Microlevel Analysis of Climate Change Adaptation and its Determinants: The Case of Smallholder Farmers in the Kembata Tembaro Zone, Southern Ethiopia
Abstract
The high vulnerability of regions that predominantly rely on rain-fed agriculture to climate change and variability underscores the importance of understanding adaptation responses as a foundation for designing targeted interventions. This study investigates microlevel climate change adaptation strategies and factors determining such choices of smallholder farmers in the Kembata Tembaro Zone, southern Ethiopia. It employed a convergent parallel mixed-method research design, incorporating quantitative data collected through household surveys of 364 randomly selected smallholder farmers across diverse agro-ecological zones. This was complemented by qualitative data obtained through interviews and focus group discussions (FGDs). A multinomial logit (MNL) regression model was used to analyze the determinants of the farmers’ adaptation decisions. Nearly all respondents perceived an increase in temperature, consistent with observed trends. However, the farmers’ perception of declining rainfall corresponded only with trends during the short rainfall season (Belg). Farmers in lowland areas demonstrated greater experience with adaptation strategies and are more resilient than those in midland and highland regions. On average, 88.2% of the farmers reported implementing adaptation measures, such as adjusting planting dates, crop diversification, soil and water conservation, adopting improved livestock breeds, engagement in off-farm activities, and agroforestry practice. The regression analysis identified several significant factors influencing adaptation decisions, including age, education level, farm size, farm income, access to credit, participation in climate change training, availability of climate information, market access, and agro-ecological context. This study offers unique, context-specific insights through inter-agroecological comparisons of smallholder farmers’ adaptation choices, highlighting the complex interplay between socioeconomic and environmental variables in determining adaptation to climate change. The findings can inform policymakers when designing development policies tailored to specific local contexts.
1. Introduction
There is general agreement that Earth’s climate is changing [1]. Climate change is a global challenge that affects humanity and their livelihoods [2, 3]. Developing countries are generally more vulnerable to the impacts of climate change than developed countries [4]. Sub-Saharan Africa is considered the most vulnerable region to climate change due to its heavy reliance on rainfed agriculture, which is highly sensitive to weather and climate variabilities [5–7]. Smallholder farmers in the region are particularly at risk, given their reliance on rain-fed agriculture, limited irrigation use, and low adaptive capacity [8]. Consequently, efforts towards climate change adaptation and mitigation are essential to minimize the impacts of climate variability and change, especially in the agricultural sector [4].
Various adaptation measures have been implemented, including crop and farm diversification, the use of improved crop varieties, soil conservation practices, adjusting planting time, small-scale irrigation applications, and income diversification [9, 10]. However, the current adaptation strategies are insufficient to combat climate change’s present and anticipated impacts, necessitating further research on climate change adaptation and its determinants.
A diverse set of factors influences smallholder farmers’ adaptation to climate change. Climatic conditions directly impact agricultural productivity [11]. Socioeconomic factors, such as income, access to resources, and education, shape adaptive capacity [12]. Furthermore, institutional factors, including access to information and support services, are crucial in facilitating adaptation [12]. Absence or inefficiencies of these services are particularly critical for smallholder farmers in developing countries, exacerbating their vulnerability due to reliance on rain-fed agriculture and low adaptive capacity [13, 14].
Ethiopia is one of the countries in Sub-Saharan Africa, particularly considered vulnerable to climate change, not only owing to its heavy reliance on rainfed agriculture but also due to severe land degradation, high population pressure, small land holdings, and limited adaptive capacity [15, 16]. The nation’s agricultural sector, which significantly contributes to the overall economy, is extremely susceptible to climate-related disasters, such as drought and floods [17]. Approximately 95% of Ethiopia’s principal crops are produced by smallholder farmers [18]. These farms face numerous challenges, including poor soil fertility, land degradation, weather-dependent crop production, and economic constraints [19].
Researchers in Ethiopia have conducted many studies focusing on smallholder farmers’ climate change adaptation choices and their determinants [14, 20–25]. These studies have reported that farmers’ adaptation responses and their determinants vary considerably across space, scale, and time. This calls for more studies to generate additional data under diverse settings to help design targeted policy and intervention strategies fitting local contexts [26, 27]. Furthermore, assessment of smallholder farmers’ choices of climate change adaptation and its determinants at the microscale provides more detailed information than large-scale studies [28, 29], which may help in designing more practical and effective adaptation strategies [16].
The Kembata Tembaro Zone, the current study area, is one area where climate change poses a significant threat to smallholder farmers who rely heavily on climate-sensitive agriculture. The climate impact in the area is further compounded by social and economic challenges such as high population density, rugged terrain, fragmented farm plots, deforestation, and extensive land degradation [30, 31], leading to a decline in agricultural production in the region [31]. The objectives of this study were to (i) explore smallholder farmers’ adaptation strategies to climate change and (ii) investigate the determinants of smallholder farmers’ choices of these adaptation strategies in the Kembata Tembaro Zone, southern Ethiopia.
2. Materials and Methods
2.1. Description of the Study Area
This study was conducted in the Kembata Tembaro Zone, SNNPR of Ethiopia. The Kembata Tembaro Zone is located in the northeastern part of the SNNPR. It is located between 7° 5’ N–7° 35’ N latitude and 37° 23’ E–38° 5’E longitude (Figure 1). The elevation of the Zone ranges between 900 and 2906 m above sea level. Various landscapes, such as mountains, plateaus, and plains, characterize the zone. In 2014, the total population of the Zone was 1,080,837, of which 536,676 were male and 544,161 were female [32]. It is one of the most densely populated areas in the country, with a crude population density of 665 people/km2, four times the estimated regional average of 164 people [32]. Land use in the zone was predominantly rain-fed cereal crop cultivation and a fragmented homestead agroforestry system. The farming systems are mostly mixed (crop and livestock). The study area was subdivided into three main agroecological zones: Lowland (<1500 masl), mid-highland (1500–2300 masl), and highland (>2300 masl), categorized as elevation range. The study area has bimodal rainfall patterns with a short rainy season (February to May) and a major rainy season (June to September). The dry period lasts from October to January. The major share of annual rainfall was received in Kiremt (rainy season) (47%) and Belg (growing season) (37%). The lowest rainfall occurred during the short rainy season (Belg) that lasted between February and May [30, 33].
