Goats are known as some of the most impactful invasive mammals on islands and are now considered the sixth greatest threat to plant communities worldwide due to uncontrolled and extensive grazing. The negative impact of the growing population of semi-feral animals on endemic plants in Socotra Island (Yemen) is well known. Yet, there is virtually no data on the proportion of individual endangered plant species in the diet of goats. It is also unknown whether and to what extent goats consume non-native plants, and what factors influence the overall composition of their diet. This study is the first to focus on the dietary habits of domestic goats on the island of Socotra using DNA barcoding of plant remains in goat excrements. We hypothesize that goats browse on woody plants, including tree seedlings, mainly during the dry season, as in the rainy season, there are sufficient grass and herbs. Therefore, the dry season is crucial for the survival of woody plant seedlings, when high mortality of saplings is not caused only by drought but also by overgrazing. Our results are consistent with this hypothesis. We found a statistically significant positive correlation between the dry season and woody plants grazing and a negative correlation between altitude and grazing of endemics. The reduction of goat’s abundance on the island is a necessary management measure to ensure the tree protection and their regeneration, which should be accompanied by a change of goat grazing system and active support of tree regeneration.
The brown hare is an important component of agro-ecosystems and is also considered a good indicator species for habitat changes or predator-control interventions. Predator management to protect Great bustards can therefore affect this small game species. Our study aimed to assess this impact in an experiment, where we estimated the abundance of the brown hare using spotlight count in three National Park study sites and an independent but similar control habitat and we surveyed the populations of red fox, Eurasian badger and hooded crow between 2021 and 2023. The effectiveness of predator management was characterised by the culling rate. The population size of the brown hare did not change over the years. No correlation was found between predator culling rate and hare population size. The population size of predators also did not vary with changes in the intensity of the culling rate, except for the red fox in the control area. Here, the population size of foxes decreased as a result of more intense predator removal. Overall, the experiment did not provide clear evidence of a positive effect of predator management, as in many other studies, but this may have been due to the small sample size (short period). It is also possible that the habitats are currently able to support this amount of hare and that further habitat development and predator management would be necessary to significantly increase the population size.
A Tiszainoka külterületén található Fekete-halom egy őskori kurgán, valószínűleg a Jamnaja-kultúra építhette a késő rézkorban. A halom felszínét a 18. században már művelték. Az 1960-as években a tetején földmérési pontot állandósítottak. A halom északi oldala erősen erodált, a roncsolás a korábban rajta áthaladó földút használata miatt alakult ki. Az európai uniós agrár-környezetvédelmi támogatásoknak köszönhetően a halom területén 2014-re felhagyták a szántóföldi művelést. A felszínt spontán jól regenerálódó parlag borítja, többek között a déli és nyugati oldalon megjelent a védett törpe mandula (Prunus tenella) is. Természetvédelmi kezelésként a kímélő kaszálás indokolt, továbbá érdemes lenne a halom körül széles gyepi pufferzónát kialakítani.
Species distribution models (SDMs) often extrapolate into regions that are ecologically and geographically unsuitable for the focal species. Using the Western Palaearctic wood mouse (Apodemus sylvaticus) as a case study and data pre-filtering, we evaluated six pseudo-absence (PA) generation strategies and four variable selection protocols aimed at improving the accuracy of predicted distribution maps while minimising spatial autocorrelation of residuals. Models were fitted using an iterative Bayesian additive regression trees algorithm (bart.step) and evaluated based on AUC, TSS, false-negative rate, Moran’s I, and Biogeographic Realism – the latter defined as the absence of predicted suitability in absolute dispersal barriers such as deserts, Arctic tundra, and northern islands.
The persistence and variety of anthropogenic factors have only accelerated the degradation process of Mediterranean forests. This study focuses on the cork oak forest of Mzirâa located northwest of El Kala in northeastern Algeria. The objective of this study is to evaluate the ecological status (floristic composition and vegetation structure) of the flora in selected areas of the Mzirâa forest. The study is based on the development of a floristic matrix using a systematic sampling method in each study locality. The findings revealed a regressive spatio-temporal dynamism of the flora and vegetation of the forest. In addition, a floristic dynamics and analysis using NDVI for the study area indicated that there is rapid degradation of the Mzirâa forest. This implies that the forest is under serious ecological threats and needs urgent management strategies to protect the forest from further degradation. This study provides crucial information for researchers, policymakers, and forest protection authorities in Algeria.
