Legume integration in smallholder farming systems for food security and resilience to climate change

PLOS One | https://doi.org/10.1371/journal.pone.0327727 August 13, 2025 6 / 20 farm productivity (aggregate farm product per year divided by land farmed in that specific year) was not surprisingly differing among the HH educational status. As the level of the HH head education increased, farm productivity did not improve. On the other hand, male-headed HHs had significantly higher mean farm productivity per year per hectare, probably since male- headed HHs had access to farm inputs, more formal interaction with agricultural extension service providers, and become better in adopting varieties & implementing agronomic management (Table 1). The study explores the relationship between family size and crop production in a region, revealing that larger families face challenges in food insecurity due to limited resources sharing, and addressing this requires enhancing cropping intensity and diversification. In terms of land possessions, some respondents had no arable land and relied on the cottage industry for a living, while the largest land possession reported was 8 ha (Table 2). The farmers annually cultivated between 0.1 and 6 hectares of the arable land mentioned, resulting in significant diversity in both agricultural practices and crop yields. Production records indicated that a maximum of 16 tons was harvested per year, contrasting with the minimal yield of 0.1 tons during the same period, highlighting the wide range of output across farms. Majority of their output was used for domestic use (in most cases up to 90%) and savings during tough times (common

for farmers with surplus farm produce) that have been strongly varied from 10–90% among the HHs. Wealthier farmers or households sold enough to generate income and have modest living and sent their children to school. Livestock possession was predominantly concentrated in lowland regions, with the highest of TLU (38.52) documented in the pastoralist territories of the South Omo zone. A limited number of individuals residing in the highland areas of the region exhibited a complete absence of livestock ownership, resulting in a pronounced variability in livestock possession at the individual household level. Consequently, the data demonstrated considerable variability, accompanied by excep- tionally high standard deviations (Table 2). Farm challenges cross the study area Most people in the study area depended on agriculture for their livelihood, and agricultural goods were their main source of income. Population increases shrunk the farm sizes and made the people farm more fragmented land, affecting their entire productivity. The growing population and its effects on development and agriculture have been the subject of con- tentious debate locally and globally. In the current study area, there was a conversion of natural forests, sacred forest, and uncultivable areas into farmlands (personal observations during survey of the study sites), indicating the challenges of land shortage and the strategy of the communities to cope with challenges. Land fragmentation

The research sites displayed a noteworthy degree of land fragmentation, despite the continual expansion of agricultural territories and the establishment of new settlements in remote regions. Land fragmentation intensity has been defined at the district and zone levels within the research area (Fig 1. tif). Table 2. Farm characteristics owned by the households. Variables Mean sd Min Max Arable land (ha) 1.71 1.21 0 8 Land cropped 1.33 0.95 0.1 6 Product in tons/farm/year 2.89 2.94 0.1 16 Farm product sold (%) 18.59 15.52 25 75 Farm product consumed (%) 68.07 23.4 10 90 Farm product saved (%) 12.02 12.78 10 90 Tropical Livestock Unit (TLU) 5.2 4.22 0 38.52 https://doi.org/10.1371/journal.pone.0327727.t002