Smallholder farmers' adaptation strategies among coffee farming household heads in the Kafa biosphere reserve, Ethiopia.

Smallholder farmers’ adaptation strategies among coffee farming household heads in the Kafa biosphere reserve, Ethiopia - PMC Skip to main content An official website of the United States government Here's how you know Here's how you know Search Log in Primary site navigation Search Log in Search PMC Full-Text Archive Search in PMC Journal List User Guide As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsement of, or agreement with, the contents by NLM or the National Institutes of Health. Learn more: PMC Disclaimer | PMC Copyright Notice Sci Rep . 2026 Apr 13;16:17244. doi: 10.1038/s41598-026-46526-6 Smallholder farmers’ adaptation strategies among coffee farming household heads in the Kafa biosphere reserve, Ethiopia Amare Amsalu 1, 6, ✉ , Weyessa Garedew 2 , Girma Alemu Melka 3 , Mehmet Ali Çelik 4, 5 , Dessalegn Obsi Gemeda 1 Author information Article notes Copyright and License information 1 Department of Natural Resource Management, College of Agriculture and Veterinary Medicine, Jimma University, Jimma, Ethiopia 2 Department of Horticulture and Plant Sciences, College of Agriculture and Veterinary Medicine, Jimma University, Jimma, Ethiopia 3 Department of Geography and Environmental Studies, College of Social Sciences and Humanities, Jimma University, Jimma, Ethiopia 4 Department of Geography, Faculty of Arts and Sciences, Igdir University, Iğdır, Turkey 5 Department of Geography, Nakhchivan State University, Nakhchivan, 7012 AZ Azerbaijan 6 Department of Natural Resource Management, College of Agriculture and Natural Resource, Bonga University, Bonga, Ethiopia ✉ Corresponding author. Received 2025 Dec 28; Accepted 2026 Mar 26; Collection date 2026. © The Author(s) 2026 Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/ . PMC Copyright notice PMCID: PMC13234277  PMID: 41974853 Abstract Climate change poses a major threat to smallholder agriculture in ecologically sensitive regions like Ethiopia. This study assess climate trends and identifies the determinant that influence adoption of adaptation strategies among coffee-farming households in the Kafa Biosphere Reserve, Ethiopia. Data was gathered through household surveys, focus group discussions, and key informant interviews. Climatic trends were analyzed using the Mann-Kendall test and Sen’s slope estimator, while a Multivariate Probit Model were used to evaluate factors driving four adaptation strategies: agronomic practices, soil and water conservation (SWC), livelihood diversification, and agroforestry. Results show a significant warming, especially in minimum and mean temperatures, with erratic rainfall. Agronomic practices were the most adopted strategy (23.1%), while 29.9% of households undertook no adaptation. The econometric analysis reveals that access to institutional resources, particularly access to credit, extension services, and early warning systems, are the key determinant factors that enable adoption. Conversely, forest dependency significantly reduces the likelihood of adopting nearly all strategies. Female-headed households are more likely to diversify livelihoods but less likely to adopt labor-intensive SWC. Farm size and agroecological zone (AEZ) also shape strategy choice. Moreover, significant complementarities exist between SWC and agroforestry. The study concludes that building climate resilience requires moving beyond the dissemination of technologies to actively strengthen institutional support, design gender-responsive programs, break forest-dependency traps through integrated livelihood approaches, and promote synergistic practice packages. These targeted interventions are essential for enhancing equitable adaptive capacity in smallholder systems. Supplementary Information The online version contains supplementary material available at 10.1038/s41598-026-46526-6. Keywords: Agroforestry, Climate change adaptation, Institutional access, Multivariate probit model, Smallholder farmers Subject terms: Climate sciences, Ecology, Ecology, Environmental sciences, Environmental social sciences Introduction Global agriculture faces unprecedented risks from climate change and variability 1 – 3 manifesting as rising temperatures, erratic rainfall, and heightened frequency of extreme weather events 4 , 5 . Developing regions, with their inherent socio-economic vulnerabilities and heavy reliance on rain-fed systems, are disproportionately affected 6 – 8 . In response, farmers employ a variety of appropriate adaptation strategies 9 , 10 . However, the effectiveness of these strategies is frequently challenged by weak institutional support, financial limitations 11 , and entrenched gender disparities that curtail access to resources and information 12 . The agricultural sector in Africa is particularly vulnerable to climate risks 13 , 14 , due to its heavy reliance on rain-fed cultivation and limited adaptive capacity 15 . In this regard, agroforestry has gained prominence as a critical climate-smart agriculture (CSA) practice 16 , 17 . Beyond enhancing farm productivity and income through diverse products (fruits, timber, non-wood forest products), agroforestry systems deliver vital ecosystem services 18 , such as improving microclimates, enhance soil fertility and water retention, and contribute to carbon sequestration 19 , 20 . Studies from South Sudan and Northern Ethiopia show that agroforestry significantly strengthens household resilience by diversifying livelihoods, improving adaptive capacity, and reducing sensitivity to climate shocks 19 , 21 . Similarly, agronomic practices, soil and water conservation (SWC), and livelihood diversification are also essential in climate change adaptation strategies for stabilizing yields, improving resource resilience, and reducing dependence on climate-sensitive livelihoods 22 – 24 . Nevertheless, the adoption of such sustainable practices is not immediate. It is shaped by a complex interplay of socio-economic, institutional, and agroecological factors 25 , which includes access to credit and extension services, land tenure security, market proximity, and farmers’ perception of climate risk. The challenge of adaptation is further compounded by households with high dependence on natural forests 26 . Research from Northeast India and Sudan illustrates that while forests provide a crucial safety net and substantial household income, heavy reliance can also signal a vulnerability trap 27 . This dependency may undermine long-term agricultural adaptation investments, as immediate subsistence needs take precedence over farm improvement, which potentially locks households in a cycle of resource depletion and diminished resilience 28 , 29 . This implies the necessity for integrated approaches that link sustainable forest management with the development of alternative, climate-resilient livelihood options. Ethiopia’s economy and food security are tied to a climate-sensitive agricultural sector dominated by smallholders who face recurrent droughts and increasing rainfall variability 30 . Despite national policies like the Climate Resilient Green Economy (CRGE) strategy, implementation gaps persist, particularly in remote and ecologically sensitive regions 31 . In this regard, Kafa Biosphere Reserve in the highland of southwestern Ethiopia is uniquely important. This region is part of Eastern Afromontane Biodiversity Hotspot and home to wild Coffea arabica 32 , which supports communities whose livelihoods are intricately