Estuaries can be found all throughout Sri Lanka's coasts, but the kinds of research that have been done on them, their relative importance, and the trends in those studies have not been systematically examined until now. Therefore, the aim of the present review is to undertake a bibliometric analysis of estuaries in the context of Sri Lanka, comprising trends in publication, most-cited articles, influential authors, collaborating nations, prominent institutions, journal types, keyword co-occurrence, and bibliographic coupling in the field.Articles on Sri Lankan estuary research published in the Web of Science (WoS) database between 1987 and 2023 were analyzed using bibliometric techniques and visual mapping. This study examined 58 papers in total, Visualization of Similarities (VOS) viewer was used to analyze data. 1987 to 2011 had one or two publications per year, however the positive quadratic coefficient (0.0074) indicates an accelerating growth trend in publications over time. Many studies have been conducted focusing on the Negombo estuary. Notably, the majority of articles focused on water quality studies. Further, fisheries, mangroves, and environmental-related research have also been undertaken equally in this regard. However, the findings emphasize the necessity of integrating spatial information techniques, geological and geographical data, coastal sediment dynamics, tourism, and further studies on sea-level rise in estuaries. Many studies have focused mainly on the southern and western parts of the estuaries. The Northern Coastal Zone's estuaries remain largely unexplored. High-security zones surrounding the estuaries, particularly in the coastal regions to the north and east, restrict researchers' freedom to conduct certain research. Fewer studies have been conducted, yet those that have are detailed. Therefore, the review indicates that future research should consider an integrated, multidisciplinary, and holistic ecosystem approach to Sri Lanka's estuaries.
Understanding how species diversity interacts with spatial environmental gradients to shape landscape-level community patterns is a central question in contemporary landscape ecology. Although predictive modelling of biodiversity is advancing globally, dry Afromontane forests remain poorly represented in analyses that link spatial heterogeneity with community structure. This study addresses this gap by integrating diversity indices with a spatially explicit Boosted Regression Tree modelling framework to predict species abundance and community patterns across the Desa'a Forest, northern Ethiopia. Data from 301 plots were used to quantify woody species richness, abundance, and community composition, and to identify the environmental controls governing the distribution of dominant species. The forest contained ninety-four woody species and 7,207 individuals, with diversity values and plot-level abundances revealing a highly heterogeneous landscape composed of species-poor and species-rich patches. Cluster analysis delineated seven ecological communities, and dominant species such as Cadia purpurea, Tarchonanthus camphoratus, and Juniperus procera contributed disproportionately to spatial variation in forest structure. Boosted Regression Tree models showed that temperature, elevation, precipitation, and slope were the strongest predictors of Juniperus procera abundance, explaining sixty-three percent of deviance and revealing a narrow elevational and climatic niche. These findings clarify how landscape-scale environmental gradients shape species distributions and community mosaics in dry Afromontane forests. By demonstrating a data-driven approach for predicting spatial community structure, this study provides a modelling framework that can enhance conservation planning and landscape-level management in biodiversity-rich mountain ecosystems worldwide.
Forests are dynamic systems, where structure, composition, and ecological function emerge from the complex interplay of altitude, soil, and biological interactions. Although the forests in the region are central to biodiversity conservation and local livelihoods, they remain poorly documented, especially in terms of ecological community structure. To address this gap, we classified the forest's woody plant communities and identified the environmental gradients that govern their distribution. Vegetation data were collected from 120 systematically arranged 20 m & times; 20 m quadrats placed along five transects. Species composition, abundance, and cover were quantified and analyzed using cluster analysis, canonical correspondence analysis, and standard diversity measures. A total of 60 woody species belonging to 55 genera and 33 families were recorded, with shrubs (35 species) more numerous than trees (25 species). The analyses revealed four distinct community types. The forest exhibited high overall diversity (Shannon H ' = 3.5) and a well-balanced distribution of species (evenness J = 0.93). Variation in altitude, aspect, soil pH, available phosphorus, and sand content emerged as the principal factors explaining differences in community composition (p < 0.05). These findings show that the Damota Forest is strongly structured by topographic and edaphic gradients and retains considerable woody-species diversity despite ongoing human pressure. Strengthening conservation efforts, restoring degraded sites with native indicator species, and aligning local forest management with climate- and carbon-focused initiatives such as Ethiopia's Green Legacy Programme are essential steps toward securing the forest's long-term resilience.
