Assessing the Effects of Climate Variability on Maize Yield in the Municipality of Dschang—Cameroon

Abstract

Evidence-based research on the effects of rainfall, temperature, and relative humidity variability on maize yield is essential for understanding the climate dynamics of, and paving the way for informed adaptive solutions to future potential negative impacts in, Dschang-Cameroon. This study employed the non-parametric Mann–Kendall and Sen’s slope method to detect trends in climate variables and maize yield in the period between 1990 to 2018. Pearson correlation and multilinear regression (MLR) analyses were also used to establish the linear relationship between climate variables and maize yield, and to explore the behavior of the response variable (maize yield) with the predictor variables (climatic variables), respectively. In addition, perceptions of climate variability and its impact on maize yield from a hundred farmers were collected through a questionnaire and analyzed in SPSS. Twenty key informants’ interviews (KII) were conducted using a semi-structured interview and analyzed by thematic analysis. The results showed that the minimum temperature exhibited a decreasing trend at a rate of 0.039 °C per annum, whereas relative humidity had an increasing trend of 0.25% per annum with statistical significance at p = 0.001. In addition, a decreasing trend of rainfall, at a rate of 4.94 mm per annum, was observed; however, this had no statistical significance. Furthermore, the MLR analysis showed that mean temperature and relative humidity have an inversely proportional but statistically significant relationship with maize yield (p = 0.046 and p = 0.001, respectively). The analysis of farmers’ perceptions confirmed the results of trend analyses of decreasing rainfall and increasing maximum temperatures. Moreover, the farmers asserted that the vulnerability of farmers to climate variability is also linked to gender and locality, where women’s outputs are more assailable and farms in low-lying areas are more prone to floods. The high price of farm inputs was also reported as a key factor, other than climate variability, hindering the flourishing of the maize sector in Dschang. Finally, an analysis of the KII indicated the inadequate implementation of flagship agricultural programs in the locality.

1. Introduction

Climate variability and change are critical issues on today’s agenda for all countries worldwide, as they threaten the achievement of the Sustainable Development Goals [1]. One study [2] showed that most of the Horn of Africa, Sahel, central, and southern Africa could experience a 3 to 6 °C increase in temperature by 2100, with a mean temperature of 4.5 °C. In Ref. [3], the authors also forecast lengthier dry spells, by around 30 to 50%, compared to those of the reference period (1976–2005) in many parts of west and central Africa for the GHG high forcing scenario RCP8.5 by the end of the 21st century. In Ref. [4], the authors further put forward that Africa is expected to be more vulnerable to climate variability and change due to its over-reliance on natural resources—particularly for rainfed agriculture—and its low economic and institutional capacity to cater for climate adaption strategies.

Cameroon is experiencing shifts in rainfall and temperature patterns that could have significant impacts on the country’s agricultural sector, which is a major contributor to the economy [5]. Maize is extensively grown and consumed as a staple food in sub-Saharan Africa and is a vital crop in Cameroon, playing a key role in ensuring food security and sovereignty [6], as cited in [7]. Most Cameroonians’ daily diet heavily depends on maize; as a result, maize and its by-products are consumed in various forms such as fufu corn, maize pudding, maize doughnuts, and corn beer. Additionally, maize is used for various other purposes, including as chicken feed [8]. The ongoing Russian-Ukraine war has impacted the production of bread in Cameroon, leading to a shift towards using maize flour to reduce dependence on wheat flour imported from Russia and Ukraine.

Despite the strong cultural preference for maize in Cameroon, there is an evident imbalance in maize supply, with domestic demand outstripping production [7]. Several factors have been identified that affect the yield of maize in Cameroon such as low institutional support, and lack of appropriate farming methods [5,9]. Most importantly, maize is typically rainfed; its production is subject to natural weather conditions for growth, which exposes maize to a variety of risks, including low crop water availability, pests, and weather variability such as high temperatures and strong winds [10].

Furthermore, in Ref. [11], it was shown that a decrease in rainfall patterns and increased temperatures between 1990 and 2010 have negatively affected maize productivity of smallholder farmers by an average of 200 kg/ha in the Western Highlands of Cameroon. Similarly, the authors of Ref. [12] asserted that the higher relative humidity reported in the Western Highlands of Cameroon is conducive to fungal infection in maize grains, and, eventually, mycotoxin contamination. Lower maize productivity is a threat to food security in the country and increases poverty among smallholder farmers [5]. Despite the great natural potential with which Cameroon, and particularly the municipality of Dschang has been endowed with for the growth of maize, maize yields are not resilient. One study [13,14,15] indicated that climate variability and change are becoming major contributors to the decrease in maize yield in Cameroon. Hence, it is crucial to understand how the variability in rainfall, temperature, and relative humidity is affecting the productivity of maize in the municipality of Dschang and to suggest solutions to adapt to these effects.

In addition to climatic factors, other health and socio-political factors at the global scale have exacerbated the risks associated with maize cultivation in Cameroon, notably the onset of the COVID-19 pandemic in 2020 and the Russia-Ukraine war in 2022 have greatly disrupted agricultural supply chains (by increasing agricultural inputs prices), aggravated the vulnerability of smallholder farmers, and have accrued the threats to food insecurity in the country.

Several studies in Cameroon have attempted to establish the effects of climate variability on the maize crop. However, none have employed both quantitative and qualitative analyses to provide a holistic understanding of the topic, and eventually to suggest policy solutions to the problems identified.

Against this background, this study, therefore, attempts to comprehensively assess the effects of climate variability on maize yield in the municipality of Dschang, Cameroon based on both quantitative and qualitative analyses and formulate effective policy solutions geared at developing the resilience of maize yield and curtailing the vulnerability of smallholder farmers in the municipality.

