3.5

Integrated pest management (IPM)

Novel IPM strategies are needed to detect diseases and prevent their spread on crop production for European organic and conventional agricultures, to increase organic farming and allow the development of horticultural systems that will use less/no pesticides. Improved IPM will require developments in the following fields:

• Smart systems based on portable real-time pest disease detection, diagnostics and monitoring platforms to provide rapid local and regional disease incidence alerts (georeferenced) e.g. weather/climate information for predictive models providing risk assessments and decision support for IPM.
• IoT devices and real-time edge AI computing platforms specialized in pests and disease measurements, such as insect traps and other systems based on image recognition and/or AI models.
• Wearable Plant Sensors for directly monitoring the plant and being able to quickly detect pest appearance.

Agro ecology based: Move from conventional to organic, regenerative agriculture

To support the EU “From Farm to Fork” implementation, smart ECS can help farmers to drastically decrease the use of pesticides and their impact on human health and the environment. This will require:

• Development of cost-effective and intelligent intra-row, herbicide-free weeding techniques using advanced robots and robot fleets for individual plant recognition with high precision, based on advanced (vision) sensor technologies and real-time edge AI computing platforms working under in-field conditions.
• Development of smart and power efficient sensors to monitor the quality of spraying, as this is essential for biocontrol products and contact pesticides. Moreover, quantum sensors and advanced photonics to monitor soil and plant health are needed such as pH, NO3 & EC, soil moisture, CO2, leaf wetness, surface temperature, airborne pathogens. They should be precise, low cost, highly miniaturized and biodegradable electronic and sensor components (printed antennas, organic batteries, biodegradable substrates, elimination of sensor/chip packaging) to avoid a negative Impact of electronic products on soil.
• Integration, into the same framework, of AI decision-support tools and precision agriculture tools to simplify farm management, improve crop quality and reduce costs.
• Advanced tools, based on real-time edge AI computing architectures, may yield on just farming 80% of the most productive soil leaving 20% for natural recovery reducing costs of fertilizers, energy to harvest and human time to collect, by this precision agriculture some complementary activities or species may be used to fix soil and reduce erosion and desertification.
• Dedicated services such as technical support (infrastructure deployment and maintenance), precision agriculture as a service, education, etc.

Plant precision breeding and plant phenotyping

The development of smart technologies can support precision plant breeding and phenotyping. This could be nanotechnology solutions or smart sensor solutions to support the following:

• Genomics and transcriptomics: DNA informed breeding, gene editing, genome prediction, breeding optimisation, phenotyping and seed sowing optimisation.
• Large scale and high precision measurements of plant growth, architecture and composition: these measurements are required to optimise plant breeding by increasing our understanding of the genetic control and response of plants to their environment. Sensor systems should allow the study of plants in relation to biotic and abiotic factors, including plant-microbiome interactions, plant-plant competition, plant diseases and exposure to a multitude of variable abiotic environmental conditions such as light quality, irradiance levels, nutrient supply, temperature, humidity, soil pH and atmospheric CO2 levels. Plant health is also related to nutrients and the 4R strategy "right source at the right dose, at the right place, at the right time", however, plant models, besides agri knowledge, are needed for nutrient content measurement and monitoring.

Livestock health is crucial for food safety. Different animal pathogens can be a serious threat for livestock management, because pathogens can be divided in those that can infect multiple species, and those that infect specific terrestrial and aquatic animal species (i.e., cattle, sheep, goats, equines, etc.). A great threat is represented by diseases that are highly transmissible, having the potential to spread rapidly across borders and cause significant socio-economic and public health consequences. All these lead to great economic loss to the farmers. Moreover, the use of antibiotics to treat animals increases the cost of production, as well as creating other problems like residues and resistance, which is of serious concern to public health.

