AI-Driven Nutrient Optimization Systems for Precision Agriculture: A Data-Integrated Approach to Reducing Nutrient Runoff and Enhancing Soil Health in Agricultural Systems

International Journal for Multidisciplinary Research (IJFMR)

E-ISSN: 2582-2160 ● Website: www.ijfmr.com ● Email: editor@ijfmr.com

IJFMR260585286 Volume 8, Issue 5, September-October 2026 1

AI-Driven Nutrient Optimization Systems for
Precision Agriculture: A Data-Integrated
Approach to Reducing Nutrient Runoff and
Enhancing Soil Health in Agricultural Systems

Obebe Olufisayo Andrew1, Akinola Ebenezer Abayomi2

1,2Kentucky State University, Frankfort, USA.

Abstract
Nutrients play a fundamental role in plant growth, soil fertility, and overall agricultural productivity.
However, inefficient fertilizer application in conventional farming systems often leads to nutrient
imbalance, reduced soil quality, and increased environmental degradation. The emergence of precision
agriculture and artificial intelligence (AI) has introduced new opportunities for improving nutrient
management through data -driven decision -making. This study investigates an AI -driven nutrient
optimization system designed to enhance fertilizer efficiency while reducing nutrient runoff and improving
soil health. Findings indicate that AI -based systems significantly improve nutrient use efficiency by
enabling site-specific fertilizer recommendations and adaptive management strategies. The integration of
multi-source agricultural data enhances prediction accuracy for soil nutrient status and crop demand, while
also supporting early identification of nutrient loss risks such as runoff and leaching. The study concludes
that AI-driven nutrient optimization provides a sustainable and scalable approach for modern agriculture,
supporting both productivity enhancement and environmental protection through intelligent, data -
integrated farming systems.

Keywords: Artificial intelligence, Digital agriculture, Nutrient, Soil health, Data integration, Machine
learning, Sustainable agriculture.

1. Introduction
   Nutrients are essential chemical elements required for plant growth, crop development, and productive
   agricultural systems. Primary macronutrients such as nitrogen, phosphorus, and potassium support root
   growth, photosynthesis, and biomass production, while micronutrients help maintain plant health and
   physiological balance (Noulas et al., 2023).
   Balanced nutrient supply is therefore critical to crop yield, food production, and long -term soil fertility.
   Nutrient management refers to the planned application and monitoring of nutrient inputs to meet crop
   demand efficiently while protecting environmental resources. Its importance has increased as agriculture
   faces the challenge of raising productivity while sustaining soil quality and ecosystem health (Khambalkar
   et al., 2025). Efficient fertili zer use is now recogni zed as vital for food security, w aste reduction, and
   environmental protection (Penuelas et al., 2023). Proper nutrient management also improves soil structure
   and biological activity (Mamatha et al., 2024).

International Journal for Multidisciplinary Research (IJFMR)

E-ISSN: 2582-2160 ● Website: www.ijfmr.com ● Email: editor@ijfmr.com

IJFMR260585286 Volume 8, Issue 5, September-October 2026 2

Conventional fertilizer practices often rely on uniform application methods that i gnore variations in soil
conditions, crop demand, and climate. This can cause over -application in some areas and nutrient
deficiency in others, lowering efficiency and increasing production costs (Cai et al., 2025). Poor nutrient
allocation may also reduce profitability through unnecessary input expenditure and lower yields
(Rodrigues, 2022). Environmental concerns linked to inefficient nutrient use have grown significantly.
Excess nitrogen and phosphorus often move beyond the crop root zone through runoff and leaching,
causing eutrophication, water pollution, and ecosystem decline (Mahmud et al., 2021). Nutrient losses can
also release greenhouse gases such as nitrous oxide, increasing climate-related risks (Wang et al., 2024).
Artificial intelligence has created new opportunities for precision agriculture and nutrient optimi zation.
Machine learning, deep learning, and predictive analytics can process complex farm data, detect patterns,
predict crop responses, and generate accurate fertilizer recommendations (Liakos et al., 2018). Kamilaris
and Prenafeta-Boldú (2018) noted that deep learning is especially effective for analyzing imagery, sensor
streams, and other large agricultural datasets. This makes AI highly suitable for adaptive nutrient decision-
making.
Digital agriculture has also expanded access to data from soil sensors, weather stations, satellite imagery,
drones, machinery, and geospatial platforms. These big data systems enable real-time collection, storage,
and analysis of farm information, impro ving management efficiency and decision quality (Wolfert et al.,
2017). Wider availability of such technologies has accelerated precision agriculture adoption across many
farming systems (Lee et al., 2021).
Conventional nutrient management methods based only on soil tests or fixed fertilizer schedules are often
inadequate for modern farming conditions. Effective decisions now require simultaneous consideration of
soil variability, climate uncertainty, crop demand, production cost, and environmental risk. AI -enabled
systems can integrate these variables into a smarter and more responsive optimization framework (Ennaji
et al., 2023). This study aims to develop an AI -driven nutrient optimi zation system for precision
agriculture using a data-integrated framework that improves soil health while reducing nutrient runoff.

2. Foundations of AI-Driven Nutrient Management Systems
   AI-driven nutrient management systems are built on the integration of agronomic knowledge, digital
   technologies, and data-based decision suppor t tools. These systems aim to improve fertilizer efficiency,
   crop productivity, and environmental sustainability through accurate nutrient planning and site -specific
   management (Penuelas et al., 2023). The growing demand for sustainable food production has made
   advanced nutrient management an important component of modern agriculture.
   Precision agriculture provides the operational foundation for these systems by using sensors, geospatial
   tools, and variable -rate technologies to match nutrient application wi th field conditions. This approach
   reduces waste, improves input efficiency, and supports better farm management decisions (Rodrigues,
   2022). Site -specific nutrient management technologies have also expanded the ability to monitor soil
   variability and crop nutrient demand across agricultural landscapes (Vullaganti et al., 2025). Artificial
   intelligence strengthens nutrient management by analy zing complex datasets and generating predictive
   recommendations. Machine learning models can evaluate soil properties, weather trends, and crop
   responses to recommend suitable fertilizer strategies (Liakos et al., 2018). Deep learning techniques are
   also valuable for processing sensor signals, imagery, and other large agricultural datasets (Kamilaris &
   Prenafeta-Boldú, 2018). Recent studies show that AI applications in nutrient management can optimi ze
   chemical fertilizer use while reducing environmental pres sure. These systems support efficient nutrient

International Journal for Multidisciplinary Research (IJFMR)

E-ISSN: 2582-2160 ● Website: www.ijfmr.com ● Email: editor@ijfmr.com

IJFMR260585286 Volume 8, Issue 5, September-October 2026 3

allocation, lower losses, and improved sustainability outcomes (Ennaji et al., 2023; Cai et al., 2025) . As
digital farming continues to expand, AI-driven nutrient management systems are becoming a core
foundation for resilient and efficient agricultural production.
2.1 Concept of Nutrient Optimization in Precision Agriculture
Nutrient optimization in precision agriculture refers to the strategic management of plant nutrients to
supply crops with the right nutrient source, at the right rate, right time, and right place in order to maximize
productivity while minimizing waste and environmental harm (Miao, 2023). It integrates agronomic
knowledge with digital technologies such as sensors, geographic information sys tems (GIS), artificial
intelligence, and data analytics to improve fertilizer decisions based on field -specific conditions rather
than generalized recommendations (Mishra & Mishra, 2024). This approach has become increasingly
important as farming systems seek to improve efficiency, profitability, and sustainability.
Macronutrients such as nitrogen (N), phosphorus (P), and potassium (K) are required in large quantities
and are fundamental for photosynthesis, root development, energy transfer, and yield formation (Gürbüz,
2025). Micronutrients, including zinc, iron, boron, and mangan ese, are needed in smaller amounts but
remain essential for enzyme activation, metabolic regulation, and crop quality (Shrivastav et al., 2020).
Deficiency or imbalance of either macro- or micronutrients can significantly reduce crop performance and
soil productivity.
A major objective of nutrient optimization is improving nutrient use efficiency (NUE), which measures
how effectively applied nutrients are converted into crop output. Low NUE remains a widespread
challenge across major global cropping systems, leading to economic losses and environmental pollution
(Liu et al., 2025). Practices such as balanced fertilization, crop -responsive nutrient scheduling, and
integrated nutrient management help increase efficiency and resilience under changing climatic conditions
(Govindasamy et al., 2023; Pierre & Regmi, 2026).
Modern agriculture is gradually shifting from conventional blanket fertilizer application, where identical
rates are applied uniformly across fields, to site-specific nutrient management. This transition recognizes
spatial variability in soil fertility, mo isture, and crop demand within farms. By using precision
technologies, farmers can apply nutrients variably according to actual field needs, thereby reducing over-
application, lowering costs, and enhancing environmental sustainability (Akhil et al., 2025; Hasan et al.,
2026).
2.2 Soil Health and Nutrient Dynamics in Agricultural Systems
Soil health refers to the continued capacity of soil to function as a living ecosystem that sustains plants,
animals, and humans while supporting environmental quality and agricultural productivity. In sustainable
agriculture, healthy soils are essential because they regulate water, cycle nutrients, store carbon, and
maintain biodiversity needed for long-term crop production (Lehmann et al., 2020). Soil health assessment
has evolved beyond crop yield measurements to include physical, chemical, and biological indicators that
determine the resilience and productivity of agricultural soils (Karlen et al., 2019). Sustainable farming
systems, therefore, depend on maintaining soil functionality rather than focusing only on short -term
nutrient inputs (Bünemann et al., 2018).
Soil organic matter is one of the most important components of healthy soils because it improves nutrient
retention, aggregation, water holding capacity, and cat ion exchange processes. It also serves as a major
energy source for soil microorganisms that drive decomposition and nutrient mineralization (Audette et
al., 2021). Higher soil organic matter levels can enhance rhizosphere nutrient cycling by influencing
microbial metabolic activity and bacterial ecological strategies (Liang et al., 2026). Microbial