Military activities represent a major driver of environmental change, leading to large-scale transformations of landscape structure and ecosystem functioning within protected areas. This study examines the impact of armed conflict on the Nature Reserve Fund (NRF) of Ukraine, with a particular focus on spatial patterns of landscape fragmentation, pyrogenic dynamics, and ecosystem degradation.An integrated methodological approach combining geoinformation (GIS) analysis, remote sensing, ecological monitoring, and descriptive assessment was applied to evaluate environmental changes during pre-war (2019-2021) and wartime (2022-2024) periods. The analysis was based on Sentinel-2 satellite imagery, FIRMS (MODIS and VIIRS) thermal anomaly data, NDVI calculations, and landscape metrics (patch density, edge density, mean patch size, and landscape shape index).The results reveal significant spatial reorganisation of protected landscapes, characterised by increased fragmentation, reduced habitat continuity, and intensified edge effects. Fire disturbance emerged as a key factor of landscape transformation, as evidenced by a sharp increase in thermal anomalies and expansion of burned areas. Vegetation dynamics indicate incomplete ecosystem recovery under conditions of repeated disturbances.In addition to structural changes, military impacts have contributed to soil degradation, water pollution, and biodiversity loss, leading to reduced ecosystem resilience. The disruption of ecological connectivity and the formation of heterogeneous landscape mosaics complicate natural recovery processes.The findings highlight the importance of applying landscape ecology approaches and geospatial tools for assessing war-related environmental impacts. The study provides a scientific basis for developing spatially explicit strategies for post-conflict ecological restoration, including the prioritisation of degraded areas and the restoration of ecological networks. The results also emphasise the broader European significance of environmental damage in Ukraine and the need for international cooperation in ecosystem recovery.
Invasive Alien Plant Species threaten the sustainability of remaining forest ecosystems including those in Protected Areas. Unfortunately, there are only a few investigations on their colonization and dispersal over time, particularly most protected areas in the Philippines. This study aimed to assess the spatial extent of areas at risk of species colonization and dispersal of T. cumingiana that was introduced in the Quezon Protected Landscape over 40 years. A tree inventory following a combined belt transect and purposive sampling was made to determine occurrences of their natural regenerations. The spatial extent and areas at risk of species colonization were determined by modeling its species distribution using species occurrence and 9 environmental predictor variables via Random Forest Classification in a Geographic Information System. The species occurrences of T. cumingiana at different growth stages were used to project areas at risk to species colonization at different time slices. Results revealed a total of 996 individuals composed of 547 seedlings, 375 saplings, 54 poles, and 20 standard trees. Projections showed an increase in spatial extent and areas at risk to species colonization from 12.54 ha (1.29 %) in the 2000s to 39.28 ha (4.05 %) by the 2020s, with distance to seed sources and topography-related features, specifically slope, aspect, and elevation, among the most important variables determining its occurrence. These results indicated slow colonization of T. cumingiana in QPL which can provide foundations for understanding biology and impacts of this invasive species, as well as identifying its optimal management and control strategies for sustaining the ecology of the landscape.
Land use land cover change (LULCC) is a complex phenomenon influenced by human activities and environmental factors that significantly affect ecosystems and sustainable development. To effectively understand this dynamic process and integrate it into planning and natural resource management, current and projected LULC information is crucial. This study aimed to assess the LULCC from 2003 to 2024 and project for the next 20 years in the Jemma sub-basin of Ethiopia. We employed random forest (RF) machine-learning algorithms for classification, and a CA-ANN model to project future LULC changes using the MOLUSCE plugin in QGIS. Elevation, distance from rivers, and distance from roads were used as inputs in the projection process. The major land use types identified in the study area were cropland, grazing land, water bodies, forest land, bare land, and built-up areas. Classification accuracy ranged from 83 % to 90 %, and the kappa coefficient ranged from 0.78 to 0.88. Between 2003 and 2024, cropland, built-up areas, and water bodies increased by 6.5, 2.3, and 0.2 %, respectively, while forest land increased by 1.7 %. Conversely, grazing and bare land decreased by 9.9 % and 0.7 %, respectively, between 2003 and 2024. In 2034, cropland, water bodies, forest land, and built-up areas are projected to increase by 1.99 %, 0.01 %, 0.56 %, and 0.18 %, respectively; however, these land use types show a slight decreasing trend from 2034 to 2044, except for built-up areas. Urban areas are expected to increase by 0.48 % between 2034 and 2044, primarily at the expense of grazing and bare lands. The findings suggest that significant land use changes are shaping the ecological and socioeconomic dynamics of the sub-basin. Therefore, promoting the wise utilization of natural resources and effective land management practices is essential to ensure ecological and environmental sustainability in the sub-basin.
