status, MCO: Member of the community, SRE: Residential situation, MOP: Member of a farmers' organization, RFA: Training in agriculture.

The results of the analysis show that 26.7% of the respondents were women. 73.3% were men. In this regard (p = 0.705), no significant differences were observed between communities. Cocoa farming is increasingly a male rather than a female activity. This is an indication that the traditional role of women in the cultivation of food crops for family consumption is changing. In their study of farmers' perceptions and practices of natural resource management in the face of climate change in Ntui, Chimi et al. (2022) observed this trend. Given the profitability of cocoa farming, women may be seeking to initiate and implement projects. Moreover, these results are consistent with those of [38], [39], [40], [41], [42], who found a strong male dominance in agricultural production. The average age of the farmers ranged from 32.7 to 47 years. The statistical test indicates that there is no significant difference in the average age between the villages (F = 0.47, p = 0.711). This means that the respondents are middle-aged. They are still fit enough to participate in agricultural production activities. In terms of household size, the mean values ranged from 2.9 to 6.9. The statistical test confirms that there is no significant difference in household size between villages (F = 1.28, p = 0.291). In terms of farming experience, the mean values for the number of years of farming experience ranged from 10 years to 20.6 years. The statistical test suggests that there is no significant difference in the length of farming experience between villages (F = 0.61, p = 0.611). The average size of the useful agricultural area varies from 2.4 to 4.5 hectares. The statistical test showed no significant difference in the values of the useful agricultural area between the different villages (F = 0.76, p = 0.521). This means that farmers in Mbangassina have been involved in agricultural production activities for a long time and operate on a small scale in the locality. Therefore, their extensive experience will help them to better understand climate change. Finally, the average annual income ranges from FCFA 2,834,000.0 to FCFA 9,089,533.3. The statistical test confirms that there is no significant difference in annual income between villages (F = 0.22, p = 0.881).

Mbangassina is located in the heart of Cameroon's Centre Region. It is much more than a geographical point. It is a microcosm of resilience. Here, farmers, families, and individuals are confronting the effects of climate change. Small-scale farmers, often unrecognized, work the land. They tend to the crops that sustain their families and communities. Their daily lives are affected by climate variability. From erratic rainfall to rising temperatures. Agriculture is the main activity that is widely practiced in the localities of Mbangassina, an essential characteristic of the rural areas of Cameroon, as highlighted by the National Institute of Statistics (NIS) in its report entitled "Annuaire Statistique du Cameroun, édition 2015". In terms of respondents' knowledge, 97.2% of the farmers in the survey indicated that they had been aware of changes in temperature and rainfall over the past 30 years. However, the dependent variable "perception of climate change" is significantly positively correlated with age (β = 0.021; p = 0.043) and annual income (β = 0.079; p = 0.046). In contrast, the number of years of farming experience (β = -0.044; p = 0.045) has a significantly negative correlation. These results are in line with those of [43], who found similar results in his work on perceptions, local knowledge, and adaptation strategies developed by producers in two communities in northern Benin. The majority of smallholder farmers in the Mbangassina ecological zone of transition from forest to savannah feel that the climate has changed in the last 10 to 30 years. Respondents reported late arrival of rains, rains in the middle of the dry season, irregular rains, and intense and prolonged droughts as climate changes. These findings suggest that the smallholder farmers are familiar with the climate of their region and are aware of the climate change that is underway. Numerous studies worldwide and in Africa have reached similar conclusions [18], [44], [45], [46], [47], [48]. Similarly, farmers in the Dodoma region of Tanzania were found to be generally well-informed about climate change by [49], and [50]. However, about whether or not climate change was occurring, some smallholder farmers remained undecided. Several factors could explain this diversity in perceptions among smallholder farmers in the Mbangassina subdivision.