2.2. Data Types and Sources
2.2.1. Research Design, Sampling Techniques, and Sample Size Determination
Based on this formula, the sample size needed for this study was 364. The probability proportional to size technique was applied to distribute the household sample sizes to each agroecology. Finally, the sample households per kebele and agroecology were selected randomly using a simple random sampling procedure. Accordingly, 119, 143, and 102 households from the lowland, midland, and highland agroecologies were involved in the study, respectively (Table 1).
| Agro-ecologies | Household size | Sample size taken |
|---|---|---|
| Lowland | 2292 | 119 |
| Midland | 2767 | 143 |
| Highland | 1970 | 102 |
| Total | 7029 | 364 |
2.3. Data Collection Methods
The primary data for this study were gathered between September 2022 and November 2022 after a pilot survey was conducted in June 2022. Participatory rural appraisal (PRA) methods were used, including household questionnaire surveys, key informant interviews (KIIs), and focus group discussions (FGDs) [35]. A structured questionnaire was used to collect data from farmers at the household level to enable the study to examine the adaptation practices of smallholder farmers and the factors that influence such practices. Thirty-six key informants, six persons from development agents, women, and elders across the three agroecologies, participated in the interview. KIIs were conducted to support the data collected from the household survey regarding climate change adaptation strategies. Moreover, we used 10-person focus group discussants in each kebele, for a total of 90 discussants from various socioeconomic statuses. FGDs were held to interact with a range of social groups and focus on group norms and dynamics around the issue being investigated. The information from FGDs and KIIs was integrated with quantitative data to determine the agroecological responses to climate change and variability. Secondary data were acquired from published articles and reports from the agricultural and kebele offices. Daily rainfall and temperature data were gathered from Ethiopia’s National Meteorological Agency (NMA) to compare farmers’ perceptions of climatic trends with actual facts. For this aim, 31 years of climate data (1989–2019) were obtained from the NMA, the country’s sole climate data provider. Outlier detection and homogeneity tests were conducted to ensure data quality. Missing values in rainfall and temperature data were filled using ArcGIS 10.4’s Inverse Distance Weighted (IDW) interpolation algorithm. Finally, the observed average yearly and seasonal data were used for trend analysis.
2.4. Data Analysis
The data was sorted, classified, coded, and analyzed using descriptive statistics (frequencies and percentages). The analysis was carried out using Stata 14 and Excel 2016. The Mann-Kendall (MK) and Sen’s slope approaches were used to analyses climate trends in the MAKESENS version 1 software. Salmi [36] provides a full explanation of the MAKESENS technique. The MK trend test was used to determine the existence and significance of monotonic (either rising or declining) trends in the observed climatic data series. Sen’s slope estimator was used to calculate the magnitude of these linear trends. Statistical significance was determined at 0.01 and 0.05 levels, representing 99% and 95% confidence intervals, respectively.
This study used a multinomial logit (MNL) model to examine factors influencing farmers’ choices of adaptation measures applied by farm households in the study area. This study used this method to analyze farmers’ choices regarding climate change adaptation strategies and the factors that determine these choices. The MNL model is suitable for estimating the likelihood that a given option is preferred to other available options, assuming that the available options are mutually exclusive [37]. The MNL model is based on previous literature on the determinants of farmers’ adaptation to climate change [38]. The MNL regression model permits the analysis of climate change adaptation decisions by analyzing decisions across more than two categories, thereby allowing for determining choice probabilities for different categories.
The unbiased and consistent parameter estimations of the MNL model in Equation (3) require the assumption of irrelevant alternative independence (IIA) to hold [22]. This indicates that the probability of using certain adaptation options by a given household must be independent of the probability of choosing another adaptation option (i.e., Pj/Pk is independent of the remaining probabilities). The MNL adaptation model was run and tested for the IIA assumption using the Hausman specification test. As a result, the test failed to reject the null hypothesis of the independence of odds of other alternatives with χ2 ranging from −4.05 to 8.88 with probabilities almost equal to 1.000, suggesting that there was no evidence against the correct specification for the adaptation model.
| Variable | VIF | 1/VIF |
|---|---|---|
| Age of HHs | 1.03 | 0.968 |
| Farm size | 1.03 | 0.968 |
| Farm size | 1.18 | 0.849 |
| Livestock ownership | 1.21 | 0.829 |
| Farm income | 1.04 | 0.961 |
| Market distance | 1.08 | 0.926 |
| Mean VIF | 1.19 | — |
| Variables | sex | ES | CI | AE | AC | AT | PT | PRF | AZ |
|---|---|---|---|---|---|---|---|---|---|
| Sex | 1 | — | — | — | — | — | — | — | — |
| ES | 0.0014 | 1 | — | — | — | — | — | — | — |
| CI | 0.0538 | 0.0084 | 1 | — | — | — | — | — | — |
| AE | 0.0236 | −0.112 | 0.0522 | 1 | — | — | — | — | — |
| AC | 0.0145 | −0.026 | −0.0619 | 0.0222 | 1 | — | — | — | — |
| AT | −0.046 | 0.0153 | −0.0193 | 0.0102 | −0.058 | 1 | — | — | — |
| PT | −0.051 | 0.0213 | 0.0993 | 0.0659 | −0.099 | −0.163 | 1 | — | — |
| PRF | 0.0546 | 0.0288 | 0.0039 | −0.03 | 0.0323 | 0.0367 | −0.084 | 0.056 | — |
| AZ | 0.1958 | −0.013 | 0.0636 | 0.0567 | 0.0142 | −0.221 | 0.085 | 1 | 1 |
- Abbreviations: AC, access to credit; AE, access to extension contact; AT, access to training; AZ, agroecological zone; CI, access to climate information; ES, education status; PRF, perceived rainfall; PT, perceived temperature.
Definition of variables and hypothesis: The response variables were climate change adaptations implemented by the farmers in the study area. As stated by Etana et al. [26], many African countries use crop diversification, shifting planting dates, livestock varieties, and soil and water conservation to adapt to climate change. The choice of independent variables was dictated by the empirical literature, its suggested behavioral hypotheses, and data availability. Explanatory variables have been hypothesized to influence farm-level adaptations [42, 43]. Table 4 describes the explanatory factors employed in this investigation, as well as the hypotheses around them or the expected signals.
