Climate and Food Insecurity Risks: Identifying Exposure and Vulnerabilities in the Post-Food Production System of Northern Ghana

Abstract

Evidence shows how food system activities, from production to consumption, underpin food security. However, studies exploring climate impacts on food security in northern Ghana have overly focused on production systems, neglecting post-production activities that loom large in food security. This paper addresses the research need to comprehensively analyze how climate change and weather variabilities affect post-production activities and exacerbate food insecurity risks in northern Ghana. It analyzes data on climate hazards, impacts, and food system vulnerabilities using questionnaires and participatory engagement with farming households in northern Ghana. Results show that climate-induced food insecurity risks in northern Ghana are not just products of persistent exposure to climate hazards and their impacts on food production in the region. Instead, risks are inextricably connected to the vulnerability contexts within which food is harvested, processed, stored, and marketed. Specifically, the results reveal that climate hazard events such as floods, extreme temperatures, and droughts damage stored grain, disrupt food supply to the market, and cause seasonal volatilities in food prices. However, these impacts are not solely externally generated circumstances. The food system is highly vulnerable; most households lack access to threshing and grinding machines, warehouse storage, post-harvest management information, and transportation services. These underlying characteristics of the post-food production system of northern Ghana, which is ultimately quite remote from climate change and weather variabilities, exacerbate household-level food insecurity risks.

1. Introduction

Climate change is increasingly recognized as a key driver in global hunger. Rising temperatures, changing precipitation patterns, and frequent extreme weather events are affecting agricultural yields, disrupting food supply chains, and reinforcing other underlying causes of food insecurity [1,2]. The increasingly erratic rainfall and prolonged dryness intensify food crop susceptibility to pests and diseases, causing yield losses [3,4]. The rise in temperatures and water shortages create problems for food quality and safety, including the outbreak of food-borne pathogens and mycotoxins, more food spoilage, and high food waste and losses [5]. It is estimated that by 2050, the global climate change phenomenon will put millions of people at risk of acute hunger, malnutrition, and poverty [6]. The brunt of this climate-induced hunger will fall on poor households in developing countries, especially across sub-Saharan Africa, South Asia, the Small Island States, and Southeast Asia [7,8,9]. Smallholder farmers in rural areas are expected to be overrepresented in climate-induced hunger situations because of their high vulnerability relating to limited access to climate information, lack of early warning systems, high levels of poverty, and dependency on climate-sensitive food systems and livelihood activities.

In Ghana, an emerging number of climate change studies are paying specific attention to the fragile and climatically vulnerable semi-arid northern region. In particular, these studies present empirical assessments of climate change impacts on the agricultural calendar, land suitability, food production diversity, and yields [3,10,11]. For example, the climate change modelling performed by Armah et al. [3] demonstrates how early onset and delayed rains in rainy season affect timelines for sowing and variety of crops cultivated. Armah and colleagues’ study further presents evidence showing that the increasingly drier conditions caused by rising temperatures and reduced rains would likely lead to a decrease in the suitability of agricultural lands for crop cultivation, with consequences for the region’s overall food production potential. Additional studies, including Antwi-Agyei et al. [12], Balana et al. [13], and Setsoafia et al. [14], have explored how climate-adaptive responses (e.g., irrigation, improved seeds and fertilizers, soil and water conservation, and flood recession agriculture) can reduce agricultural livelihoods vulnerability to climate extreme events and strengthen household food security.

Despite their robust contributions to knowledge, the available studies are overly focused on the food production-related aspects of food security [3,15,16,17]. However, climate impacts on food security extend beyond food production and cultivation systems to the other equally important post-production activities of harvesting, storing, and marketing food crops [1]. Seasonal flooding, erratic rains, and increased temperatures can create problems for food transportation, pricing, and safety. For example, flood waters overflowing narrowly constructed roads in rural communities can impede the safe transport of harvested crops and access to the market centers. Rains intermittently occurring during the harvesting and post-harvesting months can be a great disincentive to farmers preparing to harvest and store crops for future use. Crop losses due to climate impacts on transport networks, harvesting, and storage are significant food insecurity risks as they can lead to the loss of weeks’ worth of a household’s food supply. Although these post-food production-level impacts greatly influence household food security, they are largely ignored and neglected in climate change research in northern Ghana. This paper, therefore, addresses the research need to produce a detailed analysis of climate change impacts on the post-food production activities in northern Ghana. The specific objectives are to (1) identify key climatic hazards affecting post-production-level activities in rural areas of northern Ghana; and (2) assess how the combined community-level exposures to climate hazards and vulnerabilities in post-production-level activities exacerbate food insecurity risks.

The paper utilizes the concepts of food security, food systems, and climate risks and integrates them with frameworks for assessing household vulnerability to climate change and weather variabilities. Presenting these conceptual backgrounds sets the context for understanding all the key food-related activities that are critical to food security in northern Ghana and for understanding how climate change and weather variabilities, as environmental phenomena, interact with the distinctive characteristics of households and communities to induce food insecurity situations. In the sections below, this paper first discusses and reviews these key concepts and their applicability to the research topic and study context. Next, it describes the methodology, focusing on the study context and data collection approaches. It further presents the results and concludes with a discussion of the key findings and what they mean for climate resilience and food security in northern Ghana and other climatically-vulnerable localities of the world.