Mbangassina's seasons are characterized by short and long dry seasons, followed by short and long rainy seasons. However, these patterns are not as predictable as before. From generation to generation, the elders of Mbangassina have passed on their wisdom. They recognize subtle signs as indicators of weather changes, such as the call of a particular bird or the behavior of ants. The most noticeable climatic parameters have been temperature, rainfall, drought, and the beginning and end of the dry season over the past 10 to 30 years. Compared to other parameters such as relative humidity or barometric pressure, these observations are more obvious to respondents. These findings are similar to those of researchers in Benin [51], [52], Ntui in Cameroon [18], and Togo [53]. The only difference is that the beginning and end of the dry season were not recorded in their studies. It is important to note that despite global efforts to replenish the ozone layer, climate change continues to evolve. This is especially true in different localities. In agriculture, smallholder family farmers face an imprecise distribution of seasons. Previously, these seasons were well-defined. This uncertainty has a direct impact on their agricultural activities and creates anxiety among these farmers. Traditional varieties that are resistant to local conditions are struggling to thrive, despite their efforts to diversify their crops (plantains, cassava, peanuts) to protect themselves from climatic risks. As a result, crop yields are declining. Certain pests and diseases are on the rise, and new ones are emerging. Numerous studies around the world have found similar results. Faced with these challenges, farmers are often struggling. This may explain the low annual incomes of the farmers surveyed. In the hope that the rains will arrive on time, farmers adjust their planting schedules. Average temperatures fluctuate between 23°C and 25°C. Heat waves are a challenge for both people and crops. Management of water sources is essential, and shade trees are becoming a refuge. Finally, smallholder farmers are concerned about soil degradation due to erosion, deforestation, and intensive agriculture. Fortunately, in response to these challenges, innovative practices such as agroforestry and organic farming are gaining ground.

Significant changes in mean and maximum temperatures were observed in the analysis of climate trends. Thus, for precipitation, maximum, and mean temperatures, the hypothesis H₁, indicating the existence of a trend, was accepted. On the other hand, hypothesis H₀, which indicates the absence of a trend, was maintained for the minimum temperature. However, although the Mann-Kendell test confirmed H₀ for minimum temperature, Sen's slope generally shows an increasing trend in temperatures (minimum, maximum, and mean) over the last 40 years in the Mbangassina transition zone. A warming climate over time is indicated by these positive slopes. R² values are a measure of how well the linear regression lines fit the observed data. Higher R² values are an indication of a better fit. However, it is important to note that linear regression is a simplified model. Other factors such as greenhouse gases, natural variability, and regional effects also contribute to temperature changes.

Between 1981 and 2022, Mbangassina recorded an average annual rainfall of 1,491.26 mm. The yearly rainfall fluctuates during this period. The R² value (0.237) suggests that this linear model does not fit the data perfectly. The downward trend implies a decrease in annual rainfall over time. Some climate trends are influenced by complex factors such as climate variability, land use change, and regional influences. Precipitation patterns, for example, are the result of multiple interactions. These include seasonal variations, geographic factors, and climatic oscillations. It is through meteorological data that farmers perceive these climatic changes. Indeed, records show minimum and maximum temperatures increasing. Similarly, farmers' perceptions of rainfall trends are consistent with precipitation observations. However, these findings differ from some studies conducted in Tanzania. [54], [55], [56], [57], and [58] found that farmers in the semi-arid central region of Tanzania perceived climate change in terms of anomalies in total annual rainfall. These changes included increasing temperatures, extending the dry season, and changing rainfall intensity. Farmers were well informed about climate change in the Dodoma region of Tanzania. To adapt to these changes, they favored practices such as crop rotation, reduction of cultivated area, ridging, and crop rotation.

The objective of this study was to further analyze the factors that influence the perception of climate change by smallholder farmers in the sub-division of Mbangassina. By examining the goodness of fit coefficients of binary logistic regression, several indicators of model quality were identified. These indicators are about the respondents' adoption of adaptation measures. The results showed that the coefficient for the variable age was negative (-0.160). This indicates a negative relationship with the dependent variable. However, the associated p-value (0.110) is greater than 0.05. This means that this relationship is not statistically significant. For household size, the coefficient is positive (0.066). This indicates a positive relationship. The corresponding p-value (0.104) is also greater than 0.05. This means that this relationship is not statistically significant. Finally, farming experience has a negative coefficient (-0.076). This indicates a negative relationship. The corresponding p-value (0.045) is less than 0.05. This means that this relationship is statistically significant. These results could be an indication that the stakeholders may be vulnerable to future climate change despite the measures in place. However, the need for these indicators for better monitoring of changes in an environment should be in mind. Furthermore, the results of the analysis show that the agricultural area has a negative correlation with a coefficient of -0.123. This correlation is statistically significant as the p-value (0.038) is below the threshold of significance. This correlation is statistically significant. The corresponding p-value (0.038) is below the 0.05 threshold. In other words, decreasing agricultural land is associated with decreasing annual income. On the other hand, the coefficient for annual income is positive (0.202). This indicates a positive relationship. The corresponding p-value (0.037) is also less than 0.05, confirming the statistical significance of this relationship. This suggests that stakeholders may be vulnerable to future climate change despite the measures taken. In conclusion, these indicators can reveal vulnerabilities. However, their use remains essential for monitoring changes in the environment.

