Precision Agriculture in the Climate Change Era

In the first part of his lecture, he focused on the "why" behind the technology—the economic and environmental pressures facing modern agriculture. Here are the key areas Professor Comparetti discussed:

The Paradigm Shift in Management

• Scale of Management: He emphasized a move from managing by the "average" of a field to managing the specific needs of a single plant.
• Resource Efficiency: The goal is to "produce more with less," optimizing crop inputs to ensure that nothing is wasted.
• Luxury vs. Necessity: He argued that Precision Agriculture is no longer a high-tech "luxury" for wealthy farmers but has become an essential strategy for global food security.

Environmental and Climate Adaptation

• Mitigating Pressure: Technology is used to transform farms from a source of "climate pressure" into managed ecosystems that can withstand environmental shifts.
• Soil Health: A specific area of his expertise mentioned is the geo-referenced mapping of soil compaction, which is vital for maintaining soil structure and health.
• Renewable Energy: He touched upon the integration of renewable energy sources within agricultural systems to reduce the carbon footprint of food production.

Sustainability and Global Food Security

• Volatile Climate: The lecture highlighted that these technologies are the primary way farmers can adapt to an increasingly unpredictable and volatile climate.
• Ecosystem Management: By using data to "observe, measure, and respond," the farm becomes a more resilient system.

In the second part of the lecture, Professor Antonio Comparetti emphasized that the primary role of technological tools in precision agriculture is to transition management from a general "field-level" approach to a specific "plant-level" strategy. This shift is essential for maintaining global food security and environmental sustainability in the face of climate change.

The specific roles and strategies for the tools highlighted in the talk are as follows:

1. Data Collection and Monitoring

·        Satellites & GNSS: Satellites provide the Global Navigation Satellite System (GNSS) signals required for all positioning. They are used for macro-level monitoring, historical field analysis, and identifying consistent low-yield zones over many years.

·        Drones (UAVs): These provide the micro-level "granularity" needed for plant-level management. Drones can be deployed on-demand to capture high-resolution imagery for stress detection and weed identification, even under cloud cover.

• Real-Time Observation: IoT-enabled sensors are used to observe and measure crop and soil parameters in real-time, such as moisture levels, temperature, and nutrient content.
• Addressing Variability: These tools allow farmers to identify inter- and intra-field variability, ensuring that management decisions are based on the actual needs of specific areas or individual plants.
• Remote Management: Digital platforms allow farmers to monitor their fields remotely, enabling timely interventions without being physically present.

Precise Resource Application (Variable Rate Technology - VRT)

• Targeted Inputs: VRT enables the application of crop inputs (fertilizers, water, and pesticides) at varying rates across a field based on geo-referenced data.
• Waste Reduction: By applying only what is required by the crop in a specific spot, VRT reduces input waste, lowers production costs, and minimizes environmental runoff.

Intelligent Decision-Making (AI & Predictive Analytics)

• Data Interpretation: AI and machine learning process the "big data" generated by sensors to produce actionable maps for farmers.
• Forecasting: These systems are used for predictive analytics, such as forecasting crop yields under anticipated climate conditions or identifying early signs of plant stress and disease.
• Automation of Logic: AI reduces the reliance on human intervention by automating the decision-making process for irrigation and fertilization schedules.

Precision Execution (Autonomous Robotics)

• Labour Efficiency: Robots handle repetitive and labour-intensive tasks like seeding, weeding, and harvesting with higher precision than traditional machinery.
• Environmental Protection: Field robots are often lighter than traditional tractors, which helps eliminate soil compaction and reduces greenhouse gas (GHG) emissions.
• High Accuracy: Some robotic systems offer accuracy up to 2 cm, allowing every seed to be precisely placed and mapped.

Positioning and Mapping (GNSS)

• Geo-Referencing: Global Navigation Satellite Systems (GNSS) are fundamental for sensing the exact position of agricultural machines, which is required for both measuring field parameters and applying variable input rates.
• Guidance Systems: These systems support assisted guidance for machinery, ensuring rows are followed perfectly and overlapping is minimized.

Professor Comparetti emphasized that the transition to "plant-level" management is not possible without a suite of integrated technologies. These tools move the farm from a system of "averages" to a system of "precision," where every square centimeter is accounted for.