In addition, better livestock management practices must reduce the release of pathogens into the environment and substantially reduce the risk of microbial contamination of surface and groundwater. As outlined in the literature³⁷, there is relevant impact of animal pathogens in four areas: animal health, economics, food safety and security, and public health. Consequently, there is strong interest in developing new advanced solutions systems with high sensitivity and specificity, for early detection of animal diseases and minimizing antibiotics use too.

Synergistic strategies can be applied to approach these challenges. Agronomic and technological solutions must be applied to:

• Minimise risks (e.g. release of pathogens into the environment, use of antibiotics for animals);
• Surveillance of the livestock environment through an efficient process of digitalization of livestock management to prevent diseases spread.

From a microbiological perspective, microorganisms can mitigate risk because specific beneficial microorganisms can be selected to work synergistically with other microorganisms already in the environment. Beneficial microorganisms can support the nutritional requirements of plants and reduce the incidence of pathogenic microorganisms, to solubilise minerals, to conserve energy, to maintain the microbial-ecological balance of the soil, to increase photosynthetic efficiency, and to fix biological nitrogen.

Research has also studied beneficial effects when these specific microorganisms are included in animal diets³⁸. It has been shown that when they meet the organic matter that makes up the animal’s diet, they secrete beneficial substances such as vitamins, organic acids, chelated minerals, and antioxidants, influencing an antibacterial effect through a selective blocking of pathogen colonisation.

Animal welfare is also an important concern for a growing number of consumers.

We can assess that these synergistic strategies (risk reduction with agronomic solutions and surveillance) as explained above are drivers for investing in better sensing systems for animal monitoring. Combined with data analytics solutions, this will improve animal health and welfare, resulting in more animal-friendly production, higher efficiency, better quality, and improved food control safety:

• Wearable sensors at the farm/barn level, and ambient sensors during cattle transport.
• Smart sensor systems to monitor animal activity, such as individual or group behaviour, to provide useful information for early detection of diseases and to increase animal well-being.
• Smart sensor systems for rapid verification of bacterial infection together with behavioural observations to control disease spread and support clinical and veterinary stakeholders to effect suitable therapeutic interventions; body temperature can also be monitored for early disease detection to reduce antibiotics use.

A major source of GHG emissions in agriculture is related to livestock enteric emissions that must be reduced to achieve the 1.5°C pathway. The methane emissions from livestock increase atmospheric temperature approximately 80 times more than CO2 on a 20-year outlook, but methane has a shorter atmospheric lifetime than other GHGs, making it an effective target for reducing global temperatures quickly.

New methods are emerging to reduce enteric emissions in livestock, and these should be largely introduced, particularly in grassland or mixed systems, where cattle might be centrally handled only once or twice a year for weighing and treatment, and where their feed rations are unpredictable and uncontrollable. In addition to the smart sensors systems mentioned above, the following tools can be added:

• Methane verification and monitoring tools to measure methane emissions and sequestration.
• Besides GHG emissions, monitoring nitrogen emissions is crucial due to the significant impact these emissions have on the environment and public health. Nitrogen emissions primarily arise from the application of nitrogen-based fertilisers and the management of livestock, and they occur in various forms, including ammonia (NH₃), nitrous oxide (N₂O), and nitrates (NO₃⁻). Smart sensor systems to monitor nitrogen emissions will play a pivotal role in addressing environmental challenges while optimising agricultural productivity.

Taking into account the AI deployment in many sectors, the importance of a farmer-centric approach for livestock management is well highlighted within the digital livestock farming landscape³⁹.

When coupled with real-time edge computing platforms and AI models, all these data can provide real-time, objective, and holistic insights into animal welfare, but a farmer-centric approach is vital to ensure that technologies truly serve the needs of farmers and the welfare of animals. Therefore, every AI application should encourage this farmer centric approach, because the economic sustainability goal of the agricultural industry is also related to improve animal welfare and make informed decisions about disease prevention and management, leading to healthier herds and increased productivity.