The vast majority of respondents felt that the transition zone between forest and savannah in Mbangassina illustrates the community's awareness of the role of agriculture in adapting to climate change. Community-based climate change awareness programs could be based on this area. Research [59] has shown that individuals adapt to the impacts of climate change based on their observations. The decision to implement adaptation measures was most likely motivated by observed changes in climate change indicators. In Africa, studies show that the majority of farmers are aware of changes in temperature and rainfall and are already implementing adaptation measures such as diversifying crops, migrating, selling livestock, and conserving soil [18], [60], [61], [62]. This growing literature on adaptation in Africa is shedding light on how and where adaptation is taking place, while also highlighting the risks of maladaptation associated with inappropriate responses [63], [64]. In addition, farmers perceive a decrease in rainfall, an increase in temperature, and other factors [27]. Adoption of water and soil conservation techniques, ownership of manure pits, irrigation, and varietal adaptation are among the main adaptation strategies. Therefore, to preserve agricultural biodiversity in the context of climate change, adaptation measures in the Mbangassina transition zone should be based on local knowledge.

We examined the associations and contrasts between certain socioeconomic and perceptual variables using binary logistic regression. The results indicate that the variables microfinance/bank client (Pr > Wald = 0,046; Pr > LR = 0,036), gender (Pr > Wald = 0,049; Pr > LR = 0,015), community member (Pr > Wald = 0,059; Pr > LR = 0,023) and residential status (Pr > Wald = 0,079; Pr > LR = 0,001) are statistically significant at different confidence levels. Their values Pr > Wald and/or Pr > LR are less than 0.05. This means that these variables have a significant effect on the dependent variable in the logistic regression model. The estimation of the model coefficients for the variable of perception of climate change shows that microfinance/bank client (β = -0.471; p = 0.046) and gender (β = -0.627; p = 0.049) have negative coefficients. This suggests: As these variables increase, the probability of a "yes" response decreases, according to the binary logistic regression results. On the other hand, the parameters of being a member of the community (β = 0.637; p = 0.049) and living situation (β = 3.137; p = 0.043) have positive coefficients. This indicates that as these variables increase, the likelihood of answering "yes" increases. The fact that the respondent selection criteria for some of these variables did not allow for dissenting responses could explain these results. In addition, climate change is undeniable and ubiquitous. Therefore, it is likely that all social classes are aware of it. These results are different from those obtained by [18], who found that variables such as level of education and age group have an impact on perceptions. However, they also found that the status of the residence (temporary, permanent, or seasonal) plays a role in these perceptions. Therefore, it is important to seek the opinions of people who are permanent residents when it comes to perceptions of climate change in a given environment. Similarly, [65] found that a high proportion of farmers in the Kyuso district of Kenya were aware of the impacts of climate change, which was influenced by age, gender, education, farming experience, and access to extension services, among other factors.

The study examined the factors influencing smallholder farmers’ perceptions of climate change in the Mbangassina Subdivision. Initially, it assessed the farmers’ current knowledge, revealing that 97.2% perceive climate change in their locality. Their understanding of its causes and effects varies; for instance, over the past 10 to 30 years, 75.6% observed a decrease in rainfall, while average and maximum temperatures have increased. Between 1981 and 2022, Mbangassina recorded an average annual rainfall of 1491.26 mm, with significant variability influenced by seasonal changes, geographical factors, and climate variability.

The study identified vulnerability factors among smallholder farmers, showing significant associations between perceived climate change impacts and variables such as age, farming experience, annual income, microfinance status, gender, community affiliation, and socioeconomic level (p < 0.050). Additionally, adaptive capacity was influenced by factors including farming experience, usable agricultural area, annual income, microfinance client status, marital status, residential status, and membership in a farmers’ organization.

In conclusion, socio-economic and demographic factors are closely linked to smallholder farmers’ perceptions of climate change impacts. Understanding these local perceptions is crucial for developing effective adaptation and resilience strategies. The study highlights the importance of promoting sustainable and inclusive climate change solutions in Mbangassina. Advocacy for smart climate policies, land security, and disaster preparedness is gaining momentum. Integrating climate change education into school curricula ensures sustainability for future generations. Mbangassina exemplifies resilience, cultural richness, and determination, reminding us that collective action is essential to address climate challenges.