2. Literature Review

The United Nations defines food security as a situation “when all people, at all times, have physical, social, and economic access to sufficient, safe, and nutritious food that meets their dietary needs and food preferences for an active and healthy life” [18] (p. 3). According to this definition, food security rests on four pillars: availability, access, utilization, and stability [19,20]. Food availability is the quantity of appropriate food available for consumption through production output, stock levels, and net trade. Food accessibility is the ability of individuals, communities, and countries to access the required type, quality, and quantity of food for consumption through the various legal, social, economic, and political means possible. Food utilization refers to the proper biological use and benefits from food; it can be determined by care and feeding practices, food preparation methods, dietary diversity, and nutrient distribution. The fourth pillar, food stability, is a dimension of food security recently added amidst the rising impacts of weather conditions, political instabilities, and economic uncertainties on physical and economic access to food commodities. Sassi [20] describes food stability as the adequacy and reliability of food supplies for all people “at all times” without the risk of deteriorating nutritional status. Therefore, the first three components (food availability, access, and utilization) are necessary for food security, but not enough without stability; there should be adequate food for everyone, not periodically but throughout the year.

In rural communities, most households produce a significant proportion of the food they consume. To ensure stable access, however, households must thresh, dry, process, and store harvested crops for out-of-season months; they must also buy from the market to supplement production in times of crop losses or productivity decline. Thus, food production is important for rural households, but it alone cannot guarantee the stability of a household food supply as well as the access and proper biological benefits for all members. Other equally important food system activities beyond food production need critical attention.

Eriksen [21] conceptualized a food system as comprising a set of activities ranging from production to consumption. These activities are often expressed with phrases like “farm to fork,” “field to table,” and “land to mouth” [22,23]. Ericksen further broadly categorizes four groups of food system activities: (i) food production; (ii) food processing and packaging; (iii) food distribution and retailing; and (iv) food consumption. Food production involves various practices, from accessing and buying farm inputs to preparing the land, planting, caring for crops/animals, and harvesting. Processing and storage activities involve transforming food into finished or semi-finished products, together with quality control and labelling for safety in packaging. Distributing and retailing processes include transportation, trading, and selling food to consumers, while consumption consists of deciding, selecting, preparing, eating, and digesting food.

In recent years, the food system approach has become the most practical approach for analyzing household food security. This approach has been adopted in several studies to identify priority research and investment areas in food security [24,25,26,27]. Drimie et al. [24], for example, use a food system analysis to offer an insightful examination of policies for overcoming environmental constraints that undermine food security in the Southern African region. Drimie and colleagues reveal that climate change is the key environmental driver of food insecurity; it affects food availability and accessibility through impacts on production, biodiversity, prices, and household livelihoods. These findings suggest that resilience against climate change should be a priority in tackling food insecurity in the region. Despite its wide application and popularity, current use of the food system approach has been limited to analyses of the 21st-century global food systems, consisting of complex network of activities and layers of actors. This paper seeks to extend the application of the food system approach from global to micro-level analysis in rural areas of northern Ghana where the most relevant food system activities are limited to the production, harvesting, processing, storage, and marketing of maize, cereals, legumes, and vegetables. In particular, this paper puts a great deal of weight on post-production activities and how the adverse impacts of climate change and weather variabilities on these components of the food systems in northern Ghana predispose households and communities to food insecurity risks.

The concept of risk has evolved, but there is still some degree of ambiguity in the operational use of the term. In climate research, it is unclear whether researchers and practitioners should refer to risk as the probability of occurrence of a hazard event, or the probability of a particular outcome [28,29]. The IPCC provides some clarity in its special report on “Managing the Risks of Extreme Events and Disasters to Advance Climate Change Adaptation,” which defines risk as “the likelihood over a specified period of severe alterations in the normal functioning of a community or a society due to hazardous physical events interacting with vulnerable social conditions, leading to widespread adverse human, material, economic, or environmental effects…” [30] (p. 4). This definition indirectly relates the probability of negative impacts from climate change to the occurrence of a hazard and its interaction with underlying social processes (such as class differentiation, gender, and power relations). In this context, climate-induced food insecurity risk, which is widely used in this paper, is conceptualized as a function of climate hazard event and food system vulnerability (see for example, Welle and Birkmann [31]).

A climate hazard is often understood as any threatening event, or the probability of a potentially damaging event, originating from weather-related hydrometeorological sources such as droughts, floods, bushfires, windstorms, and extreme temperatures. Generally, climate and weather-related hazards occur due to natural causes, but human actions may influence their deliberate or unintended outcomes [32]. Comprehensive knowledge of the physical causal mechanisms of natural hazards is often very limited. However, the Emergency Event Database (EM-DAT) records provide enough information to specify the likelihood of a particular hazard occurring at a specific time and in a specific space. Most researchers assess threats from natural hazards by identifying the frequency and likelihood of their occurrence (see for example, Ahulu et al. [33]; Kadiri and Kijko [34]). Some also focus on an entity’s exposure and the severity of impacts, expressed in terms of the losses and damages generated (see for example, Agada and Nirupama [35]; Zhang and Zhang [36]; Prabnakorn et al. [37]). Assessing the exposure of communities, households, and resources in hazard-prone environments is particularly useful for understanding the extent of the damaging impacts they are likely to face. Such analyses also offer significant contributions to the identification of adaptation and resilience mechanisms.