Intelligent logistic systems including sensing and monitoring for food chains

Logistics are a critical component of the food chain. They not only determine the reach of distribution, but logistics delays and conditions profoundly affect the quality and safety of products reaching consumers and can result in food loss and waste in the supply chain.

Smart real-time sensing, monitoring and control systems in the food supply chain will safeguard food quality and food safety, while eventually reducing food losses in the supply chain. Therefore, technological solutions are required, but not limited to:

• Systems for monitoring and controlling the quality of food products and ingredients during transport and storage (e.g. temperature monitoring in cold chain, moisture, controlled atmosphere, ethanol, ethylene), which should be reliable, contaminant-free, secure, power efficient and interoperable along the logistics chain. Contact-less powering of the sensors placed in storage tanks could also avoid contamination of the food. Among these systems augmented reality (AR) can be also used to promote greater transparency, efficient learning, and an enriched customer experience such as:
• Training and education: Employees can learn food safety procedures or preparation techniques through interactive AR.
• Interactive marketing: Brands can offer immersive experiences, such as interactive recipes or 3D product presentations.
• Enhanced customer experience: In restaurants, AR can display 3D menus, personalised suggestions, and even virtual dish presentations.
• Predictive systems to assess quality of (perishable) food products in the supply chain, providing real-time decision-support based on actual sensor measurements, supply chain data and AI models.
• Transport route optimisation, considering not only time and cost, but also external conditions and the intrinsic properties of the products being transported.
• Logistics optimisation: Warehouse managers can use AR to visualise inventory and improve product organisation.

These needs are strongly related with traceability, as shown in the following section.

Food and beverage manufacturers and producers are faced with increasingly complex and fragmented supply chains, stricter regulation, and more demanding consumers. Regulatory compliance, competitive advantage, brand reputation and costs have made product traceability a priority and end-to-end traceability a major challenge. In today’s globalised world where people of any origin live across every country, the source of food products and ingredients, as well as their certification, are a major concern and priority for consumers, who want to be sure about the origin of food products. Therefore, traceability should also encompass certifying food origin, as well as making information available on any relevant process to which the food product has been submitted. This information should not be restricted to mere tracking across the supply chain. End-to-end traceability solutions are required, but should not be limited to:

• Integrating blockchain into current technology to increase safety while preventing fraud and counterfeiting.
• Traceability to increase alignment between production and individual consumer demands, leading to better provisioning and more personalised nutrition support.
• Product traceability: Consumers can scan packaging to learn product origins, supply chain information, and nutritional information thanks to smart tags and AR.
• Traceability to optimise distance between farm and fork – although many products are produced preferentially in specific parts of the world, there are also many examples of food that could be produced economically closer to consumers.

To this end, as IoT/IIoT solutions are increasingly being deployed, integrated hardware systems including AR devices need to deliver (apart from mobility and connectivity) long lifetime autonomous sensing and AI-based intelligence at-the-edge, as well as edge and/or cloud analytics and cybersecurity, complying with privacy regulations where applicable, on a plug-and-play, open, interoperable architecture, and platform.

Distributed Ledger Technologies (DLTs) such as blockchain allow secure storage and tracking of all kinds of information, including condensed sensing or monitoring data regarding crops and livestock. Examples include information on crop seeds, feed ingested by livestock (including medication and antibiotics), as well as recording of the whole process that any farm product is submitted to until it reaches the consumer, throughout the respective supply chain and involved actors. Such information increases the transparency of these supply chains and can reduce potential production issues, e.g. simplify the tracing of eventual product spoiling, or other issues, to the respective source, supporting possible decisions to recall product batches if necessary. Consumers can also benefit from such transparency, i.e., they can be given access to information to make better informed decisions about the offered goods they want or need to acquire, and they can use the information to provide feedback to farmers and producers incentivising their policies and practices further.