linked to coffee farming and forest resources 33 . While the area possesses rich indigenous knowledge and potential for climate-resilient practices 34 , the specific factors driving the adoption of different adaptation strategies among its coffee-farming households remain inadequately studied. Moreover, although several research in Ethiopia highlights the role of socio-economic and institutional factors for climate change adaptation 35 , 36 , how they shape interconnected adaptation strategies such as agronomic adjustments, SWC, diversification, and agroforestry remains unclear. Especially, the role of pervasive issues such as forest dependency and gender adaptation pathways within this specific socio-ecological system requires focused investigation. Therefore, this study aims to bridge these gaps by providing empirical evidence from the Kafa Biosphere Reserve. Specifically, the research addresses the following questions: (1) What are the observed trends in local temperature and rainfall, (2) What are the predominant climate adaptation strategies adopted by smallholder coffee farmers, (3) How do socio-economic, institutional, and agroecological factors determine the adoption of these different strategies, and (4) How are the decisions to adopt these strategies interrelated?. This study used a multivariate probit model to analyze the interrelationships among strategies instead of considering them in isolation. The findings are intended to inform the design of targeted, integrated, and gender-sensitive policies that enhance institutional support, break resource dependency traps, and ultimately strengthen the climate resilience of smallholder farming systems within and beyond this vital biosphere reserve. Methodology Study area This research was conducted in the Kafa Biosphere Reserve, located in the southwestern regional state of Ethiopia. Geographically, the area is located between longitudes 35°25’0"E and 36°45’0"E and latitudes 6°40’0"N and 8°0’0"N (Fig. 1 ). The landscape is topographically diverse, with elevations extending from 1,072 to 3,305 m above sea level. The mean annual temperature of the area is 19.45 °C and receives about 1,800 mm of annual rainfall 37 . Fig. 1. Open in a new tab Map of the study area. The map was generated using ArcMap version 10.3.1(Environmental Systems Research Institute [ESRI], https://www.esri.com ). Map source: Ethiopian Geospatial Information Institute (EGII) Ecologically, the region is part of the Eastern Afromontane Biodiversity Hotspot and harbors Afromontane evergreen forests, natural coffee forests, grasslands, and wetlands. It serves as a critical genetic reservoir for wild Coffea arabica 38 . The local economy is predominantly agrarian, with over 90% of households engaged in subsistence farming. Primary livelihood activities include coffee cultivation (under forest canopy and in garden plots), staple crops (maize, teff, legumes, Enset), and livestock rearing. Communities also depend on forest resources for non-timber forest products (NTFPs) such as honey, fuelwood, medicinal plants, fruits, and construction materials 39 , 40 . Research design and sampling procedure A mixed-methods research design was employed, integrating quantitative household surveys with qualitative data from focus group discussions (FGDs) and key informant interviews (KIIs). The study population comprised coffee-farming households within the Kafa Biosphere Reserve. A multi-stage sampling technique was used 41 . First, the biosphere reserve was stratified into three AEZs: highland ( Dega) , midland ( Weyina Dega) , and lowland ( Kolla) based on the climate classification by 42 (Fig. 2 ). Subsequently, two kebeles (the smallest administrative unit) were purposively selected from each AEZ to ensure representation across the elevation and climate gradient, resulting in a total of six kebeles . Fig. 2. Open in a new tab Map of agroecological zones. The map was generated using ArcMap version 10.3.1(Environmental Systems Research Institute [ESRI], https://www.esri.com ). Map source: Ethiopian Geospatial Information Institute (EGII) The sample size was determined using 43 formula for a finite population (Eq. 1), with a 95% confidence level and a 5% margin of error. 1 where n is sample size, N is total household size (84,000), and e is the margin of error (0.05). This yielded a minimum sample size of 398 households. The sample was proportionally allocated across AEZs (Eq. 2) and then distributed among selected kebeles (Table 1 ). 2 Table 1. Distribution of sample households across AEZ s Agroecological Zone Total households (N_i) Sample households (n_i) Highland 15,000 71 Midland 45,000 214 Lowland 24,000 113 Total 84,000 398 Open in a new tab where n is determined sample size and ni is the sample size taken from group i , and Ni is population size of group i and N_total is the total population size of all groups combined. Finally, within each selected kebele , respondent households were chosen using a simple random sampling technique from lists provided by local agricultural offices. Data sources and collection Primary data were collected between January and March of 2025. A semi-structured questionnaire was pre-tested and refined in a similar community outside the sample area with 30 respondents. It was administered to the 398 household heads to capture socio-economic characteristics, resource endowments, institutional access, and detailed adoption of climate adaptation strategies. Qualitative data were gathered to triangulate and enrich the survey findings. This included two FGDs with seven participants each and four KIIs with agricultural experts and development agents from each AEZs. Secondary climate data were obtained from the Ethiopian Meteorological Institute (EMI). This included historical rainfall and temperature records (1985-2023) from 12 synoptic stations distributed across the three AEZs, complemented by gridded datasets for the same period to fill gaps and validate spatial patterns. Data analysis Climate trend analysis Station and gridded data were integrated for robustness. The non-parametric Mann-Kendall test 44 , 45 and Sen’s slope estimator 46 were used to detect and quantify trends in minimum (Tmin), maximum (Tmax), and mean (Tmean) temperatures, as well as annual rainfall. The Inverse Distance Weighted (IDW) method in Python 3.12.4 was used for spatial interpolation. Model accuracy was validated using Leave-One-Out Cross-Validation, yielding low Root Mean Square Error (RMSE) and Mean Absolute Error (MAE) values. Mann-Kendall test The non-parametric Mann-Kendall test statistics ‘S’ 44,45 was employed to investigate rainfall and temperature trend using Eq. 3 3 4 5 where m = number of tied groups, t i​ = number of ties in the i th group, n is observations. For n > 10, the standardized test statistic Zs​ is computed as: 6 A positive Zs shows an increasing trend while negative Zs indicates a decreasing trend and 0 no trend. If Zs > 1.96 (at α = 0.05), the trend is statistically significantly increasing. If Zs < -1.96, the trend is statistically significantly decreasing. Sen’s slope estimator The magnitude of trend (slope Q) was estimated using 46 as indicated in Eq. 7. 