2. Study Area

The study area is the municipality of Dschang, which is located in the Menoua Division, in the Western Region of Cameroon, between a latitude of 5°25′ and 5°30′ N, and a longitude of 9°57′ and 10°05′ E, covering a surface area of 262 Km² (Figure 1). It is part of the Western Highland Agroecological Zone of Cameroon and has a humid tropical climate, with three main types, depending on the altitude, as follows: it is coastal at low altitudes like in Nteingue village near the Mbo plain, with increasingly hot weather; it is moderately cool in the mid-altitude zones (between 1200 and 1600 m); and it is temperate and increasingly cold at higher altitudes. This climate is characterized by a long rainy season from March to October, and a short dry season from November to February [13]. The mean annual rainfall for the period 1990 to 2022 is approximately 1680 mm, while the mean annual temperature stands around 20.8 °C. The municipality of Dschang consists of vast areas of arable land with various types of soil of high agricultural value. For this reason, this region’s economy is essentially driven by agriculture and relies on the following three main groups of crop types: perennial, food, and vegetable crops [13].

3. Data and Methods

3.1. Data

Climatic data (rainfall, relative humidity, minimum, maximum, and mean temperature) at annual time steps from 1990 to 2022 were obtained from the Dschang meteorological station located at the Agricultural Research Institute for Development, with coordinates at a latitude of 5°27′0″ N and a longitude of 10°04′0″ E, and an altitude of 1399 m above the mean sea level (Figure 1). The annual maize yield data (2000 to 2018) for the Dschang municipality, on the other hand, were obtained from the Divisional Delegation of Agriculture and Rural Development (DDADER-MINADER). The data were prepared using Microsoft Excel v2407.

3.2. Methods

3.2.1. Statistical Analysis

For the trend detection and quantification in this study, trend analyses were conducted for the three sets of climate variables (rainfall, temperature, and relative humidity) for the period 1990 to 2022 and the maize yield dataset from 2000 to 2018. Mann–Kendall (MK) and Sen’s slope tests which is a non-parametric trend detection method was employed as they stand out as one of the most accepted and generally used non-parametric methods [16]. Mann–Kendall (MK) and Sen’s slope tests are mainly used for data that are not normally distributed (as is the case for most hydro-meteorological time series data), they are more robust against extreme outliers compared to parametric methods and have fewer requirements and assumptions, giving them a wider range of applications [17,18,19], as cited in [16].

An MK test with a 95% confidence limit was employed as a monotonic trend test. The null hypothesis (H₀) indicates there is no monotonic trend in the tested population while the alternative hypothesis (H₁) implies a monotonic trend. The null hypothesis (H₀) is rejected if p ≤ 0.05.

The quantification of the trends, on the other hand, was conducted using Sen’s slope test, denoted by Equation (1), as follows:

$$Q_i=\frac{x_j-x_k}{j-k} \mathrm{for} i=1, 2, 3, \dots N$$

(1)

where x_j and x_k are data values at time j and k (j > k), respectively. The median of these N values of Q_i is Sen’s estimator of slope. If N is odd, then Sen’s estimator is computed by Q_med = Q (N + 1)/2; if N is even, Sen’s estimator is computed by Q_med = [QN/2 + Q (N + 2)/2]/2. Finally, Q_med is tested by a two-sided test at a 100% (1 − α) confidence interval, and the true slope is obtained.

The Pearson correlation (r) was conducted using SPSS v23.0.0.0 to determine the strength of the relationship between each of the climatic variables (rainfall, temperature, and relative humidity), and maize yield. The Pearson correlation coefficient appeared to be ideal because it is employed when dealing only with quantitative variables. Additionally, Refs. [16,20] and several other researchers have made use of the Pearson correlation to establish the linear relationship between climate variables and crop yield (Equation (2)). The outcome of this correlation is a value between −1 and 1, where −1 indicates a total negative linear correlation, +1 is a total positive correlation and 0 means no linear correlation, as follows:

$$r=\frac{\sum \left(x_i-\overline{x}\right) (y-\overline{y})}{\sqrt{\sum {(x_i-\overline{x})}^{2 } {(y_i-\overline{y})}^{2 }}}$$

(2)

where r is the correlation coefficient; x_i are the values of one of the climatic variables; $\overline{x}$ is the mean of the values of the climatic variable; y_i are the values of the maize yield data; and ȳ is the mean of the values of the maize yield data.

In addition, by making use of multilinear regression (MLR), the predictor variables (rainfall, temperature, and humidity) were correlated with the response variable (maize yield data) all together to establish a cause-effect relationship and to predict the behavior of the response variable with the predictor variables. The datasets (rainfall, T_max, T_min, T_mean, and relative humidity) run for 19 years (2000–2018). The datasets were first standardized to make the data unitless and comparable, after which the MLR was computed in Stata. The equation for computing MLR is displayed as Equation (3), as follows:

∆Y = (α × ∆Rf) + (β × ∆T_max) + (γ × ∆T_min) + (ψ × ∆T_mean) + (ω × ∆Rh) + ϵ

(3)

where ∆Y is the observed change in the yield due to rainfall, temperature and relative humidity in the same year as maize growth; α, β, γ, ψ, and ω are coefficients of the rainfall, minimum, maximum, mean temperatures, and relative humidity during the year, respectively; ∆Rf, ∆T_max, ∆T_min, T_mean, and Rh are the observed changes in rainfall, temperatures, and relative humidity over the years of the study period, respectively.

3.2.2. Survey Data Analysis

A questionnaire survey permitted us to obtain insights into how smallholder farmers perceive the effects of climate variability on their maize yield. The sample size was a hundred (100) farmers, derived from Equation (4), as proposed by [21]. The Menoua Divisional Delegation of Agriculture in Dschang [13] reported that an average of 13,500 maize farmers for the two production cycles have been registered in Dschang. The collected data were analyzed in SPSS, as follows:

$$n\ge \frac{\frac{1 }{e^2} \ast N}{\frac{1}{e^2}+N}$$

(4)

where N is the total population, e is the sampling error between 1% to 20%, and n is the number of the sample population.