| Explanatory variables | Variable type | Variable measurement | Expected effect |
|---|---|---|---|
| Sex | Dummy | 1 if male, 0 otherwise | ± |
| Age | Continuous | Year | ± |
| Family size | Continuous | Number | + |
| Education status | Dummy | Number | + |
| Farm size | Continuous | Number | − |
| Livestock ownership | Continuous | Tropical Livestock Unit | + |
| Farm income | Continuous | Birr | + |
| Access to climate information | Dummy | 1 if there is access, 0 otherwise | + |
| Access to extension service | Continuous | 1 if there is access, 0 otherwise | + |
| Access to credit | Dummy | 1 if there is access, 0 otherwise | + |
| Access to training | Dummy | 1 if there is access, 0 otherwise | + |
| Distance to the market | Continuous | Number | − |
| Rainfall trend change | Dummy | 1 if they perceive, 0 otherwise | + |
| Temperature trend change | Dummy | 1 if they perceive, 0 otherwise | + |
| Agro-ecology | Categorical | 0 if midland, 1 if lowland, and 2 if highland | ± |
3. Results and Discussion
3.1. Demographic and Socioeconomic Characteristics
The age of the household heads involved in this study ranged from 25 to 70 years, with an average age of 48 years (Table 5). Household family size varied from one to nine members, with an average of five people, similar to the findings of [44] and the national average [45]. The education level of the household heads ranged from “illiterate” to “completion of grade 12.” Of the total heads of the households, 79.9% had received formal education, while 20.1% did not. Literacy rates varied across agro-ecologies, with 21.8% in the lowlands, 21.7% in the midlands, and 15.7% in the highlands having no formal education (Table 5). The average education level of the sample households is a bit higher than the national average of 5% [46]. Smallholder farmers had an average farm size of 0.79 ha, with minimum and maximum values ranging from 0.125 to 1.5 ha. The average farm size of the households is less than the national average of 0.95 ha [47], suggesting a shortage of agricultural land in the study area. The livestock holding of respondents ranged between 1 and 7.85 tropical livestock unit (TLU), with an average of 3.32 (Table 5). The annual income of the households ranged from 2400 to 16,350 Ethiopian Birr (ETB), with an average income of 5811.1 ± 2122.6 ETB per annum. Regarding distance to market, the average distance the respondents travel to reach the nearest market center is ~6.74 km, with minimum and maximum distances of 4 and 10 km, respectively (Table 5). This finding is backed by previous studies [48–50].
| Variables | Lowland | Midland | Highland | Overall |
|---|---|---|---|---|
| μ (±SD) | μ (±SD) | μ (±SD) | μ (±SD) | |
| Age | 46.8 ± 8.49 | 47.22 ± 11.0 | 51.24 ± 11.2 | 48.21 ± 10.5 |
| Family size | 4.93 ± 1.58 | 5.15 ± 1.43 | 6.08 ± 1.29 | 5.34 ± 1.52 |
| Education status | 3.86 ± 3.08 | 4.24 ± 3.21 | 5.18 ± 3.51 | 4.3 ± 3.29 |
| Farm size | 0.76 ± 0.42 | 0.82 ± 0.37 | 0.78 ± 0.42 | 0.79 ± 0.40 |
| Tropical livestock unit | 2.65 ± 1.49 | 3.39 ± 1.69 | 3.96 ± 1.61 | 3.31 ± 1.68 |
| Household income | 4886 ± 1516 | 5383 ± 1809 | 7490 ± 2188 | 5811 ± 2123 |
| Market distance | 7.36 ± 1.155 | 6.49 ± 1.428 | 6.37 ± 1.371 | 6.74 ± 1.39 |
- Note: μ = Mean, SD = standard deviation.
As shown in Table 6, the majority of households (88.5%) were male-headed. This trend varied across agroecologies, with 93% from lowlands, 91% from midlands, and 71% from highlands. The percentage of female-headed households averaged 11.5% and ranged from 6.7% in the highlands to 21% in the lowlands. This is lower than the nationally reported 16% of female-headed households [44]. In Ethiopia, women typically become household heads only in the absence of their male counterparts, resulting from death, divorce, or seasonal migration for wage work. Farming experience among the surveyed families varied, and it was classified as low (0–10 years), medium (10–20 years), and high (more than 30 years) based on [46]. Accordingly, about 16.2% of the households possessed high farming experience, with more than 30 years of engagement in agriculture. The relatively small percentage of households with high experience suggests that a significant portion of the farmers are still in the early to mid-stage farming careers, potentially reflecting recent engagement with agriculture (Table 6). About 70.3% of the households had medium farming experience categories, with 10–20 years engagement in farming, reflecting substantial practical knowledge accumulated over time. Conversely, 13.5% of the households had lower farming experience, which ranged from 0 to 10 years, indicating that a smaller portion of the respondents are relatively new to agriculture or had fewer opportunities to engage in farming.
| Variables | Households | Lowland | Midland | Highland | Total |
|---|---|---|---|---|---|
| % | % | % | % | ||
| Sex | Male | 93.3 | 91 | 79 | 88.5 |
| Female | 6.7 | 9.0 | 21 | 11.5 | |
| Farm experience | Short | 6.7 | 22 | 8.8 | 13.5 |
| Medium | 71 | 66 | 76 | 70.3 | |
| Higher | 22 | 12 | 16 | 16.2 | |
| Livelihood practice | Crop | 19 | 17 | 49 | 26.4 |
| Livestock | 7.6 | 5.6 | 2.9 | 5.5 | |
| Mixed | 74 | 78 | 48 | 68.1 | |
| Climate information | Yes | 34 | 36 | 29 | 33.5 |
| No | 66 | 64 | 71 | 66.5 | |
| Extension contact | Yes | 87 | 86 | 84 | 85.7 |
| No | 13 | 14 | 16 | 14.3 | |
| Access to training | Yes | 59 | 47 | 30 | 46.2 |
| No | 41 | 47 | 70 | 53.8 | |
| Access to credit | Yes | 24 | 22 | 37 | 27.2 |
| No | 76 | 78 | 63 | 72.8 | |
- Note: Source: Survey result, 2022.
About 68% of the respondents engage in mixed crop and livestock rearing as their primary means of livelihood (Table 6). In addition, 66.5% of the respondents had no access to climate information. About 85.7% of respondents had access to extension services, with a frequency of extension contact ranging from 1 to 15 times per annum (Table 6). The results showed that about 53.8% of the respondents had access to agricultural training services, while 46.2% had no access (Table 6). Furthermore, 72.8% of the respondents had no access to credit services. This result is consistent with the results of previous studies [48, 51, 52]. The result demonstrates that understanding demographic and socioeconomic characteristics is vital for designing effective adaptation strategies to climate change.
3.2. Smallholder Farmers’ Perceptions of Climate Change
When asked whether they had noticed any changes in the climate within their locality over the past 31 years, 98% of respondents indicated that they had perceived climate change during this period. In contrast, about 2% of participants did not notice any climate change trends. This finding is consistent with the study by Teshome et al. [53], who reported that the majority of farmers (83%) recognized the existence of climate change in Eastern Ethiopia. However, our findings show differences in climate change perception across agroecologies. Farmers in the lowland agroecology were most likely to recognize the presence of climate change (100%), while those in the midland and highland agroecologies were slightly less aware (98% and 96%, respectively). These differences may result from more intense environmental factors, such as droughts that reduce water availability and diminish crop yields in lowland agroecology compared to the mid and highland agroecologies. It is also apparent that farmers in lowland agroecologies are more vulnerable to the impacts of climate change. This finding is consistent with the study of Teklay et al. [54], which reported the existence of differences in climate change perception across agroecologies in the Megech Watershed, Ethiopia.