International Journal for Multidisciplinary Research (IJFMR)

E-ISSN: 2582-2160 ● Website: www.ijfmr.com ● Email: editor@ijfmr.com

IJFMR260585286 Volume 8, Issue 5, September-October 2026 4

communities play a central role in converting organic residues into plant -available nutrients such as
nitrogen, phosphorus, and sulfur, thereby sustaining soil fertility and crop growth (Dincă et al., 2022). The
use of organic fertilizers has also been shown to increase microbial diversity and stabilize nutrient cycling
processes in agricultural soils (Chen et al., 2025).
Soil pH strongly affects nutrient solubil ity and plant uptake because highly acidic or alkaline conditions
can reduce the availability of essential nutrients and increase toxicity risks. Maintaining optimum pH is
therefore necessary for balanced fertilization and efficient nutrient absorption (Ha rtemink & Barrow,
2023). Soil texture also influences nutrient dynamics because clay -rich soils generally retain more
nutrients and water than sandy soils, which are more prone to leaching losses (Zheng et al., 2021). In
addition, soil structure determines aeration, root penetration, and water infiltration, while soil moisture
controls microbial activity and nutrient transport within the soil profile (Dewangan et al., 2024).
Improper fertilizer management can degrade soil productivity over time. Excessive fertilizer use may lead
to soil acidification, salinity buildup, nutrient imbalances, loss of beneficial microbes, and contamination
of surrounding ecosystems (Yanardağ, 2024). Conversely, under-fertilization can deplete nutrient reserves,
reduce soil biolo gical activity, and lower crop yields over successive seasons. Therefore, l ong-term
agricultural sustainability depends on balanced nutrient replenishment strategies that maintain soil fertility
without causing environmental damage (Bhattacharyya & Furtak, 2022).
2.3 Artificial Intelligence in Precision Agriculture
Artificial intelligence (AI) in precision agriculture refers to the use of computational systems that mimic
human decision -making to improve farming efficiency, productivity, and sustainability. Core AI
methodologies include machine learning, deep learning, and predictive analytics, which enable computers
to identify patterns, learn from historical data, and generate recommendations for farm management
decisions (Liakos et al., 2018). Machine learning models are increasingly applied to complex agricultural
datasets such as weather records, sensor outputs, and crop observations to support real-time management
decisions (Wang & Wang, 2026). In soil science, machine learning tools are also used to ana lyze soil
variability, classify soil properties, and estimate nutrient conditions with greater precision than
conventional methods (Padarian et al., 2020).
In remote sensing, one of the most important applications of AI in agriculture for crop monitoring, drones
and smart sensors generate large volumes of field data that AI systems can analyze to detect crop stress,
diseases, moisture deficiencies, and growth anomalies at early stages (Talaviya et al., 2020). This allows
farmers to respond quickly to changi ng field conditions and reduce production risks. AI also supports
automated monitoring systems that continuously track crop performance throughout the growing season,
improving farm responsiveness and operational efficiency (Mishra & Mishra, 2024).
AI has also made major contributions to crop yield prediction through the use of advanced learning
algorithms. Deep learning and reinforcement learning models can combine climate, soil, management,
and historical yield data to forecast crop output with improved a ccuracy (Elavarasan & Vincent, 2020).
Hybrid neural network models such as convolutional neural networks-recurrent neural networks ( CNN-
RNN) frameworks have proven effective in modeling temporal and spatial relationships affecting
agricultural yield performance (Khaki et al., 2020). Recent developments integrating explainable artificial
intelligence further improve trust and transparency in predictive systems used for farm planning (Mohan
et al., 2025).
In nutrient management, AI-driven recommendation systems help determine appropriate fertilizer types,
rates, timing, and placement according to site -specific crop and soil needs. These systems process data

International Journal for Multidisciplinary Research (IJFMR)

E-ISSN: 2582-2160 ● Website: www.ijfmr.com ● Email: editor@ijfmr.com

IJFMR260585286 Volume 8, Issue 5, September-October 2026 5

from soil tests, sensors, satellite imagery, and management records to optimize nutrient applications while
reducing waste and environmental losses (Miao, 2023). Predictive analytics also enables farmers to
anticipate nutrient deficiencies and make proactive fertilizer adjustments before yield losses occur (Jabed
& Murad, 2024).
Decision support systems (DSS ) powered by AI are becoming central tools for fertilizer optimization in
precision agriculture. These platforms integrate agronomic knowledge with real -time data analytics to
generate actionable recommendations for farmers and agronomists. AI -based DSS can improve nutrient
use efficiency, lower production costs, and enhance sustainability outcomes through evidence -based
fertilizer decisions (Mishra & Mishra, 2024). Despite these benefits, AI adoption in agriculture faces
several limitations and challenges. Many farming regions still lack reliable digital infrastructure, quality
datasets, and the technical expertise required to implement advanced systems. High investment costs,
limited digital literacy, and uncertainty about economic returns can also slow ad option among farmers
(Ansari et al., 2024). In addition, model accuracy may decline when algorithms trained in one region are
transferred to different soils, climates, or crop systems, highlighting the need for localized calibration and
continuous system i mprovement (Liakos et al., 2018). Figure 1 illustrates the distribution of machine
learning techniques applied in agricultural systems, highlighting their usage across different agricultural
domains such as crop production, soil management, water systems, and farmer decision support (Cravero
et al., 2026).

Figure 1: Machine Learning Techniques Employed in the Development of Agricultural Systems
(Cravero et al., 2026).
2.4 Data Integration in Smart Nutrient Management
Data integration is central to smart nutrient management because effective fe rtilizer decisions depend on
combining multiple agricultural data sources into a unified system. Important sources include field

International Journal for Multidisciplinary Research (IJFMR)

E-ISSN: 2582-2160 ● Website: www.ijfmr.com ● Email: editor@ijfmr.com

IJFMR260585286 Volume 8, Issue 5, September-October 2026 6

sensors, Internet of Things (IoT) devices, drones, satellite imagery, geographic information systems (GIS),
and farm management records that provide information on soil conditions, crop growth, weather patterns,
and management history (Soussi et al., 2024). These technologies improve the accuracy and timeliness of
nutrient recommendations by generating continuous farm-level data (Kour & Arora, 2020).
Data fusion techniques are used to combine soil, crop, and climate datasets into meaningful decision
outputs. Machine learning models can integrate heterogeneous data streams to identify nutrient
deficiencies, predict crop demand, and optimize fertilizer application across variable field conditions
(Wang & Wang, 2026). In soil management, data -driven models are increasingly used to analyze soil
variability and nutrient status for precision interventions (Padarian et al., 2020). AI -based systems also
support crop yield forecasting by linking environmental and management data for proactive nutrient
planning (Elavarasan & Vincent, 2020).
Despite these advantages, several challenges affect integrated agricultural data systems; Poor sensor
calibration, missing values, inconsistent data formats, and weak interoperability between devices can
reduce decision accuracy and system reliability (Fizza et al., 2021). Limited digital infrastructure, adoption
costs, and compatibility issues across platforms also constrain implementation, particularly in developing
farming regions (Ansari et al., 2024). Real -time interoperable systems are therefore needed to ensure
seamless communication among smart farm technologies (El-Dosuky, 2025).
Real-time analytics plays a vital role in nutrient decision -making by enabling immediate responses to
changing soil and crop conditions. AI systems can process incoming field data continuously and
recommend timely fertilizer adjustments that improve nutrient use efficiency an d reduce losses (Mishra
& Mishra, 2024). Such intelligent platforms are becoming essential tools for future on -farm planning,
productivity enhancement, and sustainable nutrient stewardship (Hasan et al., 2026). Table 1 presents
examples of IoT -based smart agriculture platforms used for data collection, monitoring, and decision
support in precision farming systems. These platforms integrate sensor technologies, cloud systems, and
analytics tools to enhance agricultural productivity and resource management (Soussi et al., 2024).