7 8 The Sen’s slope is the median of all Qi​. Econometric analysis: multivariate probit model To capture potentially correlated adoption of multiple adaptation strategies, a Multivariate Probit (MVP) model was used. This model accounts for correlations in the error terms across equations arising from unobserved household characteristics that jointly influence adaptation decisions. The MVP model consists of a system of m equations (where m = 4, corresponding to the four adaptation strategies): 9 where is the latent propensity of household i to adopt adaptation strategy j ; is a vector of observed explanatory variables; is a vector of parameters to be estimated; and is the error term. The observed outcome is defined as if , and otherwise. The error terms are assumed to follow a multivariate normal distribution with a covariance matrix containing off-diagonal correlation coefficients . A statistically significant indicates that the adoption decisions for strategies j and k are interrelated. The model was estimated to use simulated maximum likelihood techniques in R (version 4.4.2). Variable definitions The dependent variables were four binary indicators for the adoption (1) or non-adoption (= 0) of: (1) agronomic practices (improved seeds, adjusted planting dates, fertilizer use); (2) SWC: terracing, bunds, water harvesting; (3) livelihood diversification (off/non-farm income activities); and (4) agroforestry (trees integrated with crops/livestock). Independent variables were selected based on literature and categorized as demographic, socio-economic, institutional, and agroecological. Their definitions and measurements are summarized in Table 2 . Table 2. Definitions and measurements of variables in the empirical model Variable Definition Measurement/Unit Agroecology AEZ classification Midland = 1 if midland, 0 otherwise; Highland = 1 if highland, 0 otherwise; Lowland = reference Age Age of household head Years Gender Gender of household head Dummy (1 = Female) Education Years of formal schooling of head Years Family size Total number of household members Number Farm size Total cultivated land area Hectares (ha) Distance to farm Distance from home to farm plot Kilometers (km), log-transformed Distance to market Distance from home to nearest market Kilometers (km), log-transformed Livestock ownership Livestock holdings Tropical livestock units (TLU) Extension access Access to agricultural extension services Dummy (1 = Yes) Credit access Access to institutional credit Dummy (1 = Yes) Early warning access Access to climate early warning systems Dummy (1 = Yes) Forest dependency Reliance on forest resources for income/consumption Index (1 = Very low to 5 = Very high) Coffee yield Average coffee production per unit area Kilograms per hectare (kg/ha) Open in a new tab Ethical considerations The study protocol was reviewed and approved by the Institutional Review Board (IRB) of Jimma University. All methods were performed in accordance with relevant guidelines and regulations. Informed consent was obtained from all participants prior to data collection. For illiterate participants, the information sheet and consent form were read aloud in the local language, and verbal consent was obtained in the presence of a witness and appropriately documented. Participation was voluntary, and confidentiality of respondents was strictly maintained. Results Adoption of climate adaptation strategies by smallholder farmers Approximately 29.9% of surveyed households reported undertaking no adaptive measures, indicating a significant adaptation deficit. Among adopters, agronomic practices were the most prevalent strategy (23.1%), followed by agroforestry (19.8%), SWC (17.1%), and livelihood diversification (10.1%) (Fig. 3 ). Fig. 3. Open in a new tab Percentage distribution of adaptation strategies adopted by smallholder coffee farming household heads Local climate trends The analysis of historical climate data revealed a clear warming patterns across the Kafa Biosphere Reserve (Table 3 ). The Mann-Kendall test showed statistically significant increasing trends in minimum (Tmin), maximum (Tmax), and mean (Tmean) temperatures. In contrast, rainfall trends exhibited high spatial variability. The midland ( Weyina Dega ) showed slight but non-significant increasing trends, highland ( Dega ) zones significantly increasing trend, while the lowland ( Kolla ) zone experienced a significant declining trend in annual rainfall (SM Figs. 1 -5). Table 3. Trends in temperature and rainfall across AEZ s (1985-2023) Agroecological Zone (AEZ) Elevation range Tmin trend (°C/yr) Tmax trend (°C/yr) Tmean trend (°C/yr) Rainfall trend (mm/yr) Midland ( Weyina Dega ) 1,500 -2,300 m 0.021* 0.018* 0.020* + 1.84 Highland ( Dega ) 2,300 -3,350 m 0.024* 0.022* 0.023* + 3.12* Lowland ( Kolla ) 500 -1,500 m 0.019* 0.015* 0.017* -4.37* Open in a new tab Note: Sen’s slope estimates shown. Statistical significance at p < 0.05 determined by two-tailed Mann-Kendall test where |Zs| > 1.96. Determinants of adaptation strategies: multivariate probit estimates The MVP model was jointly significant (Wald χ² (60) = 186.30, p < 0.001), indicating that the explanatory variables collectively influence adaptation decisions. The coefficient estimates reveal distinct drivers for each strategy (Table 4 ). Table 4. Multivariate Probit model estimates of factors influencing the adoption of climate adaptation strategies Explanatory variable Agronomic practices SWC Livelihood diversification Agroforestry Agroecology (Ref: Lowland) Midland 0.401* (0.212) 0.574** (0.234) 0.492* (0.256) 0.283 (0.231) Highland 0.215 (0.245) 0.387* (0.221) 0.198 (0.239) 0.512** (0.248) Age of household head 0.026* (0.014) 0.002 (0.013) 0.015 (0.015) 0.022 (0.016) Gender (1 = Female) 0.839 (0.654) -0.813* (0.467) 1.503* (0.812) 0.604 (0.574) Education (years) 0.251*** (0.068) 0.123* (0.064) 0.219*** (0.072) 0.052 (0.063) Family size -0.336** (0.148) -0.006 (0.142) 0.041 (0.159) 0.099 (0.137) Log distance to farm -0.658*** (0.211) -0.433* (0.225) -0.413* (0.234) 0.111 (0.218) Log distance to market -0.215 (0.241) 0.143 (0.258) 0.064 (0.269) 0.300 (0.251) Farm size (ha) 0.525* (0.274) 0.873*** (0.236) 0.291 (0.288) 0.407* (0.229) Livestock (TLU) 0.050 (0.095) -0.101 (0.107) 0.242** (0.112) -0.040 (0.098) Extension access (1 = Yes) 1.307*** (0.342) 1.696*** (0.365) 0.808** (0.391) 0.830** (0.337) Credit access (1 = Yes) 2.051*** (0.485) 1.347*** (0.402) 1.041** (0.463) 1.234*** (0.388) Early warning access (1 = Yes) 1.807*** (0.396) 1.801*** (0.413) 1.183** (0.451) 0.547 (0.419) Forest dependency index -0.421** (0.187) -0.302* (0.174) -0.510*** (0.193) -0.185 (0.181) Coffee yield (kg/ha) 0.003* (0.001) 0.002 (0.001) 0.001 (0.001) 0.003** (0.001) Constant -3.978*** (1.356) -3.499*** (1.289) -5.826*** (1.423) -4.566*** (1.302) Model diagnostics Number of observations 398 Wald χ² (60) 186.30*** Log Pseudolikelihood -543.42 AIC / BIC 1206.84/1446.02 McFadden’s Pseudo-R² 0.238 Open in a new tab Note: Coefficients are reported with robust standard errors in parentheses. Significance levels: *** p < 0.01, ** p < 0.05, * p < 0.1. Institutional and financial factors Institutional and financial factors were the most powerful enablers. Access to credit significantly increased the probability of adopting all four strategies, with the strongest marginal effect on agronomic practices. Similarly, access to extension services had a large, positive impact, particularly on SWC and agronomic practices. Early warning system access significantly promoted the adoption of agronomic practices and SWC. Socio-economic and demographic factors Socio-economic and demographic factors also played crucial roles. Education level increased the likelihood of adopting agronomic practices and livelihood diversification. Farm size was a positive and significant determinant for SWC and agroforestry. Conversely, greater distance from the homestead to the