The incorporation of Azolla into soil was investigated in this study for its potential to enhance soil fertility by influencing key parameters, including organic carbon (Organic-C) content, total nitrogen (Total-N), and cation exchange capacity (CEC). This study was conducted in a controlled greenhouse environment using a Completely Randomized Design (CRD) with eight treatments and three replications. The primary objective was to evaluate the effects of Azolla on soil quality, particularly in improving organic matter content and nitrogen (N) retention, both of which are essential for sustainable agricultural management. The findings indicate that Azolla incorporation led to a 29% increase in soil Organic-C and a 21% increase in Total-N compared to control treatments (p < 0.05). Additionally, CEC was enhanced by 33.4%, demonstrating improved nutrient retention capacity. A strong positive correlation was observed between Organic-C content, soil pH, and CEC, suggesting that Azolla contributes to optimizing soil nutrient dynamics. These results highlight the capacity of Azolla to function as a biofertilizer, improving soil fertility and nitrogen cycling while reducing dependence on synthetic fertilizers. The potential of Azolla to serve as an eco-friendly amendment aligns with sustainable agricultural practices aimed at enhancing soil health and long-term productivity. The findings contribute to the growing body of research on biofertilizers, offering valuable insights for soil management strategies that prioritize environmental sustainability and resource efficiency.

The rice crisis represents a significant threat to food security and economic stability in Southeast Asia, a region where rice serves as the primary staple for the majority of the population. This crisis is exacerbated by a confluence of factors, including climate change, crop failures, and restrictive export policies, as exemplified by the El Niño phenomenon and India’s 2023 rice export ban. Rising rice prices have been linked to increased social unrest, with the potential to trigger widespread demonstrations across affected nations. To proactively address this issue, the restlessness indicator was introduced as a predictive tool, integrating key variables such as rice prices, consumption patterns, and per capita income. This study employs a Spatio-Temporal Autoregressive (STAR) model to forecast restlessness values across six Southeast Asian countries—Indonesia, the Philippines, Thailand, Vietnam, Malaysia, and Cambodia—from 2024 to 2028. The STAR (5,1) model was identified as the optimal framework, achieving a Mean Absolute Percentage Error (MAPE) of 15.1%. The forecasting results indicate that none of the analyzed countries are projected to enter a state of unprecedented restlessness during the specified period, suggesting that no severe rice crisis is anticipated within this timeframe. These findings provide critical insights for policymakers and stakeholders, enabling the development of preemptive strategies to mitigate potential food security challenges. The study underscores the utility of the restlessness indicator as a robust tool for monitoring and forecasting rice-related crises, contributing to the broader discourse on sustainable food systems in Southeast Asia.

Seed Quality is an important area of agriculture and directly influences crop yield and germination percentage. Visual examination forms the foundation of traditional seed testing techniques, which are cumbersome, inflexible, and inefficient for effective assessment. This study proposed an automated approach to seed quality assessment based on physical measurement using machine learning and image processing techniques. Snapshots of the new seeds were captured and underwent feature extraction, segmentation, and image improvement to explore notable morphological attributes, such as size and colour. To tag seeds as "good" or "bad" based on physical characteristics, Support Vector Machines (SVMs) are used as a reference model. Rather, Convolutional Neural Networks (CNNs) have been utilised for deep feature extraction and classification. Experimental findings indicate that CNNs perform better than conventional machine learning models, with a scalable and highly accurate method of seed quality assessment. Future use will utilise quantum machine learning to improve prediction and facilitate sustainable, precision agriculture. The improved framework, optimised with great care for onion seeds, is a major breakthrough in increasing the agricultural productivity of onion cultivation.

Shallot farming in Indonesia has significant risks, primarily due to production variability and price instability. These risks deter farmers from adopting strategies involving higher risk tolerance levels. Risk aversion varies across individuals, leading to differences in decision-making processes. The study examines income risk levels in shallot farming. It explores farmers' behaviors in response to these risks in two distinct regions: the highlands of Karanganyar Regency, Central Java, and the lowlands of Bantul Regency, Daerah Istimewa Yogyakarta. A total of 200 shallot farmers were randomly selected for structured interviews to assess their risk behavior and the factors influencing it. The analysis reveals that shallot farming entails a high degree of income risk, and the highland areas exhibit a greater coefficient of variation (0.574) compared to the lowlands (0.544). Approximately 65% of highland farmers and 80% of lowland farmers were observed to be risk-averse concerning their shallot farming activities. Key factors influencing risk behavior include land size, household size, farming experience, age, frequency of crop failure, education, income, and farming location. Notably, farming experience, education, household size, and income positively impact risk behavior, increasing farmers' likelihood of adopting risk-taking strategies. The primary source of income risk was production variability, exacerbated by staggered planting schedules. This study highlights the importance of synchronizing planting schedules and strengthening farmer group networks to improve planning, marketing, input procurement, and knowledge exchange. The findings also provide a foundation for policymakers to design regulations that optimize planting times and mitigate income risks in shallot farming.

The determinants of successful partnership models between coffee farmers and key stakeholders—comprising private enterprises, cooperatives, and governmental bodies—were investigated to enhance productivity and sustainability within the coffee sector in Mojokerto, Indonesia. A mixed-methods approach was employed, integrating factor analysis and multiple linear regression modeling to examine the predictive influence of partnership dimensions. Four core dimensions—economic, social, cultural, and agroclimatic—were evaluated through exploratory factor analysis to uncover latent structures underpinning partnership success. The analysis resulted in the identification of four principal components: socio-economic exchange dynamics, socio-economic connectivity of agriculture, capital networks and socio-economic experience, and economic and educational networks. These components were subsequently used as independent variables in a multiple linear regression model, where partnership success was operationalized through kernel weight outputs as a proxy for productivity performance. The regression model accounted for 84.19% of the variance in partnership success, indicating strong explanatory power. The findings underscore the critical role of non-economic dimensions—particularly social connectivity and education—in driving effective partnerships, alongside traditional economic considerations. Policy implications include the need to design intervention strategies that enhance farmers' access to capital, strengthen educational and training programs, and encourage participation in socio-economic networks. While the model demonstrates strong internal validity within the context of the coffee industry, its applicability to other agricultural commodities remains to be tested. Further research is recommended to validate these findings across diverse agro-industrial contexts, thereby supporting the development of inclusive and scalable partnership models. This study contributes empirical evidence to inform stakeholder decision-making and promote resilient, equity-driven frameworks for agricultural collaboration.