Evaluating the Resilience of the Cocoa Agroecosystem in the Offinso Municipal and Adansi North Districts of Ghana

Abstract

The application of the resilience concept within socioecological systems has recently received much attention. Assessing the characteristics of cocoa agroecosystems in the dry and moist semi-deciduous ecological zones has become critical for resilience analysis in this era of climate change and the constant shrinking of cocoa suitability areas. Previous studies have used one of the dimensions of resilience to analyse complex adaptive systems, excluding critical factors and variables. This study applied a multi-criteria decision-making process, the Analytic Hierarchy Process (AHP) that accommodates the three dimensions of resilience, i.e., buffer capacity, adaptive capacity and self-organisation. The AHP is a multi-criteria decision-making tool that proceeds with the design of a hierarchy system for the goal, criteria, attributes and variables. Selected cocoa farmers were assigned weights related to criteria, attributes and variables in a comparison matrix. The resilience of the cocoa agroecosystems in Offinso Municipal and Adansi North Districts was 2.75 ± 0.06 (mean ± SD) and 3.23 ± 0.10 (mean ± SD), respectively. Buffer capacity contributed the highest proportion (44.3%) in the Offinso Municipal District, followed by adaptive capacity (38.7%) and self-organisation (17%). A similar trend was recorded for the Adansi North District: buffer capacity (42.9%), adaptive capacity (42.9%) and self-organisation (14.3%). Across the two study areas, shade trees, crop diversification, soil quality, cocoa variety, farm size, farm age, alternative livelihood, annual income and co-operative membership contributed prominently to the construction of cocoa agroecosystem resilience. The assessment of agroecosystem resilience is location-specific, and the study provides a simplified methodology for evaluating resilience. The paper aims to understand the importance of the components of the cocoa agroecosystem, and a simplified methodology for evaluating its resilience to perturbations. It presents a conceptual and methodological framework for the analysis and measurement of agroecosystem resilience in a participatory manner.

1. Introduction

The resilience concept, especially its application within socioecological systems, has received much attention over the past few decades. Assessing the characteristics of cocoa agroecosystems in the dry and moist semi-deciduous ecological zones has become critical for resilience analysis in this era of climate change and the constant shrinking of cocoa suitability areas. In Ghana, a cocoa agroecosystem could be a cocoa agroforestry system with considerable shade or full-sun cultivation [1]. Whereas the latter is perceived to give a higher yield, the former is practised for particular reasons such as sustainable yield, soil erosion control and soil fertility [1,2]. The physical, biological, socioeconomic and cultural subsystems within an agroecosystem coalesce and interact within the limits of human-led agricultural processes [3]. An agroecosystem is considered beyond the biophysical space it occupies and is also influenced by symbolic, economic, political and technological relationships that interact on multiple scales and hierarchies [4]. The cocoa agroecosystem is a socioecological system conceptualised from the perspective of a “fully integrated system(s) of people and nature” [5].

Within complex adaptive systems like the cocoa agroecosystem, applying systems thinking in resilience analysis provides an approach that integrates ecological and social aspects [6,7,8]. Increasing resilience means reducing the system’s sensitivity, for example, to climate change, and finding ways to better deal with risks, shocks and uncertainties. The cocoa agroecosystem’s resilience is influenced by the nature and structure of the components of the system, the complexity of the interactions between them and external factors. By investigating the socioecological resilience of the cocoa agroecosystem from the perspective of systems thinking, the link between resilience and sustainability can be drawn without difficulty.

Both concepts, sustainability and resilience, have a long history in science and management, but the latter is rooted in engineering science. Currently, the resilience concept is explored for diverse uses and applications in natural science. This study is hinged on the evolution of resilience from a purely ecological perspective to that of complex systems analysis [9]. Until the dawn of ecological resilience studies, it was conceived as the capacity to confront, absorb and adapt to disturbances, without changing, to return to a state of normality [10,11]. Within the engineering space, the amount of time it would take a system or an object to return to that state of normality is evaluated as resilience [12]. However, the socioecological system context challenged the idea of normality, adopting an understanding of multiple equilibriums and accepting the inevitability of change [7,9,13]. It has been proposed that resilience is a system’s adaptation based on learning, planning and reorganisation to preserve function, structure and identity [14]. Resilience to climate change is an important factor because it measures the capacity of social-ecological systems (SES) to absorb recurrent disturbances to retain essential structures, processes and feedbacks [15,16]. Resilience is classified into buffer capacity, adaptive capacity and self-organisation in this study [17].

A key element is a feedback mechanism, which enables farmers to receive signals, process and interpret them, and respond with adequate changes in their management practices [8]. The environmental and anthropogenic stressors within the cocoa agroecosystem emphasise the need to improve the adaptive capacity of farmers. This can be achieved through learning to live with change and uncertainty; nurturing diversity for resilience; combining different types of knowledge about complex systems; and creating opportunities for the sustainable management of the social-ecological system [18]. In this study, resilience is referred to as the ability of a cocoa agroecosystem to absorb disturbance and retain its basic functions and structure [11]. It would therefore be able to deliver outputs in the form of goods and services for the benefit of humanity.