The following sections expand on the specific roles these technologies play in modern agriculture:

• Global Navigation Satellite Systems (GNSS) & Geo-referencing:
• The Foundation of Precision: GNSS provides the essential positioning data required to map field variability and guide machinery with sub-decimeter accuracy.
• Mapping Soil Compaction: Using geo-referenced data, sensors can create detailed maps of soil compaction, allowing for targeted aeration or mechanical intervention rather than treating the entire field.
• Assisted Guidance: Satellite positioning enables "auto-steer" capabilities, reducing operator fatigue and ensuring that rows are perfectly aligned to avoid overlapping application of seeds or chemicals.
• Internet of Things (IoT) and Proximal Sensing:
• Real-time Monitoring: IoT-connected sensors act as the "eyes" of the farm, constantly observing parameters such as soil moisture, leaf temperature, and nutrient levels.
• Variability Assessment: These tools are used to measure "inter- and intra-field variability," identifying specific zones that are underperforming compared to the rest of the field.
• Remote Management: Data is transmitted to digital platforms, allowing for immediate observation and response to environmental shifts without requiring constant physical presence in the field.
• Variable Rate Technology (VRT):
• Dynamic Application: VRT allows for the "prescribed" application of crop inputs (fertilizers, water, pesticides) where the rate changes automatically as the machine moves across the field.
• Optimized Inputs: By shifting from a uniform "field-level" application to a "plant-level" one, farmers can produce more while using fewer inputs, significantly mitigating environmental impact and runoff.
• AI, Big Data, and Predictive Analytics:
• Processing Complexity: AI systems process the massive amounts of data generated by GNSS and IoT sensors to create actionable "prescription maps" for farmers.
• Proactive Strategy: Predictive analytics allow for forecasting crop yields and identifying potential pest or disease outbreaks before they become visible to the human eye.
• Management Ecosystems: AI helps transform the farm into a "managed ecosystem," where data-driven decisions help the system withstand 21st-century environmental shifts.
• Autonomous Robotics:
• Precision Execution: Robots and drones carry out high-precision tasks, such as individual weed spot-spraying or automated seeding, with an accuracy as fine as 2 cm.
• Reducing Environmental Pressure: Lightweight autonomous robots help eliminate the soil compaction typically caused by heavy traditional tractors, while also operating on electric power to reduce greenhouse gas emissions.

Specialized Focus: Geo-Referenced Mapping and Soil Compaction

A significant portion of Professor Comparetti’s research and lecture addressed the physical health of the agricultural environment, specifically focusing on soil compaction. He explained that as machinery has traditionally become larger and heavier, the pressure on the soil has increased, leading to degraded soil structure and reduced crop yields. To combat this, he detailed the following strategies:

·        Geo-Referenced Measurement: Using GNSS and specialized sensors, researchers can measure soil resistance and compaction levels at specific coordinates across a field.

·        Mapping Variability: This data is used to create high-resolution maps that identify "hotspots" of compaction. Instead of tilling an entire field—which is energy-intensive and can damage soil health—farmers can use these maps to apply mechanical relief only where strictly necessary.

·        The Role of Robotics: Professor Comparetti highlighted that the shift toward autonomous robotics is a primary solution for preventing future compaction.

·        Weight Reduction: Unlike massive traditional tractors, autonomous field robots are significantly lighter, which helps to eliminate soil compaction issues from the outset.

·        Sustainable Ecosystems: By managing soil density with precision, the farm maintains better water infiltration and root growth, making the "managed ecosystem" more capable of withstanding the 21st century's environmental shifts.

The Role of Weed Density in Precision Agriculture

Professor Comparetti explained that traditional farming treats a field as if weed density is uniform, leading to the "broadcast" spraying of herbicides. Precision Agriculture, however, treats weed density as a spatially variable parameter:

·        Sensing and Mapping: Using RGB cameras and AI-driven machine vision, autonomous systems can distinguish between crops and weeds even in "dense, intertwined conditions". These tools generate weed density maps that identify specific "hotspots" or "weed-infested regions".

·        Targeted Treatment (Variable Rate Technology): Instead of spraying the whole field, Variable Rate Technology (VRT) allows a sprayer to automatically adjust its flow. It can increase the dosage in high-density areas and shut off entirely in areas with no weeds, which has been shown in some studies to save nearly 30% of herbicides.

·        Robotic Intervention: Professor Comparetti highlighted that Autonomous Robotics can perform "individual weed spot-spraying" or mechanical removal. These robots use real-time data to estimate weed pressure and execute precise interventions (up to 2 cm accuracy) only where weeds are detected.

Selective Harvesting Strategies

In his lecture, Professor Comparetti also highlighted the transformation of grape harvesting through "Precision Viticulture." He explained that traditional harvesting is often inefficient because it treats an entire vineyard as if the fruit is uniformly ripe.

By applying precision techniques, harvesting becomes a data-driven process focused on selective picking and quality optimization.

• Maturity Mapping: Using hyperspectral and visual imaging, sensors can map a vineyard to identify precisely where grapes have reached optimal sugar content and acidity.
• Spatially Variable Harvest: High-tech harvesters (such as those from Pellenc or Gregoire) are now equipped with multiple hoppers. These machines can automatically sort grapes into different bins based on their quality or vegetative vigour as they move through the rows.
• Yield Prediction: Machine learning algorithms analyze aerial and ground imagery to predict yields before the harvest begins, helping farmers plan logistics and storage more effectively.