Nowadays, end-consumers are more concerned about the origin of the agriculture products that they consume. A complete system to manage traceability and to offer, to the end-consumers, a complete transparency of the actions taken in the farm (for instance in the vineyard) and during the transformation process (in the example of a winery) is required: a solution could use blockchain technology. However, it is important that small-scale farmers with low technological expertise, resources, and insufficient size to integrate blockchain could be supported, eventually through a dedicated ecosystem. This can be devised through blockchains implemented to include such farmers and respective food product chains, which would also allow the support of food safety assurance, namely in respective local markets, eventually involving less costs associated with logistics and distribution, and thus also contributing to sustainability and fair food systems. Nevertheless, the lack of regulations, standardisation and interoperability are challenges for incorporating blockchain.

In-situ real-time monitoring of soil nutrients and herbicides

The optimal use of chemical fertilisers and organic manures to deliver the ever-increasing food production requires a complete understanding of the nitrogen- and phosphorous-based nutrients applied in the fields with a much greater spatial and temporal resolution than is available today. Current methods of soil analysis do not provide real-time, precise and in situ nutrient analysis in fine detail, and delays in receiving soil results are common because of backlogs in commercial labs due to high sample volumes, thus reducing the value of the soil test results for the farmer. Moreover, herbicide application is another huge problem due to their environmental and health impact. To solve these issues, the following actions must be done:

• Smart quantum sensors and bio-sensors (with miniaturised and ultra-low power consumption components allowing these sensors to harvest their own energy) need to be developed to provide accurate in situ and in real time measurements of soil quality, soil nutrient levels and soil moisture with at parts per million (ppm) concentrations to optimise fertiliser use and irrigation. Such devices must have the appropriate packaging to extract water from the soil. Ideally, they should be buried in the soil for long periods of time or at least while sustaining operation capabilities for the entire growing season. To optimise effectiveness, low proximity sensors should be combined with optical sensors and high proximity sensors to retrieve the maximum amount of information on soil health.
• Likewise, smart actuators could prove to be highly beneficial. Such miniature units could be deposited on or buried into the soil. Coupled to sensing functionalities into the same module, critical actions (e.g. release of agrochemicals) could be triggered very close to the plant roots for maximised efficiency.
• Multidisciplinary approaches for developing novel sustainable smart ECS are needed. Indeed, current ECS contain a variety of toxic materials and chemicals. As such, they cannot be left in the soil. The optimisation in the use of agrochemicals should not come at the expense of another ecological burden. New sustainable "green" ECS made of eco-friendly materials that will have benign environmental impact must be created.

IoT systems with edge and/or cloud-based AI computing architectures are also necessary to provide farmers with decision-support regarding fertilisation strategies, by translating measurements into meaningful agronomic indicators and respective measures. These strategies should prioritise the use of organic fertilisers and the gradual reduction of chemical ones until eliminated to restore the biodiversity contribution in the preservation of soil health. Furthermore, this type of system should detect weeds, preserve the “good ones” and eradicate those that are competing with the crop in question. This requires low-cost vision technologies (not only plain optical red/green/blue (RGB), but also 3D, hyperspectral imaging, etc.) and edge AI computing platforms for in situ prompt recognition and decision-making. Recent advances include the integration of AI for efficient soil carbon monitoring and digital soil mapping for comprehensive property assessment. Additionally, for crop monitoring, advanced photonics enable hyperspectral imaging to assess plant health, detect diseases early, and improve yield predictions. For pest detection, early identification of pest infestations using sensitive optical sensors facilitates targeted interventions.

Sensors and diagnostics for air quality monitoring (indoor, urban and rural)

According to the World Health Organisation, the air we breathe is becoming dangerously polluted. Nine out of ten people now breathe polluted air, which kills seven million people every year. There has been much progress on identifying and reducing the sources of air pollution at lower concentrations and with higher spatial coverage. This is necessary to provide adequate data on what people are breathing, and to provide localised as well as holistic solutions. Microsensors and/or mini-stations can be used during fieldwork campaigns in cities, but there are technical problems relating to power source, data transmission, data storage, and data handling and assessment. Besides, local measures are not always effective since local concentrations of particulate matter may be influenced by long-range transported pollutants from agricultural activities occurring outside city boundaries.

