1 Introduction

More than half of the global population faces potential exposure to natural hazard-induced disasters, resulting in economic losses of 151.6 billion US dollars [1]. The frequency and severity of these disasters have been increasing and are expected to continue rising due to environmental degradation and global warming [2]. Low-income countries bear a disproportionate burden of these hazards compared to high and middle-income countries due to their inadequate infrastructure and limited capacity to manage or withstand disasters [2, 3]. Within countries, rural areas are often more vulnerable due to their geographical location and limited resources for disaster preparedness and response [4]. Geological hazards such as landslides, earthquakes, and volcanoes pose significant threats to millions worldwide, destroying homes, food and water shortages, disruption of livelihoods, and loss of life or injury [5]. Landslide hazards, in particular, are more widespread in frequency and intensity compared to other geological events due to environmental degradation [6].

Natural hazards intensify the challenges faced by households, increasing concern and stress related to food security in their aftermath [7, 8]. It’s important to recognize that even if crops remain unscathed by such events, household food security is not guaranteed [9]. Households exposed to natural hazard shocks are more prone to food insecurity compared to those unaffected by such shocks [10]. In the East African highlands, landslide hazards significantly disrupt food security by depriving agricultural land [11]. However, the remote nature of affected areas and the relatively small scale of individual landslide events contribute to significant underreporting of their impact, leading to an underestimation of landslides’ effects on human livelihoods and food security [12, 13].

According to the Food and Agriculture Organization (FAO), food security is a complex phenomenon influenced by various factors, including food availability, access, supply stability, and biological utilization [14, 15]. Landslide hazards negatively impact both domestic food production and imports, affecting food availability within the context of food security [16]. Moreover, when an area is affected by landslide hazards, both physical and economic access to food, as well as the stability of food supplies, are compromised [2]. The consequences of landslide hazards can also impact the utilization aspect of food security, as individuals may be compelled to compromise on the quality of their diet. Consequently, all core dimensions of food security are affected [17]. Thus, there is an urgent need for research on disaster risk-lessening strategies in the highland areas of Africa driven by the region's susceptibility to landslides, population growth, anthropogenic interventions, potential climate-related changes, and the lack of comprehensive data for effective risk assessment and mitigation [18, 19].

In Ethiopia, natural hazards, such as droughts, floods, landslides, and other climatic events, significantly impact rural household livelihoods [20]. The effects are often complex and multifaceted, affecting various aspects of agriculture, livelihoods, and overall well-being [21]. It can increase the vulnerability of rural livelihoods, as they are heavily dependent on agriculture [22]. Households residing in regions prone to landslides are frequently more susceptible to food insecurity due to the heightened jeopardy posed to their livelihoods [23]. Limited access to resources, combined with the constant threat of landslides, creates a cycle of vulnerability that can be challenging to break [24]. Landslides hazard-caused disasters can lead to the loss of arable land, making it unsuitable for cultivation [25]. This directly affects the agricultural productivity of rural households, reducing the land available for growing crops and contributing to food shortages [26].

The Gamo highlands are one of Ethiopia's most affected areas by landslide hazards and their effect is significant [27]. Highlights the urgency to understand and study the risks associated with landslides in the region. Specifically, in Gacho Baba nearly 87% of the entire district is classified within the very high to moderate landslide susceptibility class [28]. However, most studies conducted on landslide hazards in the Gamo Highlands have not directly examined the impact of landslide hazards on household-level food security. Therefore, the aims of this study were (1) to assess the food security status of households affected by landslide hazards using multiple indicators, and (2) to ascertain the level of landslide hazard awareness and preparedness strategies among households in the Gacho Baba district of the Gamo Highlands in southern Ethiopia.

2 Methods and materials

2.1 Study area description

The Gamo zone, situated within the recently restructured South Regional State of Ethiopia, currently comprises a total of 20 administrative structures, with 14 rural and 6 urban divisions. Among the rural structures are Bonke, Boreda, Chencha Zuria, Arba Minch Zuira, Dera Malo, Dita, Gacho Baba, Gerese, Kamba Zuria, Kogota, Kucha Alpha, Marta Garda, Mirab Abaya, and Kucha districts. The six town administrations include Arba Minch, Chencha, Kamba, Selamber, Gerese, and Birbir [28]. This study will focus specifically on the Gacho Baba rural administrative district, situated within the Gamo highlands. The district lies between the Abaya-Chamo Lake basins and is drained by Sile and Kulfo watershed tributaries. Geographically, the Gacho Baba district spans between 5° 38′ 30" and 6° 10′ 50" N latitude and 37° 20′ 30" and 37° 31′ 30" E longitude. Positioned approximately 5041 km south of Addis Ababa and 36 km from the zonal capital, Arba Minch, the district consists of 11 kebeles, namely, Gatse, Koddo, Wusamo, Zegiti Merche, Tsayite, Zegiti Bakole, Gerbenssa-tsenkile, Kuyile, Laka, Zegiti Feriso, and Mazo Doyisa (Fig. 1).

Fig. 1
figure 1

Locational map of the study area. Source, Lemma et al. [28]

Full size image

In terms of agro-ecology, the Gacho Baba district exhibits a diverse landscape, categorized into three main zones: Temperate, comprising 37% of the area, followed by the cool zone at 39%, while the cold zone constitutes the smallest portion at 24% (Table 1). The mean annual temperature within the study district varies significantly, ranging from 16 to 25 °C. Notably, areas classified as cold agroecology register temperatures consistently below 10 °C. Moreover, the highland regions maintain an average annual temperature of 15 °C. It’s worth mentioning that temperature gradients are observable across the district, with a noticeable decrease in temperature from west to east correlating with an increase in elevation from east to west [28].

