Agricultural Decision Research Articles

No-tillage farming enhances widespread nitrate leaching in the US Midwest

Abstract Conservation tillage has been promoted as an effective practice to preserve soil health and enhance agroecosystem services. Changes in tillage intensity have a profound impact on soil nitrogen cycling, yet their influence on nitrate losses at large spatiotemporal scales remains uncertain. This study examined the effects of tillage intensity on soil nitrate losses in the US Midwest from 1979–2018 using field data synthesis and process-based agroecosystem modeling approaches. Our results revealed that no-tillage (NT) or reduced tillage intensity (RTI) decreased nitrate runoff but increased nitrate leaching compared to conventional tillage. These trade-offs were largely caused by altered water fluxes, which elevated total nitrate losses. The structural equation model suggested that precipitation had more pronounced effects on nitrate leaching and runoff than soil properties (i.e. texture, pH, and bulk density). Reduction in nitrate runoff under NT or RTI was negatively correlated with precipitation, and the increased nitrate leaching was positively associated with soil bulk density. We further explored the combined effects of NT or RTI and winter cover crops and found that incorporating winter cover crops into NT systems effectively reduced nitrate runoff but did not significantly affect nitrate leaching. Our findings underscore the precautions of implementing NT or RTI to promote sustainable agriculture under changing climate conditions. This study provides valuable insights into the complex relationship between tillage intensity and nitrate loss pathways, contributing to informed decision-making in climate-smart agriculture.

The Early Prediction of Kimchi Cabbage Heights Using Drone Imagery and the Long Short-Term Memory (LSTM) Model

Accurate and timely crop growth prediction is crucial for efficient farm management and food security, particularly given challenges like labor shortages and climate change. This study presents a novel method for the early prediction of Kimchi cabbage heights using drone imagery and a long short-term memory (LSTM) model. High-resolution drone images were used to generate a canopy height model (CHM) for estimating plant heights at various growth stages. Missing height data were interpolated using a logistic growth curve, and an LSTM model was trained on this time series data to predict the final height at harvest well before the actual harvest date. The model trained on data from 44 days after planting (DAPs) demonstrated the highest accuracy (R2 = 0.83, MAE = 2.48 cm, and RMSE = 3.26 cm). Color-coded maps visualizing the predicted Kimchi cabbage heights revealed distinct growth patterns between different soil types, highlighting the model’s potential for site-specific management. Considering the trade-off between accuracy and prediction timing, the model trained on DAP 36 data (MAE = 2.77 cm) was deemed most suitable for practical applications, enabling timely interventions in cultivation management. This research demonstrates the feasibility and effectiveness of integrating drone imagery, logistic growth curves, and LSTM models for the early and accurate prediction of Kimchi cabbage heights, facilitating data-driven decision-making in precision agriculture for improved crop management and yield optimization.

A Framework for Agricultural Intelligent Analysis Based on a Visual Language Large Model

Smart agriculture has become an inevitable trend in the development of modern agriculture, especially promoted by the continuous progress of large language models like chat generative pre-trained transformer (ChatGPT) and general language model (ChatGLM). Although these large models perform well in general knowledge learning, they still have certain limitations and errors when facing agricultural professional knowledge about crop disease identification, growth stage judgment, and so on. Agricultural data involves images and texts and other modalities, which play an important role in agricultural production and management. In order to better learn the characteristics of different modal data in agriculture, realize cross-modal data fusion, and thus understand complex application scenarios, we propose a framework AgriVLM that uses a large amount of agricultural data to fine-tune the visual language model to analyze agricultural data. It can fuse multimodal data and provide more comprehensive agricultural decision support. Specifically, it utilizes Q-former as a bridge between an image encoder and a language model to achieve a cross-modal fusion of agricultural images and text data. Then, we apply a Low-Rank adaptive to fine-tune the language model to achieve an alignment between agricultural image features and a pre-trained language model. The experimental results prove that AgriVLM demonstrates great performance in crop disease recognition and growth stage recognition, with recognition accuracy exceeding 90%, demonstrating its capability to analyze different modalities of agricultural data.