farm plot significantly reduced the adoption of management-intensive practices like agronomy and SWC. Gender-specific adaptation pathways Female-headed households were significantly less likely to adopt labor-intensive SWC but were more likely to pursue livelihood diversification. Agroecological setting Compared to the lowland reference category, midland farmers were more likely to adopt SWC and agronomic practices, while highland farmers showed a stronger preference for agroforestry. Forest dependency Forest dependency consistently exerted a negative influence, significantly reducing the adoption likelihood of agronomic practices, SWC, and livelihood diversification. Interdependence between adaptation strategies The correlation matrix of error terms from the MVP model (Table 5 ) reveals significant interdependence between adoption decisions. A strong positive correlation (ρ = 0.336, p < 0.01) was found between SWC and agroforestry. A positive correlation was also observed between agronomic practices and agroforestry (ρ = 0.198, p < 0.05). In contrast, livelihood diversification showed no significant correlation with farm-based strategies. Table 5. Correlation matrix (ρ) of error terms from the Multivariate Probit model Strategy Agronomic practice SWC Livelihood diversification Agroforestry Agronomic practices 1.000 SWC 0.284*** (0.078) 1.000 Livelihood diversification -0.132 (0.091) -0.087 (0.085) 1.000 Agroforestry 0.198** (0.082) 0.336*** (0.076) 0.054 (0.089) 1.000 Open in a new tab Note: Standard errors in parentheses. *** p < 0.01, ** p < 0.05. Marginal effects and predicted probabilities The Average Marginal Effects (AMEs) indicate the change in adoption probability (Table 6 ). Access to credit increased the probability of adopting agronomic practices by 50.1% points, while extension access increased SWC adoption probability by 41.4 points. Each additional year of schooling raised the probability of adopting agronomic practices by 6.1 points. Female-headed households were 36.7 points more likely to diversify livelihoods but 19.8 points less likely to adopt SWC. Table 6. Average Marginal Effects (AMEs) of significant determinants on adoption probabilities Variable Agronomic practices SWC Livelihood diversification Agroforestry Agroecology (vs. Lowland) Midland 0.098** (0.052) 0.140*** (0.057) 0.120** (0.062) 0.069 (0.056) Highland 0.052 (0.060) 0.095* (0.054) 0.048 (0.058) 0.125** (0.061) Age (per year) 0.006* (0.003) 0.000 (0.003) 0.004 (0.004) 0.005 (0.004) Gender (Female = 1) 0.205 (0.160) -0.198* (0.114) 0.367* (0.198) 0.147 (0.140) Education (per year) 0.061*** (0.017) 0.030* (0.016) 0.053*** (0.018) 0.013 (0.015) Family size (per member) -0.082** (0.036) -0.001 (0.035) 0.010 (0.039) 0.024 (0.033) Distance to farm (log km) -0.161*** (0.051) -0.106* (0.055) -0.101* (0.057) 0.027 (0.053) Farm size (per ha) 0.128* (0.067) 0.213*** (0.058) 0.071 (0.070) 0.099* (0.056) Livestock (per TLU) 0.012 (0.023) -0.025 (0.026) 0.059** (0.027) -0.010 (0.024) Extension access 0.319*** (0.083) 0.414*** (0.089) 0.197** (0.095) 0.202** (0.082) Credit access 0.501*** (0.118) 0.329*** (0.098) 0.254** (0.113) 0.301*** (0.095) Early warning access 0.441*** (0.097) 0.439*** (0.101) 0.289** (0.110) 0.133 (0.102) Forest dependency (per unit) -0.103** (0.046) -0.074* (0.042) -0.124*** (0.047) -0.045 (0.044) Coffee yield (per 100 kg/ha) 0.073* (0.024) 0.049 (0.024) 0.024 (0.024) 0.073** (0.024) Open in a new tab Note: AMEs represent the change in predicted probability. Standard errors in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.1. Predicted probabilities for different farmer profiles further illustrate these disparities (Fig. 4 ). Profiles with combined advantages (e.g., educated, male-headed with credit access) showed dramatically higher probabilities, especially for SWC (61.8%) and agronomic practices (52.4%). In contrast, vulnerable profiles (e.g., female-headed, forest-dependent, no credit) had the lowest probabilities. The highest adoption rates were predicted for resource-endowed farmers with early-warning information. Fig. 4. Open in a new tab Predicted probabilities of adopting adaptation strategies across different farmer profiles in the Kafa Biosphere Reserve, Ethiopia Discussions This study provides empirical evidence on climate trends and the determinants of adaptation strategies among smallholder coffee farmers in the Kafa Biosphere Reserve. The findings reveal a clear warming trend along with erratic rainfall, creating a pressing need for adaptive responses. However, the capacity to respond is shaped by a complex interplay of institutional, socio-economic, and agroecological factors. The following sections interpret these findings considering existing literature. Climatic trends and the imperative for adaptation Our analysis shows a significant warming trend across all three AEZs, with nighttime (Tmin) and average (Tmean) temperatures rising most sharply. This pattern of asymmetric warming aligns with regional and global observations, often linked to increased atmospheric moisture and changes in cloud cover that inhibit nighttime cooling 47 . Concurrently, the high spatial variability in rainfall trends, particularly the significant decline observed in the lowland ( Kolla ) zone, resonates with broader findings from East Africa that climate change manifests as an intensification of hydroclimatic variability rather than uniform shifts 22 , 49 . A recent study in the Ethiopian highlands also reported similar patterns of increasing temperatures and erratic rainfall 50 , underscoring the regional consistency of these climate pressures. This evolving climate regime, characterized by hotter nights and erratic rainfall poses a direct and severe threat to the stability of rain-fed coffee systems and subsistence agriculture in Kafa 51 , suggesting a strong stimulus for adaptation. Determinants of adaptation strategy adoption The central role of institutional and financial access The most powerful drivers of adaptation identified in this study are institutional. Access to credit increased the probability of adopting agronomic practices by over 50% points, highlighting that financial constraints are a primary barrier to investing in improved inputs, irrigation, or tree seedlings 52 . Similarly, the robust impact of extension services, especially for SWC, underscores the critical role of knowledge transfer and technical guidance in reducing uncertainty and enabling the adoption of complex practices 53 . The significant positive effect of early warning systems on in-season decisions (agronomic practices and SWC) provides empirical support for the growing policy emphasis on climate information services (CIS) 54 . It demonstrates that actionable, timely forecasts are a crucial catalyst for proactive measures, such as adjusting planting dates or implementing moisture conservation techniques ahead of predicted dry spells 55 . These findings collectively show that strengthening institutional pillars is paramount for enabling adaptation. Human capital and demographic factors The findings show the complex influence of demographic characteristics. Education emerged as a key enabler, significantly increasing the adoption of agronomic practices and livelihood diversification 56 , 57 . This aligns with the established understanding that formal schooling enhances cognitive ability, information processing, and openness to innovation, thereby facilitating the adoption of new technologies and strategies 58 . Contrary to some studies, we found that age slightly increased