Table 1 Agro-ecology of Gacho Baba district
Full size table

The rainfall distribution in the study area is bimodal, which shows a distinctive difference among the areas. The west side of the study area receives more rainfall (more than 1500 mm), while the east side receives less rainfall (less than 1000 mm) with an annual average of 800–1500 mm [28, 29].

Similar to many rural areas in Ethiopia, the economy of the Gamo Highlands relies predominantly on rain-fed agriculture, which sustains and economically supports nearly all inhabitants in the region [29, 30]. In the Gamo Highlands, the cultivation of 'enset' (Ensete ventricosum), potatoes, and cereals, particularly barley, and wheat, form the foundation of livelihoods, while maize and sorghum serve as primary food sources on the lower slopes [31]. However, productivity in the Gamo Highland areas remains low due to factors such as small landholdings, land fragmentation, poor soil fertility resulting from severe erosion, and various human and natural hazards [32]. Consequently, livelihoods in the Gamo Highlands have increasingly relied on subsistence agriculture and some off-farm economic activities in recent years [31]. The study was conducted specifically in the Gacho Baba district, one of the districts within the Gamo Highlands. According to [33], Gacho Baba district has a total population of 101,786, with 51,088 males and 50,697 females. The young dependency ratio (individuals under the age of 15) and old dependency ratio (individuals over the age of 64) are 48.3% and 2.7%, respectively, resulting in a total dependency ratio of 48.9%.

2.2 Approach, data types, and sampling

2.2.1 Research approach

The research approach serves as a strategic framework and assertion utilized by researchers to underpin the study's strategy and methods [9, 34, 35]. Essentially, it outlines a broad strategy for how researchers intend to address the research question(s), encompassing research techniques, options, and time horizons [34]. In this study, a mixed research approach was adopted, acknowledging the importance of acquiring reliable and pertinent information regarding the actual practices pertinent to the issue under investigation, while mitigating biases inherent in using a single method [36,37,38]. The chosen mixed research method design was concurrent embedded, which combines quantitative and qualitative data collection within a single phase, with greater emphasis placed on the quantitative aspect to gain a deeper understanding of the phenomenon under study [39, 40].

2.2.2 Data types and sources

This study employed a comprehensive approach, utilizing both quantitative and qualitative data collection methods. Primary data were obtained from a carefully selected sample of rural households, participants in Focus Group Discussions (FGDs), and in-depth interviews with key informants. 5 FGDs (one for each sample village) were conducted with purposively selected (6–8) individuals from village leaders, village youth and female representatives, village agricultural extension workers, and model farmers. In-depth interviews were conducted with 9 key informants purposively selected. That is from the Zonal disaster risk management office 2 experts (1 from disaster risk management and 1 from food security), the Gacho Baba district disaster risk management office 2 professionals (1 from disaster risk management and 1 from food security), and sample village 5 administrative bodies (1 from each). In addition, secondary data were sourced from various written documents, including zonal and district official reports, as well as relevant literature about the research topics. Given the focus on understanding the vulnerability of rural households to landslide hazards and their impacts on food security, the household served as the primary unit of analysis in this investigation.

2.2.3 Sampling techniques

For this study, multistage sampling techniques were employed. Firstly, purposive sampling methods were utilized to choose the study district and sample villages, based on the researchers’ expertise and field experience within the district. Furthermore, the study area was selected due to its high vulnerability to natural hazards, especially landslides, which have recurrently caused significant damage. Secondly, five rural villages, namely, Gatse, Kodo, Wusamo, Zigiti Merche, and Tsayite, were purposefully selected because these villages are repeatedly affected by landslide hazards and are among the villages in the district identified as food insecure.

All households in the above-mentioned village affected by landslide hazards, identified by the village leaders and district disaster risk prevention professionals, became the target population of this research, and the appropriate sample size for each village was selected proportionally (Table 2). Accordingly, after considering the various possibilities made by different scholars about representative sample size selection procedures and models for this research, the researchers have decided to choose the recommended sample size determination formula of [41] for a finite population, which is illustrated: \({\text{n}} = \frac{{\text{N}}}{{1 + {\text{N}}\left( {{\text{e}}^{2} } \right)}}\) Thus, \({\text{n}} = \frac{1,045}{{1 + 1,045\left( {0.05^{2} } \right)}}\) = 289 households.

Table 2 The allocation of the target population, the sample, and the methods employed for sampling
Full size table

In this context, ‘n’ represents the desired sample size, while ‘N’ indicates the total number of households studied. ‘e’ denotes the precision level, set at 0.05, with a confidence level of 95% at p =  ± 5, and 1 represents the probability of the occurrences. The total target population list, or sample frame for this research is 1045 household heads affected by landslide hazards.

2.3 Instruments for data collection

2.3.1 Survey

The study utilized questionnaires to assess food security, employing tools such as the Household Food Security Access Scale (HFIAS), the Food Consumption Scale (FCS), and the Reduced Coping Strategies Index (RCSI), as outlined in sources [42,43,44,45,46]. These tools are recognized for their efficiency, accuracy, and applicability across various contexts [47]. Additionally, standardized questions were adapted from guidelines for evaluating the human impact of disasters [48] to gather information on the effects of natural hazards on household food security status. The aim was to collect data on the impact of landslide hazards on household food security in the study area.