Propose and implement a Rule-Based System to Predict Crop Yield Production

Accurate meteorological forecasting plays a pivotal role in agricultural decision-making, particularly in determining crop yield potential and optimizing agricultural practices. This study investigates the application of data mining techniques in meteorological forecasting to enhance crop yield prediction accuracy Preliminary findings suggest that data mining techniques, including machine learning algorithms, neural networks, and ensemble methods, offer significant potential for improving the accuracy and reliability of meteorological forecasts for crop yield prediction. In conclusion, this study underscores the potential of data mining techniques in improving meteorological forecasting for crop yield potential assessment.

Simulating farm structural change dynamics in Thessaly (Greece) using a recursive programming model

Although the policy impacts on farms accumulate year by year, most farm decision models focus on short-term decisions, evaluating policies based on snapshots. Structural changes are gradually built; therefore, farm decision models should consider the sequences within the period under study. Multiyear data from the arable sector in Thessaly, Greece, have fed a newly developed recursive programming model mainly to simulate farm structural change dynamics. The proposed model incorporates new evidence on the strategic decision of arable crop farms regarding their remaining in the production system and farm expansion. Results reveal an evident gradual farmland concentration in relatively large farms, accompanied by a gradual expansion of the most profitable cropping activities, verifying the real-world survival strategy of farms.

The impact of rainfall on beef cattle growth across diverse climate zones

Genetics, animal husbandry and the feedbase all impact cattle growth. Australia’s cattle feedbase covers 40% of the continent and encompasses diverse climates and landscapes, making stocking rate decisions challenging. Of the factors contributing to climate change, rainfall is a primary determinant of feedbase growth and, with this, cattle growth. Understanding the interplay between rainfall and cattle growth across the diverse Australian landscape is thus critical to aid farmer decision-making. However, revealing such interactions between landscape and rainfall for cattle growth for such decision-making has until now been infeasible due to a lack of sufficient temporal and spatial cattle growth data. The Optiweigh (OW) system has been deployed across Australia’s extensive beef production systems as a voluntary weighing unit, opportunistically monitoring cattle liveweight (LW). This study determined the impact of rainfall on the temporal and spatial variability of beef cattle growth across three of Australia’s agro-climatic zones (grassland, subtropical and temperate), aiming to describe the diverse feedbase through patterns of LW variability. A total of 1.3 million cattle LW observations were collected from 82 026 cattle over 2 years (2020–2022). Rainfall data from the Australian Bureau of Meteorology were also collated for the closest meteorological station to each OW unit. Cattle LW average daily gain (ADG) was clustered by season and zone. A series of linear mixed models were used to examine ADG for each zone-season combination, with random effects for individual animals and farms, and fixed effects for climate zone and current and lagged rainfall. The overall mean ADG for the dataset was 0.68 kg/day, with greater growth variability between farms within a zone (SD: 0.349 ± 0.021 kg/day, estimate ± SE) than between cattle within a farm (SD: 0.229 ± 0.004 kg/day). This ADG variability can be partly attributed to the timing and amount of rainfall, with agro-climatic zones showing unique interactions between rainfall and ADG. Seasonal lagged rainfall effects were present in the grassland and temperate zones, while rainfall in the temperate zone had a year-round effect on cattle growth. Further, season-wise lagged rainfall had mixed effects on ADG, whereas rain occurring in a season reduced growth in the same season across zones (P < 0.001). These findings provide valuable initial insights into the variability of ADG across the landscape over time and markedly improve our understanding of the interplay between climate and Australia’s diverse feedbase, contributing to improved management strategies.