Strengthening the adaptive capacity and resilience of smallholder farmers requires support and interventions from governments, NGOs, and other stakeholders. Strategies include providing accessible, low-cost digital technologies, promoting local knowledge and innovation, facilitating rural-urban migration and remittances, and strengthening farmer groups and cooperatives. Achieving sustainable food production in the context of climate change hinges on addressing the factors that influence local perceptions and practices of smallholder farmers. This study contributes valuable insights for policymakers and future research directions. By integrating local knowledge into climate policies, policymakers can create more effective, inclusive, and sustainable solutions that are better suited to the specific needs and conditions of smallholder farmers.

Credit authorship contribution statement

Chimi: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Writing-original draft. Mala: Methodology, Validation, Visualization, Writing–review & editing. Fobane: Data curation, Formal analysis, Methodology, Validation, Visualization, Writing–review & editing. Feunang, Tchonang, Medoh, Endalle, Ngamsou, Batamack, Funwi, Kouoguem, Mazak, Nyonce, Tchandjie, Bell & Mbolo: Validation, Visualization, Writing-review & editing.

Ethics approval and consent to participate: The agriculture, fishing, animal husbandry, and environmental matters subdivision delegation, as well as the Institute of Agricultural Research for Development within the Mbangassina Local Government Research Ethics Committee, has confirmed that no ethical approval is required for their activities.

Informed Consent Statement: Before data collection, farmers provided informed consent.

Competing interests:The authors have no relevant financial or nonfinancial interests to disclose. The authors have no competing interests to declare that they are relevant to the content of this article.

Funding: This study received no external funding.

Data availability statement: Data that have been analyzed without participant identity may be available on request.

Acknowledgments

The populations of the villages of the Subdivision of Mbangassina for their contribution to the realization of this study. The sectors (the subdivisions delegation of agriculture, fishing, and animal industry; the environment; trade; and the Institute of Agricultural Research for Development) contributed information to the preliminary data.

Financial interests: The authors declare that they have no financial interests.