Vulnerability is the degree to which a system is likely to suffer harm from a hazard [38]. Despite the numerous interpretations of the concept, the literature consistently identifies vulnerability of a socio-ecological system or an entity as a function of three elements: exposure to a particular hazard, sensitivity to that hazard, and the capacity of the system to cope, adapt, and recover from the effects or consequences of the hazard [39,40]. Exposure denotes the degree and extent to which a system comes into contact with or falls within the geographical range of a hazard. Sensitivity describes the degree to which a system will likely suffer harm or significantly change its state when a hazard strikes. The capacity to adapt or cope is the ability to take advantage of, accommodate, or recover from the consequences of a hazard. An example of how these elements combine to determine vulnerability can be given in the case of subsistence households in semi-arid environments. Because of their location, these households are likely to experience dry spells or droughts (exposure); those highly dependent on a rainfed scheme will suffer significant losses whenever dry spells occur (sensitivity); and those with alternative livelihoods and income sources outside of the agricultural sector are in a better position to purchase food to supplement losses from their farms (capacity to adapt). All these three elements are usually incorporated into climate vulnerability assessments in one way or another (e.g., IPCC [41]; Antwi-Agyei et al. [42]; Acheampong et al. [43]; William et al. [44]; Segnon et al. [45]).

There are many frameworks and approaches for assessing vulnerability of households to climate change (see IPCC [41]; Adger [39]; Birkmann et al. [46]; Ribot [47]). One such framework is the sustainable livelihood framework, which is widely rooted in household-level poverty studies and focused on the critical aspects of poor people’s livelihoods (see Reed et al. [48]; Antwi-Agyei et al. [49]). Key components of the livelihood framework include access to assets (e.g., natural, human, social, physical, and financial capital) to pursue income-generating activities and the vulnerability contexts, i.e., exposure to shocks and threats, that are likely to affect earnings from livelihood activities. This framework is particularly relevant because it provides a basis for understanding the vulnerability of activities that poor households pursue daily, including the food production, storage, and marketing activities that underpin food security. Notably, the asset-based concept of the livelihood framework helps identify how asset profiles can influence the susceptibility of livelihood activities to harm as well as the potential for building adaptive capacities and resilience to unforeseen future climatic changes and weather variabilities (e.g., financial capital to purchase crop insurance). Various researchers, including Hahn et al. [50]; Etwire et al. [51]; Antwi-Agyei et al. [49]; Opiyo et al. [52]; and Panthi et al. [53], have applied the asset-based approach to determine household vulnerability to climate change. This research draws on these existing studies to analyze the vulnerability of the post-food production system in Kassena Nankana Municipality to climate hazards. Adopting this approach enables the research to integrate underlying socio-economic characteristics of the study communities for a nuanced understanding of climate and food insecurity risks in northern Ghana.

3. Materials and Methods

3.1. Study Context

The northern part of Ghana is administratively divided into the Upper West, Upper East, Northeast, Northern, and Savannah regions. These regions experience above-average mean temperatures, which usually reach 29 °C or more in the dry season, compared to 25–26 °C in the southern regions [54,55,56,57]. The average precipitation rate in the north is often less than 1000 mm, while it is 1200 mm in the south and can even reach up to 1500 mm in the far southeast. This study was conducted in the Kassena Nankana Municipality in the Upper East region of Ghana, where mean annual rainfall over the past few decades has been sporadic, with frequent surges and relapses (see Figure 1a). One of the highest mean annual rainfalls occurred in 2007 (1203.8 mm); this followed periods of lower rainfall from 2000 to 2006. However, in the subsequent years, the mean rainfall decreased, reaching as low as 617.3 mm in 2014, before increasing again up to 1363.4 mm in 2019. The mean annual temperature of the Municipality also increased by 0.5 °C over the past two decades, from 29 °C in 2000 to 29.5 °C in 2020. However, this increase has also been characterized by significant fluctuations (see Figure 1b). In 2005, the mean temperature increased sporadically by 0.8 °C relative to the record in 2000 and remained high until 2008, when the mean temperature dropped by 0.1 °C below the record in 2000. The mean temperature has since risen, reaching as high as 29.8 °C and 29.5 °C in 2014 and 2020, respectively.

The Kassena Nankana Municipality has a population of 99,895 people, of which 48,658 (48.7%) are male and 51,237 (51.3%) are female [58]. The average household size is approximately 4.1 persons [58]. Ethnicity is mainly Nankani and Kasem. Economic activities are predominantly agricultural—more than 80% of the households earn their livelihoods from overlapping food cropping and livestock rearing [59,60]. Households mainly cultivate maize, rice, millet, guinea corn (sorghum), beans, and groundnuts on farm plots ranging from 0.4 to 4.0 hectares (1–10 acres). Farmers cultivate for subsistence and store their harvests in heaps in their barns or in sacks packed in their rooms. Post-harvest losses are widespread because of poor storage facilities and poor handling of stored crops, which attract rodents, birds, weevils, and grain borers. In addition to agriculture, households in some communities trade in food crops, semi-processed foods, and crafts. Buying and selling mainly occur in the municipality’s only central market in Navrongo and satellite markets in a few communities.

3.2. Research Methods

The data for this research were collected over six months from January to June 2021. The Kassena Nankana Municipality comprises six major communities/townships representing the zonal divisions for local government administration: Navrongo, Doba, Manyoro, Pungu, Kologo, and Naaga (see Figure 2). A multi-stage stratified purposive sampling procedure was used to select two villages from the six communities to be representative of the whole municipality. The villages were selected based on four main criteria: (1) exposure to climate extremes; (2) food insecurity; (3) accessibility for data collection; and (4) willingness to participate.