The shaded and full-sun cultivation systems are different cocoa production systems in Ghana that differ in structure, composition and management style. Studies on the factors that promote or undermine the resilience of cocoa production systems to anthropogenic and environmental stressors are critical to the course of sustainability. In the absence of a methodological framework that considers resilience from a system’s perspective, cocoa farmers seemingly carry out all kinds of practices in an attempt to alleviate the shocks. These sometimes result in maladaptation, and the cost to cocoa farmers is enormous. Previous studies on cocoa agroecosystem resilience to climate change in Ghana did not expatiate on the three dimensions of resilience. This should be a cause for concern to stakeholders in the cocoa industry, given the contribution of cocoa to economic growth, the need to improve resilience and the goal for Ghana to regain its status as the world’s leading producer of cocoa.

This paper aims to provide a better understanding of the importance of the components of the cocoa agroecosystem and a simplified methodology for evaluating its resilience to perturbations. It presents a conceptual and methodological framework for the analysis and measurement of agroecosystem resilience in a participatory manner. The Analytic Hierarchy Process (AHP) method, allows for wider participation and detailed, context and location-specific resilience analysis. Estimating the cocoa agroecosystem’s resilience using the three dimensions can help reduce the conceptual difficulties with the resilience approach that make it difficult to translate the resilience theory into practice. The objective of this study is (1) to measure the resilience of the cocoa agroecosystem by buffer capacity, adaptive capacity and self-organisation estimates and (2) to identify context-specific indicators to determine the resilience of the cocoa agroecosystem.

Resilience and Systems Thinking

The study of cocoa agroecosystem resilience is rooted in systems theory. The approach requires the definition of goals and objectives, boundaries, and the structure and function of the system’s components [19]. The systemic perspective applies a global vision that underlines a system’s function rather than simply breaking it down into constituent parts and then re-forming it [20]. It shows the elaborate structure of the system and how changes in one component produce effects elsewhere. Systems theory is an interdisciplinary theory about every system in nature, in society and in many scientific domains as well as a framework with which we can investigate a phenomenon using a holistic approach [21]. Systems thinking comes from the shift in the paradigm from the assessment of parts to the whole [21,22] considering the observed reality as an integrated and interacting unicum of a phenomenon, where the individual properties of the single parts become indistinct. The major difficulties with the utilisation of systemic analysis revolve around (1) the time- and context-dependent nature of problems; (2) the types and role of interactions in systems theory and human system analysis (e.g., how feedback can determine system behaviour); (3) types of mechanisms of transformation and adaptation in interrelated human and natural systems (e.g., how general is the adaptive cycle model); (4) how to identify and address mismatches of scale between human actions and responsibility and natural interactions [23]. More importantly, a fundamental notion of general systems theory is its focus on interactions that differentiate between the behaviour of a single autonomous element and when the element interacts with others [20]. Social-ecological systems are regarded as complex adaptive systems, and hence the social-ecological resilience approach is used as a lens to address and understand their dynamics [24].

The study emanates from the systems thinking viewpoint, and the resilience of a socioecological system is woven around the three dimensions of resilience (buffer capacity, adaptive capacity and self-organisation). The buffer capacity (BCⁱ), adaptive capacity (ACⁱ) and self-organisation (SOⁱ) indicators have a clear relationship with the related ecosystem (REⁱ) and the social, economic and political (Sⁱ) setting. The conceptualisation of cocoa agroecosystem resilience was inspired by Ostrom’s analysis of social-ecological systems [25] and the farm resilience conceptual framework [6]. This paper presents a conceptual framework (Figure 1) that highlights feedback loops within the cocoa agroecosystem and the interaction among its components and with external systems.

Within the construct of the resilience of a socioecological system, to buffer is understood as to cushion, soften and reduce shocks, neutralise intensity, decrease variation and resist change [17]. Therefore, buffer capacity explains the capacity of the cocoa agroecosystem to cushion change, and possibly take advantage of emerging opportunities to achieve better resilience outcomes such as sustained production. It is the amount of change the system can undergo and retain the same structure, function, identity and feedback on the function and structure [26,27]. Again, the ability of a system to retain basic functions while tolerating disturbance, which, by extension, determines its ability to cope and adapt, describes its buffer capacity [15,28].

