Dr. Muhammad Mumtaz Khan
Department of Plant Sciences, College of Agricultural and Marine Sciences
Climate change, water shortages, and a continuously growing population are putting pressure on global agricultural production and demand. For decades, farmers have been using their experiences and visual assessments to monitor their crop performances. However, these traditional methods are no longer sufficient to warrant their best crop productivity due to climate variability, droughts, new pest and disease emergence, lack of real-time information, and best agricultural management practices. It is worth mentioning that arable lands throughout the globe are declining, while the population is increasing exponentially, which is expected to be 10 billion by 2050. Therefore, it is imperative to enhance crop productivity and sustainability through the integration of technologies to revolutionise agriculture to meet the global food demands.
Recently, in modern agriculture, advanced tools such as agricultural sensors and other high-tech instruments have been used to detect early plant stresses and soil fertility strata. Moreover, hyperspectral and multispectral cameras, thermal imaging LiDAR, and chlorophyll fluorescence detect biochemical and physiological changes in plants before any visible symptoms appear. These technological advances made it possible to detect drought early, pre-symptomatic diseases and insects through identification, and assess nitrogen-chlorophyll levels precisely.
Artificial Intelligence (AI) is an advanced tool that has been transforming crop monitoring and management practices robustly. For example, image analysis systems identify plant diseases, insects, and plant nutrient status from a single leaf image with high accuracy. Models such as convolutional neural networks (CNNs), MobileNet, and Vision Transformers are increasingly used on smartphones and drones, accelerating disease and other pest detection and making it more accessible while reducing reliance on manual field inspections. These high-technology sensors and devices also observe biochemical and physiological changes before disease, or any stress symptoms appear.
Furthermore, machine learning also assists with appropriate decisions by predicting yields and analysing spectral data. However, there are still limitations as many models focus on computational accuracy, incorporating plant biological processes, which makes them less efficient in real field conditions. These deficiencies could be covered by using the Internet of Things (IoT), which enables real-time monitoring across entire fields. Communication systems such as LoRaWAN, ZigBee, and NB-IoT connect systems and gather information on soil moisture, temperature, and plant microclimate to help farmers to monitor crops and respond quickly to emerging issues.
Decision Support Systems (DSS) represent another critical method through which the data can be turned into beneficial advice, as well as provide better opportunities for dealing with pests, optimising irrigation, and fertilisation. Unfortunately, at present, DSS platforms remain underdeveloped, and they cannot be used to control sensor networks and AI algorithms, reducing the practicality of this approach. On the other hand, digital agriculture is characterised by the lack of integration, as drones, satellite imaging, and sensors are independent systems and share very little data. The future of agriculture depends on the full-scale integration of smart farming systems, where sensors identify stresses, IoT networks transmit data, drones monitor fields, machine learning algorithms analyse the external environment, and decision-making platforms suggest actions. All mentioned techniques combined constitute what is referred to as "autonomous agriculture" and make farming more productive, while providing efficient resource utilisation.
Thus, the upcoming decade is expected to be a critical period for global agriculture, as the integration of biology, engineering, and artificial intelligence in agriculture might bring resilience to its production and sustainability. Furthermore, these technological advances will be able to contribute to food security, save water and fertiliser resources, reduce production risks, and ensure sustainable intensification.
Therefore, emerging agricultural technologies offer significant opportunities depending on their integration and practical application in combining AI, IoT, and decision-support tools, which are needed to address the global challenges of climate change, resource scarcity, and rising food demands. My colleagues and I at the Department of Plant Sciences, with the collaboration of the Department of Electrical and Computer Engineering, have been exploring such initiatives to enhance the food production systems and food security.