Likewise, semi-structured interviews helped to gather information from 20 key informants who were mainly executives from MINADER and the PCP-ACEFA Program. They were interviewed according to their availability and willingness to disclose information. The KII were analyzed using thematic analysis from a deductive approach (with a theory that the flagship agricultural programs in the municipality of Dschang do not accentuate support to farmers with regards to building resilience towards climatic shocks). The process of carrying out the thematic analysis was guided by Flick’s [22] list of basic questions for coding and identifying themes (Figure 2). Anonymization was also performed to keep respondents’ identities undisclosed.

The methodology employed in this study is summarized in Figure 3.

4. Results

This section provides a concise and precise description of the experimental results, their interpretation, as well as the experimental conclusions that can be drawn.

4.1. Trend Analysis of Climate Variables

4.1.1. Trend Analysis of Rainfall

Results from the Mann–Kendall test for total annual rainfall from 1990 to 2022 showed a downward trend (with no statistical significance) (

Table 1. Summary of MK and Sen’s slope estimation for total rainfall in Dschang (1990–2022).

Time Series	Test Z	Significance	Q
Annual Rainfall	−1.22	ns	−4.94
January	−0.57	ns	−0.063
February	0.36	ns	0.097
March	0.14	ns	0.119
April	−0.51	ns	−0.399
May	−1.10	ns	−0.401
June	−0.67	ns	−0.792
July	−1.50	ns	−2.803
August	−1.53	ns	−1.586
September	1.13	ns	1.458
October	0.00	ns	0.004
November	0.45	ns	0.380
December	0.43	ns	0.043

Note. Test Z = denotes trend direction; Significance = denotes the presence or absence of statistical significance; Q = denotes trend magnitude.

). The Sen’s slope (Q) test indicated that the total annual rainfall in the municipality of Dschang decreased at a rate of 4.94 mm per annum (Figure 4).

Rainfall was shown to have decreased in the months of January, April, May, June, July, and August, with the highest decrease noticed in the months of July and August (Q = −2.803 mm per annum and −1.586 mm per annum, respectively) (Table 1). However, the monthly trend results are statistically non-significant. The second maize-growing season generally begins at the end of July or the beginning of August. Thus, a decrease in rainfall within these months will jeopardize the second maize planting and growing season in the long term in the study area.

Similarly, the month of April represents the first maize-growing season. This implies that a decrease in rainfall at this point is highly detrimental to the young plant, as the plant requires enough water to develop and grow. Inadequate rainfall supply to the plant may lead to hydric stress, which will hinder the plant’s development or lead to plant death.

4.1.2. Trend Analysis of Maximum Temperature

The MK and Sen’s slope test results demonstrate that, on an annual basis, the maximum temperature is increasing (Figure 5) at a rate of 0.011 °C per annum (

Table 2. Summary of MK and Sen’s slope estimation for maximum, minimum, and mean temperatures in Dschang (1990–2022).

Time Series	Maximum Temperature			Minimum Temperature			Mean Temperature
	Test Z	Significance	Q	Test Z	Significance	Q	Test Z	Significance	Q
Annual	0.84	ns	0.011	−3.55	***	−0.039	−1.22	ns	−0.011
January	2.28	*	0.045	0.00	ns	0.000	1.67	+	0.026
February	0.84	ns	0.022	−2.60	**	−0.061	−0.03	ns	0.000
March	−1.58	ns	−0.036	−2.20	*	−0.050	−2.88	**	−0.040
April	0.31	ns	0.013	−3.10	**	−0.080	−2.03	*	−0.040
May	−2.09	*	−0.039	−3.66	***	−0.086	−4.17	***	−0.063
June	−0.59	ns	−0.009	−2.70	**	−0.047	−2.43	*	−0.032
July	1.77	+	0.043	−1.83	+	−0.049	−0.15	ns	0.000
August	−0.67	ns	−0.015	−1.95	+	−0.051	−1.55	ns	−0.023
September	−0.85	ns	−0.023	−0.33	ns	−0.006	−0.54	ns	−0.011
October	0.88	ns	0.020	−2.23	*	−0.050	−0.36	ns	−0.005
November	1.26	ns	0.029	−1.47	ns	−0.040	−0.77	ns	−0.013
December	0.39	ns	0.014	1.24	ns	0.025	0.43	ns	0.008

Note. ns = not significant; * = significant at 0.05 level of statistical significance; ** = significant at 0.01 level of statistical significance; *** = significant at 0.001 level of statistical significance; + = significant at 0.1 level of statistical significance.

), with no statistical significance. This is an indication that, in the next 10 years, the increase in maximum temperature will be 0.11 °C, implying possible negative repercussions on the stages of maize cultivation for both the first and second cycles. According to [24], although temperatures ranging between 20 and 30 °C are necessary to initiate seed germination, a uniform temperature of 20 °C is optimal for the growth of maize seedlings. Hence, a consistent increase in maximum temperature over the years may negatively affect the maize plants’ development and growth.

4.1.3. Trend Analysis of Minimum Temperature

Minimum temperatures (T_min) from 1990 to 2022 are decreasing (Figure 6) by 0.039 °C per annum (Table 2) as attested by the MK and Sen’s slope tests, with significance at 0.001 level of statistical significance. The months with the highest reduction in T_min occur to be April (0.080 °C per annum) with significance at 0.01 level of statistical significance, and May (0.086 °C per annum) with significance at 0.001 level of statistical significance, respectively. April and May correspond to the sowing, growth, and maintenance stages of the first cycle of maize cultivation. According to [25,26], cited in [27] Mbiadjeu et al. (2021), maize requires a minimum temperature of 10 °C for active germination and at least 18 °C for flowering. Therefore, a steady decrease in minimum temperature may have unfavorable effects on maize growth in the next decades. Furthermore, the fact that several months show statistical significance indicates that minimum temperature has a potentially high influence on maize yield.