A mixed method consisting of household surveys and focus groups was used to collect the data for this study. The household surveys involved administering 288 questionnaires using a mix of random and snowballing methods to reach many potential research respondents for a high response rate [61]. The sample size estimation was based on the sampling guidelines of the World Food Programme (WFP) [62]: n = (D)[Z² × P × (1 − P)]/d², where n is the sample size; D is the design effect, ideally represented by 2 for two-stage sampling; Z is the z-value for a confidence level of 95 percent, which is 1.96; P is the prevalence level of food insecurity in the municipality, which was 39.7% as per a recent assessment [63]; and d is the desired precision level of 5%—the rate recommended by the WFP for one-time research and advocacy projects. Substituting these parameters into the formula and anticipating a possible nonresponse rate of 8% yielded a sample of approximately 288, which represents 0.8% of the municipality’s total population of households. The sample was distributed equally among the six communities, that is, there were 48 research participants from each community.

The questionnaire comprised questions on household exposure to climate hazards, perceptions of climate impacts on post-production activities, and access to key livelihood assets for managing these impacts. Specifically, the research respondents were asked to provide a YES or NO response to confirm their perception of hazard occurrences and report their access to post-production resources and services for climate resilience. They were also requested to indicate whether the severity of climate impacts on post-production activities were low, moderate or high, whether there was no impact, or whether they did not know (that is, they had no knowledge). All the questionnaires were administered using Open Data Kit (ODK) (https://getodk.org/), which is open-source software that enables offline data collection and management with mobile devices in remote areas. Although the questionnaires were designed to be answered by a household head on behalf of the entire household, in some cases, especially where the household head was unavailable, adult household members willing to participate were invited to answer the questions. On average, each research participant spent 45–90 min responding to the survey questions. The data gathered on household access to assets were used to compute a vulnerability assessment which drew on existing studies, including Hahn et al. [50]; Etwire et al. [51]; Antwi-Agyei et al. [49]; Opiyo et al. [52]; and Panthi et al. [53]. Here, the asset-based approach to post-food production system vulnerability assessment is expressed as a composite of the resources and services critical for reducing adverse impacts of climate hazards on food processing, storage, and marketing activities. Ten assets were identified as critical in the post-food production system vulnerability analysis for the Kassena Nankana Municipality based on the focus group discussions with the participating households. Out of these assets, six are for food processing and storage and four are for food marketing activities. The roles of each of these assets are presented in

Table 1. Critical assets for food system activities in northern Ghana.

Food System Activities	Assets	Role/Importance
Food crop processing and storage	Threshing machines	Rice and maize threshers enable farmers to quickly process their harvested crops and prepare them for storage without them getting damp due to uncertain rains.
	2. Warehouses	Warehouses enable farmers to safely store and protect food crops against insect pests and theft. Proper warehouse handling enables food crops to be stored throughout the year and beyond. Crops stored in warehouses are usually marked for the market during off-season periods to stabilize seasonal prices and smoothen household consumption.
	3. Chemical storage	Chemicals protect food crops against insect pest attacks and suppress the growth of insects in stored grains, allowing farmers to keep crops for up to eight or more months after harvest.
	4. Storerooms	Some households have dedicated rooms for storing threshed grains and paddy. Crops may be kept in the storeroom for up to six months or more, depending on the farmer’s knowledge of proper storage techniques, the use of chemicals to protect against pests, and the conditions of the storeroom.
	5. Grinding mills	Access to a grinding mill provides a convenient way for a household to mill grains. In communities without access to grinding mills, households may spend hours pounding rice to remove the husks.
	6. Education	Access to knowledge and training on proper harvesting and storage techniques helps reduce post-harvest losses and crop damages and helps save households days’ to months’ worth of food for consumption.
Food crop marketing	Market centers	Market centers serve essential roles as converging points for buyers and sellers; households also obtain access to more diverse food crops than those they produce. More accessible market centers can help lower transaction costs when buying food crops, and this can smoothen consumption in out-of-season months.
	2. Vehicle transport	Access to roads and vehicles linking rural areas to the market centers enables poor farmers to sell their surplus produce at better prices and earn income that can supplement household food needs in the post-harvest season.
	3. Non-farm incomes	Non-farm work such as trading and day labor jobs in construction, weaving, and pottery are great opportunities for households to shift surplus agricultural labor to earn income during the dry season. Income from non-farm work helps resource-poor households secure money for investment in climate-resilient food production activities.
	4. Migrant incomes (remittances)	Remittances from migrant relatives (either in the south or agricultural-rich communities in the north) enable households to buy more food and replenish food stock to deal with the insecurities associated with yield losses from climatic hazards and disasters.

. The asset index for each food system activity is calculated as follows 1^,2:

$$\mathrm{Food} \mathrm{processing} \mathrm{and} \mathrm{storage}{ }^1=\frac{(\mathrm{P}\mathrm{S}\mathrm{t}\mathrm{a}1+\mathrm{P}\mathrm{S}\mathrm{t}\mathrm{a}2+\mathrm{P}\mathrm{S}\mathrm{t}\mathrm{a}3+\mathrm{P}\mathrm{S}\mathrm{t}\mathrm{a}4+\mathrm{P}\mathrm{S}\mathrm{t}\mathrm{a}5+\mathrm{P}\mathrm{S}\mathrm{t}\mathrm{a}6)/\mathrm{n}}{\mathrm{N}\mathrm{u}\mathrm{m}\mathrm{b}\mathrm{e}\mathrm{r} \mathrm{o}\mathrm{f} \mathrm{a}\mathrm{s}\mathrm{s}\mathrm{e}\mathrm{t}\mathrm{s} \mathrm{c}\mathrm{r}\mathrm{i}\mathrm{t}\mathrm{i}\mathrm{c}\mathrm{a}\mathrm{l} \mathrm{f}\mathrm{o}\mathrm{r} \mathrm{f}\mathrm{o}\mathrm{o}\mathrm{d} \mathrm{p}\mathrm{r}\mathrm{o}\mathrm{c}\mathrm{e}\mathrm{s}\mathrm{s}\mathrm{i}\mathrm{n}\mathrm{g} \mathrm{a}\mathrm{n}\mathrm{d} \mathrm{s}\mathrm{t}\mathrm{o}\mathrm{r}\mathrm{a}\mathrm{g}\mathrm{e} \mathrm{a}\mathrm{c}\mathrm{t}\mathrm{i}\mathrm{v}\mathrm{i}\mathrm{t}\mathrm{i}\mathrm{e}\mathrm{s}}$$