The exploration of socioeconomic characteristics is best achieved through the analysis of adaptive capacity and self-organisation variables. Adaptive capacity explains the capacity to respond by finding alternative coping strategies, involving a learning process and transformation, which brings about leadership and empowerment processes [29]. It is a component of resilience that reflects the learning aspect of system behaviour in response to disturbance [30]. The various definitions suggest that the initial adaptive capacity of a system must be examined before the disturbance and then to find out what happens after the event as the system adjusts and copes with the effects. Adaptive capacity is not the same as adaptation; adaptations are manifestations of adaptive capacity and represent ways to reduce vulnerability to change. To better distinguish between the terms, four factors have been identified as fostering adaptive capacity: (1) learning to live with change and uncertainty; (2) nurturing diversity for reorganisation and renewal; (3) combining different types of knowledge for learning; (4) creating opportunity for self-organisation towards social-ecological sustainability [31]. As impacts such as pest and disease invasion, drought and extreme weather events vary in occurrence, increasing or building the adaptive capacity of cocoa farmers improves their ability to manage the various impacts while maintaining flexibility for adjustment towards future occurrences.

Self-organisation has been conceptualised in both a general systemic and a specific autonomous form. General self-organisation in social systems refers to the spontaneous emergence or re-creation of society (rules, norms, values, and organisation) through a dialogue between social structures and human actions, without explicit control or constraints from outside the system [32]. Autonomous self-organisation, on the other hand, refers to a state where actors determine their own rules. Self-organisation is how a system or community modifies its internal structures and behaviours, often in response to internal growth and external change. Our understanding of resilience includes self-organisation, which implies that adaptation measures are organised by the actors according to their own needs and visions [2]. Self-organisation is seen as the opposite of either “lack of organisation, or organisation forced by external factors” [26].

2. Materials and Methods

2.1. Study Area

Cocoa as a tropical crop is specific in its climatic and soil requirements. In Ghana, cocoa can only be profitably grown in the tropical rainforest belt. The tropical rainforest of Ghana is divided into five major forest types (Figure 2) namely: wet evergreen; upland evergreen; moist evergreen; moist semi-deciduous (northwest and southeast subtypes) and dry semi-deciduous (fire zone, inner zone and marginal subtypes).

The study was conducted in the dry (Offinso Municipal District) and moist (Adansi North District) semi-deciduous ecological zones of Ghana, which are also separated by political boundaries (Figure 3). The two areas have diverse forest structures and resources, and different cultural norms and traditional administration; however, both are major cocoa-producing areas in the Ashanti Region of Ghana [34,35].

The Forest Services Division and Ghana Cocoa Board (COCOBOD) have district offices overseeing the sustainable management of the forest and cocoa farms, respectively. It has been widely published that the expansion of cocoa farming has led to massive deforestation in cocoa-growing areas [36,37,38]. These indications are crucial for creating a robust cocoa industry to support the growing populations in these areas.

Despite the differences in the ecology, both areas share common production and management practices. In both study areas, the cocoa cropping year begins in October, with purchases of the main crop, while the smaller mid-crop (light or lean crop) cycle is in July. Though light-crop beans have smaller sizes than the main-crop beans, both are of the same quality [39]. The average farm size of a cocoa farm is 4 Ha and is owned by families or individuals [40]. Tenure of cocoa farms may be through rent, hire (for a specific number of years), sharecropping or be farmer-owned (inherited or purchased) [41]. Largely, the heads of cocoa-growing households are men with an average age of 55 years [42].

The detailed characteristics of the two study areas are described in

Table 1. Summary description of the Offinso Municipal and Adansi North Districts.

Attribute	Offinso Municipal District	Adansi North District
Location	The municipality lies within longitude 1°45’ and latitude 6°95’ N. The municipality also shares a common boundary with the Ahafo Ano South District in the west, whilst its eastern and southern borders are shared with the Afigya Kwabre District.	The district lies between longitude 1°50’ W and latitude 6°17’ N. The district shares a boundary with Amansie Central in the northwest, Obuasi Municipality in the south, Adansi South District, and Bekwai Municipality in the north.
Demographics	A total population of 76,895 with 48% males, and a land area of about 1451 km2. Agriculture is mostly rain-fed and employs about 68% of the population. About 3455 people are engaged in cocoa farming.	A total population of 107,091 and a 1140 Km2 land size. There are 23,863 households in the district, out of which 2540 farmers are engaged in cocoa farming. Agriculture is rain-fed and employs about 77% of the labour force.
Physical features	Generally undulating topography with an elevation of about 277 metres above sea level. The three main underlying rock types are the Voltaian, Birimian and granite rock types. The soil is porous and the area is drained by four main rivers, the Offin, Anyinasu, Ode and Pro.	The topography of the area is generally a gently rolling landscape drained by the Tano River (dendritic flow pattern) and some streams, notably the Bemin, Fum, Gyimi, Kyeabo, Ankafo, Adiembra, Asabri, Subine, Konwia, Kyekye and Atraime. The elevation ranges between 213 and 274 meterst above sea level. The Birimian (mineral deposits) and Dahomeyan formations and ochrosols characterise the soil type, which is said to be rich in humus content.
Vegetation	The vegetation is characterised by a dry semi-deciduous type of forest, with interspersed thick green cover. Forest reserves including Asufu East, Asufu West, Giamaian, Kwamisa, Opro and Afram Headwaters are suffering serious degradation.	This is a moist semi-deciduous (MS) ecological zone and said to be the most productive in Ghana. Some forest reserves are the Fum Headwaters at Fumso, Adu Kofi Forest Reserve at Nyankomasu, Dampayaw Forest Reserve at Yaw Dankwakrom and the Kusa Ranges.
Climate	Bi-modal rainfall pattern. Mean annual rainfall is 103.8 mm; temperature ranges between 27 °C and 32 °C. Major and minor rains occur in April to June and September and October, respectively.	Semi-equatorial climatic region and bi-annual rainfall pattern. Major rains occur from April to July and minor in September to December. The mean annual rainfall is 1250–1750 mm, and the mean annual temperature is around 26–27 °C. During the rainy season the area records a relative humidity of 80%, and of 70% in the dry season.