Table 1: Examples of Platforms Available for Smart Agriculture (Soussi et al., 2024)
IoT Platform Application
Phytoprove (2019) Develops tools for early, rapid, and non -destructive determination of
nitrogen and water supply status of plants.
Telemetry2u Provides integrated hardware and software system for measuring
temperature and irrigation control.
Sensoterra Provides soil moisture monitoring using integrated sensor systems.
CropX (2013) Enables measurement of electrical conductivity, soil moisture, and
temperature for precision nutrient management.
ONFARM Provides dashboard -based farm data visualization for real -time
decision support.
Phytech Uses data -driven insights to optimize irrigation and improve plant
health and resource efficiency.
Semios (2020) Provides real-time monitoring of pests, diseases, and crop health for
yield optimization.

International Journal for Multidisciplinary Research (IJFMR)

E-ISSN: 2582-2160 ● Website: www.ijfmr.com ● Email: editor@ijfmr.com

IJFMR260585286 Volume 8, Issue 5, September-October 2026 7

2.5 Nutrient Runoff and Environmental Sustainability
Nutrient runoff and leaching are major environmental challenges associated with modern agricultural
production. They occur when excess nutrients, especially nitrogen and phosphorus, are transported from
farmland through surface runoff, erosion, irrigation drainage, or downward water movement beyond the
root zone. These losses ar e often caused by excessive fertilizer application, poor timing of nutrient
placement, heavy rainfall events, unsuitable irrigation practices, and limited crop nutrient uptake (Juncal
et al., 2023). In many farming systems, inefficient nutrient management increases the likelihood of
nutrients escaping into surrounding ecosystems instead of being utilized by crops (Del Rossi et al., 2023).
The environmental consequences of nutrient runoff are significant. Nutrient enrichment of rivers, lakes,
and coastal waters can trigger eutrophication, leading to algal blooms, oxygen depletion, and degradation
of aquatic ecosystems. Excessive nitrogen and phosphorus use for major crops has also been linked to
biodiversity loss in freshwater environments and wider ecologica l disruption (Jwaideh et al., 2022). In
addition, agricultural soils are an important source of greenhouse gas emissions such as nitrous oxide,
which contributes substantially to climate change (Basheer et al., 2024). Nitrogen losses through
volatilization and gaseous emissions, therefore, represent both environmental and agronomic inefficiency
(Pan et al., 2022). These concerns have increased policy and regulatory attention toward nutrient
management in agriculture. Governments and environmental agencies a re adopting stricter standards on
fertilizer use, water quality protection, and emission reduction targets. Economic analyses show that
nutrient pollution imposes substantial social and environmental costs, creating a strong case for improved
regulations, incentives, and sustainable nutrient stewardship policies (Del Rossi et al., 2023).
Predictive tools are increasingly important for protecting the environment while maintaining farm
productivity. Decision-support systems, nutrient budgeting models, runoff forecasting tools, and emission
risk assessments help farmers anticipate losses before they occur and adjust management practices
accordingly. Such tools support nutrient neutrality strategies by improving fertilizer timing, placement,
and application rate s based on environmental risk conditions (Juncal et al., 2023). Advanced predictive
approaches are therefore essential for climate -smart and environmentally sustainable agriculture (Pan et
al., 2022).

3. AI-Driven Nutrient Optimization System Design
   AI-driven nutrient optimization systems are designed to combine sensing technologies, data analytics, and
   intelligent decision tools to improve fertilizer management in precision agriculture. These systems
   typically integrate soil sensors, climate data, satellit e imagery, and farm records into a unified digital
   platform that supports accurate nutrient recommendations and efficient resource use (Soussi et al., 2024).
   A core component of system design is data fusion, where multiple data streams are combined using
   machine learning models to provide a complete understanding of field conditions. Sensor fusion
   techniques help interpret soil moisture, nutrient levels, crop health, and weather variability for timely
   nutrient planning and adaptive farm management (Wang & W ang, 2026). AI systems can therefore
   generate more precise recommendations than conventional single-source approaches. The soil intelligence
   layer focuses on collecting and analyzing soil information such as pH, organic matter, nutrient content,
   and moistu re status. Machine learning tools are increasingly used to classify soils, predict nutrient
   deficiencies, and map field variability for site-specific interventions (Padarian et al., 2020). This improves
   the accuracy of fertilizer placement and reduces unnecessary application.

International Journal for Multidisciplinary Research (IJFMR)

E-ISSN: 2582-2160 ● Website: www.ijfmr.com ● Email: editor@ijfmr.com

IJFMR260585286 Volume 8, Issue 5, September-October 2026 8

The nutrient optimization layer converts field data into practical fertilizer decisions, AI models evaluate
crop demand, soil nutrient supply, and environmental conditions to recommend suitable fertilizer type,
rate, timing, and placem ent. Precision nutrient management systems can significantly improve nutrient
use efficiency while lowering production costs and minimizing environmental losses (Miao, 2023).
Decision support systems are also central to modern nutrient optimization platforms. These tools provide
farmers with dashboards, mobile alerts, and scenario analysis for better on -farm decisions. AI-supported
systems enhance farmer decision -making by translating complex data into clear management actions
(Kaushik et al., 2025). Data-driven farm management platforms are increasingly important for improving
productivity and sustainability (Mishra & Mishra, 2024). Real -time capability is another critical design
feature, Modern precision agriculture systems continuously monitor changing field conditions and update
recommendations instantly, allowing rapid responses to crop stress, rainfall events, or nutrient
deficiencies. Real-time architectures are becoming essential for scalable and sustainable smart farming
systems (Bhat et al., 2026).
3.1 Systems-Level Framework Architecture
A systems-level framework architecture for AI -driven nutrient optimization is designed as an integrated
digital platform that connects data collection, analysis, optimization, and advisory functions for efficient
fertilizer management (Soussi et al., 2024). Such frameworks support precision agriculture by
coordinating multiple technologies into one intelligent decision environment (Bhat et al., 2026). The first
layer is the sensing layer, where soil sensors, weather stations, drones, satellite systems, and smart farm
devices collect real -time data on nutrient levels, moisture, crop health, and environmental conditions
(Soussi et al., 2024). Continuous sensing improves field visibility and creates the raw data needed f or
nutrient decisions (Miller et al., 2025).
The second layer is the analytics layer, where machine learning models process raw datasets into useful
insights such as soil classification, nutrient deficiency detection, and crop demand forecasting (Padarian
et al., 2020). Data analytics also improves accuracy by combining multiple data sources into a clearer
representation of field variability (Wang & Wang, 2026). The third layer is the optimization layer, where
AI algorithms recommend the best fertilizer source, rate, timing, and placement according to crop and soil
requirements (Miao, 2023). This layer aims to maximize nutrient use efficiency while reducing runoff,
leaching, and unnecessary input costs (Mishra & Mishra, 2024). The fourth layer is the advisor y layer,
where recommendations are delivered through dashboards, mobile applications, or automated farm
systems for practical use by farmers and agronomists (Kaushik et al., 2025). Clear decision interfaces help
transform technical outputs into actionable farm operations (Mishra & Mishra, 2024).
Data flow within the framework begins with field -level sensing and moves through transmission,
analytics, optimization, and final decision outputs in real time (Bhat et al., 2026). Feedback loops allow
the system to update recommendations whenever field conditions change (Wang & Wang, 2026).
Interoperability is essential because these systems must connect across sensors, machinery, cloud
platforms, and farm software from different providers (Soussi et al., 2024). Sta ndardized data exchange
improves compatibility, scalability, and long-term adoption of digital nutrient management systems (Bhat
et al., 2026).
3.2 Soil Intelligence and Analytics Layer
The soil intelligence and analytics layer is responsible for collecting, processing, and interpreting soil data
to support precise nutrient management decisions. It forms a core component of AI -driven nutrient
optimization systems by transforming raw soil information into actionable insights for farmers and