Land cover changes in arid zones with industrial agglomerations require a comprehensive spatiotemporal assessment of drought and soil airborne pollution. Geospatial maps were obtained for each level of the two time periods -2019-2021 and 2022-2024 based on the distribution and contextual significance of each parameter. An analysis of the identified trends in the climate change indicators (precipitations, evapotranspiration and land surface moister) allows us to speak about a meteorological type of drought complicated with large Kakhovka water reservoir disappearance in June 2023. At the same time, the advance of drought conditions was fixed in the LAI change in canopy density throughout the steppe zone of Ukraine by 2024. It was found that the progression of arid conditions has the character of horizontal zonality in the direction from south to north.The main differences in the risk of acid rain caused by industrial agglomerations are related to the spatial distribution of nitrogen dioxide in the atmosphere in the northern part of Ukrainian steppe.
Dry afromontane forests are important for climate change mitigation in Ethiopia, yet a systematic synthesis of their biomass and carbon storage capacity is lacking. This review aims to (1) quantify the biomass and carbon stocks in these forests, (2) map their geographic distribution, and (3) identify the key biophysical and anthropogenic factors driving carbon stock variation. Following the PRISMA guidelines, we systematically reviewed 72 relevant studies (2000-2025) identified from Web of Science, Scopus, PubMed, and Google Scholar. The results showed that aboveground biomass (AGB) ranged from 35.1 +/- 16.6 t ha(-)& sup1; in Desa Forest to 720.7 +/- 503 t ha(-)& sup1; in Banja Forest, with belowground biomass (BGB) following a similar pattern and generally representing 18-22 % of AGB. Soil organic carbon (0-30 cm depth) also varies substantially from 58 +/- 7.6 t ha(-)& sup1; in Gara Muktar to 277.6 +/- 11.6 t ha(-)& sup1; in Egdu Forest. Forests such as Banja, Gedo, Egdu, Ades, and Zafenigus show particularly high AGB, highlighting the capacity of well-conserved high forests to store roughly 215-425 t ha(-)& sup1;, depending on site conditions. Factors contributing to this variation include measurement errors, the choice of allometric equations for biomass and carbon estimation, species composition and community structure, and topographic factors such as altitude and slope. Additionally, human disturbances play a significant role. Future research focuses on integrating advanced remote sensing technologies, particularly LiDAR, and applying climatic and biogeochemical models (e.g., CO2Flux, BIOME-BGC) to simulate future biomass and carbon dynamics.
This study presents a comprehensive spatio-temporal analysis of nighttime Land Surface Temperature (LST) and Urban Heat Island Intensity (UHII) in Coimbatore from 2001 to 2022, highlighting statistically significant warming trends and intensifying urban heat island effects. Urban areas experienced a notable nighttime LST increase from 21.4 °C in 2001 to 23.7 °C in 2019, compared to a rural rise from 20.5 °C to 22.5 °C. The average urban–rural LST differential (~1 °C) widened post-2016, aligning with the recorded peak LST of 26.8 °C. The minimum LST dropped to 8.5 °C in 2001, indicating a reduction in cold extremes. Kendall’s tau analysis confirmed a stronger warming trend in urban areas (τ = 0.593) than rural zones (τ = 0.429). Seasonal UHII analysis showed progressive winter intensification post-2012, while summer UHII peaked in 2013 and 2015, then dipped post-2016 before rising again in 2022. Mann-Kendall tests confirmed statistically significant increasing trends in winter UHII, urban LST, and rural LST, with urban LST exhibiting the steepest rise. Spatial autocorrelation analysis using Moran’s Index revealed intensifying clustering of high LST zones: the annual Moran’s Index increased from 0.797 (2001) to 0.857 (2022), with z-scores rising from 42.253 to 45.445. Winter showed the most pronounced clustering, with Moran’s Index jumping from 0.812 to 0.903 and z-scores reaching 47.848 by 2022. Hotspots with 99 % confidence levels were primarily urban, expanding over time with temperatures between 24.8 °C and 26.7 °C, while cold spots (99 % CL) remained stable in rural areas. These findings confirm the persistent and intensifying nature of UHI in Coimbatore, driven by urban expansion, declining vegetation, and increased impervious surfaces. This study fills a critical research gap by providing one of the first long-term assessments of nighttime UHI intensity in a mid-sized Indian city, thereby contributing to the broader understanding of urban thermal dynamics beyond metropolitan regions. The study underscores the urgent need for spatially informed interventions, such as urban greening, reflective materials, and climate-sensitive planning, to mitigate urban thermal stress and enhance resilience in rapidly growing cities.