4.1.4. Trend Analysis of Mean Temperature

The outcome of the trend analyses showed that the mean temperature (T_mean) from 1990 to 2022 is decreasing (Figure 7) and the estimation of Sen’s slope (Q) gave a rate of decrease of 0.011 °C per annum (Table 2). March, April, and May exhibit the highest reduction in mean temperature with 0.04, 0.04, and 0.063 °C per annum, with significance at 0.01, 0.05, and 0.001 levels of statistical significance, respectively.

4.2. Trend Analysis of Relative Humidity

There exists an upward monotonic trend in relative humidity, as confirmed by the MK and Sen’s slope tests (Figure 8). On an annual basis, relative humidity is increasing at a rate of 0.25% (

Table 3. Summary of MK and Sens’s slope estimation for the mean of relative humidity (1990–2022).

Time Series	Test Z	Significance	Q
Annual	3.77	***	0.250
January	1.92	+	0.558
February	2.62	**	0.726
March	1.77	+	0.158
April	1.83	+	0.064
May	1.77	+	0.065
June	3.53	***	0.095
July	2.25	*	0.076
August	2.88	**	0.088
September	1.24	ns	0.024
October	2.60	**	0.088
November	2.67	**	0.232
December	3.78	***	0.703

Note. * = significant at 0.05 level of statistical significance; ** = significant at 0.01 level of statistical significance; *** = significant at 0.001 level of statistical significance; + = significant at 0.1 level of statistical significance.

). Ref. [28] suggests that high relative humidity may attenuate the effects of heat on maize seeds by sustaining pollen viability and a high silk emergence ratio. However, a persistent increase in relative humidity over the years may have negative consequences on maize yield. As demonstrated by [28,29], the number of hours of daytime air temperature between 20 °C and 30 °C, and a nighttime relative humidity ≥90% is highly correlated with gray leaf spot severity during the growing season. Furthermore, increasing relative humidity is associated with more rapid growth of molds, and insects, facilitating the spoilage of grains [30].

4.3. Trend Analysis of Maize Yield

The trend in maize data from 2000 to 2018 depicts an increasing trend with a magnitude of 29.36 tons per annum (Figure 9). However, this trend is not statistically significant (p = 0.599).

4.4. Climate–Maize Yield Relationship by Pearson Correlation

The outcome of the Pearson correlation (

Table 4. Correlation between climate variables and maize yield.

	Maize Production
	Correlation (r)	p	Significance
Rainfall	0.279	0.247	ns
Tmax	−0.181	0.460	ns
Tmin	−0.291	0.227	ns
Tmean	−0.288	0.232	ns
RH	−0.648	0.003	Significant at α = 0.01

) shows that there exists a weak positive correlation coefficient (r = 0.279) with no statistical significance between rainfall and maize production. This indicates that as rainfall increases, maize production increases, and vice versa. T_max, T_min, and T_mean, all gave a weak negative correlation (r = −0.181, −0.291, and −0.288) with no statistical significance. This implies that as T_max, T_min, and T_mean increase, the quantity of maize produced decreases. Furthermore, only relative humidity displayed statistical significance at a 0.01 level of significance with a moderate negative correlation (r = −0.648), implying that an increase in relative humidity (RH) will cause a decrease in the quantity of maize produced.

4.5. Relationship between Climate Variables and Maize Yield Using Multi-Linear Regression

The MLR analysis (

Table 5. Summary results of the MLR analysis.

Source	SS	Df	MS	Number of Obs.	=19
Model	4,878,718.75	5	975,743.749	F(5, 13)	=6.22
Residual	2,038,119.82	13	156,778.448	Prob > F	=0.0037
				R-squared	=0.7053
				Adj R-squared	=0.5920
Total	6,916,838.57	18	384,268.809	Root MSE	=395.95
Yield	Coef.	Std. Err.	t	p ˃ |t|	[95% Conf. Interval]
std_rain	38.88215	100.4923	0.39	0.705	−178.2183	255.9826
std_max_temp	1570.009	730.3051	2.15	0.051	−7.719458	3147.737
std_min_temp	1050.519	601.7203	1.75	0.104	−249.4185	2350.457
std_mean_temp	−2575.806	1169.354	−2.20	0.046	−5102.041	−49.57012
std_rel_hum	−420.816	101.4952	−4.14	0.001	−639.7486	−201.2145
_cons	2988.337	90.83775	32.90	0.000	2792.094	3184.58

) revealed that 70.53% of the model explains the relationship between all the predictor variables (rainfall, relative humidity, T_max, T_min, and T_mean) and maize yield (the response variable), at a highly statistically significant level (p = 0.0037).

The standardized regression coefficients for mean temperature and relative humidity exhibit statistical significance at p = 0.046 and p = 0.001, respectively. However, these climate variables show an inversely proportional relationship with maize yield, implying that as the relative humidity and mean temperature increase, maize yield decreases.

4.6. Farmers’ Perception of the Effects of Climate Variability on Their Maize Yield

The questionnaire administered to 100 smallholder farmers revealed that 97% of the respondents believed climate variability negatively affects their maize yields and have witnessed a drop in their maize yield over the last two decades. The respondents also pointed out a decrease in rainfall, high rainfall intensity, and irregularity as the main prevailing climatic conditions in the area over the past 20 years, followed by increased temperatures (Figure 10). A diminution in annual rainfall, and an increased irregularity in rainfall patterns coupled with a temperature rise may lead to drought which affects maize yield. The results from the survey confirm those obtained from the Mann–Kendall and Sen’s slope statistical tests, which showed that rainfall is decreasing in the municipality with a magnitude of 4.94 mm per annum, and maximum temperatures are increasing at 0.011 °C per annum.