$$\mathrm{Food} \mathrm{marketing}{ }^2=\frac{(\mathrm{M}\mathrm{k}\mathrm{t}\mathrm{a}1+\mathrm{M}\mathrm{k}\mathrm{t}\mathrm{a}2+\mathrm{M}\mathrm{k}\mathrm{t}\mathrm{a}3+\mathrm{M}\mathrm{k}\mathrm{t}\mathrm{a}4)/\mathrm{n}}{\mathrm{N}\mathrm{u}\mathrm{m}\mathrm{b}\mathrm{e}\mathrm{r} \mathrm{o}\mathrm{f} \mathrm{a}\mathrm{s}\mathrm{s}\mathrm{e}\mathrm{t}\mathrm{s} \mathrm{c}\mathrm{r}\mathrm{i}\mathrm{t}\mathrm{i}\mathrm{c}\mathrm{a}\mathrm{l} \mathrm{f}\mathrm{o}\mathrm{r} \mathrm{f}\mathrm{o}\mathrm{o}\mathrm{d} \mathrm{m}\mathrm{a}\mathrm{r}\mathrm{k}\mathrm{e}\mathrm{t}\mathrm{i}\mathrm{n}\mathrm{g} \mathrm{a}\mathrm{c}\mathrm{t}\mathrm{i}\mathrm{v}\mathrm{i}\mathrm{t}\mathrm{i}\mathrm{e}\mathrm{s}}$$

Using the above equations, the mean asset index score was estimated on a scale of 0 to 1 for food processing and storage and marketing activities in each study community. The estimates are based on the data gathered from the household surveys. An overall estimated asset index score of 0 corresponds to the lowest asset capacity and, thus, high vulnerability. Similarly, a score of 1 indicates the highest asset capacity, and hence low vulnerability. Interpreting the results in this manner enables easier understanding by a diverse audience, including policymakers and farmers. Adopting the asset-based approach helps provide a snapshot of a community’s weaknesses and strengths regarding climate resilience since asset profiles can influence the susceptibility of food system activities to the impacts of climatic changes and weather variabilities.

Following the household surveys, two focus group discussions were organized in each of the six communities. Discussions were aimed at gathering more detailed explanations for the survey results.. In all, the discussions brought together 95 participants in 12 separate focus groups, each consisting of 6 to 12 people. Discussions lasted 80–120 min and included adults over 20 years of age but not more than 70 years of age.

The quantitative data were analyzed with SPSS Version 21.0 using descriptive statistics, including frequencies and custom tables. Tables and graphs were generated from these statistical tools to show the results. The qualitative data from the focus groups were analyzed following the methods outlined by Miles and Huberman [64] and Berg [65]. This involved the use of unique alphanumeric codes to label participants’ responses, which were later hand-coded and categorized based on recurring themes. Responses relating to key themes such as climate exposure, hazard occurrence, and impacts were identified and integrated into the study results in the form of quotations.

4. Results

4.1. Perceived Exposure to Climatic Hazards

The available assessments of climate change in the northern part of Ghana identify five key hazard events to which communities are exposed: extreme temperature, drought/dryness, floods, windstorms, and erratic rains (see Barry et al. [66]; Yiran et al. [11]; Osman, [67]; Ogunrinde et al. [68]; Antwi-Agyei and Nyantakyi-Frimpong [69]).

The results presented in

Table 2. Perceived exposure to climate hazards in Northern Ghana.

Climate Hazards	Study Community						Total
	Doba	Manyoro	Kologo	Naaga	Navrongo	Pungu
Extreme temperatures	89.60%	93.80%	83.30%	87.50%	85.40%	77.10%	86.12%
Droughts/dry spells	89.60%	95.80%	97.90%	95.80%	83.30%	79.20%	90.27%
Windstorms	93.80%	93.80%	91.90%	92.00%	90.90%	89.60%	92.00%
Floods	92.80%	95.80%	95.50%	91.70%	88.80%	79.20%	90.63%
Erratic rains	89.60%	91.70%	79.20%	68.80%	97.90%	91.70%	86.48%
Total	91.08%	94.18%	89.56%	87.16%	89.26%	83.36%	89.10%

indicate that almost all the research participants reported in the affirmative that they are exposed to all five climate hazards. More households confirmed exposure to windstorms (92%), floods (90.6%), and drought/dry spells (90.2%) than extreme temperatures (86.12%) and erratic rains (86.48%). Further discussions in the focus groups revealed the frequency and intensity of exposure to these hazards. For example, the following comment by one of the participants reflects points that came up repetitively in the discussions:

Windstorms are widespread and occur throughout the year. We experience both dusty and moist windstorms. In particular, the dusty harmattan windstorms are a frequent phenomenon in the dry season, whilst the moist windstorms occur in the rainy season. The moist windstorms are usually heavy, noisy, destructive, and associated with intense but short-duration rains. Here in the north, there is always a heavy windstorm preceding the intense rains in August-September. It’s impossible for the rains to occur without the windstorm before the rain there is always a windstorm.