adoption of agronomic practices, likely reflects the value of experiential knowledge. The findings on gender reveal distinct adaptation pathways shaped by gender access to resources and labor roles 59 . Female-headed households were less likely to adopt labor-intensive SWC due to women’s dual burden of domestic and farm workload 60 . Conversely, their higher propensity for livelihood diversification underscores a strategic risk-spreading mechanism, often pursued out of necessity due to limited access to land and agricultural capital 61 . Family size showed a negative association with agronomic practices, possibly because larger households, while potentially providing labor, may also face higher immediate consumption needs, directing resources away from agricultural investments and toward off-farm labor for short-term income 62 . Resource endowments and biophysical constraints Fundamental resource endowments critically shape adaptive capacity 63 . Farm size was a strong positive determinant for land-intensive, long-term strategies like SWC and agroforestry, as larger landholdings provide spatial and economic flexibility 64 to dedicate parcels to conservation without compromising food security. In contrast, the negative influence of distance to farm on management-intensive practices (agronomy and SWC) highlights the tyranny of distance 65 . Remote plots incur higher transaction costs for monitoring, labor, and input application, discouraging investments that require regular attention. Livestock ownership (a wealth indicator) positively influenced livelihood diversification, providing capital to venture into off-farm activities 66 , but did not significantly influence on-farm practices, suggesting its role as a financial buffer rather than a direct input for farm-based adaptation. Agroecological context and the forest dependency paradox Adaptation preferences varied significantly by AEZ . Midland farmers showed a stronger affinity for SWC and agronomic practices, likely due to more moderate slopes and reliable moisture favoring such intensification. Highland farmers exhibited a greater preference for agroforestry 19 , a logical strategy for mitigating soil erosion on steeper slopes while diversifying income pattern observed in other mountainous regions 67 . This underscores the necessity for decentralized, location-specific policy design rather than blanket national programs. A critical and concerning finding is the consistent negative impact of forest dependency on the adoption of almost all agricultural adaptation strategies. This suggests that reliance on forest resources for immediate needs limits investment in long-term farm resilience. While forests provide a vital safety net and significant household income, as evidenced in Sudan, with NWFPs contributing over 50% to household economies 48 , 27 , heavy dependence can also indicate a lack of viable alternatives. This creates a cycle where environmental degradation may worsen, further undermining agricultural potential. Therefore, building climate resilience in this context requires integrated landscape approaches that combine sustainable forest management with the creation of alternative, sustainable livelihoods to break this dependency trap. Interdependencies among adaptation strategies The significant positive correlation between SWC and agroforestry adoption indicates that farmers perceive and utilize these strategies as complementary packages. This synergy is ecologically sound, as trees can enhance the biological stability of physical structures like terraces 68 . The positive link between agronomic practices and agroforestry further suggests that farmers seeking to boost productivity are also attentive to the long-term ecological health of their farms 69 . In contrast, the lack of correlation between livelihood diversification and on-farm strategies indicates it operates as an independent risk-management decision 70 , likely driven by distinct factors like local labor market opportunities or asset poverty 71 . These interdependencies indicate that adaptation involves a combination of interrelated choices, implying that promoting bundled interventions could be more efficient and effective than promoting isolated practices. Policy and practical implications The findings in the study area give clear and actionable guidance for enhancing climate resilience for smallholder coffee farmers in the study area and other similar agroecology. Beyond mere awareness, effective action requires addressing structural and institutional barriers such as: Strengthen institutional enablers as a foundational priority. Financial and information barriers are the most significant constraints to adaptation. Policy must prioritize: Scaling inclusive financial products, such as low-interest microloans or input-credit schemes synchronized with coffee harvest cycles, to overcome the capital constraints inhibiting investment. Investing in decentralized, farmer-centric extension systems that provide hands-on training in integrated climate-smart practices, particularly SWC and agroforestry. Expanding access to localized climate information services (CIS), disseminating tailored agro-meteorological advisories via mobile platforms or community radio to support critical in-season decisions. Design for gender equity in adaptation. The pronounced gender disparities in strategy choice necessitate targeted programming: Promote labor-saving SWC technologies (e.g., drip irrigation, biological stabilizers) and facilitate women’s access to group-based labor-sharing arrangements to reduce the burden of physically intensive practices. Actively support women’s propensity for livelihood diversification through skills training, micro-enterprise development, and enhanced market linkages for off-farm activities, thereby validating and strengthening an existing risk-coping strategy. Integrate conservation with livelihood security to break dependency traps. The negative link between forest dependency and agricultural adaptation signals a need for integrated landscape approaches. Initiatives should combine sustainable forest management with the creation of alternative income sources, such as developing value chains for certified NTFPs like honey, medicinal plants. Promote on-farm agroforestry schemes that provide tree-based products, thereby reducing extraction pressure on natural forests while enhancing farm productivity and resilience. Promote synergistic practice packages. The strong complementarity between strategies like SWC and agroforestry indicates that bundled interventions are more efficient and effective. Extension programs and subsidy schemes should be designed to support integrated practice bundles (e.g., providing tree seedlings alongside training on terrace construction) to leverage natural synergies for greater impact. Locally tailored adaptation strategies. The significant variation in adaptation preferences across AEZ s ( Kolla , Weyina Dega , Dega ) renders blanket national programs inefficient. Local agricultural offices and development agents must be empowered to tailor recommendations, incentives, and support based on zone-specific biophysical constraints and socio-economic opportunities. Limitations of the study While this study provides valuable insights, its findings should be interpreted considering certain constraints. The cross-sectional design captures a snapshot of adaptation behavior, limiting causal inference and the ability to track changes over time. Data on yields, practices, and perceptions rely on farmer recall and