FAO, « Organisation des Nations Unies pour l’alimentation et l’agriculture: Le Cameroun actualise sa stratégie de développement du secteur rural | FAO au Cameroun | Organisation des Nations Unies pour l’alimentation et l’agriculture ». Consulté le: 24 janvier 2024. [En ligne]. Disponible sur: https://www.fao.org/cameroun/actualites/detail-events/fr/c/1306909/
FAO, « La Situation mondiale de l’alimentation et de l’agriculture 2017 - Mettre les systèmes alimentaires au service d’une transformation rurale inclusive », 2017.
C. Lambert, « Le monde rural : réalités, difficultés et espoirs », Constructif, vol. 55, n^o 1, p. 20‑24, 2020, doi: 10.3917/const.055.0020.
B. A. Mushagalusa, B. S. Patient, L. N. Albert, M. M. Didier, I. K. Fifi, et N. M. F. Jules, « Genre et exploitations agricoles familiales en milieu rural au Katanga: étude de cas de Kipushi [Gender and family farms in rural area in Katanga: case study of Kipushi] », Int. J. Innov. Appl. Stud., vol. 11, n^o 2, p. 367‑375, 2015.
E. Sanz Sanz, C. Napoléone, et B. Hubert, « Caractériser l’agriculture périurbaine pour mieux l’intégrer à la planification urbaine : propositions méthodologiques », L’Espace Géographique, vol. 46, n^o 2, p. 174‑190, 2017, doi: 10.3917/eg.462.0174.
UN, « Climat : « les enjeux n’ont jamais été aussi élevés », alerte le chef du GIEC | ONU Info ». Consulté le: 9 avril 2024. [En ligne]. Disponible sur: https://news.un.org/fr/story/2022/02/1114252
FAO, « The State of Food Security and Nutrition in the World 2021. The World Is at a Critical Juncture ». FAO Rome, 2021.
FAO, « L’Action de la FAO face au changement climatique », 2015.
CIRAD, « L’agroforesterie peut jouer un rôle clé dans la lutte contre le changement climatique | Cirad ». Consulté le: 9 avril 2024. [En ligne]. Disponible sur: https://www.cirad.fr/espace-presse/communiques-de-presse/2023/agroforesterie-potentiel-d-attenuation-climatique
IFAD Annual Report 2012 », IFAD. Consulté le: 29 octobre 2023. [En ligne]. Disponible sur: https://www.ifad.org/fr/web/knowledge/-/ifad-annual-report-2012-1?p_l_back_url=%2Ffr%2Fweb%2Fknowledge%2Fannual-reports
C. M. Lambi et S. S. Kometa, « Climate Change in Cameroon and Its Impacts on Agriculture », in Implementing Adaptation Strategies by Legal, Economic and Planning Instruments on Climate Change, vol. 4, E. Albrecht, M. Schmidt, M. Mißler-Behr, et S. P. N. Spyra, Éd., in Environmental Protection in the European Union, vol. 4. , Berlin, Heidelberg: Springer Berlin Heidelberg, 2014, p. 237‑254. doi: 10.1007/978-3-540-77614-7_15.
A. A. Diendéré et D. Ouédraogo, « Women farmers and climate change: Empirical evidence from Burkina Faso », Agric. Resour. Econ. Rev., p. 1‑25, 2023.
K. E. Giller et al., « The future of farming: Who will produce our food? », Food Secur., vol. 13, n^o 5, p. 1073‑1099, Oct. 2021, doi: 10.1007/s12571-021-01184-6.
S. B. Bedeke, « Climate change vulnerability and adaptation of crop producers in sub-Saharan Africa: a review on concepts, approaches, and methods », Environ. Dev. Sustain., vol. 25, n^o 2, p. 1017‑1051, févr. 2023, doi: 10.1007/s10668-022-02118-8.
P. M. Chimi et al., « Vulnerability of family farming systems to climate change: The case of the forest-savannah transition zone, Centre Region of Cameroon », Res. Glob., vol. 7, p. 100138, 2023.
D. Dalle, Y. Gecho, et S. B. Bedeke, « Spatiotemporal Variability and Trends in Rainfall and Temperature in South Ethiopia: Implications for Climate Change Adaptations in Rural Communities », Adv. Meteorol., vol. 2023, 2023, Consulté le: 1 février 2024. [En ligne]. Disponible sur: https://www.hindawi.com/journals/amete/2023/1939528/
A. Aich, D. Dey, et A. Roy, « Climate change resilient agricultural practices: A learning experience from Indigenous communities over India », PLOS Sustain. Transform., vol. 1, n^o 7, p. e0000022, 2022.
P. M. Chimi et al., « Climate change perception and local adaptation of natural resource management in a farming community of Cameroon: A case study », Environ. Chall., vol. 8, p. 100539, 2022.