Source: [34,35,43,44].

according to their demographics, physical features, vegetation and climate.

2.2. Research Design, Sampling Technique and Size

A multistage sampling approach was adopted to study the resilience of cocoa production systems in the study areas with different climatic descriptions and conditions. The first stage involved the selection of the dry and moist semi-deciduous ecological zones to reflect the two major cocoa production zones in Ghana. In the second stage, the Offinso Municipal and Adansi North Districts were purposively sampled to represent the dry and moist semi-deciduous ecological zones, respectively. The selection was based on the contribution to cocoa production, accessibility to cocoa farms, distribution of cocoa agroforestry systems [45] and willingness to participate and contribute to the project. The individual study communities were also purposively sampled at the third stage according to their proximity to the district capital, accessibility to the cocoa farms of the respondents and the existence of viable Farmer Co-operative Groups. In consultation with the COCOBOD Extension Officers in the two study areas, ten communities, five from each study area, were selected—Abofour, Camp 31, Koforidua, Sampronso and Aburokyire in the Offinso Municipal District and Ayokoa, Akrofuom, Brofoyedu, Atetem and Anwiaso from Adansi North District. The final stage involved the selection of farmers and Cocoa Extension Officers to proffer their judgement on the factors that shape the resilience of cocoa agroecosystems.

2.3. The AHP and the Weighting of Criteria, Attributes and Variables

The AHP technique is a multi-criteria decision-making tool that breaks down complex problems into a hierarchical structure to enhance their solution. The tool simplifies complex problems into partial elements, which have hierarchical relationships to each other. Following the hierarchy formation, the criteria are ranked via paired comparisons. The advantage of a paired comparison is that the decision-maker prioritises only two options under comparison, irrespective of the other options. A weighting matrix was built with a hierarchical structure composed of one goal (cocoa agroecosystem resilience), three criteria (buffer capacity, adaptive capacity and self-organisation) that represent the dimensions of socioecological resilience, six attributes (productive practices, farm attributes, production and experience, livelihood diversity, socioeconomic demographics and financial factors) and fifteen variables (crop diversification, shade trees, soil quality, cocoa variety, farm size, farm age, alternative livelihood, annual income, farming experience, multiple farms, farmer age, co-operative membership, educational status, gender, and access to credit (Figure 4).

2.4. Data Collection

The assessment of cocoa agroecosystem resilience in the Offinso Municipal and Adansi North Districts was carried out using both qualitative and quantitative methods. The data collection instruments employed for this study were focus group discussion (FGD) and semi-structured interviews.

Two focus group discussions were conducted in each study area. There were five (5) experienced cocoa farmers (with more than 20 years of farming) in each group, with fair gender representation. Each group had at least a female cocoa farmer. The farmers were selected in consultation with the Cocoa Extension Officers in the respective Districts. During the FGD, the Analytic Hierarchy Process (AHP) was explained to the respondents as a participatory data collection tool that accepts deliberated values (consensus) as scores and weights for the criteria, attributes and variables. All technical terms were broken down into simple, clear language to help the respondents’ scoring process. Key informant interviews were conducted to clarify and improve the understanding of the resilience of the cocoa agroecosystem to climate change. The procedure for evaluating the resilience of the cocoa agroecosystem in the Offinso Municipal and Adansi North Districts using the Analytic Hierarchy Process (AHP) consisted of three phases: (i) selection and weighting of criteria, attributes and variables, (ii) scoring of variables, (iii) assigning quantified values to resilience. The use of AHP to evaluate cocoa agroecosystem resilience, modified from previous studies, proceeded by determining the hierarchical system for the goal, criteria, attributes and variables [9,46] (Figure 4).