The survey also pinpointed the vulnerability of farmers’ production to climate variability. It appears that women’s maize outputs are the most assailable (Figure 11); 23% of women have reported being highly vulnerable while 27% reported a moderate vulnerability compared to only 14% of men who reported being highly vulnerable and 3% not vulnerable. This is most probably a result of women’s limited capacity to adapt to climate variability given that women in general are less economically well-off compared to male farm operators.

The level of farmers’ vulnerability is also linked to the locality where their farming activities are carried out. The municipality of Dschang comprises 14 agricultural stations; each station is found in a different locality with varying altitudes. Based on this, the vulnerability of farmers to climate variability in the localities differs (Figure 12). The highest vulnerability is reported to be in the locality of Nteingue (a low-lying area). All the surveyed farm operators (8% of respondents) in this locality have reported their activities as being highly vulnerable to climate variability, as this locality has been hit by floods every 9 months for the past 10 years.

Crop water requirements are not met through rainfall, as evidenced by eighty-nine percent (89%) of the respondents. However, only 7% (Figure 13) of the respondents have an existing irrigation infrastructure (in the form of canal irrigation and rainwater harvesting) to supplement their crop water requirements and to respond to episodes of drought, especially at the beginning of the planting seasons. It can be inferred that irrigation practices are not much rooted in farmers’ cultivation habits in the municipality of Dschang.

4.7. Farmers’ Adaptation Measures to Climate Variability

Crop diversification (a soil conservation technique) is the most applied adaptation measure to climate variability in Dschang, with over 98% of respondents having recourse to it. The second highest adaptation measure is the shift in planting dates, comprising 96% of the surveyed farmers, followed by other soil conservation techniques (93%), particularly mulching and tied ridges. Irrigation, the use of improved maize seeds, and pest control are the least practiced adaptation measures by the farmers, with 7, 15, and 21%, respectively (Figure 14).

4.8. Other Issues Encountered in Farmers’ Maize Cultivation Activities

According to the respondents, climate variability is not the sole cause of maize yield decrease. Other factors also influence farmers’ productivity. These factors include high fertilizer prices, lack of appropriate farming methods, post-harvest loss, low institutional support, lack of access to improved seeds, pest invasion, land degradation, and the state of road infrastructure (Figure 15). The most constraining issues encountered in maize farming activities, according to the respondents, are high fertilizer prices, post-harvest loss, land degradation, and low institutional support. The respondents unanimously agreed to the fact that these factors, in addition to climate variability (decreased and irregular rainfall, and increased temperature), are the ones that greatly hamper their maize crop productivity, with the price of fertilizer perhaps having more negative impacts on productivity than climate variability.

4.9. Assessment of Agricultural Policies/Programs

The KII revealed that, presently, the flagship agricultural programs in the study area are the PCP-ACEFA (Program for the Consolidation and Sustainability of Agro-pastoral Counseling) and the PARPAC (Cameroon Agricultural Production Support Programme) established in 2022, all operating under MINADER. Several themes emerged from the analysis of these interviews: four themes from the interviews with PCP-ACEFA executives and five themes from the interviews with MINADER executives (office heads and Agricultural Station Managers (ASMs)) (Figure 16a,b).

4.9.1. Thematic Analysis of the PCP-ACEFA Program According to Themes

• Theme 1: Agricultural Sustainability

All the participants described the PCP-ACEFA program as a nationwide program that is enshrined in Cameroon’s National Development Strategy to improve food security across the country since its onset in 2008. As such, the participants pinpointed that it was envisioned for the program to achieve agricultural sustainability by the end of 2024 (its expected closing date) through the consolidation and improvement of agro-pastoral counseling systems for producers. Agro-pastoral counseling entails helping producers make informed decisions from production to marketing to improve their competitiveness and income;

• Theme 2: Stakeholders

This theme reflects the participants’ view of the flow of information among some stakeholders. The program’s active stakeholders according to the participants are the French Development Agency (AFD) through the Debt Reduction Development Contract (C2D) and the government of Cameroon through the Ministry of Finance (MINFI) (they provide resources for the project’s unfolding); the MINADER and its officers, all the PCP-ACEFA officers and advisories, as well as the program’s partners (such as Ambre Conseil, IPAVIC, and PLANOPAC), which provide information about the program; and the registered farmers who are the beneficiaries.

One of the participants from MINADER reported the limited flow of information between them and the PCP-ACEFA program officers, and the feeling of being sidelined from the program: “We hardly receive monthly reports of the PCP-ACEFA program’s unfolding of activities. Sometimes they may send us trimestral reports but very little data on the account of their activities can be found in these reports”;

• Theme 3: Achievements; Financing

The top PCP-ACEFA executives empowered to give an account of the program’s progress and statistics revealed that the program has supported 435 clients across the Menoua Division (in terms of durable goods, i.e., goods that cannot be destroyed in a production cycle) since the onset of its third phase in 2018. However, the executives reported that maize farming was not considered a major production group under the program in the production basin of Dschang. Thus, only 11 groups of maize farmers have been supported by the program to date. Seven of these groups were financed for wheelbarrows and sprayers; two groups were financed for motor pumps and sprayers; one group was financed for the construction of a corn crib; and another group was financed for a maize storage warehouse.