Similarly, while indoor air quality has been shown to unambiguously impact the wellness, health and performance of people as shown for instance with Covid-19 in schools due to lack of indoor air quality measurement, there is also a lack of spatial granularity and a significant lag between exposures and sensing, actuation and management interventions for risk mitigation. In addition to indoors, air quality is made more complex by the interaction between indoor and outdoor air, emissions from buildings and their contents (paints, furniture, heating, and cooling systems, etc.), human activities (breathing, cooking, cleaning, etc.) and the effects of long-term exposure to low concentrations of volatile organic compounds. These issues necessitate development and deployment of real-time intelligent multi-sensor technologies with high selectivity and embedded (re-)calibration techniques. These should be combined with a monitoring network (edge-based) as part of the indoor infrastructure to provide the spatial and temporal information needed for specific, targeted and appropriate actions. A required ingredient of this monitoring solution is a digital twin platform able to predict in real-time the pollutant diffusion, considering the complex environment monitored. Such actions should also include public awareness and the promotion of behavioural changes.

Smart systems for controlling and preventing GHG emissions

Strong evidence has been accumulated on the climate emergency resulting from human activities that add GHGs to the Earth’s atmosphere. The EU is the world's third biggest GHG emitter after China and the US. Although several measures have been taken since the Paris Agreement, breakthrough technologies and state-of-the-art deployment are still needed across the transport sector and other industries with a high emission footprint to achieve a further reduction in emissions. These would be facilitated by the following:

• Smart systems and digitalisation to improve industrial processes performance and energy/resource efficiency towards a low-carbon economy, while reducing the impact of mobility and agricultural processes on the environment and human health, thereby controlling and preventing GHG emissions.
• A focus on the GHG emissions from animals by investigating microbiological sensing technologies on or in animals (in their rumens or breath, for instance) to increase efficiency while reducing environmental impact, as well as performing analysis of the gathered data to support decision-making for mitigation measures (for instance, leading to change in feed).
• GHG verification and monitoring tools to measure GHG emissions and sequestration.

Currently smart quantum sensors allow air quality monitoring by measuring pollutants such as CO2, NOx and particulate matter with great accuracy, aiding in environmental regulation compliance. Additionally, advanced photonics provide real-time monitoring of industrial emissions, enabling rapid response to pollution spikes.

Integrated waste systems

Despite proactive European policies and regulations⁴⁴, effective bio-waste management remains a challenge. Reducing, recycling and reusing food/kitchen waste requires significant progress in technological solutions along with strong policymaking and shifting community behaviour. These solutions could be based on the following:

• Smart monitoring, controlling waste treatment units in real-time as well as gas emissions in landfills and anaerobic digestion monitoring. Data analytics should include gamification for behavioural triggers.
• Smart real-time quality control systems of the sorted waste (i.e., measure of the “purity” of the sorted waste) to ensure their proper recyclability
• Smart waste collection bins (radio-frequency identification (RFID) tags, self-compacting bins, fullness level sensors, automated waste segregation), even in remote locations, and without access to power supply, including automated robotic systems and optimised separation systems, which can be complemented by the upcycling of waste streams into usable resources and optimal routing systems, as well as vehicle tracking. These solutions should be integrated and interconnected into the product life cycle “from cradle to grave” to enable circular and resource-efficient methodologies.
• Edge or Cloud AI computing platforms could serve as the basis for building a complete system including the proposed solutions.