Each focus group from the study areas was guided to complete the comparison matrix designed by SCB Associates (https://www.scbuk.com/ahp.html; accessed on 20 May 2023). At each level of the hierarchy, values were assigned to the criteria, attributes and variables in a comparison matrix. For each pair of criteria, attributes and variables, the farmers indicated which was the most important for enhancing resilience and to what extent, using a 1–9 ordinal scale. For example, under variables, if the preference of crop diversification over shade trees was absolute, then the entry (Crop Diversification, Shade Trees) of the matrix was set at 9, and the entry (Shade Trees, Crop Diversification) was set at 1/9. If crop diversification and shade trees were of equal importance, then (Crop Diversification, Shade Trees) and (Shade Trees, Crop Diversification) were set at 1, and so on. The preference for one variable over another can be “absolute” (9), “very strong” (7), “strong” (5), “moderate” (3) or “equal” (1). Intermediate values are possible (2, 4, 6, 8) [47].

This methodological approach is mixed method and participatory since cocoa farmers are consulted to determine both qualitative and quantitative values of factors, criteria and variables. The participatory nature of the AHP allows the methodology to be applied to or reproduced in other territorial contexts because the variables are location-specific and are evaluated by persons well familiar with the study area and subject matter. The matrices developed by the cocoa farmers were scrutinised by experts who were predominantly Cocoa Extension Officers. The cocoa farmers and experts interviewed played a critical role in the planning and development of the methodology, as well as in the analysis and discussion of the results. To satisfy the ethical consideration of scientific research, the consent of the interviewees was sought and the aim of the research was explained before commencing the exercise. Interviewees were assured of the confidentiality of their views and the information given.

2.5. Assigning Scores to Variables

The variables considered for the evaluation of cocoa agroecosystem resilience were reported in different measurement units. For instance, the ages of farmers and farms were expressed in years, while the total cocoa yield was measured in kilograms, and others such as cocoa variety or gender were purely qualitative. Given these variations, all measurement units were transformed to a standard scale of 0 to 5, where 0 represents the lowest level of contribution to resilience and 5 is the highest. This methodological strategy has been utilised and validated in several studies [47,48,49,50]. The values were negotiated in a participatory manner by the respondents.

2.6. Data Analysis

The weights assigned to each criterion and attribute as well as the variables were entered in a Microsoft Excel sheet using the SCB Associates AHP Excel Template. Once the comparison matrix was obtained, the eigenvector and criteria weights were automatically calculated by the SCB Associates AHP Excel Template. To reduce the subjectivity of the estimation and potential personal biases of the cocoa farmers, the comparison matrices were checked for consistency. According to [47], the consistency index (CI) is used to verify the consistency of the comparison matrix. If the CI is less than 0.1 or 10%, the set of judgments can be regarded as reliable, otherwise, it is reviewed to meet the consistency condition. The percentage contribution (weight) of each variable, attribute and criterion was retrieved. The scores assigned to the variables on a scale of 0 to 5 to standardise the measurement unit of each variable were averaged. The value of cocoa agroecosystem resilience was computed from the sum of the product of the 15 weighted variables and their average scores, where AgRe is agroecosystem resilience; Vi is the average variable score and Wi is the weight of the variables.

$$AgRe={\displaystyle \sum }_{i=1}^{i=x}\mathrm{V}i*\mathrm{W}i$$

(1)

A t-test analysis was conducted on the agroecosystem resilience values to determine whether or not there was a statistical difference between the two study areas. The p-value, mean and standard deviation values of the variables were calculated. A radar diagram was used to illustrate cocoa agroecosystem resilience in the Offinso Municipal District and Adansi North District.

3. Results

The evaluation of the resilience of cocoa agroecosystem in the Offinso Municipal and Adansi North Districts carried out in a participatory manner revealed the final values that represent the findings and the thinking of cocoa farmers about the characteristics and farming conditions in the two study areas.

Table 2. Weighting matrix of criteria, attributes and variables for the assessment of cocoa agroecosystem resilience in the Offinso Municipal District.

Criteria	Attributes	Variables
Buffer Capacity (44.3%)	Productive Practices (69%)	Food Crops (6%)
		Shade Trees (10.7%)
	Farm Attributes (31%)	Soil Quality (10.1%)
		Cocoa Variety (7.3%)
		Farm Size (8%)
		Age of Farm (9.7%)
Adaptive Capacity (38.7%)	Livelihood Diversity (50%)	Alternative Livelihood (6%)
		Annual Income (9.1%)
	Production and Experience (50%)	Farming Experience (3.9%)
Self-Organisation (17%)	Socioeconomic Demographics (80%)	Multiple Farms (5.2%)
		Age of Farmer (4.4%)
		Co-operative Membership (6.7%)
		Educational Status (3.8%)
		Gender (3.7%)
	Financial Factors (20%)	Access to Credit (5.3%)

and

Table 3. Weighting matrix of criteria, attributes and variables for the assessment of cocoa agroecosystem resilience in the Adansi North District.