The findings reveal that almost all households perceive changes in rainfall, indicating widespread awareness of climate variability and change. A majority (51%) perceive declining rainfall, and 32% note shifts in the timing of rains, reflecting concerns about rainfall unpredictability, which could threaten agricultural productivity. Perceptions of increased rainfall (15%) contrast with the overall negative outlook, suggesting varied experiences among households, while only 2% reported no noticeable change. Consistent with household perceptions, discussants reported a decline in rainfall over the study area. This reduction has adversely affected agricultural productivity and water availability, decreasing crop yields and water resources. Additionally, participants highlighted that these changes have contributed to the increased prevalence of human and animal diseases, ultimately exacerbating food insecurity in their respective communities.
Regarding temperature, a higher percentage of the research participants (83%) reported experiencing a rise in temperature over the past 31 years (Table 7). Some participants (12%) perceived a decrease in temperature. On the other hand, a few (4%) said they have not noticed any change, while 1% were unsure about the trend. Most farmers across all the agroecological zones observed an increasing temperature trend, although the degree of perception varied by zone. A greater proportion of farmers in the lowlands (91%) perceived rising temperatures compared to those in the highlands (80%) and midlands (78%). Similarly, focus group discussants reported a rising temperature trend across the study area, negatively affecting crop and livestock productivity. Geographical differences significantly influenced local perceptions of temperature changes. In lowland areas, the warmer climate is primarily attributed to lower elevation and proximity to bodies of water, contributing to higher average temperatures. Conversely, highland regions, situated at higher altitudes, generally experience cooler conditions. Additionally, lowland areas may exhibit varied microclimatic conditions shaped by topography, vegetation, and land use. These factors can make temperature fluctuations more noticeable to local farmers, further highlighting the contrasts experienced across the three geographic settings. This finding corresponds with other similar studies [55, 56], while slightly differing from some other studies conducted in Ethiopia and other countries, confirming variability in perceptions of climate change depends on social and geographic context [53, 57].
| Parameters | Trend | Respondents (%) in respective agro-ecology | |||
|---|---|---|---|---|---|
| Lowland | Midland | Highland | Total | ||
| Climate change | Perceived | 100 | 98 | 96 | 98 |
| Not perceived | — | 2 | 4 | 2 | |
| Rainfall | Increasing | 7 | 23 | 15 | 15 |
| Decreasing | 61 | 54 | 55 | 51 | |
| Altered change | 39 | 21 | 22 | 32 | |
| I do not know | — | 2 | 3 | 2 | |
| Temperature | Increasing | 91 | 78 | 80 | 83 |
| Decreasing | 2 | 20 | 15 | 12 | |
| Altered change | 8 | 2 | 2 | 4 | |
| I do not know | — | — | 3 | 1 | |
- Note: Source: Survey result, 2022.
3.3. The Trend of Meteorological Data and Participants’ Understanding
3.3.1. Rainfall Trends
Conducting climate trend research necessitates a significant amount of rainfall and temperature data, with a minimum of 30 years necessary. The MK trend test was utilized for this research because it is widely used in hydrology and climatology [58]. The annual rainfall in the Kembata Tembaro Zone showed an increasing trend (p < 0.05), with a rate of change of 3.58 mm/year (Table 8). In this regard, the Kiremt (main rainy season) rainfall increased significantly (p < 0.05) at a rate of 3.822 mm/year. Conversely, the Belg (short rainy season) rainfall showed a non-significant decreasing trend (p > 0.05), with a rate of change of −0.778 mm/year. Overall, annual rainfall showed an increasing trend, indicating a general trend towards higher precipitation over the past three decades. Specifically, the Kiremt season experienced a statistically significant increase in rainfall, while the Belg season showed a slight decline, but this was non-significant (Table 8). The 31-year trend analysis of rainfall in the study area shows divergent trends between meteorological data and farmer perception, necessitating careful interpretation. Specifically, the study shows an increase in yearly and Kiremt (main rainy season) rainfall trend, but farmers report a downward trend in rainfall during the Belg season, which is consistent with meteorological data for this season. This discrepancy occurs because meteorological data shows broad-scale, long-term trends that may not fully convey seasonal or localized swings affecting farmers’ daily experiences. Farmers’ perspectives are directly influenced by current and localized weather conditions, particularly during crucial agricultural seasons such as Belg. Despite an average increase in annual rainfall, the reduction or unreliability of Belg rains can substantially impact farmers’ livelihoods, giving the impression that rainfall is decreasing during that season. Recognizing this divergence highlights the significance of combining scientific data with local experiences for developing effective climate adaptation programs. Other researchers have also reported the inconsistency of meteorological data and farmers’ observations [59–61]. In these studies, respondents felt a reduction in rainfall while meteorological records show increasing trends. The inconsistency could be explained as resulting from the inherent limitation of mean value analysis in the meteorological data, extreme value observation of farmers, and the farmers’ water demand. The farmers’ water demand could be explained as resulting from the increasing water scarcity due to population pressure and land degradation.
| Scale | No | Min | Max | μ | Z | Q |
|---|---|---|---|---|---|---|
| Annual | 31 | 1044 | 1521 | 1271 | 1.67 | 3.583 |
| Kiremt | 31 | 373 | 732 | 542 | 2.92 ∗∗ | 3.822 |
| Belg | 31 | 319 | 661 | 470.6 | −0.306 | −0.778 |
- Note: No = Number, Min = Minimum, Max = Maximum, μ = Mean, Z, test statistic, Q = Sen’s slope.
- ∗, ∗∗, and ∗∗∗ indicate statistical significance at the 0.1, 0.05, and 0.01 Alpha levels, respectively.