A high-ranked executive of the program stated that: “It may be surprising that a program like the PCP-ACEFA has not included the cultivation of maize as a major production group in Dschang. This is because the maize cultivation and commercialization in this locality have proven not to be very profitable… Our objective is to improve farmers’ competitiveness and income… Nevertheless, we have some groups of maize producers that are being followed up and have their projects financed by the program”;

• Theme 4: Capacity-Building and Advisory

This theme highlights the lack of proper support in terms of capacity-building and/or advisory from the program to the farmers as regards climate-related issues. All the participants acknowledged that farmers’ productivity is affected by climate variability, yet the PCP-ACEFA program does not emphasize supporting farmers in building resilience in farming. One of PCP-ACEFA’s representatives in the Menoua Divisional Technical Unit in Dschang indicated: “For now, emphasis on climate resilience in food systems is not yet a priority for the program in the municipality of Dschang. This is because climate variability or change here is less felt than in other regions... Presently, we train the farmers in Dschang in biofertilizer production. We also capacitate them on how to farm following contour lines. Again, in case of a natural disaster such as a flood that devastates maize fields, all we can do is speak to the producers to bring them solace because we cannot control nature”.

4.9.2. Thematic Analysis of the MINADER’s Interviews According to Themes

• Themes 1 and 2: Agricultural advisory; Capacity-building and Training

These themes were grouped because they depict the means employed by MINADER to support producers in the municipality (i.e., through agricultural advisory and capacity-building). The respondents assert that the advisory, counseling, and other agricultural extension services provided to the producers are ensured by the Agricultural Station Managers (ASMs), while the top executives oversee the process. Overall, there are 14 agricultural stations in the municipality and each station is headed by an ASM. One of the participants—a top MINADER Officer explained: “In the case of maize yield, we provide the farmers with advisory on the technical itinerary of production and the ideal sowing periods following the agricultural calendar that our ministry has established. Depending on climate conditions, farmers may be advised to make a shift in planting dates or observe other measures to adapt to prevailing climatic conditions. In terms of capacity-building, producers are trained by our ASMs on techniques to produce biofertilizers to lessen their dependence on mineral fertilizers which are very costly”;

• Themes 3 and 4: Subventions; Agricultural Programs

These themes also mirror how MINADER undertakes its support to farmers. According to the respondents, MINADER makes subventions through agricultural programs like the PARPAC (French acronym meaning Cameroon Agricultural Production Support Programme). In September 2022, a 30–70% subvention to make granular and foliar fertilizers readily available for producers in Cameroon was established. However, some participants opined that although encouraging, this program is facing some bottlenecks, as decried by one of the participants: “The 30–70% fertilizer policy of MINADER is already a step forward, though a weak one. Not only do I perceive the 30% subvention to be small given the high prices of fertilizers and that most of the farmers are poor, but also the mechanisms for the ordering, payment, and acquisition of the fertilizers are very complex, time and money-consuming. In addition to that, the fertilizers are usually not delivered on time by the supplier firms, thus farmers cannot make use of them at the planned time… In the end, this 30–70% fertilizer policy is not seen to play its intended role of helping poor and smallholder farmers acquire fertilizers at affordable prices to boost their productivity and enhance their competitiveness”.

The participants also pointed out that another significant support of MINADER to farmers is the distribution of pure and improved seeds. These seeds are obtained from the research institute IRAD and are distributed free of charge to the producers at the beginning of the planting season. The participants assert that a total number of 152.72 tons of maize seeds were distributed by MINADER in 2021, representing only 2.72 tons of improved seeds and that farmers who use improved maize seeds tend to have better yields with less vulnerability to climatic shocks compared to pure seeds;

• Theme 5: Performance

The respondents’ opinions concerning the performance of MINADER in implementing their advisory and capacity-building activities towards the farmers are generally tepid. MINADER officers who do not work in direct contact with the farmers believe that the Ministry is performing very well. A top executive at the Sub-Divisional Delegation of the Ministry of Agriculture and Rural Development (DAADER-Dschang), brought forward that: “Agricultural advisory and counseling have been very helpful to the producers. The farmers can better plan their farming activities and make provisions for the future based on the agricultural calendar made available every year. However, the fact that producers still face challenges in improving their maize outputs is due to external factors like climate variability, which are things we cannot control. However, MINADER has established the PARPAC program on fertilizers, a policy that will help lessen the woes of farmers”.

On the other hand, the ASMs who are in constant contact with the farmers do not share this opinion. According to several ASMs, the staff at MINADER is reduced, implying that they have a lot of workloads that they cannot effectively accomplish within the generally short, allotted time. Another issue the ASMs have raised is the lack of work equipment to efficiently carry out their agricultural extension services. One of the ASMs, indicated: “Our job seems so easy in the eyes of the outside world. But the truth is that we face a lot of difficulties. There is a very limited number of ASMs to serve all the farmers in the municipality of Dschang. In my opinion, 14 ASMs for 14 agricultural posts are very meager. The ratio ASM/farmer is 1:7611. You can imagine how much work each ASM can have. Additionally, the lack of equipment to perform our tasks is a great restriction. We do not have adequate measuring instruments to measure crop production. Again, every one of us normally needs to have a motorcycle for ease of movement across their area of jurisdiction including the follow-up of activities or even for conflict resolution among farmers. But this is not the case. Another issue is the fact that many of the agricultural stations are not built. This makes it difficult for the ASM to have a precise spot to serve as head office for meetings with farmers, data keeping, and archiving. Therefore, my honest opinion about our performance towards farmers in the provision of agricultural extension services is that we try our best to fulfill this task, but we need to be more equipped to do a better job”.

4.10. Policy Implications

The study provides an avenue to rectify the severe lack of irrigation facilities among maize farmers, to cater for periods of drought in this locality. The promotion of irrigation, the use of concerted water management in low-lying areas and watersheds, and the development of water conservation techniques to prolong agricultural campaigns, as prescribed by the NCCAP and RSDS/NAIP national policies, are shown to be failing. A study by [9] established that a rise of 2.5 °C in Cameroon will lead to a loss of $0.65 billion from agriculture. Furthermore, a $4.56 billion loss in revenue is projected in the case where precipitation is reduced by 14%. Additionally, in Ref. [31], it was found that a 1 °C increase in temperature led to a decrease in a net farm income of 2200.20 FCFA ($3.67) per hectare. Hence, it is crucial for smallholder farmers to not solely rely on rainfall for their crop water requirements, but to be granted the necessary support to turn to irrigation schemes as crop water supplements.