Agent-Based Modeling and Simulation for Farmer Decision Making

Shallots are a highly demanded, perishable commodity whose fluctuating prices pose significant risks to farmers, affecting their income stability. This research aims to model and simulate the selling decisions of shallot farmers using Agent-Based Modeling (ABM). The model represents the individual behaviors of farmers and buyers, focusing on decisions to either delay or immediately sell harvested shallots. Key variables such as market prices, inventory levels, and profit expectations were incorporated into the model. Primary data were collected from observations of shallot farmers in Nganjuk, East Java. In contrast, secondary data were sourced from the Indonesian Central Statistics Agency (BPS). The simulation results indicate farmers maximize their profits by selling shallots when the minimum sales profit margin reaches 20%. Farmers tend to postpone sales when market prices fall by drying the shallots, reducing market supply, and elevating prices. The study concludes that delaying sales can be an effective strategy to stabilize prices and enhance profitability, particularly when dried shallots command higher prices than fresh ones. These findings offer valuable insights for developing agricultural policies to optimize farmers' income while mitigating price volatility risks.

Variability of Winter Frosts in Central South America: Quantifying Mechanisms with Decision Trees

AbstractAgricultural production in Central South America (CSA) is substantially influenced by frost events. This study characterises and quantifies the physical processes leading to frost conditions in CSA from 1979 to 2022, focusing on three innovative aspects: regional frost properties, a novel multi-parametric upper-level jet description, and the quantification of underlying mechanisms through decision trees (DTs). The regionalisation analysis identifies five homogeneous frost regions in CSA. In all regions, the events tend to occur more frequently during the La Niña phase. Moreover, a significant increase in the frequency of widespread frost events has been observed in the Argentinean Pampas during the study period, primarily due to negative trends in minimum temperatures. Furthermore, the synoptic mechanisms triggering frosts, such as cold fronts and post-frontal anticyclones enhanced by subsidence near the subtropical jet (STJ) entrance, have not shown major long-term changes. To describe the jets, we compute six parameters for the STJ and seven for the polar front jet, including latitude, intensity, height, tilting, longitudinal extent, and branch number. DTs are used to identify key jet parameters linked to frost events, such as the latitude, longitudinal extent, and tilt of the Atlantic STJ. Frost likelihood increases when the STJ is north of 31°S, and the extension of the Atlantic STJ is longer than 35° and has a negative tilt. Finally, DTs focused on the onset and end of events highlight geopotential height anomalies and STJ extension as critical variables. These DTs provide concise and accessible information for agricultural decision-makers in CSA.

Exploring Female Agripreneurship: Insights from Grey Literature Using TCCM Framework

Women agripreneurs' expertise, abilities, and passion are some of the most effective strategies to improve their financial and social status. There is a wealth of systematic literature on women in entrepreneurship, as well as bibliometric reviews on women in entrepreneurship and sustainability. However, the systematic literature on women in agriculture or agribusiness on grey literature receives relatively little attention. The present study applies SPAR-4-SLR and TCCM frameworks to validate the research questions. Women frequently encounter disempowerment in terms of their participation, influence, and control over agricultural decision-making and income, which leads to the existence of gender disparity. Access to financial services could potentially reduce rural-to-urban migration by providing enhanced local economic opportunities. Suggestions provided for a supportive environment that values and respects the contributions of women in agriculture, facilitating the adoption of various technologies to break down barriers and reduce gender disparities. The investigation highlights the need for further research on the topics of female agripreneurship, female empowerment and digital technology, women's perspectives on technology and innovations in agriculture, and gender disparities in agripreneurship, all with the aim of achieving sustainable agriculture. The novelty of the study is the usage of the TCCM framework in the grey literature on women in agriculture and a proposed framework for technology-enhanced women's empowerment in agriculture.