2.3.2 Focus group discussions (FGD)

Following the recommendations of [49, 50], the Focus Group Discussions (FGDs) were meticulously organized. This involved defining the primary objectives of the sessions, formulating key questions, planning the session recording method, and selecting and inviting appropriate participants, typically ranging from six to eight individuals. The FGD participants were village leaders, youth and female representatives, agricultural extension workers, and model farmers who were not included in the statistically representative selected samples. Structured questions and checklists guided the Focus Group Discussions (FGDs), enabling participants to share their experiences and perspectives on the topics being investigated. Five FGDs were conducted (one for each sample village), and the researcher himself played a moderating role by introducing the topics for discussions to help the group members participate in a lively and normal temper.

2.3.3 In-depth interview

An in-depth interview is among the various qualitative research methods employed to collect data on participants’ subjective experiences. It serves to obtain comprehensive information that elucidates an individual’s viewpoint and the interpreted significance of a specific subject, concern, or procedure [37, 51, 52].To make the process manageable, an In-depth interview was conducted with nine purposefully selected key informants from the Gamo zone disaster risk management office 2 experts (1 from disaster risk management and 1 from food security), Gacho Baba district disaster risk management office 2 professionals (1 from disaster risk management and 1 from food security), and sample village 5 administrative bodies (1 from each village). The interviews addressed the impact of landslide hazards on the food security status of rural households within the study region.

2.3.4 Field observation

Field observation is frequently employed in conjunction with other data collection techniques and provides a deeper and more dynamic understanding of situations that may not be as readily captured through alternative methods [53]. It involves immersing oneself in the community or social system under investigation while maintaining a degree of separation from the observed activities [37, 39]. Field observation served as an additional tool in the data collection process, encompassing direct observation of the vulnerability conditions to landslide hazards in the selected villages and their impact on household-level food security status.

2.4 Data analysis methods

The researchers employed a blend of quantitative and qualitative data analysis techniques, utilizing appropriate statistical and econometric analysis software. The quantitative data gathered through food insecurity measurement scores, underwent coding, editing, entry, and analysis using Statistical Products and Service Solutions (SPSS) version 26. In assessing food security status, the study relied on the Household Food Security Access Scale (HFIAS), the Food Consumption Scale (FCS), and the Reduced Coping Strategies Index (rCSI). HFIAS was chosen for its comprehensive dimensionality compared to alternative measures, while FCS was selected to grasp the utilization aspect of food security, with its constituent questions tailored to capture dietary diversity as indicated in Table 3.

$$FCSi\, = \,\sum (wXj)(fXji)$$
(1)
Table 3 Nutrition-based weight, food consumption groups, and their description used for this study
Full size table

Were,

FCS i = Food consumption score for i household.

wXj = Weight based on the nutritional content for food group Xj.

fXji = Frequency of consumption (number of days a food group Xj is consumed) by household i.

The inclusion of rCSI serves as a supplementary measure to corroborate the findings derived from HFIAS and FCS, employing a simple bivariate correlation analysis utilizing the continuous scores of rCSI, FCS, and HFIAS. Thus, Household CSI scores were calculated by multiplying the number of days each strategy was utilized in the previous week by its respective severity weight and summing the resulting totals [57].

$$rCSI\,i\, = \,\sum {\left( {fXji} \right)} \left( {wXj} \right)$$
(2)

Were,

rCSI i = reduced coping strategy index for household i.

fXji = number of times (frequency) coping strategy Xj was used by household i within 7 days preceding the date of the survey.

wXj = severity weight attached to coping strategy Xj.

Quantitative data, acquired through standardized questions sourced from guidelines for evaluating the human impact of disasters [48] regarding the effects of natural hazards on household food security status, were also compiled and tabulated according to relevant sub-topics related to the research questions. These data were subjected to analysis employing both descriptive and inferential statistics. Qualitative data obtained from key informants, participants in Focus Group Discussions (FGDs), field observations, and transect walks were documented and analyzed textually to validate the statistical findings from the structured questions. These qualitative findings were then triangulated with data obtained from quantitative analysis.

3 Results and discussions

3.1 Characteristics of sample households

Before delving into the data pertinent to the fundamental inquiries, an overview of the subjects' characteristics was provided. Six demographic variables household age, family size, gender, marital status, and educational attainment were selected based on their potential to impact the outcomes of this investigation. Household age and family size were treated as continuous variables, while the remaining variables were categorical. The mean age of households in the study was 46.02 years, with a standard deviation of 10.23, encompassing individuals of working age and older adults in particular. The average family size in the study area was 6.81 members per household, with a standard deviation of 2.51 (Table 4).

Table 4 Descriptive statistical results for continuous variables (N = 289)
Full size table

The surveyed data reveals the majority of respondents were male (96.9%), indicating a gender imbalance in the sample population this is due to the availability as the sample households were included only those affected by landslide hazards in the past 5 years. The overwhelming majority of respondents were also married (99.7%). The distribution of educational statuses among respondents reveals a significant proportion with no formal education (55.7%), followed by primary and secondary education levels (Table 5).

Table 5 Descriptive statistical results for dummy variables (N = 289)
Full size table

3.2 Food security status among households affected by landslide hazards

Addressing food security presents a multifaceted challenge due to its diverse dimensions, making both its operationalization and the identification of food-secure households complex [47]. Acknowledge the difficulty in independently measuring the various dimensions of food security and the near impossibility of capturing the entirety using a single metric. Therefore, in this study, the Household Food Insecurity Access Scale (HFIAS) and Food Consumption Score (FCS) were utilized as primary indicators of food security, while the Reduced Coping Strategy Index (rCSI) was employed as an additional measure to validate the findings derived from HFIAS and FCS.