The study has also shown that there is a very limited use of fertilizers and other phytosanitary products due to high prices, and farm mechanization is completely non-existent among smallholder farmers. As put forward by [32], a reduction in fertilizer use in Cameroon is a serious threat to food security, which may lead to starvation and malnutrition in millions of people in Cameroon and its neighboring countries. In Ref. [32], the authors reinforce this point by mentioning that low fertilizer use in Cameroon will cause increased degradation of the environment through deforestation, soil erosion, and desertification. Furthermore, the assured benefits of fertilizer use to the environment significantly surpass the potential, but uncertain, adverse effects. Hence, the fact that fertilizer use per hectare in Cameroon is lower compared to several other developing countries, opens an avenue for efficient measures to be taken toward improving fertilizer use [32].

Additionally, the study highlights the fact that access to improved seeds as well as to the dissemination of research on seed varieties remain a challenge to many farmers in the municipality of Dschang. The quantities of improved maize seeds distributed annually by MINADER are greatly insufficient to satisfy the growing demand; the PCP-ACEFA program finances only durable goods that do not include seeds; and the research results conducted by research institutes are not adequately disseminated to enhance knowledge sharing among farmers. Thus, in the face of a a varying climate, farmers can no longer solely rely on farm seeds as they are more vulnerable and less resistant to climatic shocks. Lack of access to improved seeds has been identified by [33] as a key obstacle repeatedly reported by farmers in rural areas around the world. The use of improved seeds provides an avenue to resist climatic disturbances such as sporadic rainfall and droughts.

Furthermore, the lack of good road infrastructure to rally farm produce from production basins to external markets is another critical point that the study highlights. The policy of opening up production areas to enable access to the market was instigated by the NCCAP and RSDSP/NAIP policies; however, the municipality of Dschang does not have good roads on which to channel goods. In the case of maize, this situation is a high contributor to the post-harvest losses experienced every year by maize farmers. This is backed up by [34] who opined that access to roads is critical to ensure the fast and effective conveyance of farm produce to nearby or far-off markets in good conditions. Additionally, with an available and accessible road, information about new technologies and climate change manifestations can easily be disseminated. Road accessibility, which may come from rehabilitation or maintenance, can also enhance other income opportunities and decrease farmers’ over-reliance on agriculture as a source of livelihood.

This research has also demonstrated the lack of a consistent series of maize yield data (starting from the 90′s) at the DDADER-MINADER. This calls for the promotion of methodical and continuous crop data collection to guide evidence-based research.

4.11. Proposed Policy Options and Framework

In Ref. [35], it was shown that the development of policy options needs to take into account all considerations about social, economic, environmental, and political feasibility. The assessment of policy options is also necessary to come out with the best policy options. According to [35], the steps in policy options assessment are, as follows: (i) to identify the problem and all possible impacts; (ii) stakeholder analysis; (iii) monetizing the impacts; (iv) defining policy options/solutions; (v) assessing and monetizing these options; and (vi) selecting the best option(s). For this research, the scope of policy options assessment will be limited to (i), (ii), and (iv).

 (i) 

Problem Identification and Possible Impacts

The problem is the lack of resilience in maize farming due to climate variability and other factors in the municipality of Dschang. The main impact reported is the decrease in maize productivity.

 (ii) 

Stakeholder Analysis

It is important to engage stakeholders, as this process can determine the social and institutional factors that can influence the overall success or failure of a policy option or solution. Engaging or involving stakeholders before defining policy options can be beneficial in creating a common understanding of the situation and in providing avenues for effective cooperation among all stakeholders, which will greatly influence the success of the proposed policies [36]. Stakeholder engagement also helps reduce misunderstanding and overlapping of roles, increases credibility, and generates trust among the stakeholders [36]. Stakeholder analysis equally represents a tool with which to identify the social acceptability level of a policy solution. Stakeholder analysis, as proposed by [36] is carried out in the following two steps: (i) identifying the stakeholders; and (ii) analyzing and categorizing the engaged stakeholders. In Ref. [36], it was further suggested that all stakeholders can be identified according to their POWER and INTEREST, from which four categories of stakeholders can be established (

Table 6. Categorization of stakeholders.

Engaged Stakeholders	Stakeholders’ Interest/Role	Category
MINADER	Programs executors: are primary stakeholders as they have a strong interest in achieving the programs’ goals.	B
MINEE, AMEE	Provides technical support for water-related projects	A
MINFI	Provides financial resources	A
Dschang Council	Oversees programs’ execution and supports their sustainability through decentralized powers	B
FEICOM	Supports in the funding of community projects	A
Agricultural Banks	Grant loans and credit to farmers	A
The Private sector (AFD, AfDB, GIZ, UNDP)	Provide technical and/or financial support and advisory	D
Fertilizer suppliers, Agro dealers	Serve as a link between farmers and the government for fertilizer supply	C
IRAD, FASA	Research to enhance scientific knowledge on the effects of climate variability on agriculture	D
Rural engineers, climate experts, agricultural engineers	Propose solutions for rural infrastructure development (e.g., roads)	D
Farmers	Policies’ direct beneficiaries	D
Consumers	Policies’ indirect beneficiaries	D
The media (local TV, Radio, newspapers)	See to the communication and dissemination of information on the policies, and carry out programs’ advocacy	C

Note. Category A: High power low interest; Category B: High power high interest; Category C: Low power low interest; Category D: Low power high interest.

).

 (iii) 

Proposed Policy Options

The policy options developed from this research are, as follows: (i) the development of a low-cost solar-powered rainwater harvesting system for small-scale drip irrigation in farms; (ii) increasing the government’s subvention of fertilizers to at least 50%; (iii) improving road infrastructures; and (iv) facilitating seed multiplication and enhancing the distribution of improved seeds.