Blockchain-Enhanced Traceability Framework for Smart Farming with Integrated Ontology-Based Data Standardization

This paper introduces a new conceptual framework to enhance crop traceability within smart farming environments. The framework merges blockchain technology with ontology-based data standardization to address persistent challenges in traditional traceability systems, such as data fragmentation and inconsistencies. The primary goal is establishing a transparent, reliable, and efficient traceability process from farm to table. The methodology employed in this research follows the Design Science Research Methodology (DSRM), involving iterative cycles of designing, building, and evaluating the proposed framework to ensure its effectiveness in real-world applications. The framework offers a promising solution to improve food safety, quality, regulatory compliance, and data-driven agricultural decision-making by capitalizing on blockchain's immutable and decentralized attributes and leveraging standardized data formats facilitated by ontologies. The proposed framework comprises three key components: ontology development, system integration, and blockchain implementation. Ontology development addresses data fragmentation and inconsistencies by providing a common language and structure for data exchange. Meanwhile, system integration ensures seamless data exchange among stakeholders, and blockchain implementation secures and verifies traceability records to maintain data integrity and trust. This integrated system is expected to foster trust and collaboration among all participants in the agricultural supply chain, thereby advancing the efficiency and sustainability of smart farming practices. Through continuous refinement and validation, this research aims to significantly contribute to the practical implementation of advanced traceability solutions in the agricultural sector, ensuring that these systems are robust, scalable, and adaptable to various farming contexts.

Feasibility of machine learning-based rice yield prediction in India at the district level using climate reanalysis and remote sensing data

CONTEXTYield forecasting, the science of predicting agricultural productivity before the crop harvest occurs, helps a wide range of stakeholders make better decisions around agricultural planning. OBJECTIVEThis study aims to investigate whether machine learning-based yield prediction models can capably predict Kharif season rice yields at the district level in India several months before the rice harvest takes place. METHODOLOGYThe methodology involved training 19 machine learning models such as CatBoost, LightGBM, Orthogonal Matching Pursuit, and Extremely Randomized Trees on 20 years of climate, satellite, and rice yield data across 247 of India's rice-producing districts. In addition to model-building, a dynamic dashboard was built understand how the reliability of rice yield predictions varies across district. RESULTS AND CONCLUSIONSThe results of the proof-of-concept machine learning pipeline demonstrated that rice yields can be predicted with a reasonable degree of accuracy, with out-of-sample R2, MAE, and MAPE performance of up to 0.82, 0.29, and 0.16 respectively. This performance outperformed test set performance reported in related literature on rice yield modelling in other contexts and countries. In addition, SHAP value analysis was conducted to infer both the importance and directional impact of the climate and remote sensing variables included in the model. Important features driving rice yields included temperature, soil water volume, and leaf area index. In particular, higher temperatures in August correlate with increased rice yields, particularly when the leaf area index in August is also high. Building on the results, a proof-of-concept dashboard was developed to allow users to easily explore which districts may experience a rise or fall in yield relative to the previous year. The dashboard show that the model may perform better in some regions than in others. For instance, the absolute percentage error for predicted versus actual yields ranged from an average of 7.1 % in districts in Uttarakhand to an average of 14.7 % in Uttar Pradesh. SIGNIFICANCEThis study underscores the potential for policymakers to consider scaling and operationalizing machine learning approaches to rice yield prediction in the context of agricultural early warning systems to deliver timely crop yield forecasts on a rolling basis throughout the season, thereby equipping agricultural decision-makers with the ability to make informed choices on irrigation scheduling, fertilizer application, and harvest planning to optimize crop output and resource use.

Motivation and opportunity may drive Tunisian farmers to reduce chemical pesticides in horticulture

This study investigates the sociopsychological factors influencing Tunisian farmers’ adoption of sustainable pest management techniques by employing the motivation-opportunity-ability framework and using partial least squares structural equation modelling. It is the first empirical study to test this framework in predicting farmers’ readiness to reduce pesticide usage in agriculture, distinguishing between intrinsic and extrinsic motivations. The findings reveal that both intrinsic and extrinsic motivations significantly impact farmers’ readiness to reduce pesticide use. Particularly, extrinsic motivations, such as peer influence and prevailing practices within the community, play a crucial role, highlighting the importance of social dynamics in sustainable agricultural decisions. Additionally, opportunities to adopt non-chemical pest control methods, enhanced by access to information and technology, positively influence farmers’ intentions to minimize pesticide use. These results suggest that enhancing farmers’ motivation through targeted educational programs and fostering opportunities via supportive policy environments are key strategies to promoting sustainable pest management practices in the region.