The loss of potential storage space due to vegetation succession is a significant issue, particularly in unmanaged farmlands connected to drainage channels, reducing the effectiveness of the land in mitigating flood risk. This study aimed to discover unmanaged spaces within the alluvial plain of the Lower Tone River Basin, Japan, that have undergone vegetation succession. By monitoring the distribution of vegetation succession, the aim was to develop a technical method to identify and repurpose vegetation succession sites as storage spaces through proposed interventions to mitigate inland flooding within the alluvial plain of the Lower Tone River Basin. The analytical framework involved extracting alluvial plain areas, monitoring vegetation succession by integrating optical and synthetic aperture radar satellite images over the 2017-2024 period, analyzing the relationship between land-use types and vegetation succession patterns, extracting targeted vegetation succession sites, and classifying targeted vegetation succession patches based on height above the nearest drainage, distance to the nearest drainage channel, and other parameters. The classification resulted in three functional classes. Class 1 spaces were closely linked to water channels, making them critical areas for immediate floodwater storage. Class 2 spaces served as intermediate storage areas, situated at moderate distances from the water channels. Class 3 spaces provided final storage spaces for dispersed floodwater at the highest elevation and the furthest distance from the drainage channels. The sequential impact of inland flooding characterizes the functional hierarchy among Classes 1, 2, and 3, considering the on-site conditions. Overflowing from the drainage channels first impacts the Class 1 spaces. As the floodwater dispersed, it reached Class 2 spaces. Finally, Class 3 spaces served as the last line of mitigation, storing floodwater spread widely across the landscape. By identifying and categorizing vegetation succession patches for conversion into storage spaces, this study provides a basis for prioritizing areas for intervention and developing targeted flood mitigation strategies.
Flooding is one of the most frequent and destructive environmental hazards globally, and Indonesia is among the most affected countries. Understanding land use land cover (LULC) changes and flood distribution is essential for effective mitigation strategies. However, conventional methods using multiple processing environments are time-consuming, whereas integrating multi-sensor data within a single platform such as Google Earth Engine (GEE) improves efficiency and accuracy. This study aims to analyzes flood distribution and LULC changes in the Konaweha watershed from 2015 to 2024 using multi-temporal Sentinel-1 SAR and Landsat-8 optical imagery. Flooded areas were mapped using the Otsu thresholding combined with change detection, while LULC changes were identified using the Random Forest algorithm. The result reveal that flood inundation expanded from 6,709.24 ha in 2015 to 16,295.35 ha in 2020, before declining to 9,243.28 ha in 2024. Major LULC transitions included reductions in wetlands (12.82 %), primary forest (1.94 %), and agriculture (28.82 %), alongside increase in built-up areas (80.48 %), secondary forest (8.13 %), and water bodies (41.76 %). This finding indicate a strong correlation between flood occurrence and LULC changes, emphasizing the influence of environmental degradation on flood dynamics. The study contributes to global discourse on flood risk assessment by demonstrating the effectiveness of integrating multi-sensor remote sensing data for near real-time flood monitoring and sustainable land management planning.
Forest fragmentation poses a critical threat to ecosystem service functionality, particularly in rapidly urbanizing regions. This study investigates the spatio-temporal dynamics of forest fragmentation and its impacts on two essential ecosystem services, water provisioning and recreational value, within the Petaling District, Selangor, Malaysia, across the years 2005, 2014, and 2023. Using Landsat satellite imagery, supervised classification, and key landscape metrics (patch size, edge density, and connectivity), the research identified clear signs of intensified fragmentation. Although there was partial forest recovery post-2014, increased patchiness and elevated edge densities indicated persistent structural degradation. To understand how fragmentation affects ecosystem services at a local scale, the study used Geographically Weighted Regression (GWR). GWR is a spatial econometric approach that captures local variations and spatial non-stationarity. This is in contrast to conventional global models which assume uniform relationships across space. Significant spatial heterogeneity was found in the influence of fragmentation metrics on water provisioning and recreational services. The results can be interpreted as site-specific and thus important for adaptive conservation planning. Fragmentation metrics were found to influence water provisioning and recreational services unevenly across the landscape, with proximity to water bodies and developed areas serving as proxy indicators. Notably, areas with high edge density and reduced patch sizes were more vulnerable to hydrological service loss, while recreational value exhibited a spatially complex relationship, influenced by accessibility and landscape aesthetics. The findings underscore the urgent need for spatially adaptive planning and proactive forest governance, especially in urbanizing districts. By integrating ecological indicators with spatial econometrics, this research offers nuanced, site-specific insights that support conservation prioritization and sustainable land use strategies. The methodological framework and results contribute to broader discussions on urban ecological resilience and ecosystem service preservation under accelerating land-use pressures in tropical regions.