According to [37], a policy options matrix shows the projected outcomes of different policy options. A policy option matrix also helps to facilitate decision-making by selecting the best available option. The policy option matrix for this research has been developed based on the farmers’ and key informants’ perceptions of the economic, social, environmental, and political feasibility of each policy option (

Table 7. Policy options matrix.

Policy Option	Effectiveness	Economic Viability	Social Acceptability	Environmental Soundness	Political Acceptability
Irrigation from RWH	Very good	Very good	Excellent	Very good	Good
50% fertilizer policy	Good	Very good	Very good	Fair	Fair
Roads	Very good	Good	Excellent	Good	Good
Improved seeds	Very good	N/A	Excellent	Very good	Fair

Note. N/A: Not applicable.

).

Following the aforementioned policy options, it is essential to suggest a policy framework for building maize resilience in the municipality of Dschang that will bring all stakeholders together to work towards the achievement of this common goal (Figure 17).

5. Discussion

This study highlights the effects of climate variability on maize yield in Dschang-Cameroon and provides an avenue for informed decision-making. The decrease in rainfall and increase in maximum temperature depicted in the statistical analysis and backed by the farmers’ perceptions poses a concern to maize yield in the locality. This is consistent with the results in a study [11] demonstrating similar findings.

Maize yields exhibiting an increasing trend is paradoxical to the views of farmers, who have reported a decreasing trend in maize yield over the last two decades. This might be explained by the onset of agricultural programs in Cameroon, such as the Program for the Consolidation and Sustainability of Agropastoral Counseling—PCP-ACEFA) and the National Agricultural Extension and Research Programme (PNVRA, French acronym) after the 2008 food crisis, which might have had a positive impact on maize yields [13].

In addition to climate variability and insufficient climate adaptation strategies, the study identifies other pre-existing socio-economic and political issues encountered in maize cultivation activities in Dschang, including high fertilizer prices, low institutional support, and bad road infrastructure, seed quality, and seed prices, all of which add to the burden of smallholder farmers in their struggle for agricultural resilience. According to [8], the high prices of farm inputs, particularly fertilizers and improved seed, hinder smallholder farmers’ access to the right quantity and quality of farm inputs. Bad road infrastructure also impedes the flow of farm produce from farm communities to market outlets, leading to post-harvest losses and further deterring the sustainability of smallholder farmers [5].

Furthermore, access to irrigation facilities is limited and innovations to foster adaptation to climate variability are restrained, as farmers rely on rudimentary adaptive measures including shifts in planting dates, crop diversification, and other soil conservation methods like mulching and tied ridges. This is backed by [5], where it was pinpointed that climate change adaptative measures are generally traditional and devoid of innovation. In Ref. [5], the authors also stress that the lack of innovation for climate-smart options in Cameroon increases the vulnerability of the local farming sector to climate shocks, affecting agricultural outputs.

In their zest to sustain the local agricultural sector, the Cameroon Government has set up robust sectoral policies, including the Rural Sector Development Plan/ National Agriculture Investment Plan (RSDS/NAIP) with implementation programs such as the PCP-ACEFA and the PARPAC operating through the MINADER [13]. The key informants in this study have argued that the support, in terms of adaptive capacity, for climate variability and shocks is still lukewarm. Additionally, the programs face some bottlenecks at the monitoring and implementation levels. This is in accordance with [5], in which it was affirmed that the high rate of corruption in Cameroon is an obstacle to the implementation of agricultural sectoral policies, thus affecting advancements in food security, rural livelihoods, and environmental and smallholder sustainability.

6. Conclusions

The study has revealed that climate variability affects maize yields in Dschang, particularly through decreased minimum temperatures and increased relative humidity. The trend in maize yields was observed to have increased by 29 tons per annum, contrary to the perceptions of farmers, who have noted a reduction in maize yields over the past two decades. The study, therefore, indicates that there is a discrepancy between the trend in maize yield data (increasing) and the trend reported by the farmers (decreasing) over the past two decades. Although the farmers attribute their perceived yield decrease to climate variability, they also point to other factors such as the high price of farm inputs and low institutional support. In addition to rainfall and temperature, this study has also highlighted relative humidity as an influential factor in maize yields, which, has not to date been prominent in Cameroon’s literature. In addition, this research has taken a step ahead to address a research gap in Cameroon’s scientific research by providing an in-depth qualitative analysis of the flagship agricultural programs on the ground. Programs aimed at supporting maize farmers are inadequately implemented (in the case of PARPAC) and do not emphasize improving farmers’ adaptive capacities to build resilience in a setting of climate variability (in the case of the PCP-ACEFA).

Leveraging nature-based solutions such as solar-powered rainwater harvesting drip irrigation systems to improve irrigation is essential. Furthermore, supporting farmers to have better access to agricultural inputs (fertilizers and improved seeds) is also imperative. Again, assisting farmers in the effective practice of soil conservation is essential to achieve the expected outcome of increasing maize resilience. Overall, good political will, constant monitoring, and evaluation of agricultural programs are fundamental.

The study was mainly hindered by the lack of availability of a longer set of maize yield data. Thus, the study recommends the scrupulous and consistent collection of maize data by MINADER and research institutes to facilitate subsequent research. Future studies should focus on considering other environmental variables, such as evapotranspiration, wind speed, insolation, and dew point, to forecast the impact of climate change on maize and other crops yields in the locality as well as the use of more advanced models for trend analyses. Future studies can also center on providing an in-depth assessment of the proposed policy options disclosed in the framework of this research. Furthermore, exploring how Cameroon’s situation can be interpreted in relation to neighboring countries and non-African countries with similar climatic conditions is also a valuable basis for future studies.