X. Feng et al., « Climate-smart agriculture practice promote sustainable maize production in northeastern China: Higher grain yield while less carbon footprint », Field Crops Res., vol. 302, p. 109108, 2023.
J. Gray et J. Dunn, « Optimizing Crop Plant Stomatal Density to Mitigate and Adapt to Climate Change », Cold Spring Harb. Perspect. Biol., p. a041672, 2023.
B. Ateba Henri Aurélien, O. N. Bruno Emmanuel, M. A. E. Hervé William, et M. A. Thierry, « “Does climate change influence conflicts? Evidence for the Cameroonian regions” », GeoJournal, vol. 88, n^o 4, p. 3595‑3613, janv. 2023, doi: 10.1007/s10708-023-10824-7.
F. Ngum, D. Alemagi, L. Duguma, P. A. Minang, A. Kehbila, et Z. Tchoundjeu, « Synergizing climate change mitigation and adaptation in Cameroon: An overview of multi-stakeholder efforts », Int. J. Clim. Change Strategy. Manag., vol. 11, n^o 1, p. 118‑136, 2018.
E. L. Molua, « Turning up the heat on African agriculture: The impact of climate change on Cameroon’s agriculture », Afr. J. Agric. Resour. Econ., vol. 2, n^o 311-2016‑5519, p. 45‑64, 2008.
E. L. Molua, « The economic impact of climate change on agriculture in Cameroon », World Bank Policy Res. Work. Pap., n^o 4364, 2007, Consulté le: 1 février 2024. [En ligne]. Disponible sur: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1016260
M. Flourens, « Le réchauffement climatique : causes, conséquences et solutions », Oxfam France. Consulté le: 9 avril 2024. [En ligne]. Disponible sur: https://www.oxfamfrance.org/climat-et-energie/comprendre-les-changements-climatiques/
World Bank, « Cameroun Rapport National sur le Climat et le Développement 2022 - Cameroon | ReliefWeb ». Consulté le: 9 avril 2024. [En ligne]. Disponible sur: https://reliefweb.int/report/cameroon/cameroun-rapport-national-sur-le-climat-et-le-developpement-2022
P. N. Kabore, B. Barbier, P. Ouoba, A. Kiema, L. Some, et A. Ouedraogo, « Perceptions du changement climatique, impacts environnementaux et stratégies endogènes d’adaptation par les producteurs du Centre-nord du Burkina Faso », VertigO, vol. 19, n^o 1, 2019, Consulté le: 12 octobre 2023. [En ligne]. Disponible sur: https://www.erudit.org/en/journals/vertigo/1900-v1-n1-vertigo04936/1065432ar/abstract/
G. Karam et J. A. Gérard, « Changement climatique et agriculture : Entre perception, réalité et pratiques dans la région semi-aride de la plaine de la Beqaa (Liban) », VertigO - Rev. Électronique En Sci. Environ., n^o Volume 23 Numéro 2, Art. n^o Volume 23 Numéro 2, sept. 2023, doi: 10.4000/vertigo.41649.
B. A. Mushagalusa et al., « Perception et stratégies d’adaptation aux incertitudes climatiques par les exploitants agricoles des zones marécageuses au Sud-Kivu », VertigO - Rev. Électronique En Sci. Environ., n^o Volume 21 Numéro 1, Art. n^o Volume 21 Numéro 1, mai 2021, doi: 10.4000/vertigo.31673.
A. Traore, « Changement climatique et agriculture en Afrique subsaharienne. Perception des agriculteurs et impact de l’association entre une céréale et une légumineuse sur les rendements des deux espèces et leur variabilité inter-annuelle sous climat actuel et futur. | Knowledge for policy ». Consulté le: 9 avril 2024. [En ligne]. Disponible sur: https://knowledge4policy.ec.europa.eu/publication/changement-climatique-et-agriculture-en-afrique-subsaharienne-perception-des_en
PCD, « Plan communal de développement de NTUI ». Consulté le: 22 mars 2023. [En ligne]. Disponible sur: https://docplayer.fr/74101125-Plan-communal-de-developpement-de-ntui.html
A. Alemayehu et W. Bewket, « Determinants of smallholder farmers’ choice of coping and adaptation strategies to climate change and variability in the central highlands of Ethiopia », Environ. Dev., vol. 24, p. 77‑85, 2017.
T. Alemayehu et al., « Farming Systems, Food Security and Farmers’ Awareness of Ecosystem Services in Inland Valleys: A Study From Côte d’Ivoire and Ghana », Front. Sustain. Food Syst., vol. 6, 2022, Consulté le: 19 mars 2023. [En ligne]. Disponible sur: https://www.frontiersin.org/articles/10.3389/fsufs.2022.892818
A. Ali, R. Akhtar, et J. Hussain, « Unveiling High Mountain Communities’ Perception of Climate Change Impact on Lives and Livelihoods in Gilgit-Baltistan: Evidence from People-Centric Approach », Environ. Commun., vol. 17, n^o 6, p. 602‑617, août 2023, doi: 10.1080/17524032.2023.2229044.