This article investigates the cultural significance of Siamese herbal rice (khaoyam) in Kelantan, Malaysia, with a focus on its role in shaping the relationship between people, plants, and place. Using an ethnographic approach, the research involved fieldwork in six villages across four districts and utilized focus group interviews and key informant discussions to compile a comprehensive list of fifty-nine edible plants used in herbal rice preparation and herbal rice narratives. The analysis considers both the plant knowledge embedded in plant selection and the ways herbal rice practices are performed, including khaoyam shops and temple-based festivals. The study introduces the new concept of "Eating in a shared landscape" to articulate how the preparation and sharing of herbal rice materialize deep-rooted ecological knowledge and foster adaptation, negotiation within a multicultural context. Findings demonstrate that herbal rice acts as a living foodscape that reflects Siamese identity, transmits ecological wisdom across generations, and mediates relationships with both landscape and neighboring communities.
Soil quality is a key determinant of ecosystem functions and land sustainability in rapidly changing coastal regions. This study evaluated spatial variation in soil quality across five land-use types (agriculture, urban green spaces, built-up areas, mangroves, and barren land) in the Puducherry region of India. A total of 225 surface soil samples (0-20 cm) were collected and analyzed for fifteen physicochemical parameters. Principal component analysis (PCA) identified four key indicators: electrical conductivity (EC), organic carbon (OC), available nitrogen (Av. N), and manganese (Av. Mn). A weighted linear scoring system was used to compute the Soil Quality Index (SQI), and spatial interpolation was performed using ordinary kriging. SQI values ranged from 0.45 in built-up areas to 0.60 in mangroves, showing the order: mangroves > urban green spaces > agriculture > barren > built-up areas. Mangrove soils showed higher SQI due to greater organic carbon and nutrient content, while inland and built-up areas reflected lower quality linked to soil disturbance and reduced fertility. These findings highlight spatial patterns of soil degradation and provide a practical framework for improving land management in tropical coastal environments.
It is difficult to separate coffee cultivation from the social and cultural life of the Ethiopian people, as the two are deeply interconnected in multiple ways. Improvements in coffee production should not be evaluated solely in terms of yield or economic returns; rather, they depend strongly on sociocultural determinants embedded within coffee production systems. Accordingly, this study examines the socio-economic characteristics of coffee-producing households, the dominant coffee production systems (CPS), and their implications for sustainability. Data were collected from 148 coffee-producing households (60 % men and 40 % women) using a semi-structured questionnaire administered through Kobo Toolbox. Statistical analyses were conducted using SPSS software. Land allocation to coffee cultivation differed significantly among households (p < 0.05), with an average of 0.70 ha allocated to coffee out of a total mean farm size of 2.05 ha, reflecting the predominance of smallholder production systems. The study areas more domain by semi-forest (54 %), followed by garden (36 %) production system. Forest coffee shared only 1 % that under risk of extinction, but plantation is not practiced. Around 97.4 % farmers uses trees for shade purpose, while obtaining its ecosystem service. About 68 % of households grew coffee inter-cropped, which made food security and land use more efficient. Inter-cropping practices differed significantly among CPS (p < 0.05) and were ranked as garden > semi-plantation > semi-forest coffee. Forest coffee systems were unsuitable for inter-cropping but exhibited greater shade-tree diversity, distribution, and evenness. Dominant shade tree species included Albizia, Erythrina, Milletia, and Acacia. The socio-economic characteristics of coffee producers in southwestern Ethiopia are shaped by production environment, cultural heritage, and technological access. The sustainability of future coffee production systems depends on improved land management, adoption of appropriate technologies, and better market integration. Limited access to improved technologies remains a major constraint to coffee productivity and long-term sustainability among smallholder farmers.