(Male Farmer, Doba)

Most of the participants reported exposure to floods because of the frequent occurrence of torrential rains, which leave behind excess water in land areas beyond the crests of river plains, canals, and dugouts. During the discussions with the communities located near the Tono Irrigation Dam canals, the participants reiterated the increasing rate of flood occurrences. As one farmer reported:

Heavy rains and floods are very common in the rainy season. There is always a flood any time it rains, especially in the months of August and September—every little rain can even cause flooding. Last year, there was three day of consecutive rains that left behind excess water on our farms and even our homes for days. We are used to it; whenever the rainy season is approaching, we know that floods are also ready to occur.

(Male Farmer, Navrongo)

Additionally, the results in Table 2 show variations in overall exposure to climate hazards across the study communities. More households in Manyoro (94.2%) and Doba (91.3%) reported greater exposure to climate hazards than those in other communities. The general biophysical characteristics of Kassena Nankana Municipality, which consists of large tracts of bare lands with sparse vegetation cover and non-porous soils, explain the study communities’ overall exposure to climate hazards. However, the higher exposure of households in Manyoro and Doba to climate hazards can be attributed to the increasing rates of land degradation and desertification, as well as their relative geographical location. During field observations, it was noticeable that these two communities have tracts of dry land without vegetation cover. Some pieces of land consist of baked soil surfaces which form hardpans that render the land uncultivable. Additionally, Manyoro lies in the northernmost part of the district, which borders the Sahel region of Burkina Faso. This means it experiences relatively more extreme heat conditions than the other study communities.

4.2. Impacts of Climate Hazards on Post-Production Activities

Climate hazards affect food security through their impacts on food system activities, including those relating to food processing, storage, and marketing. From the review of the literature and from interactions with the research participants, seven main directions of climate impacts on post-food production system activities were identified: four for food processing and storage, and three for food marketing. During the surveys, participant farmers were asked to indicate which one of the identified challenges they perceived as having the greatest adverse impact on post-food production system activities. The results of their responses are presented below in Figure 3a,b.

Figure 3a presents the results concerning the impacts on food crop processing and storage activities. Over one-third of the respondents reported crop spoilages/losses (42%). This impact is significantly felt by the majority of the farmers because the unpredictable rains that occur in September–November coincides with the harvesting season, and this often leads to dampness in the ready-to-harvest crops and consequently to the increased susceptibility of the stored grains to mould attack. Additionally, about one-fourth of the participants (24%) reported damages to crop storage facilities due to climate change and weather variabilities. According to the focus group participants, damage to barns is very common when windstorms occur, and this can cause further grain losses. The following concern from one of the focus group discussion participants summarizes the farmers’ perceptions of the impacts of climate hazards on storage facilities:

When windstorms occur, they can easily blow away roofs, create cracks, and collapse the barns where we mostly store our millet and guinea corn. So, if the structure is not well roofed or properly checked and repaired, there can be a huge problem. As soon as it rains, water can easily leak into the barn. This can cause stored grains to regerminate and then rot. One can lose good seeds from the rainwater leaking into the barn and storage rooms.

(Male Farmer, Manyoro)

Figure 3b illustrates the results concerning climate impacts on food marketing activities. Nearly half of the participants (43%) indicated that they had encountered poor food pricing due to climate hazards. This impact on food prices is often experienced by farmers in their roles as sellers in the harvesting season and buyers purchasing food to supplement own production in the dry season. Across the study communities, the farmers generally face weather-induced losses which often translate into low stocks during the out-of-season months and food price increases to levels unfordable to poor households. Nonetheless, the research participants seemed more concerned about the fact that weather-induced damages limit the ability of farmers to bargain for better farmgate prices. One of the focus group participants indicated the following:

The unpredictable rains occurring in the early harvesting period usually cause cosmetic damage and rot in food crops, especially beans and maize. When you take a bag of black-eyed beans to the market and it has a lot of moisture-damaged pieces, you cannot bargain for fair farmgate prices. You just have to accept the low-price offers. Sometimes, the prices are as low as half the price of the good-looking ones. If moisture-damaged pieces are too much, no one will even buy them; you are likely to bring them back home for household consumption.

(Woman Farmer, Pungu)

4.3. Post-Food Production System Vulnerability to Climate Hazards

The food crop processing and storage vulnerability analysis for the study communities presented in

Table 3. Mean asset scores for post-production activities across communities in the Kassena Nankana Municipality.

Community	Food Processing and Storage			Food Marketing Vulnerability
	Mean	N	Std. Deviation	Mean	N	Std. Deviation
Doba	0.3507	48	0.16927	0.4219	48	0.20077
Manyoro	0.2014	48	0.13734	0.4219	48	0.17225
Kologo	0.5660	48	0.16751	0.4688	48	0.20385
Naaga	0.5451	48	0.15275	0.3385	48	0.25521
Navrongo	0.6354	48	0.19946	0.5417	48	0.22081
Pungu	0.5208	48	0.22182	0.3698	48	0.21257
Total	0.4699	288	0.22959	0.4271	288	0.22062

below shows that Manyoro has the lowest mean asset index score for food processing and storage (0.2014) and, therefore, the highest vulnerability. In practical terms, Manyoro has a lower asset capacity than the other communities, limiting its ability to protect ready-to-harvest and stored crops from the dampness and cosmetic damages often associated with unpredictable and out-of-season rains.