3.2.1 Household food security status assessed via household food insecurity access scale (HFIAS)

The Household Food Insecurity Access Scale (HFIAS) serves as a tool for gauging the food security status of households, offering structured and standardized methods for measuring and comprehending the level of food security within a household [23]. Through its assessment of factors concerning food accessibility, HFIAS aids in pinpointing the degree of food insecurity, which is vital for devising focused interventions and policies to meet the distinct requirements of vulnerable households [54, 55]. Monitoring and analyzing households' food security status with HFIAS contribute to informed decision-making and effective resource allocation in the realm of food security and nutrition [56].

Table 6 offers a comprehensive depiction of food insecurity prevalence among households, elucidating responses to fundamental inquiries outlined in the Household Food Insecurity Access Scale (HFIAS). These inquiries encompass concerns about food availability, limitations in dietary diversity, and instances of hunger experienced within the 30 days preceding the survey date. The analysis reveals that a considerable majority, comprising 77.5% (228) of the households surveyed, expressed apprehensions regarding the likelihood of encountering food shortages or inadequate food supply. Within the framework of food security assessment, this dimension underscores the stability, accessibility, and availability facets, as posited by previous research [47]. Thus, the substantial proportion of households (77.5%) reporting varying degrees of worry underscores the pronounced apprehension surrounding the consistent access and availability of food resources.

Table 6 Summary of the response frequencies to the HFIAS questions
Full size table

This finding resonates with the conclusions drawn in a similar context by [57] in the Sidama region of the Borecha district, where 71% of rural households voiced concerns about potential food shortages. Such consistency across different geographic contexts underscores the pervasive nature of food insecurity challenges and emphasizes the urgent need for targeted interventions to bolster food access and stability within vulnerable communities.

Additionally, the data indicates that a considerable portion of households faced challenges, such as being unable to consume their preferred foods (78.9%), having to settle for a limited variety of foods (78.5%), and even consuming foods they did not desire (78.9%). These findings highlight the difficulties many households encounter in accessing and affording nutritious foods that align with their preferences and dietary requirements [47]. Moreover, as noted by [54, 58], the inability to eat preferred foods, limited food variety, and consumption of undesirable foods are recognized as key indicators of insufficient food quality.

Another salient aspect highlighted by the data is the prevalence of insufficient meal sizes and frequency stemming from food scarcity. A substantial proportion of households were observed to grapple with consuming reduced meal portions (76.1%) and fewer meals throughout the day (73%) due to constraints in food availability. Moreover, the data underscores instances of acute food deprivation, with a noteworthy percentage of households recounting episodes where they lacked any food to consume whatsoever (57.8%). Additionally, some households reported members having to retire to bed on an empty stomach (18%), while others endured an entire day and night without sustenance (13.8%). These stark figures underscore the severity of food insecurity experienced by these communities.

These empirical findings echo the conclusions drawn in prior studies [47, 59, 60], which emphasize the deleterious repercussions of food insecurity on dietary adequacy and nutritional intake. Such circumstances can exacerbate the risk of malnutrition and precipitate various health ailments, particularly among vulnerable demographics such as children and the elderly. In essence, the data underscores the urgent need for targeted interventions aimed at alleviating food insecurity and enhancing food access and stability within these communities. Efforts to address these challenges must prioritize strategies that bolster household food security and safeguard the nutritional well-being of vulnerable residents.

Food security is divided into four categories: (1) Food security, which assesses the extent to which households have access to sufficient, safe, and nutritious food, with consistent access to enough food for a healthy life; (2) Mild food insecurity, characterized by some uncertainty or limited access to adequate food that is not severe; (3) Moderate food insecurity, indicating reduced access to adequate food, more severe than mild insecurity; and (4) Severe food insecurity, where households face chronic limitations in accessing enough food, leading to hunger [1, 4, 5, 43].

Table 7 provides a breakdown of food security status among surveyed households, categorized into four levels: food secure, mildly food insecure, moderately food insecure, and severely food insecure. Notably, a substantial portion of households experience some level of food insecurity, with mildly, moderately, and severely food-insecure households collectively comprising 62% of the total sample. This underscores the pervasive nature of food insecurity in the area, with moderately and severely food-insecure households accounting for a combined total of 50% of the sample.

Table 7 Households’ food in/security status (HFIAS) based categorization scheme
Full size table

Also, the data reveals disparities in food security status among households. While 38% are classified as food secure, indicating consistent access to enough food, 12% are mildly food insecure, 34% are moderately food insecure, and 16% are severely food insecure. These findings emphasize the necessity for targeted interventions to address the specific needs of households experiencing varying levels of food insecurity. Overall, the survey results indicate a significant (62%) proportion of rural households falling within the food insecurity spectrum in the study area.

3.2.2 Assessment of household food security status using food consumption score (FCS)

Household food consumption refers to the eating habits of households over the previous 7 days, serving as an indicator of current food security. The Food Consumption Score (FCS) is a composite measure based on dietary diversity, food frequency, and the nutritional significance of various food groups. Dietary diversity reflects the variety of individual foods or food groups consumed within the past 7 days, while food frequency indicates how often specific food items were consumed by the household during this period [45, 61]. The FCS is determined by recalling the household’s food consumption over the past 7 days, and categorized into three groups: poor consumption (FCS = 1.0–21), borderline (FCS = 21.5–35), and acceptable consumption (FCS > 35). It is calculated as a weighted sum of food groups, where the score for each food group is derived by multiplying the number of days the items were consumed by their relative weight [61].

Table 8 shows a comprehensive breakdown of household intake frequency for various food groups, offering valuable insights into consumption patterns and dietary habits among the surveyed households. Beginning with staple foods, a striking observation emerges: a significant majority of households, comprising 77.2%, rely on these items for sustenance daily. However, it's noteworthy that a substantial proportion, accounting for 17%, only consume main staples 6 days a week, indicating a consistent dependence on staple foods for daily sustenance, underscoring their pivotal role in local diets.