I. Khan et al., « Farm households’ risk perception, attitude and adaptation strategies in dealing with climate change: Promise and perils from rural Pakistan », Land Use Policy, vol. 91, p. 104395, 2020.
L. Mwadzingeni, R. Mugandani, et P. L. Mafongoya, « Perception of climate change and coping strategies among smallholder irrigators in Zimbabwe », Front. Sustain. Food Syst., vol. 7, p. 1027846, 2023.
B. Koneswarakantha et T. Ménard, « Statistical Modeling for Quality Risk Assessment of Clinical Trials: Follow-Up at the Era of Remote Auditing », Ther. Innov. Regul. Sci., vol. 56, n^o 3, p. 433‑441, mai 2022, doi: 10.1007/s43441-022-00388-y.
B. Hauwa, H. S. Umar, E. S. Salau, I. Shehu, et S. Bako, « Gender Analysis of Labor Contribution Among Maize Farming Household in Agricultural Zone C of Kogi State, Nigeria », Eur. J. Agric. Food Sci., vol. 4, n^o 2, p. 1‑6, 2022.
I. Tafida, B. Nazifi, et K. Muhammad, « ASSESSMENT OF SESAME FARMERS’PERCEPTION AND ADAPTATION STRATEGY TO CLIMATE CHANGE IN MACHINA LOCAL GOVERNMENT AREA, OF YOBE STATE, NIGERIA », Food Sci., vol. 7, n^o 1, p. 1‑10, 2024.
B. B. Umar, « Adapting to climate change through conservation agriculture: a gendered analysis of Eastern Zambia », Front. Sustain. Food Syst., vol. 5, p. 748300, 2021.
B. B. Umar, P. H. Nyanga, D. Chibamba, et W. S. Nchito, « Women’s empowerment, land and donor-driven agricultural interventions in Eastern Zambia », World Dev. Perspect., vol. 19, p. 100208, 2020.
Y. A. Umar, I. A. Abu, M. H. Sani, S. Idi, et E. O. Owoade, « Gender Analysis of Cattle Farmers’ Participation in Cattle Production in Bauchi and Yobe States, Nigeria », Asian J. Agric. Ext. Econ. Sociol., vol. 41, n^o 9, p. 637‑646, 2023.
R. Dimon, « Adaptation aux changements climatiques: perceptions, savoirs locaux et stratégies d’adaptation développées par les producteurs des communes de Kandi et de Banikoara, au Nord du Bénin », PhD Thesis, UAC, 2008. Consulté le: 8 avril 2024. [En ligne]. Disponible sur: https://agritrop.cirad.fr/573169/
S. A. Abossolo et J. A. A. E. M. Tchindjang, Perturbations climatiques et pratiques agricoles dans les zones agroécologiques du Cameroun: changements socio-économiques et problématique d’adaptation aux bouleversements climatiques. Connaissances & Savoirs, 2017.
L. A. Bougma, M. H. Ouédraogo, N. Sawadogo, M. Sawadogo, D. Balma, et R. Vernooy, « Perceptions paysannes de l’impact du changement climatique sur le mil dans les zones sahélienne et soudanosahélienne du Burkina Faso », Afr. Sci., vol. 14, n^o 4, p. 264‑275, 2018.
J.-L. Kouassi, « Variabilité climatique, dynamique des feux de végétation et perceptions locales dans le bassin versant du N’Zi (Centre de la Côte d’Ivoire) », PhD Thesis, Institut National Polytechnique Felix Houphoët Boigny-Yamoussoukro, 2019. Consulté le: 8 avril 2024. [En ligne]. Disponible sur: https://theses.hal.science/tel-02183044/
W. A. Mala, C. J. Geldenhuys, et R. Prabhu, « Perceptions of Climate Variability and Pest-Disease Incidence on Crops and Adaptive Forest-Agricultural Practices », Cult. Agric. Food Environ., vol. 34, n^o 1, p. 53‑67, juin 2012, doi: 10.1111/j.2153-9561.2012.01064.x.
A. D. Tidjani et al., « Perceptions de la variabilité climati que et stratégies d’adaptation dans le système oasien de Gouré (Sud-est Niger) », Agron. Afr., vol. 28, n^o 2, p. 25‑37, 2016.
H. K. Mayaya, G. Opata, et E. C. Kipkorir, « Understanding climate change and the manifestation of its driven impacts in the semi-arid areas of Dodoma Region, Tanzania », Ethiop. J. Environ. Stud. Manag., vol. 8, n^o 4, p. 364‑376, 2015.
A. B. Chitimbe et E. T. Liwenga, « Agricultural Commercialization and Adaptation to Climate Change and Variability in Semi-Arid Central Tanzania », J. Agric. Ecol. Res. Int., vol. 3, n^o 3, p. 119‑130, 2015.
D. Callo-Concha, « Farmer perceptions and climate change adaptation in the West Africa Sudan Savannah: reality checks in Dassari, Benin, and Dano, Burkina Faso », Climate, vol. 6, n^o 2, p. 44, 2018.