Research of real-time corn yield monitoring system with DNN-based prediction model.

The real-time monitoring of corn yield by a combine harvester is a critical data source for constructing the yield histogram, which significantly benefits precision management and decision-making in modern precision agriculture. While widely used, the current photoelectric sensor-based yield monitoring method has limitations. It detects the corn height on each scraper and calculates the yield through a geometric formula. However, it neglects the noticeable difference in the corn stacking patterns affected by factors such as feeding volume, terrain, and driving speed. This oversight often results in low accuracy and poor stability in the prediction of corn yield, highlighting the need for a more advanced approach. To resolve this, we employ EDEM discrete element simulation to demonstrate the large difference of corn stacking patterns on the scraper of the elevator corresponding to feeding volume. Then, we develop a real-time monitoring system on our self-developed double elevator testing rig for carrying out a composite dataset for training three machine learning algorithm-based models, namely Deep Neural Networks (DNN), Gradient Boosting Machine (GBM), and Random Forest (RF). Importantly, these models have undergone rigorous validation under various feeding volumes, ensuring their robustness and reliability. The auxiliary elevator speed is meticulously set at 150r/min, 225r/min, and 450r/min, providing a comprehensive performance assessment. The results denote that the DNN model performs best and is stable, with a coefficient of determination (R2) of 0.998, root mean square error (RMSE) of 0.526, and mean absolute error (MAE) of 0.425. The paper also performs field experiments to test the proposed three prediction models and the system. The results also denote the DNN-based prediction model's best performance for the lowest relative error of 2.29% and the highest average accuracy of 97.85%. Consequently, the proposed real-time corn yield monitoring system achieves high accuracy and reliability for the combine harvester applications.

Enhancing food crop classification in agriculture through dipper throat optimization and deep learning with remote sensing

Remote sensing images (RSIs), a keystone of modern agricultural technology, refer to spectral or visual data captured from drones, satellites, or aircraft without direct physical contact with the Earth's surface. These images provide a wide-ranging view of agricultural landscapes, providing valuable insights into land use, crop health, and environmental conditions. Agricultural food crop classification, a vital application within precision agriculture, includes the detection and classification of different crops cultivated in a certain region. Traditionally reliant on manual techniques, the development of technologies, particularly the incorporation of RSIs, has revolutionized this process. Agricultural food crop classification has become more sophisticated and automated by harnessing the wealth of data received from RS, which facilitates precise management and monitoring of crops on a large scale. Deep learning (DL), a branch of artificial intelligence, plays a more effective role in these synergies. The incorporation of DL into the RSI analysis enables high-precision and efficient detection of various crop types, assisting more informed decision-making in agriculture. This study proposes a new Dipper Throat Optimization Algorithm with Deep Learning based Food Crop Classification (DTOADL-FCC) algorithm using Remote Sensing Imaging for Agricultural Resource Management. The DTOADL-FCC method aims to apply DL algorithms for the classification of different crop types. In the DTOADL-FCC method, fully convolutional network (FCN) based segmentation process is performed. Next, the DTOADL-FCC method exploits the SE-ResNet model for learning intrinsic and complex features. The DTOADL-FCC method makes use of DTOA for the hyperparameter tuning process. Lastly, the classification of crop types takes place using the extreme learning machine (ELM) model. The study utilizes mathematical formulations including activation functions, loss functions, fitness calculations, and iterative update processes. A brief set of simulations showcases that the DTOADL-FCC method achieves remarkable performance over other techniques with much improved results.