Criteria	Attributes	Variables
Buffer Capacity (42.9%)	Productive Practices (54%)	Food Crops (6%)
		Shade Trees (11.5%)
	Farm Attributes (46%)	Soil Quality (10.7%)
		Cocoa Variety (7.7%)
		Farm Size (7.9%)
		Age of Farm (10.6%)
Adaptive Capacity (42.9%)	Livelihood Diversity (48%)	Alternative Livelihood (6.5%)
		Annual Income (8.8%)
	Production and Experience (52%)	Farming Experience (3.8%)
Self-Organisation (14.3%)	Socioeconomic Demographics (70%)	Multiple Farms (5.4%)
		Age of Farmer (3.6%)
		Co-operative Membership (5.7%)
		Educational Status (3.3%)
		Gender (3.6%)
	Financial Factors (30%)	Access to Credit (4.9%)

summarise the weighting matrix completed by the respondents from Offinso Municipal and Adansi North Districts, respectively, from each community as well as including expert opinion from several disciplines (agriculture, agroforestry, forestry and social science) (Table 2 and Table 3). In

Table 4. Resilience scoring matrix.

Criteria	Attributes	Variables	Definition/Question	Answers	Offinso Municipal (Score)	Adansi North (Score) 1
Buffer Capacity	Productive Practices	Crop Diversification	Other food crops apart from cocoa	Yes	5	5
				No	1	3
		Shade Trees	Trees growing on the farm for shade	Yes	5	5
				No	0	1
	Farm Attributes	Soil Quality	Fertility of soil to support cocoa production	Yes	5	5
				No	0	4
		Cocoa Variety	Breed of cocoa planted	Hybrid	5	5
				Amazonia	4	4
				Tetteh Quarshie	3	3
				Does not know	0	0
		Farm Size	Total size of the farm in Ha	6.1–12	5	5
				0.1–6	0	0
		Age of Farm	Total age of the farm in years	<20	5	5
				>20	0	1
Adaptive Capacity	Livelihood Diversity	Alternative Livelihood	Alternative livelihoods plied by farmer	Yes	5	5
				No	1	1
	Production and Experience	Annual Income	Total revenue accrued over the year in USD 2	>15,000	5	5
				12,001–15,000	4	4
				9001–12,000	3	3
				6001–9000	2	2
				3000–6000	2	1
				<3000	1	1
		Farming Experience	Experience in cocoa farming	Yes	5	5
				No	0	0
Self-Organisation	Socioeconomic demographics	Multiple Farms	Owner of two or more cocoa farms	Yes	5	5
				No	1	3
		Age of Farmer	Age of farmer in years	<40	5	5
				>40	3	4
		Co-operative Membership	Member of Farmer Co-operative Association	Yes	5	5
				No	0	0
		Educational Status	Level of Education	Literate	3	4
				Illiterate	0	0
		Gender	Gender of farmers	Male	5	5
				Female	4	4
	Financial Factors	Access to Credit	Ability to obtain financial services	Yes	5	5
				No	0	1

¹ 0—Poor; 1—Low; 2—Average; 3—High; 4—Higher; 5—Highest. ² 1 GHS = USD 7.1122.

, the consolidated matrix of criteria, attributes, variables and scores is captured.

Quantitative Evaluation of Cocoa Agroecosystem Resilience.

Agroecosystem resilience (AgRe) computed from the weighted variables (Wi) and the average scores (Vi) is summarised in

Table 5. Results of the quantitative evaluation of cocoa agroecosystem resilience in the Offinso Municipal and Adansi North Districts in Ghana.

Variables	Average Score for Offinso Municipal District	Average Score for Adansi North District	Weighted Average for Offinso Municipal District	Weighted Average for Adansi North District	AgRe of Offinso Municipal District ${\displaystyle {\displaystyle \sum }_{\mathit{i}=1}^{\mathit{i}=\mathit{x}}}\mathit{V}\mathit{i}*\mathit{W}\mathit{i}$	AgRe of Adansi North District ${\displaystyle {\displaystyle \sum }_{\mathit{i}=1}^{\mathit{i}=\mathit{x}}}\mathit{V}\mathit{i}*\mathit{W}\mathit{i}$
Food Crops	3	4	0.06	0.06	0.18	0.24
Shade Trees	2.5	3	0.11	0.12	0.27	0.35
Soil Quality	2.5	4.50	0.10	0.11	0.25	0.48
Cocoa Variety	3	3	0.07	0.08	0.22	0.23
Farm Size	2.5	2.5	0.08	0.08	0.20	0.20
Age of Farm	2.5	3	0.10	0.11	0.24	0.32
Alternative Livelihood	3	3	0.06	0.07	0.18	0.20
Annual Income	2.83	2.67	0.09	0.09	0.26	0.23
Farming Experience	2.5	2.5	0.04	0.04	0.10	0.10
Multiple Farms	3	4	0.05	0.05	0.16	0.22
Age of Farmer	4	4.5	0.04	0.04	0.18	0.16
Co-operative Membership	2.5	2.5	0.07	0.06	0.17	0.14
Access to Credit	2.5	3	0.05	0.05	0.13	0.15
Educational Status	1.5	2	0.04	0.03	0.06	0.07
Gender	4.5	4.5	0.04	0.04	0.17	0.16
Agroecosystem Resilience					2.75	3.23
Mean					0.18	0.22
SD					0.06	0.10
p-value					0.07

.