3.3.2. Temperature Trends
Regarding temperature, the observed average annual temperature showed a considerable increase between 1989 and 2019 (Table 9). The study indicated that the average yearly temperature ranged from 24.9 to 27.0 °C, with a mean of ~26.2 °C. The minimum annual temperature averaged around 11.2 °C, with observed values ranging from 10.1 to 12.0 °C. The MK statistical test detected a statistically significant upward trend in both annual maximum and minimum temperatures. The average rates of increase were 0.034°C/year for maximum temperature and 0.020 °C/year for minimum temperature, indicating a warming trend in the area over the past 31 years (Table 9). The analysis also revealed increasing trends in seasonal minimum and maximum temperatures during the Kiremt and Belg seasons. Overall, annual minimum and maximum temperatures exhibited a highly significant increasing trend, with rates of change of 0.02 and 0.034 °C/year, respectively. Locally, Kiremt showed rates of change of 0.014 and 0.028 °C/year for minimum and maximum temperatures, while Belg exhibited a higher rate of change of 0.026 and 0.055 °C/year. The most substantial increases occurred in maximum and minimum temperatures during annual, Kiremt, and Belg periods, reflecting a strong warming trend from 1989 to 2019. These increases are much higher than the global average temperature rate of 0.0074 °C/year and align with the broader warming trend observed during this period [62]. The findings of this study corroborate previous results that indicate a faster rate of increase in temperatures reported for the Upper Blue Nile Basin [63], providing further evidence of climate change and variability. Thus, the participants’ perceptions of rising temperatures were consistent with the recorded meteorological data. IPCC [64] highlighted that increasing concentrations of greenhouse gases in the atmosphere have been the dominant cause of the observed change in temperature since 1950. These temperature changes in the study area could cause stress to water resources, a reduction in crop productivity, more floods and droughts, and an increase in vector- and waterborne diseases. Information on climate variables is important in designing and implementing area-specific adaptation and mitigation actions, such as water resource management and climate-smart choices [65]. This study provides similar evidence that smallholder farmers in the Kembata Tembaro Zone have accurately perceived the temperature changes occurring from 1989 to 2019. Similar observations were reported for other regions of Ethiopia [30, 65].
| Minimum temperature | Maximum temperature | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Scale | No | Min | Max | μ | Z | Q | No | Min | Max | μ | Z | Q |
| Annual | 31 | 10.1 | 12 | 11.2 | 2.82 ∗∗ | 0.02 | 31 | 24.9 | 27 | 26.2 | 3.20 ∗∗ | 0.034 |
| Kiremt | 31 | 11.6 | 13 | 12.8 | 2.41 ∗∗ | 0.014 | 31 | 23 | 25.4 | 24 | 2.79 ∗∗ | 0.028 |
| Belg | 31 | 10.2 | 12.9 | 11.4 | 1.87 ∗∗ | 0.026 | 31 | 25.4 | 28.5 | 27.4 | 3.16 ∗∗ | 0.055 |
- Note: No = Number, Min = Minimum, Max = Maximum, μ = Mean, Z, test statistic, Q = Sen’s slope.
- ∗, ∗∗, and ∗∗∗ indicate statistical significance at the 0.1, 0.05, and 0.01 Alpha levels, respectively.
3.4. Climate Change Adaptation Strategies
The majority of research participant farmers (88.2%) reported that they have been implementing measures to adapt to climate change, while the remaining 11.8% did not adopt any such measures (Table 10). Similar findings were reported by [66, 67] in their respective studies in the Adama District (Oromia Region) and Ambassel District (Northern Ethiopia), where ~82.9% and 93.9% of respondents, respectively, implemented various adaptation strategies to counter the adverse impacts of climate change. The main strategies adopted by farmers in this study included adjusting planting dates (68.4%), crop diversification (62.4%), soil and water conservation (57.7%), using improved livestock varieties (48.9%), engaging in off-farm activities (45.6%), and agroforestry practice (44.2%) (Table 10).
| Adaptation strategies | Respondents (%) in respective agro-ecology | |||
|---|---|---|---|---|
| Lowland | Midland | Highland | Total | |
| Adjusting planting dates | 79.8 | 67.1 | 56.9 | 68.4 |
| Crop diversification | 70.6 | 60.8 | 54.9 | 62.4 |
| Soil and water conservation | 49.6 | 58.7 | 65.7 | 57.7 |
| Using a livestock variety | 54.6 | 52.4 | 37.3 | 48.9 |
| Off-farm activity | 55.5 | 46.9 | 32.4 | 45.6 |
| Agroforestry practice | 57.1 | 42.7 | 31.4 | 44.2 |
- Note: Source: Survey result, 2022.
Adjusting planting dates emerged as the most commonly practiced adaptation strategy, practiced by 68.4% of the households. In the study area, smallholder farmers have utilized planting earlier or later than the conventional methods traditionally followed as an alternative adaptation strategy to reduce the increasing temperature and rainfall variability that they observed. The findings revealed that farmers synchronized crop growth stages with appropriate rainfall and temperature conditions, increasing the possibility of a better harvest.
In addition to adjusting planting dates, crop diversification was also popular, with 62.4% of the households practicing such an adaptation strategy, which reduces farmers’ reliance on a single crop and spreads the risk associated with climatic unpredictability. It entails producing multiple crop varieties with varying climate stress tolerances and growing cycles. Smallholder farmers expanded their farming operations by cultivating alternative crops such as maize, sorghum, finger millet (Teff), bean, cowpea, and pea. Farmers who plant a range of crops can better adapt to changing weather patterns and offer a more secure food supply. Furthermore, the use of hybrid seeds with shorter maturity times is an essential adaptive approach, as noted by Williams et al. [68]. These seeds are meant to be more resistant to adverse weather conditions and mature faster, allowing farmers to harvest their crops before extreme weather strikes. This technique is especially useful in areas where the growing season is becoming shorter or more erratic due to climate change. Furthermore, farmers in the research area have used root crops such as sweet potato and yam varieties to lessen the negative impact resulting from the temperature and rainfall changes [68]. Accordingly, it is reasonable to suggest that farmers in the Kembata Tembaro zone can increase their resilience to the effects of climate change by focusing on adaptable crops and applying appropriate agricultural practices.
In addition to crop diversification, 57.7% of smallholder farmers used soil and water conservation measures such as mulching and cover crop planting [69]. These techniques are critical for retaining soil moisture and safeguarding crops during periods of drought or unpredictable rainfall. Soil water conservation methods have long been regarded as useful strategies for minimizing the effects of climate change, especially in areas where water availability is becoming increasingly uncertain.
Additionally, 48.9% of farmers reported practicing livestock variety as an adaptation strategy to climate change (Table 10). This involves selecting breeds that are more resistant to heat stress and disease, which also require less or more adaptable feed sources as alternative adaptation strategies to climate change. Accordingly, farmers in the study area have utilized livestock varieties such as Zebu cattle (Bos indicus), which perform well under low-input systems and survive on sparse grazing and limited water, as an alternative adaptation strategy; the ability of such varieties to withstand environmental stresses ensures sustainable livestock production in the region. In addition, the farmers have adopted the Doyogena sheep breed, which is well-suited to the local environment, as an alternative adaptation strategy that helps to respond to the observed climate change in the Kembata Tembaro zone. Such a livestock management system contributes to sustained production while easing pressure on scarce resources, thereby improving smallholder farmers’ resilience to climate change.