International Journal for Multidisciplinary Research (IJFMR)

E-ISSN: 2582-2160 ● Website: www.ijfmr.com ● Email: editor@ijfmr.com

IJFMR260585286 Volume 8, Issue 5, September-October 2026 9

agronomists (Padarian et al., 2020). Smart sensing systems now enable continuous digital monitoring of
soil conditions across agricultural fields (Soussi et al., 2024).
This layer gathers soil physical data such as texture, bulk density, moisture, and temperature using in-field
sensors and remote sensing tools (Miller et al., 2025). It also captures chemical data including pH,
nitrogen, phosphorus, potassium, salinity, and organic carbon , needed for fertilizer planning (Hossen et
al., 2021). Biological indicators such as microbial activity and soil health status are increasingly monitored
through AI-enabled soil restoration systems (Pandey et al., 2025). Whereas, before analysis, raw soil
datasets require preprocessing to improve quality and consistency. Common steps include noise removal,
handling missing values, normalization, scaling, and integration of multi -source data from sensors,
satellites, and farm records (Wang & Wang, 2026). Effective soil data sharing frameworks also improve
interoperability and access across institutions and digital platforms (Padarian & McBratney, 2020).
Artificial intelligence models are then applied for soil classification and nutrient deficiency detection.
Machine learning algorithms can classify soils according to fertility zones, texture gr oups, and
management capability classes with higher speed and accuracy than conventional methods (Padarian et
al., 2020). Deep learning systems are also being used to detect nutrient problems through soil image
analysis and nutrient profiling tools (Pandey et al., 2025). Soil properties often differ significantly within
the same farm, making uniform fertilizer application inefficient. In line with this, machine learning models
using satellite, weather, clay, and yield data can map field heterogeneity and pr edict spatial nutrient
patterns for site -specific nutrient management (Kammerlander et al., 2025). Such analytics improve
fertilizer targeting, reduce waste, and enhance environmental sustainability (Soussi et al., 2024).
3.3 Nutrient Optimization Layer
The nutrient optimization layer is the decision engine of an AI-driven nutrient management system, where
field data are converted into fertilizer recommendations that improve productivity and sustainability. This
layer uses computational models to determin e the right nutrient source, rate, timing, and placement for
specific crop conditions (Miao, 2023). Data -driven nutrient planning is increasingly recognized as
essential for efficient farm management under changing environmental conditions (Mishra & Mishra ,
2024). Machine learning-based fertilizer recommendation systems analyze historical and real -time farm
data to generate accurate nutrient prescriptions. These systems can learn relationships between soil
fertility, crop response, and management practices to improve fertilizer efficiency over time (Wang &
Wang, 2026). AI -based farm decision tools are also being developed to guide crop and nutrient
management through automated recommendations (Kaushik et al., 2025).
Additionally, effective nutrient optimization requires integration of soil status, crop nutrient demand, and
climatic variables. Soil nutrient levels, moisture conditions, crop growth stage, temperature, and rainfall
patterns all influence nutrient availability and crop uptake (Elavarasan & Vincent, 2020). Combining these
variables allows the system to make more adaptive and location -specific fertilizer decisions (Mishra &
Mishra, 2024).
Multi-objective optimization is important because fertilizer decisions must balance yield improvement,
production cost, and environmental protection. Studies show that low nutrient use efficiency remains
widespread across major crops, highlighting the need for smarter allocation strategies (Liu et al., 2025).
Improving nitrogen use efficiency is especially critical for sustaining crop productivity while reducing
nutrient losses under climate stress (Govindasamy et al., 2023). The nutrient optimization layer also
supports variable-rate and adaptive nutrient application strategies. Instead of uniform fertilizer application,
nutrients are supplied according to within -field variability and changing crop needs during the season

International Journal for Multidisciplinary Research (IJFMR)

E-ISSN: 2582-2160 ● Website: www.ijfmr.com ● Email: editor@ijfmr.com

IJFMR260585286 Volume 8, Issue 5, September-October 2026 10

(Akhil et al., 2025). Precision nutrient technologies can therefore enhance agronomic efficiency while
lowering environmental risks associated with fertilizer misuse (Pierre & Regmi, 2026).
3.4 Environmental Risk Prediction Layer
The environmental risk prediction layer focuses on forecasting nutrient losses from agricultural systems,
particularly runoff, leaching, and gaseous emissions. It uses AI and hydrological models to estimate how
nutrients move through soil and water systems under different management and environmental conditions
(Juncal et al., 2023). Such predictive approaches are increasingly important for supporting nutrient
neutrality and sustainable land management strategies (Shao et al., 2021). Risk prediction models
incorporate key environmental drivers such as rainfall intensity, terrain slope, soil properties, and land
management practices. These variables strongly influence nutrient transport pathways, especially under
extreme weather conditions that increase runoff and erosion risk (Yang et al., 2025). Machine learning
approaches further improve prediction accuracy by capturing complex interactions between hydrological
and agricultural factors (Elsayed et al., 2024).
Nutrient leaching and runoff risks are also influenced by fertilizer application type and timing. Studies
show that different fertilization strategies significantly affect nitrogen and phosphorus losses in
agricultural watersheds (Reisinger et al., 2025). Similar findings confirm that nutrient loss patterns vary
across soil types, crop systems, and seasonal biomass conditions (Assa et al., 2024).
AI-based models are increasingly used to simulate nutrient concentr ations in surface water and predict
pollution loads in agricultural catchments. These models integrate spatial and temporal data to estimate
how nutrients move from fields into surrounding water bodies (Lee et al., 2024). Machine learning
approaches have d emonstrated strong performance in identifying key drivers of nutrient runoff,
particularly precipitation and land-use intensity (Elsayed et al., 2024). Threshold -based alert systems are
an important component of environmental risk prediction. These systems trigger warnings when nutrient
levels or environmental conditions exceed safe limits, enabling timely intervention to reduce pollution
risks (Juncal et al., 2023). Such systems support compliance with environmental regulations and promote
sustainable fertilizer use practices (Shao et al., 2021).
3.5 Decision Support and Advisory Layer
The decision support and advisory layer transforms AI -driven nutrient model outputs into practical
guidance for farmers, ensuring that complex analytics are usable at the field level. It typically operates
through farmer-oriented dashboards and mobile-based platforms that present real-time recommendations
on fertilizer use, crop management, and soil conditions (Soussi et al., 2024). These systems improve
accessibility by deliv ering actionable insights directly to end users in simplified formats (Miao, 2023).
Explainable AI (XAI) plays a key role in this layer by improving transparency and trust in nutrient
recommendations. XAI techniques help users understand why specific fertilizer or management decisions
are suggested, making AI systems more interpretable and reliable for agricultural decision-making (Mishra
& Mishra, 2024). This transparency is essential for increasing the adoption of AI-based advisory tools in
farming systems (Kaushik et al., 2025).
Scenario analysis is another important function of the advisory layer, allowing farmers to evaluate different
management options before implementation. AI -based decision support systems can simulate outcomes
such as yield response , fertilizer efficiency, and environmental impact under different input strategies
(Elavarasan & Vincent, 2020). Such scenario -based tools support proactive planning and improved
resource allocation in agriculture (Hasan et al., 2026). Human expertise rema ins essential in AI -driven
advisory systems, particularly through the integration of agronomists and extension officers into decision

International Journal for Multidisciplinary Research (IJFMR)

E-ISSN: 2582-2160 ● Website: www.ijfmr.com ● Email: editor@ijfmr.com

IJFMR260585286 Volume 8, Issue 5, September-October 2026 11

workflows. These hybrid systems combine machine intelligence with expert validation to improve the
accuracy and contextual relevance of recommendations (Mishra & Mishra, 2024). Studies show that
human-in-the-loop designs enhance trust, usability, and adoption of agricultural AI systems (Soussi et al.,
2024).