Cocoa agroecosystem resilience (AgRe) of the Offinso Municipal and Adansi North District are 2.75 ± 0.06 (mean ± SD) and 3.23 ± 0.10, respectively. AgRe is higher in Adansi North District than in Offinso Municipal District. The results are graphically presented in the Radar Chart (Figure 5).

Discussion

Buffer capacity and adaptive capacity were adjudged the most important factors in building the resilience of the cocoa agroecosystems in the study areas. Whilst buffer capacity contributed 44.3% in the Offinso Municipal District, it scored the same percentage (42.9%) as adaptive capacity in the Adansi North District. Across the two study areas, shade trees (10.7% and 11.5%) food crops (6% and 6%), soil quality (10.1% and 10.7%), cocoa variety (7.3% and 7.7%), farm size (8% and 7.9%), farm’s age (9.7% and 10.6%), livelihood diversification (6% and 6.5%), annual income (9.1% and 8.8%) and co-operative membership (6.7% and 5.7%) contributed more than 5% to the construction of the agroecosystem’s resilience in Offinso Municipal District and Adansi North District, respectively. A representative from the Federated Commodities Licenced Buying Company stressed the need to integrate desirable shade trees in cocoa farms to provide a microclimate for sustainable cocoa production, control soil erosion and maintain soil moisture. A cocoa farmer shared that “The variation in leaf colour, number of cherelles and fruiting potential of cocoa trees growing under shade and unshaded cocoa trees gives a clear indication of the importance of shade trees to cocoa trees”. Another believed that “despite the established relationship between shade and cocoa yield, failure to control the shade through timely pruning will result in black pod diseases”. These expert judgements and cocoa farmers’ testimonies confirm the extensive discussion on the relevance of shade in cocoa farms [51,52,53]. It was expected that the cultivation of other food crops would record higher figures to signal their unique relevance if cocoa farmers were to be resilient to stressors. A farmer from the Offinso Municipal District reiterated that “Banana is cultivated to initially provide support for the planted cocoa seedlings and later as a source of shade to the farm. The root exudates from banana stems moisturise the soil which stimulates the growth of the planted cocoa seedlings during the dry season. At maturity, they serve as food to the family and are seasonally harvested and sold to earn income to fund our cocoa farming operations”. Per the weighting matrix, another variable that is instrumental in the construction of resilience in the cocoa agroecosystem is the quality of the soil. Reduced soil organic carbon in cocoa production landscapes affects cocoa yield drastically [54,55,56]. A study by [55] revealed that about 50% of farmers draw a relationship between soil fertility improvement and the integration of trees in cocoa fields due to litter fall. Cocoa farmers refer to farmlands that do not produce bumper harvests as “barren lands”. They indicated that numerous efforts and practices such as fallowing, manuring and compost application are low-cost practices that rejuvenate the soil nutrients and structure. In already established cocoa farms, farmers in Ayokoa indicated “the essence of applying compost and manure from poultry droppings but emphasised that skills must be developed through training before applying them in the farms”. Another common practice in such fields that needs no training and is widely practised among cocoa farmers is “proka” (cleared weeds are left on the farmland to decompose and enrich the soil instead of burning or disposing of them).

The resilience of a farm to climate change to any extent is influenced by the breed of crop planted. The cocoa farmers interviewed generally cited productivity and longevity as the main factors to be considered when selecting a cocoa breed for propagation. In a related study, it was reported that factors that influence farmers’ choice of cocoa variety included time to maturity (90%), resistance to pests and diseases (67%), expert advice from extension agents (62%) and availability (49%) [57]. Known to mature early and resistant to drought, pests and diseases, farmers in both study areas scored the hybrid variety 5 (highest) on the 0–5 scale. A farmer noted that “in the unlikely event of the invasion of pest and disease, it is always easier to control and manage an affected hybrid cocoa farm”. Similarly, farmers were critical of the contribution of the hybrid cocoa variety to resilience building as the breed records higher yields at the end of the season. Referred to as “Akokora bedi” (the aged will enjoy the proceeds) in the local parlance, a 62-year-old farmer mentioned that “at her age, she has no option than to plant the hybrid cocoa for her enjoy the proceeds before her death”. Cultivating a cocoa variety characteristic of higher yield and resistance to stressors increases farmers’ purchasing power, granted the farm is well managed and maintained.