Engaging in off-farm activities was adopted by 45.6% of the households (Table 10). This income diversification reduces dependance on agriculture alone and provides alternative livelihood options during periods of agricultural uncertainty. Smallholder farmers in the Kembata Tembaro zone have utilized petty trading of commodities, such as salt, coffee, and cereals, as alternative adaptation strategies to climate change, contributing to sustainable livelihood. This strategy helps diversify income sources and build resilience in the face of climate change, as noted by [70, 71].
Agroforestry practice was also practiced by 44.2% of the households as a mitigation and adaptation measure. Smallholder farmers in the study area have utilized agroforestry practices such as home gardens and parkland agroforestry as alternative adaptation strategies. These practices provide multiple benefits, including diverse food sources and income opportunities, shade, soil moisture conservation, windbreaks, and carbon sequestration. Additionally, they contribute to ecosystem restoration and reduce deforestation.
Most of these measures were consistently adopted across all agroecological zones in the study area. Respondents from the lowland, midland, and highland areas were compared in terms of their use of different adaptation strategies in agricultural practices (Table 8). Accordingly, the priority in the lowlands was adjusting planting dates (79.8%), followed by crop diversification (70.6%) (Table 10). Similar to lowland farmers, adjusting planting dates was also midland farmers’ priority (67.1%), followed by crop diversification (60.8%), soil and water conservation practices (58.7%), and improved livestock species (52.4%). In highland agroecology, soil and water conservation was the priority of farmers (65.7%), followed by adjusting planting dates (56.9%), crop diversification (54.9%), and using livestock variety (37.3%). These results align with those of other studies conducted in various regions of Ethiopia [28, 56, 66, 67]. However, the level of implementation of climate change adaptation techniques reported in this study’s lowlands was higher than that observed in prior Ethiopian research, except for soil and water conservation (Table 10). Lowland areas may have unique environmental characteristics that demand more immediate adaptation techniques due to their increased vulnerability to climate change impacts, such as warming temperatures or rainfall pattern changes, observed in the study area. This sense of urgency may encourage farmers to adopt alternative adaptation strategies. Conversely, previous studies have revealed that the adoption and implementation of adaptation strategies in highland and midland agroecologies are greater and more sustainable than in lowlands [70, 72]. Overall, this could be because the specific environmental and socioeconomic conditions in the Kembata Tembaro zone differ from those in other previously examined regions, making some adaptation mechanisms more applicable or effective in this setting. However, in the Kembata Tembaro zone, about 11.8% of smallholder farmers failed to implement adaptation strategies because of various factors, such as insufficient information, lack of money, shortage of labor, limited land size, and poor potential for irrigation (Table 10). This shows the need for institutional support to overcome these challenges, such as access to credit, extension contact, and climate change training. The findings align with the insights gained through FDGs and KIIs. Yet, the majority of the discussants noted that the existing adaptation measures were inadequate to fully mitigate the adverse impacts of climate change. These findings underscore the proactive actions taken by smallholder farmers in adopting a range of adaptation methods to protect the sustainability of their livelihoods in the face of climate uncertainty. However, they highlight the importance of ongoing support and education for farmers to successfully apply these measures. Providing farmers with access to innovative agricultural research, training in new practices, and resources like high-quality seeds.
3.5. Determinants of Smallholder Farmers’ Adaptation Strategies to Climate Change
The results of the MNL model showed that the age of the household heads had a positive and significant effect on the adaptation options for crop diversification and adjusting the planting dates (Table 11). A unit increase in the age of the household head results in a 0.3% increase in the probability of crop diversification and a 0.1% increase in adjusting planting dates. The increasing age of families, representing increased farmer experience, is critical in implementing adaptation techniques, allowing them to reduce the effects of climate change. This result is consistent with previous findings [22, 73]. A household head’s education level increases the probability of adapting to climate change. As shown in Table 11, a 1-year increment in the level of education positively increased the chance of employing crop diversification, adjusting planting dates, and soil and water conservation by 2.5%, 10.3%, and 1.9%, respectively. This implies that literate farmers will likely react to changes by evaluating the choices that best fit their knowledge, inclination, and capabilities. This result was consistent with [22, 24].
| PV | CD | APD | ULV | SWC | NA | ||||
|---|---|---|---|---|---|---|---|---|---|
| Coef. | dy/dx | Coef. | dy/dx | Coef. | dy/dx | Coef. | dy/dx | dy/dx | |
| sex | −0.789 | −0.08 | −0.934 | −0.045 | 0.339 | 0.097 | −0.604 | −0.018 | 0.046 |
| Age | 0.034 ∗∗∗ | 0.003 | 0.034 ∗∗ | 0.001 | 0.025 | 0 | 0.02 | −0.001 | −0.002 ∗∗ |
| FS | 0.121 | −0.011 | 0.14 | −0.002 | 0.255 | 0.012 | 0.213 | 0.015 | −0.014 |
| ES | 1.037 ∗∗∗ | 0.025 ∗ | 1.791 ∗∗∗ | 0.103 ∗∗ | 0.69 | −0.025 | 0.865 ∗∗ | 0.019 ∗ | −0.083 ∗∗∗ |
| FS | −1.814 ∗∗∗ | −0.091 | −1.922 ∗∗∗ | −0.044 | −1.534 ∗∗ | −0.004 | −1.524 ∗∗∗ | −0.003 | 0.138 ∗∗∗ |
| TLU | 0.219 ∗∗ | 0.026 ∗∗ | 0.184 ∗∗ | 0.005 ∗∗ | 0.033 | −0.012 | 0.113 | −0.007 | −0.012 |
| HI | 0.000 ∗∗∗ | 5.26E−06 ∗∗ | 0.000 ∗∗∗ | 5.71E−06 ∗∗∗ | 0.000 | 0.002 | 0.000 ∗∗ | 6.24E−07 ∗ | −0.001 ∗∗∗ |
| CI | 0.583 ∗∗ | 0.083 ∗ | 1.290 ∗∗∗ | 0.058 ∗∗ | 0.830 ∗ | 0.002 ∗∗ | 1.206 ∗∗∗ | 0.097 ∗∗ | −0.074 ∗∗∗ |
| AE | 0.45 | −0.022 | 0.758 | 0.03 | 0.757 | 0.029 | 0.549 | 0.009 | −0.046 |
| AC | 1.040 ∗∗∗ | 0.117 ∗∗∗ | 0.477 | −0.03 | 0.578 | −0.011 | 0.634 ∗∗ | 0.013 | −0.063 ∗∗ |
| AT | 1.381 ∗∗∗ | 0.184 ∗∗∗ | 0.731 | −0.018 | 0.606 | −0.023 | 0.547 | −0.069 | −0.074 ∗∗ |
| MD | −0.23 | −0.014 | −0.05 | 0.018 | −0.14 | 0.006 | −0.299 ∗ | −0.028 ∗∗∗ | 0.018 |
| PT | 0.26 | 0.104 | −0.345 | −0.038 | 0.185 | 0.026 | −0.432 | −0.096 | 0.004 |
| LL | −0.541 ∗∗∗ | −0.084 ∗∗∗ | 0.96 | 0.150 ∗∗ | 0.679 | 0.200 ∗∗ | −1.79 ∗∗∗ | −0.232 ∗∗∗ | 0.099 |
| HL | 0.035 ∗∗ | 0.048 ∗ | 0.329 | 0.023 ∗∗ | −0.325 | −0.048 ∗∗ | 0.541 ∗∗ | 0.089 ∗∗∗ | −0.016 |
| PF | 0.138 | 0.112 | 0.105 | 0.038 | −0.303 | −0.01 | −0.846 | 0.161 ∗∗∗ | 0.021 |
| _cons | −2.436 | — | −4.734 ∗∗ | — | −3.621 | — | −0.43 | — | — |
- Note: dy/dx for factor levels represents the discrete change from the base level, PV, Predictor variables, BC, No adaptation, Number of observations: 364, LR chi-square:107.4, Log-likelihood: −484.71, Pseudo-R2 : 0.393, Prob > χ2 : 0.001.