To understand this level of vulnerability, detailed information on the percentage of households with access to and use of the six identified assets critical for food processing and storage activities is presented in Figure 4a. Again, the results here demonstrate that none of the households in Manyoro reported using threshing machines or warehouse services for food crop harvesting and storage, respectively. Additionally, 31.3% of households had access to grinding machines and 8.3% had access to training on post-harvest management; access to both was far more than 50% in the other communities. This lower level of access and use of timesaving harvesting and post-harvesting resources in Manyoro explains its high vulnerability to climate hazards. Overall, the results here suggest that Manyoro is prone to post-harvest losses, which can directly hinder the ability of households to achieve food availability in the out-of- season months.

Concerning the food marketing vulnerability analysis, the results presented in Table 3 above show that Naaga recorded the lowest mean asset score for food marketing (0.3385) and that, therefore, it is the community with the highest food marketing vulnerability. This implies that, compared to those in the other communities, households in Naaga have limited capacity to deal with food market price, affordability, and accessibility challenges in the dry season.

Additional statistics to help identify the assets which significantly affect and explain food marketing vulnerability are presented in Figure 4b, which shows the percentage of households with access to the four key assets critical for food marketing. Primarily, vehicle transport and non-farm incomes contribute more to food market vulnerability across all the communities than the other assets. The impacts of these resources are even more pronounced in the higher vulnerability of Naaga, where none of the households (0%) reported access to vehicle transport, and only a few (10.4%) had income opportunities to buy food in the out-of-season months.

Overall, the food system vulnerability analysis presented here reveals that the post-food production activities of the Kassena Nankana Municipality are generally vulnerable to climate hazards. On a scale of 0–1, the communities’ food processing and storage vulnerability ranged from 0.20 to 0.63, while their food marketing vulnerability ranged from 0.33 to 0.54. These scores are far less than 1, indicating that the post-production activities in the study communities are highly susceptible to weather-induced hazards such as flooding, windstorms, and extreme temperatures. In practical terms, more households are unable to manage climate impacts on food processing, storage, and markets, and this leads to potential food insecurity risks. The variations in the vulnerability scores further show the differences in community asset profiles for climate resilience; those communities with the most households lacking access to threshers, warehouses, market centers, vehicular transport services, and non-farm incomes, for example, have the lowest asset scores and thus are the most vulnerable to climate hazards, and vice versa. This suggests that policy interventions to improve food security must target specific communities or households over others.

4.4. Intensity of Climate Hazard Impacts on Post-Food Production System

Climate hazards affect food security through their impacts on food crop processing, storage, and marketing activities. Figure 5a presents the results concerning the intensity of climate impacts on food crop processing while the Figure 5b focuses on storage activities.

Figure 5a shows that many households in Manyoro (49.6%), Doba (44.2%), Naaga (43.8%), and Kologo (42.9%) reported high impacts of climate hazards on food crop processing and storage. The high impacts in these communities can be attributed to the very limited access of households to proper threshing and storage services (as was shown earlier in Figure 4a). At one of the focus group discussions, a participant discussed how challenges limiting access to threshers and harvesters lead to high climate impacts, explaining that:

The harvesters only work on rice farms and not all crops. It is also just a few people who can afford the services because there is no money during the harvesting months. The harvesting period is the most challenging time for everyone in our community—no money until we sell the crops. So, we harvest our crops manually, just cut the stalks and stook in heaps. Most farmers rely on organized labour and help one another in rotation. This manual labour is not effective; it takes a longer time to finish the whole farm. So, anytime the rain occurs, it soaks into the crops; they become damp and difficult to thresh.

(Female Farmer and group leader, Bonia)

Figure 5b shows variations in the severity of the impacts of climate hazards on food marketing activities across the study communities. Many of the households in Naaga (52.9%), Kologo (44.2%), and Manyoro (32.9%) reported high impacts from climate hazards, while those in Pungu, Navrongo, and Doba reported lower impacts. The high impacts of climate hazards on food marketing in Naaga, Kologo, and Manyoro could be attributed to the fact that these three communities are distant from the central markets in Navrongo and Bolgatanga. On average, farmers in these communities must travel 25 km over rugged, dusty lateritic roads to reach the central markets. In addition, there is limited access to vehicular transport services, as is shown in Figure 4b, and as was repetitively expressed by the households. A comment from one of the focus group participants summarizes the concerns of the households:

Our roads are very bad to travel on, but they are even better now (referring to the dry season). When the rains set in it gets worse; our pothole-ridden roads become completely filled with water after every small downpour, consequently blocking farmers from transporting their crops home and market. Also, the bridges we have along the roads are very narrow. The rainwater always overflows the main bridge linking our community to Navrongo. When it happens, no one is willing to travel, not even the tricycles or trucks from Bolgatanga and Navrongo. So, we are unable to transport our produce from the farm and even to the market.

(Male Farmer, Naaga)

5. Discussion and Conclusions

Climate hazards have a range of adverse impacts on northern Ghana’s micro-level food system activities. The results of this research empirically demonstrate the high exposure of the study communities to climate hazard events. Almost all farming households (89%) confirmed the occurrence of windstorms, droughts/dry spells, erratic rains, extreme temperatures, and floods. The evidence presented here is consistent with that of past studies which have highlighted the persistent climate hazards in the northern regions of Ghana [11,54,70]. For instance, Yiran and Stringer’s [11] spatial time-series analysis of climate hazards from 1983–2012 identified constant occurrences of high and increasing temperatures, frequent but varied seasonal droughts, dry spells, floods, and windstorms in the Upper East region of Ghana. Dumenu and Obeng’s [70] study on the social impact of climate change revealed that respondents in northern Ghana frequently mentioned prolonged drought, erratic rainfall, and flooding as the most observed climate change events in their communities. Indeed, the geographical location of the study communities at the margins of West Africa’s environmentally fragile Sahel region explains their high exposure to climate hazards [71] 3.