Table 8 Summarized frequency of households’ food consumption
Full size table

Upon closer examination of pulse/legume consumption, a concerning trend surfaces: a considerable 54.3% of households do not consume these nutritious items at all. This may signify challenges related to accessibility or affordability, given the recognized nutritional value of pulses. Furthermore, among those households that do incorporate pulses into their diets, the majority consume them fewer than 5 days a week, suggesting potential obstacles in integrating these vital sources of nutrients into daily meals.

These findings shed light on the dietary landscape within the surveyed communities, emphasizing both strengths and areas for improvement. While staple foods remain a consistent dietary mainstay, efforts may be warranted to enhance access to and utilization of nutritious alternatives such as pulses. Addressing barriers to pulse consumption could contribute to improving overall dietary diversity and nutritional adequacy within these households.

The analysis of milk consumption patterns highlights a significant trend, wherein a majority of households (59.2%) abstain from consuming milk altogether. This observation, corroborated by findings from interviews and focus group discussions (FGDs), underscores the primary barrier of limited access to dairy products. Among those who do consume milk, the frequency of intake remains relatively low, suggesting that it does not constitute a staple in the daily diets of many households within the study area; only 40.8% reported consuming milk in one to 3 days. In contrast, meat consumption appears to be less common, with a striking 90.3% of households reporting no consumption of meat before the survey. This data reflects a significant absence of meat from the dietary habits of the surveyed population.

Turning to vegetable consumption, the data presents a varied picture, with a notable portion of households (18.3%) reporting daily consumption of vegetables. However, a considerable proportion (32.2%) of households still refrain from consuming vegetables altogether. This reluctance is primarily attributed to factors such as land scarcity and environmental concerns, as highlighted by participants in the FGDs. These factors underscore the complexities influencing dietary choices and access to nutritious foods within the studied community.

Similarly, fruit consumption patterns show a concerning trend, with 82.3% of households not consuming fruits in general. As observed during field study and response from FGD, the climatic conditions and other related factors are not suitable for fruit growth in the area. Oil/fat/butter consumption is widespread, with a majority of households (78.6%) using these items. However, the low frequency of consumption indicates that these food items are not necessarily consumed daily or its daily consumption was minimal.

Finally, sugar consumption appears to be relatively similar for all, with only 85.4% of households not consuming sugar. However, the intake frequency is relatively minimal, indicating that sugar is not heavily relied upon as a dietary staple. In general, the data underscores notable differences in the consumption habits of various food groups among the households surveyed, with staple foods being more frequently consumed compared to other types of food. This situation raises concerns regarding meeting dietary nutritional needs and its potential implications [56] emphasizing the severe consequences of lacking legumes, livestock products, and fruit consumption, mainly for children, as it can adversely affect their physical and mental development due to diet deficiency.

The evaluation of household food security via the Food Consumption Score (FCS) offers categorical insights into the level of food consumption attained, distinguishing between poor consumption, borderline consumption, and acceptable consumption [62, 63]. Consistent with this method, [45, 61] provides two sets of cutoff points: one for consumption patterns featuring frequent intake of oil and sugar, and another for patterns where daily consumption of these food groups is minimal. Under the first set, households with an FCS of ≤ 28, between 28.5 to 42, and ≥ 42.5 are categorized as having poor, borderline, and acceptable consumption, respectively. For the second pattern, the cutoff points are ≤ 21 for poor, 21.5–35 for borderline, and ≥ 35.5 for acceptable consumption. In this study, the second pattern was adopted to establish the cutoff point, given the survey findings indicating minimal daily consumption of both oil and sugar food groups in the study area. For example, nearly all surveyed households either did not consume oil/fat and sugar food items at all or consumed them for only one to four days within a week. Consequently, the study opts to employ thresholds of ≤ 21 for poor, 21.5–35 for borderline, and ≥ 35.5 for acceptable consumption as indicated in Table 3 above.

Table 9 displays the distribution of the surveyed households among various consumption categories determined by their Food Consumption Score (FCS). Among the 289 households surveyed, 51.3% are categorized as having poor consumption. This indicates a significant portion of households experiencing challenges in achieving adequate food consumption levels. Meanwhile, 34.9% of households are categorized as borderline, implying a precarious situation where their food consumption is at a borderline sufficiency. Conversely, only 13.8% of households are classified as having acceptable consumption levels, signifying a smaller proportion of households successfully meeting recommended dietary standards. These findings suggest a concerning prevalence of inadequate food consumption among surveyed households, with a considerable portion teetering on the edge of food insecurity.

Table 9 Food consumption score derived from the food intake groups of the households
Full size table

The documented prevalence of 51.3% in the poor consumption category within this research underscores a profound level of food insecurity among the surveyed households. These finding sheds light on their challenges in accessing a diverse and nutritious diet, reflecting the utilization aspect of food security. Such a high prevalence resonates with previous studies employing the Food Consumption Score (FCS), which consistently highlights elevated levels of food insecurity.

For instance, a report on resilience, vulnerability, and adaptability in agricultural systems [64] echoed similar trends, indicating a persistent struggle with food insecurity. Moreover, [65] identified a prevalence of 67.5% in the Kobo District of Amhara National Regional, further reinforcing the severity of the issue. In contrast, [66] documented a lower prevalence of food insecurity at 25.6% in their study conducted in the Yeka Sub-city of Addis Ababa. However, this variance in prevalence could be attributed to the urban setting, suggesting that food insecurity might be more pronounced in rural Ethiopia. Therefore, the outcomes of the current study, in conjunction with those focusing on rural areas, substantiate the widespread nature of food insecurity across diverse regions and contexts. This underscores the urgent need for targeted interventions and policy measures to address the underlying factors contributing to food insecurity and enhance food access and nutritional outcomes for vulnerable populations.