attitudes, which is very subjective to misinformation. Although the multivariate probit model accounts for key household-level factors, the analysis may not fully capture the influence of broader institutional and market conditions (e.g., detailed land tenure, governance quality, market volatility) or important behavioral factors like risk perception and social norms. The site-specific socio-ecological context of the Kafa Biosphere Reserve, while a strength for depth, also limits the direct generalizability of findings to other farming systems. These limitations underscore that adaptation is a complex process shaped by a wide array of interacting factors, meriting further longitudinal and contextually detailed research. Conclusions This study examined climate trends and the determinants of adaptation strategies among smallholder coffee farmers in Kafa Biosphere Reserve, Ethiopia. The analysis indicates a significant local warming trend, with declining rainfall in the lowlands AEZ, which creates a clear imperative for adaptive action. However, the translation of climate risk into resilient practice is neither immediate nor uniform. The multivariate probit analysis identifies access to institutional resources: such as credit, extension services, and climate information, as the most powerful catalyst for adoption, significantly increasing the probability of implementing key strategies. Conversely, the absence of these enablers, compounded by resource constraints like small landholdings or remote farms deepens vulnerability. This research reveals critical disparities in adaptive capacity. Female-headed households leaning toward livelihood diversification over labor-intensive conservation. Furthermore, high forest dependency consistently reduces the adoption of agricultural adaptation strategies, suggesting a poverty trap that undermines long-term resilience. Importantly, farmers’ decisions are interdependent; strong complementarities between practices like SWC and agroforestry indicate they are perceived and used as synergistic packages. In general, while the climate threat is clear in the study area, the capacity to respond is unevenly distributed. Therefore, sustainable adaptation requires a fundamental shift from merely disseminating technologies to actively building the institutional, financial, and social infrastructure that enables adaptation. Finally, we conclude that resilience depends not only on recognizing climate risks but on possessing the tangible capacity and support to address them. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1 (1.4MB, docx) Acknowledgements The authors express their gratitude for the financial assistance received for this research by the Ethiopian Ministry of Education (MoE), Bonga University (BU), and Jimma University College of Agriculture and Veterinary Medicine. We also extend our sincere thanks to the EMI for supplying the climatic data essential for this study. We also like to acknowledge Ethiopian Geospatial Information Agency for proving Ethio shapefile. Author contributions Amare Amsalu: Led the conceptualization and design of the study. Undertook data collection, processing, and analysis. Interpreted the findings and prepared the initial draft of the manuscript. Provided research materials and resources.Dessalegn Obsi Gemeda, Weyessa Garedew, Girma Alemu Melka, and Mehmet Ali Çelik: Contributed to the study’s design and methodology, assisted data analysis and interpretation, supervision, reviewing and editing. Funding This research was funded for a PhD grant awarded to the first author. Financial support was provided by the Ethiopian Ministry of Education (MoE), Jimma University College of Agriculture and Veterinary Medicine, and Bonga University (BU). Data availability The data supporting the findings of this study are available from the corresponding author upon reasonable request. Declarations Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References 1. Khatri, P. et al. Understanding the intertwined nature of rising multiple risks in modern agriculture and food systems. Environ. Dev. Sustain. 26 , 24107–24150. 10.1007/s10668-023-03638-7 (2024). [ Google Scholar ] 2. Sivakumar, M. Climate change, agriculture adaptation, and sustainability. In Climate resilience and environmental sustainability approaches (eds Kaushik, A., Kaushik, C. P. & Attri, S. D.) 10.1007/978-981-16-0902-2_6 (Springer, 2021) 3. Thornton, P. K., Ericksen, P. J., Herrero, M. & Challinor, A. J. Climate variability and vulnerability to climate change: A review. Glob Change Biol. 20 , 3313–3328. 10.1111/gcb.12581 (2014). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 4. Kumar, S., Chatterjee, U., Raj, A. D. & Sooryamol, K. R. Global warming and climate crisis/extreme events. In Climate crisis: Adaptive approaches and sustainability (Sustainable Development Goals Series) (eds Chatterjee, U., Shaw, R., Kumar, S., Raj, A. D. & Das, S.) (Springer, Cham, 2023). 10.1007/978-3-031-44397-8_1. [ Google Scholar ] 5. Mulla, S. et al. Climate change effect on climate parameters like temperature, rainfall and water resources sectors in India. In Climate change impacts in India (eds Pande, C. B., Moharir, K. N. & Negm, A.) 10.1007/978-3-031-42056-6_2 (Springer, 2023) 6. Abebaw, S. E. A global review of the impacts of climate change and variability on agricultural productivity and farmers’ adaptation strategies. Food Sci. Nutr. 13 , e70260. 10.1002/fsn3.70260 (2025). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 7. Sertse, S. F., Tamire, L. C. & Liu, Y. Modeling climate change risk perception, vulnerability, and adaptation in farming communities: Evidence from Raya Azebo district, Ethiopia. Theor. Appl. Climatol . 156 , 663. 10.1007/s00704-025-05893-2 (2025). [ Google Scholar ] 8. Teku, D. Navigating climate uncertainty: A comprehensive review of climatic variabilities and extreme events on environmental, socio-economic, and livelihood dimensions in Ethiopia with adaptation strategies. All Earth . 37 , 1–30. 10.1080/27669645.2025.2514418 (2025). [ Google Scholar ] 9. Rijal, S., Gentle, P., Khanal, U., Wilson, C. & Rimal, B. A systematic review of Nepalese farmers’ climate change adaptation strategies. Clim. Policy . 22 , 132–146 (2022). [ Google Scholar ] 10. Sinore, T. & Wang, F. Impact of climate change on agriculture and adaptation strategies in Ethiopia: A meta-analysis. Heliyon 10 , 4 (2024). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 11. Lee, S., Paavola, J. & Dessai, S. Towards a deeper understanding of barriers to national climate change adaptation policy: A systematic review. Clim. Risk Manag . 35 , 100414. 10.1016/j.heliyon.2024.e26103 (2022). [ Google Scholar ] 12. Li, H. & Nguea, S. M. Protecting natural resources, empowering women: Investigating the effects of natural resource protection on women’s empowerment in Sub-Saharan Africa. Sustain Dev. 34 (S1), 1341–1361 (2026). 13. Ghio, D. et al. Assessing populations exposed to climate change: A focus on Africa in a global context. Popul. Environ. 45 , 28. 10.1007/s11111-023-00439-y (2023). [ Google Scholar ] 14. Richardson, K., Calow, R., Pichon, F., New, S. & Osborne, R. Climate risk report for the East Africa region. Met. Office ODI FCDO: UK . 10.1016/j.gloenvcha.2024.102912 (2022). [ Google Scholar ] 15. Ayompe, L. M. & Epie, W. N. Building Africa’s climate resilience: Understanding the impacts and future strategies in the face of climate change. Front. Clim. 7 , 1619799. 