4. Implementation Strategies and Operational Considerations
   Successful implementation of AI -driven nutrient optimization systems requires robust digital
   infrastructure capable of collecting, storing, and processing large volumes of agricultural data. Smart
   sensors, IoT networks, cloud platforms, and edge computing systems are essential for enabling continuous
   data exchange and real -time decision support in farming environments (Soussi et al., 2024). Real -time
   system architectures further improve responsiveness and operational efficiency under changing field
   conditions (Bhat et al., 2026). Reliable model deployment also depends on high-quality datasets and strong
   analytical frameworks. Machine learning systems require accurate soil, climate, crop, and management
   data for training, calibration, and validation before field application (Cravero et al., 2026). Soil parameter
   prediction models demonstrate the importance of integrating satellite, weather, and yield data to strengthen
   model accuracy and robustness (Kammerlander et al., 2025).
   Interoperability is another major operational consideration because agricultural technologies often operate
   on different hardware and software platforms. Seamless integration among sensors, machinery, farm
   software, and advisory systems improves usability and scalability of AI solutions (Wang & Wang, 2026).
   Standardized data-sharing systems are therefore necessary for efficient smart farming ecosystems (Soussi
   et al., 2024). Human capacity and user adoption are equally important for successful implementation.
   Farmers need practical interfaces such as dashboards and mobile tools that convert complex analytics into
   clear management actions (Kaushik et al., 2025). Data -driven advisory systems become more effective
   when combined with farmer knowledge, extension support, and agronomic expertise (Mishra & Mishra,
   2024). Ultimately, l ong-term sustainability depends on economic viability, system maintenance, and
   adaptability across crops and regions. Precision nutrient management systems must remain cost-effective
   while continuously updating models as new seasonal data become available (Miao, 2023). The integration
   of IoT and AI technologies is increasingly viewed as essential for scalable and future -ready agricultural
   systems (Miller et al., 2025).
   4.1 Data Infrastructure and Governance
   Effective implementa tion of AI -driven nutrient optimization systems depends on a strong data
   infrastructure capable of storing, processing, and exchanging large volumes of agricultural information.
   Big data architectures such as cloud platforms, edge computing systems, and agricultural data warehouses
   are increasingly used to manage sensor, satellite, climate, and farm management datasets in real time (Ngo
   et al., 2019). Modern intelligent agricultural ecosystems also rely on scalable digital infrastructures that
   support continuous analytics and automated decision -making (Cravero et al., 2026). Cloud computing
   systems improve accessibility by allowing farmers, agronomists, and institutions to access shared data and
   decision tools remotely. These systems enable rapid processing of high-volume datasets generated from
   IoT devices and precision agriculture technologies (Soussi et al., 2024). Real -time cloud -edge
   architectures also improve system responsiveness under changing field conditions (Bhat et al., 2026).
   Data ownership, pri vacy, and cybersecurity remain major governance concerns in digital agriculture.
   Farmers often generate valuable operational data, yet uncertainty over ownership rights, access control,
   and commercial use can limit willingness to share information (Rouzky et al., 2025). Secure governance

International Journal for Multidisciplinary Research (IJFMR)

E-ISSN: 2582-2160 ● Website: www.ijfmr.com ● Email: editor@ijfmr.com

IJFMR260585286 Volume 8, Issue 5, September-October 2026 12

frameworks are therefore necessary to protect sensitive farm data from misuse, unauthorized access, and
cyber threats (Wang & Wang, 2026). Standardization and interoperability are essential because
agricultural technologie s often use incompatible data formats and disconnected platforms. Shared
standards improve communication between sensors, machinery, farm software, and analytics tools across
different providers (San Emeterio de la Parte et al., 2023). Open -data and open -source approaches can
further enhance transparency, innovation, and system integration in precision agriculture ecosystems
(Jeppesen et al., 2022).
Governance models for shared agricultural data ecosystems should balance innovation, collaboration, and
user protection. Federated platforms and cooperative data -sharing systems are increasingly proposed to
allow multiple stakeholders to benefit from common datasets while retaining local control of sensitive
information (Pincheira et al., 2025). Such governance approaches are important for building trust and
supporting long-term adoption of AI-enabled nutrient management systems (Rouzky et al., 2025)
4.2 Model Development and Validation
Effective nutrient optimization systems depend on selecting AI and optimizati on models that match the
nature of agricultural data and decision objectives. Machine learning techniques such as random forests,
support vector machines, neural networks, and reinforcement learning are widely used because they can
model complex relationships between soil, crop, and climate variables (Liakos et al., 2018). Model choice
should also consider computational efficiency, interpretability, and scalability for real farm environments
(Cravero et al., 2026).
Model development requires structured training, testing, calibration, and validation procedures to ensure
accurate predictions. Historical datasets are commonly divided into training and testing subsets, while
cross-validation methods are used to reduce overfitting and improve generalization performance (Padarian
et al., 2020). Calibration with field observations further strengthens reliability of nutrient recommendation
and soil prediction models (Kammerlander et al., 2025). According to Mishra and Mishra (2024),
managing uncertainty and bias is essential because agricultural datasets may contain noise, missing values,
seasonal anomalies, or sampling imbalance. Poor -quality data can reduce model accuracy and create
biased recommendations that do not reflect field realities (Mishra & Mishra, 2024). Therefore, robust
models require preprocessing, sensitivity analysis, and explainable AI methods that help users understand
prediction confidence and decision logic (Mohan et al., 2025).
Continuous learning is increasingly important in dynamic farming sys tems where weather, soil fertility,
and management practices change over time. Updating models with seasonal observations and historical
farm records allows systems to improve prediction performance across successive production cycles
(Wang & Wang, 2026). This adaptive learning approach supports more resilient and location -specific
nutrient management decisions.
4.3 Policy, Regulatory, and Institutional Support
Effective implementation of AI -driven nutrient optimization systems requires supportive agricultu ral
policies that encourage adoption of precision farming technologies. Public policies that promote digital
agriculture, smart machinery, and data-driven input management can improve productivity while reducing
waste and environmental pressure (Cravero et al., 2026). Financial support measures such as subsidies and
innovation grants are also important for helping farmers overcome high initial investment costs (Sanyaolu
& Sadowski, 2024).
Environmental regulations play a central role in guiding fertilizer use and nutrient management practices.
Regulatory frameworks increasingly emphasize nutrient -use efficiency, reduced runoff, and compliance

International Journal for Multidisciplinary Research (IJFMR)