Intelligent sustainable / biodegradable packaging

Intelligent and biodegradable packaging concepts have been gaining traction in the food industry to improve product safety and reduce environmental impact. Smart sensors in an IoT system can monitor environmental conditions and product quality, while communication devices can store and convey data throughout the product life cycle. While these concepts need to be further advanced for efficient, safe food production and waste management, intelligent packaging itself needs to become more sustainable. Novel ideas are required to solve the problem of the amount of plastic packaging produced by food manufacturers. The definition of biodegradable packaging should lead to a new generation of food packaging. Such novel ideas include:

• A synergetic interdisciplinary approach to cross the boundaries of novel materials for food packaging and smart sensors associated with analytical methods for the detection of harmful substances that can infiltrate into food, cause water contamination, etc.
• Fabrication and hybrid integration of eco-friendly nanostructured electrodes, sensors, energy harvesting and storage devices on rigid and flexible biodegradable substrates to reduce the waste from embedded electronics in smart packaging.

Efficient smart networks for remediation and decarbonisation

Remediation processes aimed at converting harmful molecules into benign ones can be undertaken in different ecosystems, such as water bodies (e.g. biotic, and abiotic farming by-products), air (e.g. GHGs) and soil (e.g. pesticides). Remediation processes are mainly carried out in wastewater treatment plants. Although some pollution sources are static and sufficiently well-known such that treatment can be undertaken effectively, other pollution sources are more mobile in both time and/or space, making treatment at single points unsatisfactory. Another limiting issue is that remediation technologies are often power-intensive and cannot be deployed for long in remote locations. Alternative high-efficiency remediation methods are needed, such as to transform/reduce the levels of CO2 in chemical products. Current devices are also prone to fouling. This means remediation processes cannot be run constantly in remote locations, and there is thus a necessity to undertake them only when and where they are most required. In this regard:

• IOT edge or/and cloud AI computing platforms that can monitor relevant status in real time, and inform on the necessity of remediation, would provide unique decision support invaluable for efficient water, air, and soil management.
• To achieve net zero emissions (decarbonisation) in different agriculture areas, carbon verification and monitoring requires quantum sensors to measure pollutants such as CO2, NOx with high accuracy for air quality monitoring, advanced photonics to provide real-time emissions monitoring, and tools are required to measure carbon emissions and sequestration enabling rapid response to pollution spikes.
• Techniques used in the measurement and analysis of carbon sequestration by soils could also investigate the current potential of soils as a remediation mechanism to improve the sequestration capacity – such investigation should include the initiative of “four per 1000”⁴⁵ presented at COP21 in Paris.
• Likewise, tools and methods able to evaluate the performance of the carbon sequestration techniques employed should be developed to guarantee their efficiency.

Healthy Water

With the aim of reducing pollution-related health problems, water utilities, water associations, academia and private industry have focused on developing new methods, policies and procedures to secure drinking water distribution by (1) water quality monitoring by detecting in real-time any compound, contaminant (heavy metals, pathogens, chemical pollutants at ultra-low concentrations) or anomaly that may represent a health risk for the end-users, (2) leak detection by identifying leaks and reduce water loss and (3) taking the required measures to mitigate these issues. This necessitates online information on the status of water sources at a scale larger than ever before. To mitigate both accidental and intentional contamination of freshwater sources, the deployment of sensors and diagnostic and decision support systems with rapid communication technologies and data analysis capabilities are needed to secure water quality and its distribution over the network. Such actions shall provide:

Connected and highly integrated low power (or self-powered for maintenance/battery-free system) multi-parameter diagnostic quantum-based optical sensors for real-time physico-chemical analysis (temperature, ionic electrical conductivity, pH, turbidity, inorganic pollutants as nitrates or heavy metals, etc.) in water distribution network and wastewater treatment plants, and biofilm growth monitoring in water pipes. Online monitoring systems at the edge, including devices with embedded AI for data analytics. The presence of a mesh of intelligent devices in drinking water networks will make it possible to identify and deal very quickly with drifts or anomalies (e.g, leaks, contamination events, etc.) while reducing the amount of data sent to the servers. Prediction in real-time of pollutant diffusion with simplified compact models, considering non-dense measurements from the multi-parameter diagnostic sensors and online monitoring that are developed and deployed.