This study demonstrates the adverse impacts of climate hazard events on post-food production activities that loom large for household food security in northern Ghana. Previous studies (including Armah et al. [3]; Atanga and Tankpa [16]; Baffour-Ata [17]) have reported the adverse impacts of climate change on food crop production. Recognizing the need for a comprehensive analysis to form a nuanced understanding of household-level food security, this study extends its attention to the post-production level by revealing the significant impacts of climate change on smallholder processing, storage, and marketing activities. Regarding the impacts on processing and storage, the study findings show that climate change causes damage to smallholder storage facilities and spoilage in stored crops. Likewise, climate change and weather variabilities have direct impacts on food crop marketing, including disruptions to transport networks, increases in the cost of transport, and rising food prices. In the context of these climate impacts, households and communities in northern Ghana are predisposed to food insecurity risks relating to high post-production losses, reductions in the quality and quantity of food crops, declining food stores, disruptions in food pricing, and income losses. These risks make it virtually impossible for poor households to maintain their physical and economic access to sufficient food to meet consumption needs. The findings of this study support the IPCC’s high confidence that observed climate change and frequent extreme events already affect food security in drylands, particularly in Africa, including northern Ghana [8]. In addition, the evidence of food insecurity risks presented here reinforces previous studies of climate impacts on food production [42,70,72] and offers additional pioneering insights into how other food system activities are equally adversely affected by climate change.

The results of this study show that generally low asset endowment influences food system vulnerability to climate change across communities in northern Ghana and other marginal areas of the West Africa Sahel and Savannah regions. These findings resonate with climate vulnerability assessment studies which demonstrate that asset endowment shapes the impacts of climatic variations, shocks, and trends on household livelihoods and food security [26,42,52,67,70,73]. This study would have failed to reveal the above-explained results and findings if it did not focus on the differences in critical resources and services for food system activities in the selected communities in northern Ghana. Notwithstanding this, this study did not reveal the differential vulnerability of various social groups, thus limiting the potential of the analysis and results to show demographic, economic, and social differences in food insecurity risks. Overall, the findings provide compelling evidence for policymakers to facilitate interventions that foster asset building for poorer communities and enable them to build safe conditions to protect food system activities against extreme climate events and impacts.

Additionally, this study presents evidence demonstrating how the “pressures” (basically, the opposing forces) from both climate hazards and food system vulnerability converge to underpin food insecurity risks in northern Ghana. The results show that climate hazards adversely affect all post-food production system activities. However, the severity of the impact is greater in communities with higher levels of vulnerability, and vice versa. As per the sustainable livelihood framework, a lack of asset endowment may limit the capacity of a household or community to manage and cope with threatening events. Across the study communities, critical assets for managing climate-induced hazards affecting post-food production activities include threshers, grinding machines, warehouses, and market centers. The results of this research show that most households in Manyoro harvest food crops without threshing machines. Also, those in Kologo and Naaga are generally far from the district capital, with impassable roads and a lack of vehicular services to connect households to Navrongo town, where they can access the market and purchase food at competitive prices, especially in the dry season. Together, the “pressures” from the high vulnerability situations of the households in these three communities and the generally high climate hazard exposure of the entire Kassena Nankana Municipality converge to produce adverse climate change-induced impacts on food system activities, and consequently high food insecurity risks, in the three communities.

Based on a detailed analysis of the results and discussions presented above, this study concludes that climate change and weather-induced food insecurity risks in northern Ghana are not just products of persistent exposure to climate hazards and the subsequent impacts on food production. Instead, risks are also inextricably connected to the vulnerable situations within which food is harvested, processed, stored, and marketed in rural communities. These findings reveal that climate hazard events such as flooding and droughts damage stored grain, disrupt food supply to the market, and cause seasonal volatilities in food prices. However, these impacts are not solely externally generated circumstances. The food system is highly vulnerable; most households lack access to threshing and grinding machines, warehouse storage, post-harvest management information, and transportation services. These underlying characteristics of the post-food production system of northern Ghana, which is ultimately quite remote from climate change, exacerbate household-level food insecurity risks.

Lastly, this research is not intended to understate the direct impacts of climate hazards on food production activities. Instead, it seeks to contribute knowledge to the broader scholarly field that critically analyzes the concepts of climate risks and food security by revealing some of the climate change impacts on the post-food production systems that underpin household-level food security. Additionally, it affirms the importance of integrating varied perspectives and complementing solutions to address climate-induced food insecurity. As policymakers formulate solutions to the social impacts of climate change and hunger, far-reaching results could be achieved with an integrated approach that recognizes the interlinked challenges of climate change and post-food production system vulnerabilities. This paper, therefore, recommends focusing on the “new Green Revolution” in Africa, emphasizing climate-resilient and high-input food production systems which are to be matched with increased investment in quality and climate-proof storage and market facilities for smallholder farmers. In particular, governments and development partners could repurpose agricultural policies and investments to build more decentralized farmer-managed warehouses, community-based food reserves, and rural transportation networks to help poorer communities adapt the entire food system to increasing climate change and weather variabilities.