- Abbreviations: AC, access to credit; AI, access to extension; APD, adjusting planting dates; AT, access to training; CD, crop diversification; CI, climate information; ES, education status; FS, family size; FS, farm size; HI, household income; HL, highland; LL, lowland; MD, market distance; NA, no adaptation; PFC, perceived rainfall change; PT, perceived temperature; SWC, soil and water conservation; ULV, using livestock variety.
- ∗ significant at 10%, ∗∗ significant at 5%, ∗∗∗ significant at 1%.
Farm size was negatively and significantly correlated with almost all adaptation options (Table 11). A unit increase in a hectare of cultivated land would decrease the likelihood of choosing crop diversification, adjusting planting dates, using livestock variety, and soil and water conservation by 9.1%, 4.4%, 0.4%, and 0.3%, respectively. This means that a unit decrease in hectares of cultivated land increases the likelihood of choosing adaptation strategies. The finding demonstrates that the larger the farmers’ land size, the more reluctant they become to implement adaptation strategies. This could be because implementing adaptation strategies on larger land sizes requires more labor and cost. The findings of this study contrast with other studies, which have revealed a positive and significant relationship between farm size and adaptation strategy choices [22, 23].
Tropical livestock unit (TLU) positively and significantly impacted crop diversification and the adjustment of planting dates. A unit increase in TLU would increase the likelihood of using crop diversification and adjusting planting dates by 2.6% and 0.5%, respectively (Table 11). This means that farmers with many animals are more likely to use crop diversification and adjust planting dates as adaptation techniques than farmers with a few animals. This is probably because cattle are a sign of wealth in many rural communities, and farmers with more cattle are expected to have more resources and better access to adaptive information. This finding agrees with previous empirical studies [24, 74].
The model results indicate that the farm income of the respondents had a positive and significant impact on crop diversification, adjusting planting dates, and soil and water conservation practices. A unit increase in farm income increased these probabilities by less than 0.01% (Table 11). This could be because farmers with high farm incomes have a higher chance of investing more time in agricultural tasks to reduce the influence of climate change. Moreover, when the main source of income is farming, and the amount of land for farming is limited, farmers tend to invest in productivity-smoothing options (adaptation options in this case) such as soil conservation, the use of different crop varieties, and changing planting dates instead of planting trees that compete with the limited land available. These findings are consistent with earlier research [74].
Access to climatic information is a significant factor that influences adaptation strategies. Smallholder farmers who had access to climate information had a positive and significant impact on the likelihood of implementing climate change adaptation strategies, such as adjusting planting dates, soil and water conservation measures, and using livestock variety at 0.001%, 0.05%, and 0.01% levels of significance, respectively (Table 11). Being well-informed about rainfall and temperature variability increased the likelihood of crop diversification, adjusting planting dates, using livestock varieties, and soil and water conservation by 8.3%, 5.3%, 0.2%, and 9.7%, respectively (Table 9). This could be because farmers with better access to weather information made better-informed adaptation decisions. These findings are consistent with prior investigations [73, 75].
Access to affordable credit increases farmers’ financial resources and ability to incur transaction costs related to different adaptation measures [74]. Access to credit positively and significantly impacts the likelihood of choosing crop diversification and soil and water conservation at the 0.001% and 0.05% significance levels, respectively, relative to the base category (Table 11). Increasing the number of households that received credit increased their likelihood of adapting to crop diversification and soil and water conservation by 11.7% and 1.3%, respectively. This result suggests the importance of increased institutional support to promote adaptation options to reduce the negative impacts of climate change on smallholder farming communities. Access to climate change training was positive and significant (p < 0.001) in determining smallholder farmers’ choices of crop diversification compared with the base category (Table 11). The marginal values in Table 11 show that increasing the number of households that received climate change training increased the likelihood of adopting crop diversification by 18.4%. The marginal effect shows significant effects of training on crop diversification options, among other options. This result indicates the importance of increased institutional support in promoting adaptation options to reduce the negative impacts of climate change. This result supported the findings of [76].
As hypothesized, the distance to the main market center had a negative and significant (p < 0.05) effect on the likelihood of choosing soil and water conservation relative to the base category (Table 11). The computed marginal effect indicated that a unit increase in walking distance from the home of a household to the main market would decrease the likelihood of using soil and water conservation by 2.8%. This may be because farmers far from the market have limited access to agricultural technologies, undermining the potential benefits of soil and water conservation in reducing the high flood risks imposed by climate change. This suggests that farmers’ proximity to the market is an important determinant of adaptation, presumably because it serves as a means of exchanging information with other farmers. The findings of this study contrast with those of studies that revealed a positive and significant relationship between distance to the main market and the choice of adaptation strategies [77, 78].