E-ISSN: 2582-2160 ● Website: www.ijfmr.com ● Email: editor@ijfmr.com

IJFMR260585286 Volume 8, Issue 5, September-October 2026 13

with sustainability targets related to soil and water protection (Juncal et al., 2023). Precision nutrient
technologies can help farmers meet these standards through more accurate timing, placement, and rates of
fertilizer application (Maaz et al., 2025).
Extension services and public institutions are essential for translating policy goals into farm -level
adoption. Adv isory agencies, research institutes, and agricultural ministries often provide training,
technical guidance, and demonstration programs that help farmers understand and apply precision
agriculture tools (Nkosi et al., 2025). Strengthening extension capacit y is particularly important in
developing regions where digital skills and technical awareness may be limited (Onyango et al., 2021).
Well-designed incentive structures can accelerate adoption of sustainable agricultural technologies.
Evidence shows that p ayments, tax incentives, cost -sharing schemes, and sustainability rewards can
encourage farmers to invest in environmentally beneficial innovations (Piñeiro et al., 2020). When aligned
with nutrient stewardship goals, such incentives support wider use of A I-enabled nutrient optimization
systems and long-term agricultural sustainability.
4.4 Stakeholder Engagement and Adoption
Stakeholder engagement is a central requirement for the successful deployment of digital and AI -driven
agricultural technologies because system effectiveness depends on collaboration among farmers,
agronomists, researchers, and agri -tech companies. Precision agriculture adoption is strongly influenced
by how well technologies align with farmer needs, contextual realities, and institutional support systems
(Soussi et al., 2024). Therefore, i ntegrated smart agricultural ecosystems depend on coordinated
participation across all actors in the value chain (Cravero et al., 2026).
Farmers play the primary role as end-users, while agronomists and extension agents act as intermediaries
who translate technical outputs into practical field decisions. Research shows that adoption is higher when
farmers receive continuous tec hnical support and when innovation systems connect them directly with
experts and service providers (Mishra & Mishra, 2024). Agri -tech companies and researchers contribute
by developing tools that integrate sensor data, machine learning models, and decision support systems for
real-world use (Wang & Wang, 2026).
However, adoption is often constrained by barriers such as high cost, limited digital literacy, inadequate
infrastructure, and lack of trust in automated systems. Studies indicate that financial lim itations and
insufficient training significantly reduce farmers’ willingness to adopt precision technologies (Cravero et
al., 2026). In addition, weak digital skills and limited access to infrastructure continue to slow down
technology diffusion in many fa rming communities (Miller et al., 2025). Capacity building and training
initiatives are essential to overcome these constraints and improve adoption outcomes. Continuous farmer
education, demonstration farms, and extension -led training programs help bridge the gap between
innovation and practice (Kaushik et al., 2025). Evidence also shows that digital literacy and hands -on
training significantly improve farmers’ confidence in using smart agricultural systems (Soussi et al., 2024).
Participatory design and a co-creation approach further enhance adoption by involving farmers directly in
the development of agricultural technologies. Co -design ensures that tools are user -friendly, context-
specific, and aligned with real farm challenges (Bhat et al., 2026). This collaborative model strengthens
trust, improves usability, and increases long-term sustainability of precision agriculture solutions (Mishra
& Mishra, 2024).
4.5 Scalability and Long-Term Sustainability
Scalability of AI-driven nutrient optimization systems depends on their ability to function across diverse
crops, farm sizes, and agro -ecological zones. Modern precision agriculture frameworks are increasingly

International Journal for Multidisciplinary Research (IJFMR)

E-ISSN: 2582-2160 ● Website: www.ijfmr.com ● Email: editor@ijfmr.com

IJFMR260585286 Volume 8, Issue 5, September-October 2026 14

designed as integrated ecosystems that combine sensing, analytics, and decision systems to support large-
scale deployment across heterogeneous environments (Cravero et al., 2026). These systems rely on
flexible digital architectures that can adapt to varying climatic, soil, and management conditions (Soussi
et al., 2024).
Economic feasibility is a key determinant of long-term adoption, particularly for smallholder farmers who
face cost and infrastructure constraints. Studies show that while AI-based precision agriculture improves
efficiency and productivity, its adoption is strongly influenced by afforda bility and access to technology
(Bhat et al., 2026). Scalable and low -cost digital solutions are therefore essential to ensure inclusivity
across different farming systems (Wang & Wang, 2026). System maintenance and continuous updating
are necessary to maintain model accuracy over time. Agricultural AI systems must be regularly retrained
using new seasonal and historical farm data to account for changing climate conditions, soil variability,
and management practices (Mishra & Mishra, 2024). Continuous learn ing frameworks help ensure that
predictive performance remains reliable across production cycles (Kammerlander et al., 2025).
Long-term sustainability of nutrient optimization systems is closely linked to their contribution to
regenerative and climate-smart agriculture. AI-enabled precision farming can improve resource efficiency,
reduce environmental degradation, and support soil restoration practices when properly implemented
(Miao, 2023). Evidence suggests that integrating AI with sustainable nutrient ma nagement strategies
enhances both productivity and environmental resilience in agricultural systems.

5. Conclusion
   AI-driven nutrient optimization systems represent a major advancement in modern agriculture by
   integrating data analytics, machine learning, and precision sensing into a unified decision -support
   framework. These systems enable more accurate nutrient management by combining soil, crop, and
   environmental information for improved fertilizer recommendations and resource efficiency.
   The transition f rom conventional fertilization practices to data -driven precision agriculture enhances
   productivity while reducing environmental impacts such as nutrient runoff, leaching, and soil degradation.
   By optimizing nutrient application at the right time and place, these systems improve soil health and long-
   term agricultural sustainability. Integrated data systems and real -time analytics strengthen decision -
   making by providing continuous monitoring of field conditions and enabling adaptive nutrient
   management. This improves responsiveness to climate variability and supports more efficient use of
   agricultural inputs.
   Effective implementation of AI -based nutrient management requires collaboration among farmers,
   researchers, policymakers, and technology developers. Strong institutional support, digital infrastructure,
   and capacity-building initiatives are essential for scaling these technologies across diverse agricultural
   systems. Overall, AI-enabled nutrient stewardship offers a sustainable pathway for improving global food
   production while protecting environmental resources, supporting climate-smart agriculture, and enhancing
   long-term soil fertility management.

References
3. Akhil, K., Renukaswamy, N. S., Sowjanya, J., & Aruna, K. S. (2025). Site -specific nu trient
management: Revolutionizing fertilization for future farming. INTERNATIONAL JOURNAL, 9(3),
544-549.

International Journal for Multidisciplinary Research (IJFMR)

E-ISSN: 2582-2160 ● Website: www.ijfmr.com ● Email: editor@ijfmr.com

IJFMR260585286 Volume 8, Issue 5, September-October 2026 15

4. Anas, M., Liao, F., Verma, K. K., Sarwar, M. A., Mahmood, A., Chen, Z. L., Li, Q., Zeng, X. P., Liu,
   Y ., & Li, Y . R. (2020). Fate of nitrogen in agriculture and environment: agronomic, eco-physiological
   and molecular approaches to improve nitrogen use efficiency. Biological research , 53(1), 47.
   https://doi.org/10.1186/s40659-020-00312-4
5. Ansari, M. N., Gupta, S., Kumar, D., & Hussain, B. (2024). Adoption challenges and incentives for
   IoT and AI in agriculture. In Agriculture 4.0 (pp. 355-372). CRC Press.
6. Assa, B. G., Bhowmick, A., & Cholo, B. E. (2024). Assessing Nitrate Leaching and Runoff
   Coefficients in the Dynamic Interplay of Seasonal Crop Biomass: A Study of Surface and Groundwater
   Nitrate Contamination in the Bilate Cropland Watershed. Environmental Advances, 16, 100528.
7. Audette, Y ., Congreves, K. A., Schneider, K., Zaro, G. C., Nunes, A. L., Zhang, H., & V oroney, R. P.
   (2021). The effect of agroecosystem management on the distribution of C functional groups in soil
   organic matter: A review. Biology and Fertility of Soils, 57(7), 881-894.
8. Basheer, S., Wang, X., Farooque, A. A., Nawaz, R. A., Pang, T., & Neokye, E. O. (2024). A review of
   greenhouse gas emissions from agricultural soil. Sustainability, 16(11), 4789.
9. Bhat, M. H., da Silva Rickiel, F., Sameer, B., Aeshna, S., & Moore, K. J. (2026). Precision Agriculture
   Through a Real-Time Systems Perspective: A Narrative Review. Agronomy, 16(5), 552.
10. Bhattacharyya, S. S., & Furtak, K. (2022). Soil –Plant–Microbe interactions determine soil biological
    fertility by altering rhizospheric nutrient cycling and biocrust formation. Sustainability, 15(1), 625.
11. Bünemann, E. K., Bongiorno, G., Bai, Z., Creamer, R. E., De Deyn, G., De Goede, R., ... & Brussaard,
    L. (2018). Soil quality–A critical review. Soil biology and biochemistry, 120, 105-125.
12. Cai, B., Shi, F., Geremew, B. A., Addis, A. K., Abate, M. C., Dessie, W., & Bayu, T. (2025). Precision
    agriculture techniques for optimizing chemical fertilizer use and environmental sustainability: a
    systematic review. Frontiers in Agronomy, 7, 1665444.
13. Chen, X., Ma, X., Liu, Z., Gu, H., Fang, H., Shen, Z., ... & Liu, J. (2025). Organic fertilizers increase
    microbial community diversity and stability slowing down the transformation process of nutrient
    cycling. Environmental Microbiome, 20(1), 130.
14. Coble, K. H., Mishra, A. K., Ferrell, S., & Griffin, T. (2018). Big d ata in agriculture: A challenge for
    the future. Applied Economic Perspectives and Policy, 40(1), 79-96.
15. Cravero, A., Sepúlveda, S., Gutiérrez, F., & Muñoz, L. (2026). From Precision Agriculture to
    Intelligent Agricultural Ecosystems: A Systematic Review of Machine Learning and Big Data
    Applications. Agronomy, 16(5), 516.
16. Del Rossi, G., Hoque, M. M., Ji, Y ., & Kling, C. L. (2023). The economics of nutrient pollution from
    agriculture. Annual Review of Resource Economics, 15(1), 105-130.
17. Dewangan, S. K., Kashyap, N., Tigga, D., & Tirkey, N. (2024). The role of physico-chemical properties
    in soil functionality: a literature review. EPRA International Journal of Research & Development,
    9(12), 167-173.
18. Dincă, L. C., Grenni, P., Onet, C., & Onet, A. (2022). Fertilization and soil microbial community: a
    review. Applied Sciences, 12(3), 1198.
19. Elavarasan, D., & Vincent, P. D. (2020). Crop yield prediction using deep reinforcement learning
    model for sustainable agrarian applications. IEEE access, 8, 86886-86901.
20. El-Dosuky, M. (2025). Real-time Framework for Interoperable Semantic-driven Internet-of-Things in
    Smart Agriculture. arXiv preprint arXiv:2510.05187.