3.2.3 Households’ food consumption assessment using the reducing coping strategies index (RCSI)

During periods of crises or adverse shocks impacting livelihoods, households may employ unconventional tactics to address diminishing access to food resources [67]. Serving as an indicator of household food insecurity, the Coping Strategy Index (CSI) combines the frequency and severity of adopted coping mechanisms, with a higher CSI indicating more significant food security challenges [68]. The Reduced Coping Strategy Index (RCSI) is utilized, focusing on a concise set of five food-related coping strategies implemented up to the survey date, along with their respective severity weights. The maximum RCSI score is 56, representing the daily implementation of all five strategies over the past 7 days, with higher RCSI scores indicating more severe coping strategies employed by a household. While there are no universal RCSI thresholds for this study, those used by [23] were employed because these thresholds would have been established through a combination of expert judgment, statistical analysis, and possible consultation with stakeholders familiar with the context making it similar with the methodology employed in this study.

Households were surveyed regarding the frequency with which they employed a series of five short-term food-related coping strategies when faced with insufficient food or lack of funds to purchase food during the week preceding the interview (Table 9). Initially, each of these strategies is assigned a standardized weight proposed by [68] based on its level of severity. These weights are as follows: opting for less preferred and cheaper food options (= 1.0); reducing portion sizes during meals (= 1.0); decreasing the number of meals consumed per day (= 1.0); resorting to borrowing food or seeking assistance from relatives or friends (= 2.0); limiting adults' consumption to allow small children to eat (= 3.0). The data collected from the survey sheds light on the coping mechanisms employed by households during a period of food scarcity in the week leading up to the survey. Five primary coping strategies were identified: opting for less preferred food, borrowing food, reducing portion sizes at meals, limiting adult consumption, and decreasing the number of meals consumed. The subsequent discussion delineates the findings and their consequences (Table 10).

Table 10 Summary of households’ consumption coping strategies
Full size table

Initially, the most commonly adopted coping strategy was opting for less preferred food, with 44.6% of households resorting to this approach at least once during the 7 days. This indicates a widespread willingness to compromise on food preferences in times of scarcity. However, it's notable that a substantial portion of households (20.9%) relied on less preferred foods for four or more days, suggesting a prolonged period of food insecurity for a significant segment of the population. Secondly, borrowing food emerged as another common coping strategy, although to a lesser extent compared to relying on less preferred food. Nonetheless, 3.1% of households depend on borrowing food for 4 days, indicating a reliance on social networks or external sources to alleviate food shortages. This finding highlights the importance of social capital and community support systems during times of crisis.

Thirdly, a significant number of households resorted to strategies related to portion control and meal reduction. Around 27% of households limited portion sizes during meals, while 25% reduced the number of meals eaten within the survey period. These findings underline the adaptive behaviors adopted by households to stretch limited food supplies over an extended period, albeit at the cost of potential nutritional deficiencies and compromised health outcomes. Fourthly, the data indicates that a significant proportion of households imposed drastic measures such as restricting adult consumption. While this strategy was less common overall, with 7% of households implementing it at least once, its occurrence underscores the severity of food insecurity faced by some families.

In conclusion, the finding underscores the resilience and adaptive capacity of households in coping with food insecurity through a variety of strategies. Nevertheless, it emphasizes the pressing requirement for comprehensive interventions to tackle the underlying reasons for food insecurity and alleviate its negative impacts on susceptible communities. Subsequently, the total CSI score was calculated, and the coping level was classified into three severity categories using the classification scheme proposed by [68]: low coping (CSI = 0–3), medium coping (CSI = 4–9), and high coping (CSI ≥ 10) (refer to Table 11). The investigation of the household data reveals that a significant portion of households exhibit high coping severity with 53.3% falling into this category. This suggests that a substantial proportion of households are facing significant challenges in managing and mitigating food insecurity.

Table 11 Consumption-based coping tactics and their corresponding severity weights for RCSI
Full size table

Conversely, 15.6% of households are classified as having low coping severity, indicating a relatively lower degree of resilience in dealing with adverse circumstances. While this proportion is smaller compared to those with medium or high coping severity, it still represents a notable segment of the sample households that may require interventions and support mechanisms to enhance their coping capacity. The middle category, representing households with medium coping severity, comprises 31.1% of the total. Although these households may not be experiencing the most severe levels of stress, they still face significant challenges in coping with adverse circumstances. Overall, the data highlights the heterogeneity in coping severity among households, underscoring the complex and multifaceted nature of food insecurity and must take into account this diversity and tailor interventions to the specific needs and circumstances of different households.

3.2.4 Cross-validation of HFIAS, FCS, and rCSI scores

When collecting data from the same community or empirical setting, it is expected that FCS would show a negative correlation with both the HFIAS and the rCSI. Conversely, the scores for households on HFIAS and rCSI are anticipated to exhibit a positive correlation. This expectation arises because higher scores on HFIAS and rCSI indicate increased food insecurity, while higher scores on FCS suggest better food consumption [13, 47, 56, 69]. To verify this, a Pearson correlation analysis was conducted using the continuous scores of the three parameters (rCSI, FCS, and HFIAS) for cross-validation (Table 12). The correlation analysis of these scores in the study aligns well with these academic expectations. Specifically, FCS demonstrates a significant negative correlation with both HFIAS (r = − 0.624, p < 0.01) and rCSI (r = − 0.625, p < 0.001). Additionally, as expected, HFIAS shows a significant positive correlation with rCSI (r = 0.614, p < 0.001), indicating strong associations among the variables. This finding is consistent with [42, 43, 47] research, which also examined various food security indicators.

