International Journal of Environment and Climate Change

Phenomics, which involves the holistic analysis of plant phenotypes using cutting-edge sensing and data processing technologies, has emerged as an AI-powered technique in plant breeding. Genomics tells us about a plant's genetic potential, whereas phenomics describes the dynamic manifestation of the genes under different environmental conditions. By combining high-throughput phenotyping technologies with AI and machine learning (ML) techniques, complex agronomic traits can be assessed rapidly, accurately and non-destructively, thus speeding up breeding. Advancements in phenotyping technologies, including hyperspectral and multispectral imaging, thermal sensors, LiDAR (Light Detection and Ranging), unmanned aerial vehicles (UAV) and ground robotic systems, provide vast amounts of information about plant growth, physiology, stress, disease, yield potential and other traits. The AI methods such as computer vision and deep learning process the data to discern patterns, forecast traits and detect stress or disease. This helps overcome the phenotyping bottleneck in conventional breeding and improves selection accuracy. AI-based phenomics integrates genomics, phenomics and environmental information to facilitate predictive breeding and mapping of genotype to phenotype to breed climate-adapted, high-yielding varieties. It also helps in precision agriculture for monitoring crops and resource allocation. Phenomics studies are being used in crops like rice, wheat, maize, barley, sorghum and soybean. In India, rice and wheat are extensively studied for high-throughput phenotyping for drought and heat tolerance. Addressing data standardisation and infrastructure issues, phenomics is a disruptive technology, bringing together the genetic potential and field performance for sustainable crop improvement.