Integrated systems for demand reduction and conservation of water

According to the UN Development Programme⁴⁷, dwindling drinking water supplies is affecting every continent. On the one hand, increased urbanisation and farming have amplified the demand of water for human consumption and for domestic and agricultural use. On the other hand, an increasing number of countries are experiencing water stress due to longer drought periods and the spread of desertification. In addition, approximately 25% of all urban drinking water is being lost forever⁴⁸ in global water distribution systems, before it even reaches the end-user. Therefore, there is an urgent need to prevent losses from water abstraction as climate effects intensify. Leak localisation is currently very time-consuming, labour-intensive, and costly. Operators must manually place equipment that “listens” to the water flow during the night. Smart integrated systems can significantly contribute to key measures aiming at affecting consumer practices in water usage, delivering greater efficiency in detecting leaks and ultimately reducing water waste. Developments are needed in the fields of:

• Smart metering, time-of-use pricing and gamification to change habits in water consumption and control appliances, along with interoperable solutions for a truly connected smart household (taps, lavatories, showers, appliances).
• Low-cost self-powered/ low-power quantum sensor nodes for flow control, leak detection and auto shut-off, along with inexpensive actuators to remotely control valves for limiting water usage by volume/time. IoT edge or cloud AI computing platforms can optimise the control of household, agricultural and industrial infrastructure/equipment in water-intensive processes.
• Smart systems able to automate leak localisation, and to respond promptly and cost-effectively. This can be a combination of in-pipe inspection (to locate the leak) and a network of low-cost, fine-grained sensors to allow predictive maintenance of distribution systems.

Efficient and intelligent water distribution

The main challenge for improving the use of water is to guide its distribution depending on its final application (drinking water, water for industry, water for the cooling of data centers, etc.). However, the existing sanitary regulations always look to optimise water safety regardless of its final use. To apply the most effective measures to make water distribution more efficient, it is necessary to thoroughly review the different supply protocols and quality criteria for each sector. Moreover, by continuously monitoring the quality and availability of water, it would be possible to better regulate its distribution depending on the final use and to adjust the price accordingly. Intelligent systems connected to smart grids will allow water inputs to the network to be made at the right times, optimising the energy cost as a result.

To address these challenges, there is a need for developing:

• Novel smart metering solutions that include electrochemical multi-parameter sensors (pH, chlorine, conductivity, etc.) with high stability, anti-fouling, high accuracy capabilities and cost-effectiveness, as well as optical sensors based on different principles (fluorescence, absorbance, etc.) integrated into miniaturised systems at a low cost.
• Robust IoT systems with adequate data analysis processing power and AI capabilities to handle the large volume of data generated by the different water management processes to satisfy quality, usage type and associated pricing.
• Efficient year-round water management in terms of storage to deal with some of the most urgent shortages, with better forecasting and warning systems based on extensive measurements – e.g. intentionally flooded areas could be used to store water in times of expected scarcity.
• Prediction of pollutant diffusion in water distribution systems.

Smart systems for irrigation management

At a global level, agriculture consumes 69% of the world’s freshwater⁴⁹. Because of this, precise control of irrigation is essential to guarantee water and food security for all. Irrigation water management is the practice of monitoring and managing the rate, volume, and timing of water applications according to seasonal crop needs, considering the soil intake and water holding capacities with the objective of using water in the most profitable way at sustainable production levels. To this end:

• Smart sensors, with low-cost and self-powered characteristics, are increasingly required as tools to implement irrigation management and monitor water levels. Sensors should be more intelligent to support real-time applications and/ or reduce latency. Optimisation of the power consumption of the overall system shall be done to enable ultra-low power, ideally self-powered battery-free solutions that are cost effective and seamlessly deployed at the edge for both outdoor and indoor use.
• Integration of systems monitoring water deficiency or surplus is also required. These could be based on narrow-band spectral reflectance of water and land surfaces for vegetation/habitat mapping, along with UAV utilisation in remote areas.
• Smart low cost actuators will provide the “right dose (of water) at the right place”.
• Appropriate simplified AI models should allow to limit the number of sensors spread over a given landscape and integration of various sources of data, included satellite ones, will expand the analysis capability of decision support systems. Also, leak detection in the irrigation management system shall be considered to ensure that water is not lost.