10.3389/fclim.2025.1619799 (2025). [ Google Scholar ] 16. Fadl, K. E. M. & Sheikh, S. E. E. Effect of Acacia senegal on growth and yield of groundnut, sesame and roselle in an agroforestry system in North Kordofan state, Sudan. Agrofor. Syst. 78 , 243–252. 10.1007/s10457-009-9243-9 (2010). [ Google Scholar ] 17. Sileshi, G. W., Dagar, J. C., Nath, A. J. & Kuntashula, E. Agroforestry as a climate-smart agriculture: Strategic interventions, current practices and policies. In Agroforestry for sustainable intensification of agriculture in Asia and Africa (Sustainability Sciences in (eds Africa, J. C., Gupta, S. R. & Sileshi, G. W.) in Asia and (Springer, Singapore, 10.1007/978-981-19-4602-8_18 (2023). [ Google Scholar ] 18. Mlambo, D. Agroforestry, climate change, and carbon sequestration. In Agroforestry for monetising carbon credits (eds (eds Raj, A., Jhariya, M. K. & Meena, R. S.) (Springer, Cham, 10.1007/978-3-031-97151-8_3 (2025). [ Google Scholar ] 19. Zeratsion, B. T. et al. Agroforestry practices for climate change adaptation and livelihood resilience in drylands of Ethiopia. Sci. Technol. 20 , 47–57. 10.1080/21580103.2023.2292171 (2024). [ Google Scholar ] 20. Tebkew, M., Asfaw, Z. & Worku, A. The role of agroforestry systems for addressing climate change livelihood vulnerability of farmers of Northwestern Ethiopia. Heliyon 10 , 17 (2024). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 21. Mayele, J. M. & Sakurai, T. The adoption of agroforestry practices: The determinants and constraints among smallholder farmers in Juba County, South Sudan. Agrofor. Syst. 99 , 237. 10.1007/s10457-025-01316-3 (2025). [ Google Scholar ] 22. Mohammed, K., Batung, E., Kansanga, M., Adusah-Poku, F. & Abdulai, A. Livelihood diversification strategies and resilience to climate change in semi-arid northern Ghana. Clim. Change . 164 , 53. 10.1007/s10584-021-03034-y (2021). [ Google Scholar ] 23. Demem, M. S. Impact and adaptation of climate variability and change on small-holders and agriculture in Ethiopia: A review. Heliyon 9 , e18972 (2023). 10.1016/j.heliyon.2023.e18972 [ DOI ] [ PMC free article ] [ PubMed ] 24. Beyene, B., Tilahun, M. & Alemu, M. The impact of livelihood diversification as a climate change adaptation strategy on poverty level of pastoral households in southeastern and southern Ethiopia. Cogent Soc. Sci. 9 , 2277349. 10.1080/23311886.2023.2277349 (2023). [ Google Scholar ] 25. Osei Boateng, B., Mensah, H., Simpeh, E. K. & Opoku, P. Amponsaa Boateng, L. Tackling local environmental challenges with livelihoods strategies: Insights from small-scale farmers in the vulnerable region of Ghana. Int. J. Agric. Sustain. 22 , 2399407. 10.1080/14735903.2024.2399407 (2024). [ Google Scholar ] 26. Furo, G., Tifo, K. & Feng, M. Socio-economic factors influencing household dependency on forest resources: A case of the Munessa–Shashemene Forest, Ethiopia. SN Soc. Sci. 3 , 7. 10.1007/s43545-022-00594-3 (2023). [ Google Scholar ] 27. Musa, F. I. et al. Contribution of non-wood forest products for household income in rural area of Sudan: A review. J. Agric. Food Res. 14 , 100801. 10.1016/j.jafr.2023.100801 (2023). [ Google Scholar ] 28. Thangjam, U., Thong, P. & Sahoo, U. K. Climate change threat on socio-economic condition of agroforestry managers: A vulnerability study in eastern Himalayan state of Mizoram, Northeast India. Clim. Risk Manag . 40 , 100519. 10.1016/j.crm.2023.100519 (2023). [ Google Scholar ] 29. Tittonell, P. Livelihood strategies, resilience and transformability in African agroecosystems. Agric. Syst. 126 , 3–14. 10.1016/j.agsy.2013.10.010 (2014). [ Google Scholar ] 30. Alemu, T. & Mengistu, A. Impacts of climate change on food security in Ethiopia: Adaptation and mitigation options: A review. In Climate change resilient agriculture and agroforestry (eds (eds Castro, P., Azul, A., Leal Filho, W. & Azeiteiro, U.) (Springer, Cham, 10.1007/978-3-319-75004-0_23 (2019). [ Google Scholar ] 31. Mdleleni, L. et al. A synthesis of environmental policies and identification of critical gaps in critical zones of South and East Africa. Environments 12 , 326. 10.3390/environments12090326 (2025). [ Google Scholar ] 32. Senbeta, F. The ecology and management of the coffee forests of Ethiopia. In Environmental change and biodiversity conservation in sub-Saharan Africa (Vol. 75) (eds Hambira, W. L., Abdeta, A. A., Moalafhi, D. B., Muposhi, V. K. & Mosepele, K.) (Springer, Cham, 10.1007/978-3-031-73136-5_12 (2025). [ Google Scholar ] 33. Amsalu, A. et al. Coffee farming households’ vulnerability to climate variability in the Kafa Biosphere Reserve, Ethiopia. Earth Syst. Environ. 10.1007/s41748-025-00847-0 (2025). [ Google Scholar ] 34. Teka, T. K. & Bender, S. Climate adaptive agriculture: A smallholders case study of the Southwestern Highlands of Ethiopia. In Sustainable agriculture and food security (World Sustainability Series) (eds Leal Filho, W., Kovaleva, M. & Popkova, E.) (Springer, Cham, 10.1007/978-3-030-98617-9_5 (2022). [ Google Scholar ] 35. Tofu, D. A., Haile, F. & Tolossa, T. Livelihood vulnerability and socio-economic determinants of households to climate change–induced recurrent drought in Ethiopia. GeoJournal 88 , 5043–5067. 10.1007/s10708-023-10904-8 (2023). [ Google Scholar ] 36. Zagre, I., Akinseye, F. M., Worou, O. N., Kone, M. & Faye, A. Climate change adaptation strategies among smallholder farmers in Senegal’s semi-arid zone: Role of socio-economic factors and institutional supports. Front. Clim. 6 , 1332196. 10.3389/fclim.2024.1332196 (2024). [ Google Scholar ] 37. Kassahun, T. & Bender, S. Food security in the face of climate change at Kafa Biosphere Reserve, Ethiopia. In Handbook of climate services (Climate Change Management) (eds Leal Filho, W. & Jacob, D.) (Springer, Cham, 10.1007/978-3-030-36875-3_23 (2020). [ Google Scholar ] 38. Moat, J. et al. Resilience potential of the Ethiopian coffee sector under climate change. Nat. Plants . 3 , 17081. 10.1038/nplants.2017.81 (2017). [ DOI ] [ PubMed ] [ Google Scholar ] 39. Bender, S. & Tekle, M. Community action for biodiversity and forest conservation and adaptation to climate change in the wild coffee forests (CAFA). In Handbook of climate change and biodiversity (eds Leal Filho, C. C. M., Barbir, W. & Preziosi, R.) J. (Springer, Cham, 10.1007/978-3-319-98681-4_5 (2019). [ Google Scholar ] 40. NABU, T. BCU NABU’s biodiversity assessment at the Kafa Biosphere Reserve . (Berlin, Addis Ababa, 2017). 41. Ahmed, S. K. How to choose a sampling technique and determine sample size for research: A simplified guide for researchers. Oral Oncol. Rep. 12 , 100662. 10.1016/j.oor.2024.100662 (2024). [ Google Scholar ] 42. Hurni, H. Agroecological belts of Ethiopia: Explanatory notes on three maps at a scale of 1:1,000,000 (1998). 43. Yamane, T. Statistics: An Introductory Analysis 2nd edn (Harper & Row, 1967). 44. Mann, H. B. Nonparametric tests against trend. Econometrica 13 , 245–259 (1945). [ Google Scholar ] 45. Kendall, M. G. Rank Correlation Methods 4th edn (Charles Griffin, 1975). 46. Sen, P. K. Estimates of the regression coefficient based on Kendall’s tau. J. Am. Stat. Assoc. 63 , 1379–1389 (1968). [ Google Scholar ] 47. Cox, D. T., Maclean, I. M. D., Gardner, A. S. & Gaston, K. J. Global variation in diurnal asymmetry in temperature, cloud cover, specific humidity and precipitation and its association with leaf area index. Glob Change Biol. 26 , 7099–7111. 10.1111/gcb.15336 (2020). [ DOI ] [ PubMed ] [ Google Scholar ] 48. Mohammed, E. M. I. et al. Linking woody plant species richness with selected ecosystem services and dendrometric features in Okalma natural forest reserve. Sci. Rep. 15 , 28751. 