Table 12 Correlation between HFIAS, FCS, and rCSI scores
Full size table

Hence, in this investigation, both primary metrics, HFIAS and FCS, exhibit a robust and meaningful correlation with the third metric, rCSI. Consequently, the variations noted in food insecurity rates, with HFIAS at 62% (including mildly, moderately, and severely food-insecure households) and FCS at 51% (poor borderline), can be ascribed to the inherent differences in the metrics employed. While HFIAS proves more responsive to food insecurity, FCS tends to be more accommodating, thereby serving the opposite end of the spectrum.

3.3 Effect of landslide hazard on rural households’ food security status

3.3.1 The level of households’ landslide hazard awareness

Evaluating households' awareness of landslide hazards is vital for enhancing disaster preparedness and resilience [70, 71]. Hence, data collected from 7 items, along with inputs from open-ended questionnaires, interviews, and FGD, are presented (Table 13).

Table 13 Households' level of landslide hazard awareness (N = 289)
Full size table

The assessment of households' landslide hazard awareness reveals varying levels of understanding and preparedness within the households (Table 13). That is analysis of the data, based on mean scores, indicates that respondents generally lean towards an undecided stance (mean = 2.79), suggesting a lack of clear consensus regarding their awareness and preparedness for landslide hazards. Notably, a significant portion of households express uncertainty about the occurrence of landslide hazards in their locality (mean = 2.7), indicating a potential gap in knowledge regarding early warning signs or risk factors. Moreover, while many respondents acknowledge the absence of prevention mechanisms for landslide hazards (mean = 2.57), a considerable proportion also expresses readiness to engage in preparedness activities such as attending meetings or receiving training (mean = 2.91 and 2.61 respectively). However, there appears to be a disparity in perceptions regarding the likelihood of future landslide hazards, with some expressing confidence (mean = 3.19) while others maintain uncertainty (mean = 2.67).

During both the interview and focus group discussion (FGD) sessions, households were queried regarding their level of awareness regarding landslide hazards. The responses yielded a spectrum of perspectives. Some participants demonstrated a dearth of knowledge and awareness concerning the mechanisms and prevention of landslides. Conversely, others underestimated the significance of landslides, attributing it to the perceived infrequency of such events in their locale. Another group acknowledged the potential for landslides but confessed uncertainty about identifying warning signs or knowing the appropriate responses during such occurrences, indicating a lack of preparedness. Additionally, there was a prevailing sentiment among participants regarding the inadequacy of education or training on landslides, contributing to the overall low levels of awareness.

These findings align with previous research reports [71, 72] that delve into the socio-economic ramifications of landslides. It underscores the critical need for targeted educational and awareness campaigns to address knowledge gaps and dispel misconceptions, thereby fostering a culture of proactive measures for mitigating landslide risks. In essence, the study underscores the necessity of empowering communities with accurate information and practical skills to enhance their resilience against landslide hazards.

3.3.2 The level of households' landslide hazard preparedness strategies

Evaluating the extent of households' readiness to deal with landslide hazards is essential for gauging their capacity to respond effectively to such events, thus minimizing risks to life and property [62, 73, 74]. Therefore, information regarding these matters was gathered through 9 items, and the results are outlined below (Table 14), complementing data acquired through open-ended questionnaires, interviews, and focus group discussions. Thus, data on households’ landslide hazard preparedness strategies highlight a mixed level of awareness and readiness within the households. While respondents generally express moderate levels of preparedness, with a grand mean of 2.47, certain areas show room for improvement. For instance, while a mainstream of respondents agree or strongly agree with certain aspects, such as being aware of government training and information provision (mean = 2.92) and having emergency kits ready (mean = 2.19), other areas show lower levels of preparedness. Notably, there seems to be uncertainty regarding crucial elements like awareness of early warning systems (mean = 2.30) and coordination with agencies for help during and after disasters (mean = 2.29).

Table 14 The level of households' landslide hazard preparedness strategies (N = 289)
Full size table

Throughout both the interview sessions and focus group discussions (FGDs), respondents were probed with a broad question regarding the existence of disaster prevention, response, or mitigation strategies for nature-related hazards in their village. The responses shed light on several critical issues. First and foremost, there was a recurring theme of limited access to essential resources such as funding, expertise, and governmental support. Consequently, communities often relied more on informal networks and traditional knowledge rather than formalized plans or resources, highlighting a significant gap in institutional support for disaster management at the local level.

Furthermore, the findings underscored the necessity for improved communication regarding evacuation plans, early warning systems, and coordination mechanisms. Many respondents expressed a lack of clarity or awareness regarding these crucial aspects of disaster preparedness. Strengthening communication channels and disseminating information effectively could significantly enhance the community's resilience and reduce the adverse impacts of landslide hazards on households [58, 75,76,77]. Moreover, the study highlighted the importance of raising awareness about disaster-proof infrastructure and shelter options. It was noted that mean scores indicating knowledge or understanding of these concepts were lower in the study areas, suggesting a need for targeted educational initiatives. By equipping residents with the knowledge of constructing or accessing safer infrastructure and shelters, the overall vulnerability of communities to landslide hazards can be substantially mitigated.

In conclusion, the study underscores the urgency of addressing systemic deficiencies in disaster preparedness and response at the local level. By bridging gaps in resources, enhancing communication channels, and promoting awareness and education, communities can better withstand the challenges posed by nature-related hazards such as landslides.