Cost-effectiveness of temporary electric fencing for preventing wild boar intrusion into small-scale paddy farmlands

Crop damage by wild boar is a serious problem globally. Temporary electric fences, installed only during periods when the damage occurs, are widely used in Japan to prevent wild boar intrusion. The exclusion method is effective, but the cost burden involved makes considering cost-effectiveness necessary, especially for small-scale paddy rice farmers. Group fences may be used to reduce the cost burden. Although enclosing an area this way is effective, it may lead to free-riding when the scale of the project becomes large. Measures like reducing the number of people involved are suggested, but few indicators exist regarding the appropriate scale of management. Therefore, in Chiba Prefecture, where the distribution of wild boar is expanding, we conducted i) a six-year survey on the amount of damage to paddy rice fields in the Awa region; ii) an evaluation of the cost-effectiveness of temporary electric fences; and iii) an extrapolation of the cost-effectiveness to the entirety of Chiba Prefecture. The results show that the optimal installation length can be calculated by considering the maximum damage area, but without a government subsidy, the fence cost-effectiveness will be low because the installation cost will exceed the maximum damage amount. A group enclosure of small farmlands, which is not currently cost-effective in Chiba Prefecture, is recommended as a management decision for planners and farmers in future damage control plans. Although the empirical data were specific to small-scale farmers in Japan, the validation process was considered broadly applicable.

Gendered implications for climate change adaptation among farmers in Madagascar

ABSTRACT Climate change is impacting farmers worldwide; none more so than the millions of smallholder farmers who rely entirely on rain-fed agriculture. Adapting agricultural practices is an important strategy to mitigate climate change’s harmful impacts to food security and livelihoods. Researchers have increasingly sought to understand the factors influencing farmer decision-making on adaptation practices but application of important theories in low-income country contexts is lacking. Therefore, we apply the Protection Motivation Theory (PMT) to examine farmer intention to adopt agricultural practices in response to climate change based on a survey of 328 smallholder farmers in rural Madagascar. Using structural equation modeling (SEM) to describe the relationship between social connectedness (e.g., interactions with others about farming-related issues), PMT constructs (threat and coping appraisal), and intention to adapt agricultural practices to climate change, we find that coping appraisal is a stronger predictor of intention to adapt than threat appraisal. Further, SEM results show a significant negative relationship between social connectedness and threat appraisal – farmers with greater social connectedness perceive climate change to be less of a threat. Additionally, as female farmers are significantly less socially connected than male farmers, this study provides further evidence of the gendered vulnerabilities to climate change among smallholders. This research has implications for the design of policies and interventions that consider farmer social networks, critical for reducing threat and building coping capacity, in supporting adaptation to climate change among vulnerable farmer populations.

Stevia rebaudiana: Reporting data from the sweetener crop for researchers and smallholder farming assistants

In science, technology and assistance to smallholder farming or large crop acreages of Stevia rebaudiana is necessary to report data using information components to facilitate investors to understand the local field conditions to make assertive economic decisions for long periods. Stevia rebaudiana produces consistent natural sweeteners with safe performance as reported in a multitude of scientific publications. The leaf chemicals components are suitable to humans with healthy disorders affected by the consumption of sugarcane or many other industrial sweeteners. Otherwise, aspects related to the agriculture of sweeteners production still require additional discussion to improve qualitative and quantitative traits necessary for cultivating cost-effective crops. The current text highlighted some agriculture decisions necessary to make the crop more profitable to investors as researchers, agriculture influencers, farmer assistants and consultants of companies looking for alternatives within agro-ecological systems.

Optimizing Rice Production Forecasting Through Integrating Multiple Linear Regression with Recursive Feature Elimination

Rice is a staple food for most Nigerians, making accurate yield prediction is crucial for food security. This study addresses the limitations of traditional forecasting methods by employing Multiple Linear Regression (MLR) coupled with Recursive Feature Elimination (RFE) to predict rice yield in Adamawa and Cross River states, characterized by distinct agroclimatic conditions. Utilizing climatic data and historical yield records from 1990 to 2022, we trained and evaluated MLR and compared the MLR results with two other machine learning models (XGBoost, and K Nearest Neighbours). RFE-optimized feature selection identified All-sky Photosynthetically Active Radiation (PAR) as a key factor. MLR demonstrated a very stable prediction performance with R² values of 0.90 and 0.92 for Adamawa and Cross River, respectively, after RFE. This research contributes to developing advanced Agro-information systems, supporting informed agricultural decision-making, and enhancing Nigeria's food security.