The age of a cocoa farm was adjudged a key contributor to the farm’s resilience in both study areas but was higher in Adansi North District (10.6%) than in Offinso Municipal District (9.7%). While yield reduces with the age of the cocoa farm [58], there is a myth about farmers owning old cocoa farms. The Cocoa Extension Officer in the Offinso Municipal District reiterated that “the prestige farmers attach to owning aged cocoa farms is a setback to improving resilience to stressors in many cocoa farms”. The Chief Farmer of Ayokoa recounted that “cocoa trees are like human beings, they lose strength and form after some years and must be replaced with improved breeds” On the other hand, the age of a cocoa farmer noted in other studies as critical to productivity recorded lower weights in both study areas in this research. Previous studies have reported that people in their prime years least participate in the cocoa farming business, and the age of cocoa farmers averages over 60 years [58], 36–60 years [59], and 51.5 years [45,60]. Due to the accumulation of knowledge and experience, the age of a cocoa farmer is an important factor in resilience building. However, the involvement of the youth in cocoa farming gives a promising future to the industry.

The minor and major seasons in Ghana’s cocoa production industry offer unique opportunities for cocoa farmers to ply other trades to gain extra income. The weighting matrix completed by the cocoa farmers explains the need for capital investment in cocoa production if high yields are expected. Aside from the income from other livelihoods that support farm operations and family upkeep, the mere fact that a farmer has diversified his/her income sources reduces risk in the case of any eventuality. Annual income of farmers was equally assigned high values for the same reason, if a resilient and productive cocoa agroecosystem is the ultimate goal of cocoa farmers. The resilience of the cocoa agroecosystem cannot be constructed without the self-organisation capacities of cocoa farmers. Though self-organisation was adjudged the least of the factors that contributed to the evaluation of the resilience criteria, membership of Farmer Co-operative Unions shapes the capacity of cocoa farmers to be resilient to stressors. The Farmer Co-operative Unions form the basis for refresher courses on new and modern methods of cocoa farming, organisation of Farmer Field Schools, farmer–farmer interaction, monitoring of farms by Extension Officers and distribution of farm inputs and other material benefits. Farmers who do not belong to Co-operative Groups forfeit all these and other vital benefits from the Government.

A t-test analysis of the resilience values of the variables produced a p-value of 0.07 > 0.05. There is no statistical difference between the AgRe of the two study areas. However, this study contends that combining the three resilience dimensions for evaluating agroecosystem resilience is more robust than considering just one or two of the dimensions of resilience.

4. Conclusions

The research aims to provide decision-makers in the cocoa industry with a simplified framework for determining the criteria and variables critical to evaluating cocoa agroecosystem resilience. By this means, policymakers will be guided during the formulation and implementation of specific future programmes and interventions necessary to boost cocoa production in the area. The AHP tool used in this study to determine the weights of the resilience criteria, attributes and variables involved farmers in a participatory manner. This approach allows farmers to own, use and transfer knowledge to other farmers. During the weighting process, the deliberated values assigned by the cocoa farmers were based on their relevance and contribution to building cocoa agroecosystem resilience. In all, the study used three (3) criteria, six (6) attributes and fifteen (15) variables for the evaluation of cocoa agroecosystem resilience in the Offinso Municipal and Adansi North Districts.

The findings revealed that the buffer capacity and adaptive capacity dimensions of resilience contributed more than 80% to building cocoa agroecosystem resilience in both study areas. It can be interpreted that managers of cocoa production landscapes and researchers should pay attention to productive practices in cocoa farms apart from efforts by government and non-governmental organisations to assist smallholder cocoa farmers adapt to climate change and other stressors. Diversification of income sources is a decisive factor in the framework for building agroecosystem resilience, but it takes a robust cocoa farm to sustain resilience. The results showed clear similarities for some variables but also differences for others. For instance, the cocoa farmers from both study areas consistently assigned higher values to the practice of maintaining shade trees on farms, improving soil quality, age of farm and annual income of the farmer.

The AHP tool used in the study was subjective, yet the participatory nature and the consideration of many factors improved the interaction and learning process between the researcher, respondents and experts. The consistency index feature of the tool reduced the biases and subjectivity of the assigned values. This makes the replicability of the AHP methodology in other contexts possible, including other indicators and weights that represent the variables valued by a group of respondents, along with their knowledge and perceptions. The use of buffer capacity or adaptive capacity to evaluate the resilience of farming systems is a common practice; however, the AHP tool employed for this study estimated the quantitative contributions of the three dimensions of resilience. This study invites researchers to consider the three dimensions of resilience in future assessments of socioecological systems, if the Analytic Hierarchy Process is to be used. This is because it is participatory and encompasses all components of the system—systems thinking.