K. Sabaï, D. S. Dolorès, H. G. Castro, et D. Siaka, « Influence of climate change and breeding mode on the reproductive parameters of cattle herds in the agro-pastoral cotton production zone of the commune of Banikoara in North Benin », Int. J. Vet. Sci. Anim. Husb., vol. 5, n^o 6, p. 55‑63, 2020.
K. Sanou, S. Amadou, K. Adjegan, et K. D. Tsatsu, « Perceptions et stratégies d’adaptation des producteurs agricoles aux changements climatiques au nord-ouest de la région des savanes du Togo », Agron. Afr., vol. 30, n^o 1, p. 87‑97, 2018.
Y. W. Githiora, M. A. Owuor, R. Abila, S. Oriaso, et D. O. Olago, « Perceptions, trends and adaptation to climate change in Yala wetland, Kenya », Int. J. Clim. Change Strategy. Manag., vol. 15, n^o 5, p. 690‑711, nov. 2023, doi: 10.1108/IJCCSM-07-2022-0089.
L. S. Jackson et al., « The effect of explicit convection on couplings between rainfall, humidity, and ascent over Africa under climate change », J. Clim., vol. 33, n^o 19, p. 8315‑8337, 2020.
M. Ojoyi, O. Mutanga, J. M. Kahinda, J. Odindi, et E. M. Abdel-Rahman, « Scenario-based approach in dealing with climate change impacts in Central Tanzania », Futures, vol. 85, p. 30‑41, 2017.
M. M. Ojoyi et J.-M. Mwenge Kahinda, « An analysis of climatic impacts and adaptation strategies in Tanzania », Int. J. Clim. Change Strategy. Manag., vol. 7, n^o 1, p. 97‑115, 2015.
A. N. Temba et M. K. Said, « Small-holder Farmers’ Perceptions of the Impacts of Climate Change on Maize Crop in Dodoma, Tanzania », Ghana J. Geogr., vol. 15, n^o 2, p. 184‑211, 2023.
P. D. Howe, J. Thaker, et A. Leiserowitz, « Public perceptions of rainfall change in India », Clim. Change, vol. 127, n^o 2, p. 211‑225, nov. 2014, doi: 10.1007/s10584-014-1245-6.
C. O. Ajuang, P. O. Abuom, E. K. Bosire, G. O. Dida, et D. N. Anyona, « Determinants of climate change awareness level in upper Nyakach Division, Kisumu County, Kenya », SpringerPlus, vol. 5, n^o 1, p. 1015, déc. 2016, doi: 10.1186/s40064-016-2699-y.
N. P. Awazi et N. M. Tchamba, « Determinants of small-scale farmers adaptation decision to climate variability and change in the North-West region of Cameroon », Afr. J. Agric. Res., vol. 13, n^o 12, p. 534, 2018.
J. S. Juana, Z. Kahaka, et F. N. Okurut, « Farmers’ perceptions and adaptations to climate change in sub-Sahara Africa: A synthesis of empirical studies and implications for public policy in African agriculture », J. Agric. Sci., vol. 5, n^o 4, p. 121, 2013.
L. Berrang-Ford et al., « A systematic global stocktake of evidence on human adaptation to climate change », Nat. Clim. Change, vol. 11, n^o 11, p. 989‑1000, 2021.
A. J. Sietsma, J. D. Ford, M. W. Callaghan, et J. C. Minx, « Progress in climate change adaptation research », Environ. Res. Lett., vol. 16, n^o 5, p. 054038, 2021.
H. K. Ndambiri et al., « Analysis of farmers’ perceptions of the effects of climate change in Kenya: the case of Kyuso district », 2012, Consulté le: 20 avril 2024. [En ligne]. Disponible sur: https://ageconsearch.umn.edu/record/159405/

No competing interests reported.

Navigating Climate Challenges: Insights from smallholder of Mbangassina’s Local Communities in the Centre Cameroon Region

Status:

Version 1

Abstract

Figures

Introduction

Materials and methods

Study area

Methodology

Methodological framework

Literature review

Quantitative survey

Qualitative interviews

Data Analysis

Quantitative Data Analysis

Qualitative Data Analysis

Analysis of climatic trends

Results

Socio-demographic characteristics

Variations in weather indicators

Temperature trends from 1981 to 2022

Precipitation trends from 1981 to 2022

Farmers' perceptions of climate-related changes and factors influencing perceptions

Farmers' adaptation to climate change

The impact of cooperation among small-scale farmers on their resilience to climate change

Binary logistic regression

Binary logistic regression on climate change perception

Quantitative data

Qualitative data

Binary logistic regression on climate change adaptation measures

Quantitative data

Qualitative data

Discussion

Conclusion

Declarations

References

Additional Declarations

Status:

Version 1