International Journal for Multidisciplinary Research (IJFMR)

E-ISSN: 2582-2160 ● Website: www.ijfmr.com ● Email: editor@ijfmr.com

IJFMR260585286 Volume 8, Issue 5, September-October 2026 16

21. Elsayed, A., Rixon, S., Levison, J., Binns, A., & Goel, P. (2024). Machine learning models for
    prediction of nutrient concentrations in surface water in an agricultural watershed. Journal of
    Environmental Management, 372, 123305.
22. Ennaji, O., Vergütz, L., & El Allali, A. (2023). Machine learning in nutrient management: A review.
    Artificial Intelligence in Agriculture, 9, 1-11.
23. Fizza, K., Jayaraman, P. P., Banerjee, A., Georgakopoulos, D., & Ranjan, R. (2021). Evaluating sensor
    data quality in Internet of Things smart agriculture applications. Ieee Micro, 42(1), 51-60.
24. Govindasamy, P., Muthusamy, S. K., Bagavathiannan, M., Mowrer, J., Jagannadham, P. T. K., Maity,
    A., ... & Tiwari, G. (2023). Nitrogen use efficiency —a key to enhance crop productivity under a
    changing climate. Frontiers in Plant Science, 14, 1121073.
25. Gürbüz, M. A. (2025). Identification of Multiple Extraction Methods to be Used in Extraction of Macro
    and Micronutrients of Neutral and Alkaline Soils by a Multi -Criteria Decision -Making Technique
    (TOPSIS). Journal of Soil Science and Plant Nutrition, 25(2), 4669-4686.
26. Hartemink, A. E., & Barrow, N. J. (2023). Soil pH-nutrient relationships: the diagram. Plant and Soil,
    486(1), 209-215.
27. Hasan, T., Mudgal, G., Ramasamy, V ., Kumar, A., & Kugonza, D. R. (2026). Precision Futures:
    Leveraging Modern Technology for Intelligent On-Farm Planning and Production Efficiency. In Smart
    Sustainable Materials Advancing Re -manufacturing and Technology: Smart and Sustainable
    Development in Agricultural Technology: Ecological and Practical Applications (V olume IV) (pp.
    247-275). Cham: Springer Nature Switzerland.
28. Hossen, M. A., Diwakar, P. K., & Ragi, S. (2021). Total nitrogen estimation in agricultural soils via
    aerial multispectral imaging and LIBS. Scientific Reports, 11(1), 12693.
29. Howari, F. (2025). AI –Remote Sensing for Soil Variability Mapping and Precision Agrochemic al
    Management: A Comprehensive Review of Methods, Limitations, and Climate -Smart Applications.
    Agrochemicals, 5(1), 1.
30. Jabed, M. A., & Murad, M. A. A. (2024). Crop yield prediction in agriculture: A comprehensive review
    of machine learning and deep learning approaches, with insights for future research and sustainability.
    Heliyon, 10(24).
31. Jeppesen, J. H., Jacobsen, R. H., Jørgensen, R. N., & Toftegaard, T. S. (2022). Towards data -driven
    precision agriculture using open data and open source software. arXiv preprint arXiv:2204.05582.
32. Johnston, A. M., & Bruulsema, T. W. (2014). 4R nutrient stewardship for improved nutrient use
    efficiency. Procedia Engineering, 83, 365-370.
33. Juncal, M. J. L., Masino, P., Bertone, E., & Stewart, R. A. (2023). Towards nutrient neutrality: A review
    of agricultural runoff mitigation strategies and the development of a decision -making framework.
    Science of the Total Environment, 874, 162408.
34. Jwaideh, M. A., Sutanudjaja, E. H., & Dalin, C. (2022). Global impacts of nitrogen and phosphor us
    fertiliser use for major crops on aquatic biodiversity. The International Journal of Life Cycle
    Assessment, 27(8), 1058-1080.
35. Kamilaris, A., & Prenafeta-Boldú, F. X. (2018). Deep learning in agriculture: A survey. Computers and
    electronics in agriculture, 147, 70-90.
36. Kammerlander, C., Kolb, V ., Luegmair, M., Scheermann, L., Schmailzl, M., Seufert, M., ... & Schön,
    T. (2025). Machine learning models for soil parameter prediction based on satellite, weather, clay and
    yield data. arXiv preprint arXiv:2503.22276.

International Journal for Multidisciplinary Research (IJFMR)

E-ISSN: 2582-2160 ● Website: www.ijfmr.com ● Email: editor@ijfmr.com

IJFMR260585286 Volume 8, Issue 5, September-October 2026 17

37. Karlen, D. L., Veum, K. S., Sudduth, K. A., Obrycki, J. F., & Nunes, M. R. (2019). Soil health
    assessment: Past accomplishments, current activities, and future opportunities. Soil and tillage
    research, 195, 104365.
38. Kaushik, P., Yadav, S., Kumar, R., Kiran, K., Balakrishnan, M., & Abraham, K. (2025). AI -Driven
    Agriculture System for Crop Recommendation and Farmer Decision -Making Support. In 2025
    Modern Electronics Devices and Intelligent Communication Systems (MEDCOM) (pp. 1013 -1017).
    IEEE.
39. Khaki, S., Wang, L., & Archontoulis, S. V . (2020). A CNN-RNN framework for crop yield prediction.
    Frontiers in plant science, 10, 1750.
40. Khambalkar, P. A., Agrawal, S., Dhaliwal, S. S., Yadav, S. S., Sadawarti, M. J., Singh, A., ... & Afreen,
    N. (2025). Sustainable nu trient management balancing soil health and food security for future
    generations. Applied Food Research, 5(2), 101087.
41. Kour, V . P., & Arora, S. (2020). Recent developments of the internet of things in agriculture: a survey.
    Ieee Access, 8, 129924-129957.
42. Lee, C. L., Strong, R., & Dooley, K. E. (2021). Analyzing precision agriculture adoption across the
    globe: A systematic review of scholarship from 1999–2020. Sustainability, 13(18), 10295.
43. Lee, S., Bae, J. H., Lee, G., Shin, H., Yang, J. E., Kim, J., & Lim, K. J. (2024). Predicting Nutrient
    Runoff Load in Cultivated Lands using Best Management Practices with Machine Learning Models.
    Journal of Agricultural, Life and Environmental Sciences, 36(2), 152-163.
44. Lehmann, J., Bossio, D. A., Kögel -Knabner, I., & Rill ig, M. C. (2020). The concept and future
    prospects of soil health. Nature Reviews Earth & Environment, 1(10), 544-553.
45. Liakos, K. G., Busato, P., Moshou, D., Pearson, S., & Bochtis, D. (2018). Machine learning in
    agriculture: A review. Sensors, 18(8), 2674.
46. Liang, S., Ma, A., Zhang, T., Fan, X., Qian, D., Shao, J., ... & Xun, W. (2026). Soil organic matter
    enhances rhizosphere nutrient cycling by shaping bacterial life-history strategy and carbon metabolic
    preference. Applied Soil Ecology, 223, 107018.
47. Liu, J., Wang, H., Penuelas, J., Mou, J., Delgado -Baquerizo, M., Sardans, J., ... & Fang, L. (2025).
    Global-scale prevalence of low nutrient use efficiency across major crops. Nature Communications,
    16(1), 11036.
48. Lyu, J., Wang, X., Hou, S., Zeb, A., Zhu, H., & Xu, Y . (2023). Content variation and potential runoff
    loss risk of nutrients in surface water of saline-alkali paddy in response to the application of different
    nitrogen fertilizer types. Sustainability, 15(9), 7040.
49. Maaz, T. M., Dobermann, A., Lyons, S. E., & Thomson, A. M. (2025). Review of research and
    innovation on novel fertilizers for crop nutrition. npj Sustainable Agriculture, 3(1), 25.
50. Mahmud, K., Panday, D., Mergoum, A., & Missaoui, A. (2021). Nitrogen losses and potential
    mitigation strategies for a sustainable agroecosystem. Sustainability, 13(4), 2400.
51. Mamatha, B., Mudigiri, C., Ramesh, G., Saidulu, P., Meenakshi, N., & Prasanna, C. L. (2024).
    Enhancing soil health and fertility management for sustainable agriculture: A review. Asian J. Soil Sci.
    Plant Nutr, 10(3), 182-190.
52. Miao, Y . (2023). Precision nutrient management. In Encyclopedia of Digital Agricultural Technologies
    (pp. 1054-1061). Cham: Springer International Publishing.