10.1038/s41598-025-13640-w (2025). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 49. Pricope, N. G., Husak, G., Lopez-Carr, D., Funk, C. & Michaelsen, J. The climate-population nexus in the East African Horn: Emerging degradation trends in rangeland and pastoral livelihood zones. Glob Environ. Change . 23 , 1525–1541. 10.1016/j.gloenvcha.2013.10.002 (2013). [ Google Scholar ] 50. Amsalu, A., Garedew, W., Melka, G. A. & Gemeda, D. O. Assessing trends and variability of rainfall and temperature in the Kafa Biosphere Reserve, southwest Ethiopia. Discover Sustain. 7 , 82. 10.1007/s43621-025-02499-6 (2026). [ Google Scholar ] 51. Amsalu, A., Garedew, W., Melka, G. A., Çelik, M. A. & Gemeda, D. O. Trends in climate extreme indices and implications for coffee farming in the Kafa Biosphere Reserve, southwest Ethiopia. Clim. Serv. 42 , 100637. 10.1016/j.cliser.2026.100637 (2026). [ Google Scholar ] 52. Olabanji, M. F. & Chitakira, M. The adoption and scaling of climate-smart agriculture innovation by smallholder farmers in South Africa: A review of institutional mechanisms, policy frameworks and market dynamics. World 6 , 51. 10.3390/world6020051 (2025). [ Google Scholar ] 53. Danso-Abbeam, G. Do agricultural extension services promote adoption of soil and water conservation practices? Evidence from Northern Ghana. J. Agric. Food Res. 10 , 100381. 10.1016/j.jafr.2022.100381 (2022). [ Google Scholar ] 54. Inkani, A. I. et al. Towards enhanced resilience, adaptation, and sustainability of farming ecosystems: Assessment of the efficiency of specialized public institutions in weather and climate services delivery to smallholder farmers. Mitig Adapt. Strat Glob Change . 31 , 10. 10.1007/s11027-025-10276-z (2026). [ Google Scholar ] 55. Mkoga, Z. J., Tumbo, S. D., Kihupi, N. & Semoka, J. Extrapolating effects of conservation tillage on yield, soil moisture and dry spell mitigation using simulation modelling. Phys. Chem. Earth A/B/C . 35 , 686–698. 10.1016/j.pce.2010.07.036 (2010). [ Google Scholar ] 56. Habib, N., Ariyawardana, A. & Aziz, A. A. The influence and impact of livelihood capitals on livelihood diversification strategies in developing countries: A systematic literature review. Environ. Sci. Pollut Res. 30 , 69882–69898. 10.1007/s11356-023-27638-2 (2023). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 57. Raghavendra, K. J. et al. Unraveling determinants of integrated farming systems adoption for sustainable livelihood and dietary diversity. Front. Nutr. 11 , 1264658. 10.3389/fnut.2024.1264658 (2024). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 58. Yan, D. & Li, G. A heterogeneity study on the effect of digital education technology on the sustainability of cognitive ability for middle school students. Sustainability 15 , 2784. 10.3390/su15032784 (2023). [ Google Scholar ] 59. Jain, V., Tewathia, N. & Barik, K. Gender-differentiated labor and adaptation effects of climate change in rural areas: A systematic literature review. Gend. Issues . 40 , 168–184. 10.1007/s12147-023-09310-8 (2023). [ Google Scholar ] 60. Pierotti, R. S., Friedson-Ridenour, S. & Olayiwola, O. Women farm what they can manage: How time constraints affect the quantity and quality of labor for married women’s agricultural production in southwestern Nigeria. World Dev. 152 , 105800. 10.1016/j.worlddev.2021.105800 (2022). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 61. Porsani, J., Börjeson, L. & Lehtilä, K. Land concessions and rural livelihoods in Mozambique: The gap between anticipated and real benefits of a Chinese investment in the Limpopo Valley. J. South. Afr. Stud. 43 , 1181–1198. 10.1080/03057070.2017.1377932 (2017). [ Google Scholar ] 62. Tadele, E. Land and heterogeneous constraints nexus income diversification strategies in Ethiopia: A systematic review. Agric. Food Secur. 10 , 37. 10.1186/s40066-021-00338-1 (2021). [ Google Scholar ] 63. Armitage, D. Adaptive capacity and community-based natural resource management. Environ. Manage. 35 , 703–715. 10.1007/s00267-004-0076-z (2005). [ DOI ] [ PubMed ] [ Google Scholar ] 64. Matocha, J., Schroth, G., Hills, T. & Hole, D. Integrating climate change adaptation and mitigation through agroforestry and ecosystem conservation. In Agroforestry – The Future of Global Land Use (Advances in Agroforestry, vol. 9) (eds Nair, P. & Garrity, D.) (Springer, Dordrecht, 10.1007/978-94-007-4676-3_9 (2012). [ Google Scholar ] 65. Blainey, G. The tyranny of distance: How distance shaped Australia’s history (Sun Books, 1966). 66. Kassie, G. W., Kim, S. & Fellizar, F. P. Determinant factors of livelihood diversification: Evidence from Ethiopia. Cogent Soc. Sci. 3 , 1369490. 10.1080/23311886.2017.1369490 (2017). [ Google Scholar ] 67. Schick, A., Wieners, E., Schwab, N. & Schickhoff, U. Sustainable disaster risk reduction in mountain agriculture: Agroforestry experiences in Kaule, mid-hills of Nepal. In Climate Change, Extreme Events and Disaster Risk Reduction (Sustainable Development Goals Series) (eds Mal, S., Singh, R. & Huggel, C.). (Springer, 2018). 10.1007/978-3-319-56469-2_17 68. Zabala, A., Pascual, U., García-Barrios, L. E. & Mukherjee, N. Drivers to adopt agroforestry and sustainable land-use innovations: A review and framework for policy. Land. Use Policy . 151 , 107468. 10.1016/j.landusepol.2025.107468 (2025). [ Google Scholar ] 69. Amare, D., Wondie, M. & Mekuria, W. Agroforestry of smallholder farmers in Ethiopia: Practices and benefits. Small-Scale For. 18 , 39–56. 10.1007/s11842-018-9405-6 (2019). [ Google Scholar ] 70. Hossain, M. R. Evaluating households’ risk management strategies considering heterogeneities in livelihood diversification: Evidence from farm households of Bangladesh. Soc. Sci. Humanit. Open. 9 , 100785. 10.1016/j.ssaho.2024.100785 (2024). [ Google Scholar ] 71. Salam, S. & Bauer, S. Rural non-farm economy and livelihood diversification strategies: Evidence from Bangladesh. GeoJournal 87 , 477–489 (2022). 10.1007/s10708-020-10269-2 Associated Data This section collects any data citations, data availability statements, or supplementary materials included in this article. Supplementary Materials Supplementary Material 1 (1.4MB, docx) Data Availability Statement The data supporting the findings of this study are available from the corresponding author upon reasonable request. Articles from Scientific Reports are provided here courtesy of Nature Publishing Group ACTIONS View on publisher site PDF (2.1 MB) Cite Collections Permalink RESOURCES Similar articles Cited by other articles Links to NCBI Databases Follow NCBI NCBI on X (formerly known as Twitter) NCBI on Facebook NCBI on LinkedIn NCBI on GitHub NCBI RSS feed Connect with NLM NLM on X (formerly known as Twitter) NLM on Facebook NLM on YouTube National Library of Medicine 8600 Rockville Pike Bethesda, MD 20894 Web Policies FOIA HHS Vulnerability Disclosure Help Accessibility Careers NLM NIH HHS USA.gov Back to Top

Related documents

• Smallholder farmers' adaptation strategies among coffee farming household heads in the Kafa biosphere reserve, Ethiopia.#9852
• Household Food Handling Practices and Associated Factors Among Adult Women in Gondar City, Northwest Ethiopia: A Community-Based Cross-Sectional Study Design.#481106
• Food-based production of smallholders in northwest Ethiopia: the extent and determinants#244683
• Co-Creating Evidence for Kenya&apos;s Agri-Food Transformation: Launching the CGIAR Policy Innovations Hub#346737
• Coffee agroforestry and associated reforestation in Arabica coffee’s native range#466992