Specialized Focus: Robotics and Automation

Professor Comparetti highlighted the role of autonomous machinery in reducing human error and environmental impact:

·        High Precision: Robotic systems achieve an execution accuracy of up to 2 cm, allowing for individual plant management and spot-spraying of weeds.

·        Weight and Energy: These robots are typically lightweight, often electric-powered, which significantly reduces the carbon footprint of the farm.

·        Assisted Guidance (Autopilot): For larger machinery, GNSS-based "electric pilot" or assisted guidance systems allow tractors to follow precise paths, reducing overlaps and operator fatigue.

Centrifugal and pneumatic spreaders

In Professor Antonio Comparetti’s research and lecture materials, specific mention is made of both centrifugal and pneumatic spreaders, primarily in the context of their adaptation for Spatially Variable Rate (SVR) fertilization. The roles and differences between these two types of spreaders are summarized as follows:

1. Centrifugal (Disc) Spreaders

• Mechanism: These spreaders typically use spinning discs to distribute granular fertilizer by gravity or mechanical force.
• Research Focus: Professor Comparetti has conducted extensive tests using centrifugal spreaders (such as the Amazone ZA-M model) to measure adjustment time and positional offset.
• Modifications for Precision: To enable variable rate application, these spreaders were modified with actuators connected to the hopper shutter slides, allowing an on-board computer to change the flow rate in real-time based on GPS data.
• Performance: His research indicates that centrifugal spreaders are highly effective for site-specific fertilization, though their accuracy is affected by tractor speed and the physical characteristics of the fertilizer granules.

2. Pneumatic (Boom) Spreaders

• Mechanism: These spreaders use a "cell wheel" dosing system and air pressure to convey fertilizer through tubes to a full-width boom, where it is distributed via reflectors.
• Comparison to Centrifugal: In his comparative studies, pneumatic spreaders were noted for having a different positional lag than disc spreaders.
• Lag Characteristics: Because the fertilizer must travel through the length of the boom, there is a specific "conveying time" that must be accounted for in the software to ensure the fertilizer hits the soil at the correct geo-referenced coordinate.

Xenon Light (Active Sensors) vs. Satellite Maps (Remote Sensing)

The most startling advancement is the conquest of darkness. Sensors equipped with Xenon lamps act as an "active light source." Unlike standard sensors that require sunlight, these lamps emit artificial light to measure leaf reflectance, allowing for 24-hour precision fertilization in the pitch black or the dense fogs of Northern Italy.

The fundamental difference lies in the source of light and the type of data collected:

·        Satellite Maps (Passive Sensing): These rely on sunlight reflecting off the plant. Satellites measure how much red or near-infrared light the leaves bounce back. While useful for seeing "greenness" (NDVI), they are "passive" because they can't control the light source and are blocked by clouds or night.

·        Xenon Light (Active Sensing): These sensors carry their own "sun." A Xenon lamp emits a high-intensity flash of light that "excites" the chlorophyll in the leaves. The sensor then measures the fluorescence—a tiny amount of light the plant re-emits as a byproduct of photosynthesis.

This example illustrate the move toward "plant-level" management:

• Pre-visual Stress Detection: Xenon-based sensors can detect "silent" stress. For example, if a deer-grazing pasture is suffering from a lack of nitrogen or early water stress, the photosynthetic activity measured by the Xenon flash will drop days before the leaves actually turn yellow on a satellite map.
• Real-time Weeding: Because Xenon lights work regardless of external lighting, they can be mounted on autonomous robots to identify and spray weeds at night. The specific "fluorescence signature" of a weed differs from that of the crop, allowing for 2 cm precision in herbicide application.
• Soil and Plant Interaction: This active lighting is often used in the "proximal sensing" he mentioned—sensors mounted directly on tractors or robots that scan the crop as they drive over it, making instant adjustments to fertilizer flow based on the plant's actual light-use efficiency.

Concluding Thoughts: Precision Agriculture as a Necessity

In his concluding remarks, he emphasised that Precision Agriculture is no longer a choice but a requirement for the 21st century. The transition from "field-level" to "plant-level" management represents a fundamental shift in how we view food production.

The Future of Global Food Security

• From Pressure to Solution: By adopting these data-driven technologies, the farm evolves from a source of climate pressure into a managed ecosystem that can withstand environmental shifts.
• Climate Resilience: Using IoT, AI, and Predictive Analytics allows farmers to adapt to an increasingly volatile climate by making proactive, rather than reactive, decisions.
• Sustainability Goal: The ultimate role of these strategies is to produce more food with significantly fewer inputs, protecting both the environment and global food supplies.

Impact of Academic Exchange

The collaboration between the University of Palermo and the University of Mauritius, through the ERASMUS+ Mobility Programme, highlights the importance of sharing specialized research—such as Professor Comparetti's work on GNSS and soil compaction—to address local and global agricultural challenges