3.3.3 Impact of landslide hazards on the food security status of affected households

Examining how landslide hazards impact the food security status of affected households holds significant importance for various reasons. Primarily, landslides can disrupt agricultural activities, leading to crop damage, soil erosion, and loss of livestock, thereby directly impacting food production and availability. Understanding these effects is crucial for assessing the extent of food insecurity and identifying vulnerable populations. Additionally, landslides can damage infrastructure such as roads and bridges, hindering transportation and access to markets, and further exacerbating food shortages [58, 70, 78,79,80]. In pursuit of this objective, this study sought to examine the impact of landslide hazards on the food security status of rural households in landslide-prone regions, utilizing seven specific criteria (Table 15).

Table 15 Effect of landslide hazards on rural households’ food security status (N = 289)
Full size table

The data indicates significant findings regarding the impact of landslides on food security and household welfare. Firstly, all participants (100%) acknowledged the tangible threat landslides pose to their food security. This high level of awareness underscores the perceived risk and potential ramifications of landslides on food availability and access in the affected areas. With a mean score of 1.67, there appears to be a strong consensus among respondents regarding the severity of this threat. Secondly, 32.5% of households reported being compelled to relocate to new areas due to the collapse of their homes caused by landslides (Fig. 2). Such displacement could entail additional challenges for food security, including loss of access to cultivable land and disruption of established food distribution networks.

Fig. 2
figure 2

Displaced households resettled in protected hilly areas of Gaste Village. Photo, by researchers, 2023

Full size image

Furthermore, a significant majority of households (92.7%) recognized the detrimental effects of landslides on their food and nutritional security, indicating a strong awareness of the connection between environmental hazards and food availability. This perception likely reflects direct experiences such as crop damage and loss of food stocks during landslide events. Moreover, almost all households (99.3%) replied acknowledging the property or household asset loss due to house collapse, emphasizing the substantial economic toll of landslide hazards. Loss of assets can hinder recovery efforts and exacerbate food insecurity by limiting households’ ability to purchase food or invest in agricultural inputs. Finally, a considerable proportion (37%) of households experienced house collapse directly attributable to landslide hazards, further underscoring the physical destruction and safety risks posed by such events. The findings underscore the urgent need for comprehensive risk reduction strategies and support systems to bolster the resilience of vulnerable households to natural disasters and protect food security amidst environmental risks.

Furthermore, to gauge the impact of landslide hazards on household food security status, comprehensive data were gathered utilizing a combination of methodologies including open-ended questionnaires, individual interviews, and focus group discussions (FGDs). Respondents were prompted with inquiries aimed at elucidating how landslide hazards influenced their households’ food availability, access, and utilization. The responses underscored the profound ramifications of landslide events, particularly those occurring in January and May of 2021. These natural disasters wrought extensive damage to residential properties and agricultural lands. Houses were obliterated, rendering families homeless and incredibly vulnerable. Simultaneously, agricultural fields were submerged under debris, leading to substantial crop loss and livestock fatalities for farmers as depicted in Fig. 3. The dual devastation of homes and farmlands exacerbated the socio-economic vulnerabilities of affected households, accentuating the dire need for immediate assistance and support to facilitate recovery efforts.

Fig. 3
figure 3

Landslide hazard caused disaster in Kodo village, Photo credit, Gamo Zone Disaster Risk Management Department, 2021

Full size image

This empirical finding resonates with prior research endeavors [72, 81, 82], which also underscore the critical importance of understanding the multifaceted impact of landslide hazards on food security. Such insights are indispensable for guiding the development of efficient disaster response and recovery strategies, thereby safeguarding the welfare and livelihoods of affected households. In essence, the study underscores the imperative of integrating considerations of food security into disaster management frameworks, recognizing its pivotal role in ensuring the resilience and well-being of communities facing the threat of landslide hazards.

4 Conclusion and recommendations

This study sought to evaluate the impact of landslide hazards on rural household food security in the Gacho Baba District, located in Southern Ethiopia's Gamo Highlands. Despite the considerable attention given to landslides in Ethiopia, their ramifications for rural household food security have often been overlooked in prior research. The findings of this study, as indicated by the Household Food Insecurity Access Scale (HFIAS), underscore widespread concerns among households regarding food shortages and limited access to preferred foods, revealing significant food insecurity challenges. The Food Consumption Score (FCS) assessment further highlights insufficient food consumption levels, particularly among households categorized as having poor consumption, indicating difficulties in accessing diverse and nutritious diets. Moreover, the analysis of coping strategies using the Reduced Coping Strategies Index (RCSI) underscores both the adaptive measures adopted by households to manage food insecurity and the urgent need for targeted interventions to bolster coping capacity and address underlying vulnerabilities.

Furthermore, this study evaluates the levels of landslide hazard awareness and preparedness among rural households, revealing varying degrees of understanding and readiness within communities. While some respondents expressed readiness to participate in preparedness activities, such as attending meetings or undergoing training, uncertainty persists regarding the occurrence of landslide hazards and awareness of early warning systems. Additionally, the study highlights the substantial impact of landslide hazards on food security, with all participants acknowledging the threat to their food availability and access.

In light of these findings, it is imperative for regional, zonal, and local authorities, along with stakeholders, to prioritize educational and awareness initiatives aimed at enhancing landslide hazard awareness among rural households. This should include providing training, disseminating information about early warning systems, and fostering community participation in disaster preparedness activities. Additionally, comprehensive risk reduction strategies must be developed and implemented to strengthen household resilience to landslide hazards. This involves bolstering infrastructure, devising evacuation plans, and providing support for recovery and livelihood restoration. Furthermore, efforts to safeguard food security should be seamlessly integrated into disaster risk management efforts, ensuring that vulnerable populations receive timely access to food aid and support in the aftermath of landslide events.