As expected, farmers living in different agroecological settings employed different adaptation strategies. For instance, farming in the lowland zone significantly increases the probability of adjusting planting dates and using livestock variety by 15% and 20% at 5%, respectively, at the 5% significance level, compared with farming in the midland (Table 11). Conversely, farming in the lowlands significantly reduces the probability of using crop diversification and soil and water conservation by 8.4% and 23.2%, respectively, compared with farming in the midlands. Moreover, farming in the highlands significantly improved the chances of implementing crop diversification, adjusted planting dates, and soil and water conservation by 4.8%, 2.3%, and 8.9%, respectively. Living in highland agroecology increases the chances of employing adaptation strategies. The difference might be due to differences in soil, climate, other natural resources, and experiences of climate-related stress. This disparity confirms the need for research at the local level, that is, in different agroecological zones, to develop efficient and effective adaptation strategies for the agriculture sector. Similarly, Desta [24], revealed that living in highland agroecology increases the likelihood of adapting to climate change compared to others.
4. Policy Implications of the Study
To increase farmer awareness and perception of climate change, focused awareness campaigns focusing on specific climate trends and their localized implications should be implemented, with special attention paid to the unique peculiarities of different agroecological zones. Such a specific strategy will allow for more prompt and suitable adaptation responses. Although previous climate-related policies in Ethiopia were mostly focused on the national and subnational levels, this study highlights the need to add microlevel and context-specific characteristics into policymaking. Addressing the diverse views of farmers from various agroecologies is crucial for improving the efficacy and inclusivity of adaptation efforts.
Given the widespread use of adaptation strategies such as changing planting dates, crop diversification, soil and water conservation, and livestock management, policies should prioritize increasing access to agricultural resources, technical training, and extension services. This is especially crucial in sensitive areas, such as the lowlands, where farmers are more vulnerable to climate change hazards. Improved access to credit, timely climatic information, and climate change training are critical for improving farmers’ adaptive capacity. This can be accomplished by establishing dependable weather forecasting systems, improving dissemination methods (such as radio, mobile platforms, and community-based networks), and creating affordable and accessible financial products targeted to smallholder needs.
Recognizing the importance of education in determining adaptation options, policies should require investments in farmer education programs and training initiatives on climate change to encourage educated, proactive, and inventive responses. Improving rural infrastructure and market access is also crucial. Transportation, storage, and market connectivity investments will allow farmers to acquire inputs and sell their goods more effectively, promoting adaptive agricultural methods.
Smallholder farmers with limited land and resources deserve special attention, as smaller farm sizes have been shown to have a detrimental impact on adopting adaptation measures. Targeted support, such as land use planning guidance, incentives for sustainable land conservation, and community-based resource management, can assist resource-constrained farmers in increasing productivity and promoting sustainable land use.
Furthermore, creating location-specific and agroecology-responsive adaptation strategies is critical, as acceptance of adaptation measures varies greatly across environmental and social contexts. As a result, policymakers should adopt localized, bottom-up initiatives that consider farmers’ specific needs, vulnerabilities, and talents in different circumstances. Facilitating inclusive, multi-stakeholder collaboration, which brings together government agencies, research institutions, non-governmental organizations (NGOs), and local people, is critical for planning, implementing, and evaluating adaptation strategies that are contextually relevant and sustainable.
Implementing such comprehensive and microlevel-responsive policy measures can significantly improve smallholder farmers’ resilience to current and future climate variability, ensuring sustainable agricultural productivity and the long-term viability of rural livelihoods.
5. Conclusion
This microlevel analysis of climate change adaptation measures among smallholder farmers in the Kembata Tembaro zone of southern Ethiopia reveals the existence of a strong perception of climate change by farming households. The 31-year trend analysis shows an increase in annual and Kiremt (main rainy season) rainfall, while farmers report a decrease in rainfall during the Belg season, highlighting the importance of careful interpretation due to the divergent trends between meteorological data and farmer perceptions. Furthermore, the perception of increased temperature changes is substantiated by long-term meteorological data, which shows statistically significant upward trends in both annual and seasonal temperatures. Such alignment between local perceptions and scientific observations underscores the reliability of farmers’ experiential knowledge, its value in localized climate adaptation, and its importance in designing context-specific adaptation interventions. Farmers have responded to these changes by adopting a range of adaptation strategies, the most common being adjusting planting dates, crop diversification, and soil and water conservation. These strategies are shaped by agroecological context, indicating that adaptation is not uniform but tailored to the environmental realities and livelihood systems of different areas within the zone. Several determinants of adaptation decisions at the household level are also identified, such as age, education level, farm size, income, livestock holdings, access to credit, exposure to climate information, and participation in climate-related training. Moreover, spatial variables such as agroecological setting and market proximity also play critical roles, highlighting the importance of geography in shaping both exposure and adaptive capacity. Despite adopting various measures, many farmers reported that their current adaptation strategies are insufficient to address the full extent of climate-related risks, especially in light of increasing variability and intensity of climatic events. This perception points to systemic limitations in institutional support, access to resources, and the scalability of existing practices. These findings underscore the urgent need for policies that move beyond national or subnational aggregates to embrace microlevel heterogeneity. Tailored interventions that address specific agroecological and socioeconomic contexts are crucial for enhancing the resilience of vulnerable farming communities. Overall, this study provides context-specific insights as to adaptive responses of smallholder farmers, reinforcing the need for climate adaptation policies that integrate local perceptions, empirical climate data, and the diverse socioeconomic determinants influencing decision-making. In doing so, it lays a foundation for more responsive, equitable, and sustainable adaptation planning, which truly reflects the lived realities of smallholder farmers at the frontline of facing the impacts of climate change in Ethiopia.
Disclosure
This manuscript is part of a Ph.D. thesis on “Microlevel analysis of climate change adaptation and its determinants: The case of smallholder farmers in the Kembata Tembaro zone, southern Ethiopia.”
Conflicts of Interest
The authors declare no conflicts of interest.
Author Contributions
Getachew Tadesse: conceptualization and design, data collection, statistical analysis, interpretation of results, drafting of the manuscript. Mulugeta Lemenih: conceptualization and design, supervision, editing, review. Teshale Woldeamanuel: conceptualization and design, supervision, editing, review. Menfese Tadesse: conceptualization and design, supervision, editing, review.
Funding
No funding was received for this manuscript.
Acknowledgments
The authors would like to sincerely thank the Bule Hora University and Wondo Genet College of Forestry and Natural Resources for giving Getachew Tadesse the support for data collection to complete this research for his Ph.D study. The authors would like to humbly appreciate the farmers for providing enough information during data collection. The author is also thankful to Fikre Tinsae, Abayneh Legesse, Mengistu Teshome (Ph.D.), Mekuria Guye (PhD), Amsalu Abich (PhD), and Tesfaye Dejene for their valuable comments. The author has confirmed the formal permission from the individuals mentioned in the acknowledgement part to be acknowledged.
Endnotes
Open Research
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.
- 1Keble is the smallest administrative division in Ethiopia.