International Journal for Multidisciplinary Research (IJFMR)

E-ISSN: 2582-2160 ● Website: www.ijfmr.com ● Email: editor@ijfmr.com

IJFMR260585286 Volume 8, Issue 5, September-October 2026 18

53. Miller, T., Mikiciuk, G., Durlik, I., Mikiciuk, M., Łobodzińska, A., & Śnieg, M. (2025). The IoT and
    AI in agriculture: The time is now —A systematic review of smart sensing technologies. Sensors,
    25(12), 3583.
54. Mishra, H., & Mishra, D. (2024). AI for data-driven decision-making in smart agriculture: from field
    to farm man agement. In Artificial intelligence techniques in smart agriculture (pp. 173 -193).
    Singapore: Springer Nature Singapore.
55. Mohan, R. N. V ., Rayanoothala, P. S., & Sree, R. P. (2025). Next-gen agriculture: integrating AI and
    XAI for precision crop yield predictions. Frontiers in Plant Science, 15, 1451607.
56. Ngo, V . M., Le-Khac, N. A., & Kechadi, M. T. (2019, June). Designing and implementing data
    warehouse for agricultural big data. In International conference on big data (pp. 1-17). Cham: Springer
    International Publishing.
57. Ngo, V . M., Le-Khac, N. A., & Kechadi, M. T. (2020). Data warehouse and decision support on
    integrated crop big data. International Journal of Business Process Integration and Management, 10(1),
    17-28.
58. Noulas, C., Torabian, S., & Qin, R. (202 3). Crop nutrient requirements and advanced fertilizer
    management strategies. Agronomy, 13(8), 2017.
59. Padarian, J., & McBratney, A. B. (2020). A new model for intra-and inter-institutional soil data sharing.
    Soil, 6(1), 89-94.
60. Padarian, J., Minasny, B., & McBratney, A. B. (2020). Machine learning and soil sciences: A review
    aided by machine learning tools. Soil, 6(1), 35-52.
61. Pan, S. Y ., He, K. H., Lin, K. T., Fan, C., & Chang, C. T. (2022). Addressing nitrogenous gases from
    croplands toward low-emission agriculture. Npj Climate and Atmospheric Science, 5(1), 43.
62. Pandey, U., Vashisth, A., Mishra, A., & Kumari, M. (2025). AI and IoT-driven soil health restoration:
    A machine learning approach for sustainable agriculture.
63. Pandey, V ., Das, R., & Biswas, D. (2025). AgroSense: An Integrated Deep Learning System for Crop
    Recommendation via Soil Image Analysis and Nutrient Profiling. arXiv preprint arXiv:2509.01344.
64. Penuelas, J., Coello, F., & Sardans, J. (2023). A better use of fertilizers is needed for global food
    security and environmental sustainability. Agriculture & Food Security, 12(1), 5.
65. Pierre, J. F., & Regmi, N. (2026). Fertilizer form influences agronomic efficiency and environmental
    sustainability in nutrient management. Discover Soil, 3(1), 50.
66. Pincheira, M., Antonelli, F., & Vecchio, M. (2025). AGRICLIMA: Towards a Federated Platform for
    Spatiotemporal Risk Analysis in Agriculture. Agriculture, 15(23), 2450.
67. Reginald, P. J. (2024). Integrating IoT and AI for Precision Irrigation and Intelligent Nutrient
    Management in Smart Agriculture. National Journal of Plant Sciences and Smart Horticulture, 34-39.
68. Reisinger, A. J., Bean, E. Z., Clark, M., Levine, A. J., & Wilson, P. C. (2025). Fertilizer management
    approaches influence nutrient leaching from residential landscapes (V ol. 54, No. 1, pp. 289-302).
69. Rodrigues, G. C. (2022). Precision agriculture: Strategies and technology adoption. Agriculture, 12(9),
71. Rouzky, R., Sarrafzadeh, A., Sowells-Boone, E., Pasandi, H. B., & Goins, G. (2025). Data governance
    in smart farming: A systematic review of challenges and gaps. Information Processing in Agriculture.
72. San Emeterio de la Parte, M., Martínez -Ortega, J. F., Hernández Díaz, V ., & Martínez, N. L. (2023).
    Big Data and precision agriculture: a novel spatio-temporal semantic IoT data management framework
    for improved interoperability. Journal of Big Data, 10(1), 52.

International Journal for Multidisciplinary Research (IJFMR)

E-ISSN: 2582-2160 ● Website: www.ijfmr.com ● Email: editor@ijfmr.com

IJFMR260585286 Volume 8, Issue 5, September-October 2026 19

72. Shao, F., Tao, W., Wang, Q., Wu, J., Su, L., Sun, Y ., & Zeng, S. (2021). A modified model for predicting
    nutrient loss in runoff using a time-varying mixing layer. Journal of Hydrology, 603, 127091.
73. Shrivastav, P., Prasad, M., Singh, T. B., Yadav, A., Goyal, D., Ali, A., & Dantu, P. K. (2020). Role of
    nutrients in plant growth and development. In Contaminants in agriculture: Sources, impacts and
    management (pp. 43-59). Cham: Springer International Publishing.
74. Soussi, A., Zero, E., Sacile, R., Trinchero, D., & Fossa, M. (2024). Smart sensors and smart data for
    precision agriculture: a review. Sensors, 24(8), 2647.
75. Talaviya, T., Shah, D., Patel, N., Yagnik, H., & Shah, M. (2020). Implementation of artificial
    intelligence in agriculture for optimisation of irrigation and application of pesticides and herbicides.
    Artificial intelligence in agriculture, 4, 58-73.
76. Vullaganti, N., Ram, B. G., & Sun, X. (2025). Precision agriculture technologies for soil site-specific
    nutrient management: A comprehensive review. Artificial Intelligence in Agriculture, 15(2), 147-161.
77. Wang, Q., & Wang, T. (2026). Machine learning-driven sensor fusion for precision agriculture: current
    applications and future perspectives. Sensor Review, 1-26.
78. Wang, Z., Xiao, S., Wang, J., Parab, A., & Patel, S. (2024). Intelligent Agricultural Management
    Considering N $ _2 $ O Emission and Climate Variability with Uncertainties. arXiv preprint
    arXiv:2402.08832.
79. Wolfert, S., Ge, L., Verdouw, C., & Bogaardt, M. J. (2017). Big data in smart farming –a review.
    Agricultural systems, 153, 69-80.
80. Yanardağ, A. B. (2024). Damage to Soil and Environment Caused by Excessive Fertilizer Use.
    Agriculture, Forestry and Aquaculture Sciences. Ankara: Serüven Publishing, 203.
81. Yang, Y ., Ge, X., Feng, W., Xian, L., Shao, G., Liao, J., & Zeng, S. (2025). Risk and driving factors of
    nitrogen and phosphorus runoff losses under extreme rainfall conditions: sludge application strategies
    in forest soils. Catena, 259, 109318.
82. Zheng, M., Zhu, P., Zheng, J., Xue, L., Zhu, Q., Cai, X., ... & Zhang, J. (2021). Effects of soil texture
    and nitrogen fertilisation on soil bacterial community structure and nitrogen uptake in flue -cured
    tobacco. Scientific Reports, 11(1), 22643.