Smart systems for flood management

Flood management has been gradually integrating smart sensors. IoT systems with water-level sensors can also play a significant role in real-time monitoring and natural hazards predictive/forecasting capacity models. This requires:

• The monitoring of water levels and devising prediction models to identify areas at a high risk of flooding. This is possible through the development and deployment of more intelligent low power sensor nodes in combination with smart predictive algorithms that will act as digital twins to integrate information from various sources, such as weather forecasts and regional georeferenced data.
• IoT interoperable systems are key for provision of real-time information to first responders, civilians and companies to proactively take countermeasures.

Smart water treatments fostering circular use (wastewater, rainwater, storm water)

Around 80% of all wastewater is currently being discharged into the world’s waterways, where it creates health, environmental and climate-related problems. Water from industrial, agriculture and domestic use contains organics, phosphates, nitrogen, cellulose, rare earth elements and other substances (including residues of medicinal products). In addition to its domestic use, purifying, distilling, or deionising water is essential for many agricultural and industrial uses – to ensure the consistency of products and to meet strict safety regulations. The global market for water and wastewater technologies reached US $347.9 billion in 2024 and is expected to rise to US $652.30 billion by 2034⁵⁰. Commercial technologies that allow resource recovery from wastewater to be commercially feasible are increasingly being developed, making transitioning to a circular economy an opportunity to accelerate and scale-up the most recent scientific and technological advances that support greater efficiency in the water sector. However, this requires further advances in technologies such as:

• A range of sensors in water systems to monitor water levels, the flow of water through different channels, temperature changes, chemical leakage, pressure level, chemical residues, and biological residues, pH, etc., associated to smart decision support systems that allow to monitor, take action in real-time and forecast the water treatment plant, considering multiple sources of data (from the sensors deployed on-site but also e.g., weather forecast).
• IoT-enabled water purifiers that can predict potential system failures to reduce downtime in water treatment plants, and to enable remote sensing for mapping groundwater resources and monitoring sustainable extraction levels.
• IoT-enabled increase of water recycling and development of the urban circular water economy. As examples waste water from industry and data centers can be used for district heating or treated and recycled to yield not only water but also energy, fertiliser, and organic inputs.
• IoT-enabled increase of water recycling in water-demanding industries (e.g., mining, semiconductor manufacturing) with reduction of the water footprint in manufacturing. Water-efficient processes must be implemented such as closed-loop systems, which capture and recycle the water used in the production process.

.

Among the key focus areas Agriculture is one of the economic activities that has the highest dependence on nature and biodiversity⁵³. On average, global mean crop yields of rice, maize and wheat are projected to decrease between 3% and 10% per Celsius degree of warming above historical levels. All crops depend directly on soil health and fertility, and more than 75% of global food crop types rely on animal pollination. However, the impact of agriculture activity on the environment must be as low as possible to preserve biodiversity. Efforts to conserve existing land resources (e.g. forests) and expand natural-based solutions (e.g. peatlands restoration) are required to reduce the GHG emissions, or to improve resistance by microbial biofertilisers ^{54 55}. In this regard, the EU Biodiversity Strategy 2030 establishes several objectives⁵⁶. To address these objectives, there is a need to develop:

• Precision farming systems and services for optimal use of fertilisers and pesticides.
• Sensing and monitoring systems for in situ measurement of soil nutrients, connected insect traps and landscape monitoring.