Ratio of physical model parameters can retrieve aggregate size from different types of soil in cultivated regions

Soil aggregate size, which is a proxy used to guide agricultural decisions and tillage management, can be estimated using optical remote sensing techniques. However, limited investigation has been conducted into the potential of using a physical model to retrieve soil aggregate size (< 2 mm) from different types of soil. This study is based on the multi-angular spectral measurements of seven soil samples from five soil types with 14 aggregate size fractions collected in Northeast China, three versions of the Hapke model were inverted using the Bayesian method. The findings confirmed that all three versions of the Hapke model can well characterize the angular reflection characteristics of all soil samples with different aggregate sizes. The inverted photometric parameters such as single scattering albedo ω, shape parameter b, and asymmetry parameter c were found to be sensitive to soil aggregate size, but the relationships rely on soil types because of the dependence of parameters related to soil composition. In order to obtain a general model that can be applied to different types of soil, the ratio of parameters (RoP) = (b + c)/ω, which is controlled by the external structure of soil aggregates, was proposed to retrieve the aggregate size from different soil types. Results show that the RoP can robustly capture the aggregate size of the soil with high accuracy and is insensitive to soil types. The combination of photometric parameters related to soil aggregate size provides a new method for retrieving the structural properties of the soil by eliminating interfering factors.

Understanding how governance arrangements within agricultural supply chains influence farmers' SAP adoption for adaptation and mitigation practices

CONTEXTIn order for farmers in developing countries to combat the effects of climate change, sustainable agricultural practices (SAP) have been promoted, but their adoption rate remains modest. Prior research examining psychological and socio-economic issues has identified farmers' limited knowledge and capital contribute to this problem. However, supply chain systems that require farmers to follow formal and informal rules (governance arrangements) have received little attention in previous studies. OBJECTIVEFrom a system thinking standpoint, this article seeks to comprehend farmers decision regarding SAP adoption by examining the influence of governance arrangements on the socioeconomic and psychological aspects of farmers. This paper concentrates specifically on governance arrangements between farmers and midstream actors, to whom farmers sell their products directly. METHODThis study utilises a qualitative research methodology, specifically employing multiple case studies. A system thinking approach utilising a Causal Loop Diagram was applied to comprehend farmers decision regarding SAP adoption. The case studies cover supply chain systems that have different governance arrangements. The study focused on examining the rice supply chains in Indramayu District and the fresh vegetable supply chains in West Bandung District, located in West Java, Indonesia. RESULTS AND CONCLUSIONThis paper found that different governance arrangements exhibit distinct influences, which in turn have varying impacts on farmers' decisions to adopt SAP. Governance arrangements that foster greater market integration and exert influence on the psychological (e.g., knowledge and information) and socio-economic (e.g., provision of production inputs and price assurance) aspects are regarded as the most effective approach to promoting the adoption of SAP by farmers. Furthermore, this paper revealed that midstream actors play a crucial role in promoting the adoption of SAP among farmers. SIGNIFICANCEThis paper contributes to the body of knowledge on how governance arrangements within agricultural supply chains influence farmers decision regarding SAP adoption. Specifically, it investigates the influence of midstream actors in facilitating farmers' adoption of SAP by addressing their psychological and socioeconomic requirements. The findings imply that government supports must be extended to midstream actors that have made efforts to persuade their farmers to implement SAP as a means of addressing climate change, as well as implementing midstream actors-to-farmer extension initiatives alongside the government's existing farmer-to-farmer extension programmes.