The science of interpreting plants

By Mafalda Reis Pereira, research assistant at the Centre for Robotics in Industry and Intelligent Systems (CRIIS)

Plants play a crucial role to human life, as an important source of food and nutrients. The statistics of the Food and Agriculture Organization (FAO) of the United Nations (UN) estimate an increase of the world’s population over the next 40 years, predicting that it will reach approximately 9.1 billion people by 2050. Consequently, there will be an increase in food demand, as well as the need for a continuous increase in agricultural production by about 70%. Experts believe that said growth will be achieved essentially through higher yields and agricultural intensification. Thus, the optimisation of plant protection conditions and practices has been considered a key goal for the sustainable development by the UN. The UN General Assembly declared 2020 as the “International Year of Plant Health” (IYPH) to raise awareness of how protecting crop health can help end hunger, reduce poverty, protect the environment, and boost economic development. Moreover, 2021 was elected as “International Year of Fruits and Vegetables” (IYFV) to emphasise the key role in human nutrition, food security and health, as well as to achieve the UN Sustainable Development Goals. 

During the production process, plants are affected by different types of biotic and abiotic stresses that can influence productivity and quality. This leads to income losses for producers and reduced availability and higher prices for consumers. Overall, biotic agents are estimated to be responsible for crop yield losses around 20-40%. Current agricultural practices also promote the spread of plant disease epidemics and the fast evolution of pathogens, since they favour the monoculture intensification in larger areas, the use of genetically uniform plant varieties and the development of supply chains and global logistical activities (enabling the circulation of plant material worldwide). Phytosanitary products are usually applied to prevent and combat these agents; however, the undiscriminating application of said products (often carried out at the wrong time, dosage and area) could damage the environment (affecting soil, water and air quality). 

This way, there are many agronomic, environmental, economic, and humanitarian reasons that justify the development of new methods of early diagnosis of plant diseases, as well as their mapping in the field, in line with precision agriculture. 

Recently, so-called indirect diagnostic methods (proxy) have emerged. These are based on the occurrence of interactions between the plant and the pathogen that affects it, which could lead to changes in the internal and biochemical structure of the host. These physiological aspects promote changes in the optical properties of plants, specifically in their reflectance and emittance, which can be detected by proximity optical sensors (e.g., through multi/hyperspectral sensors, fluorescence sensors, or thermography). This leads to a hypothesis focused on following the spatio-temporal pattern of development of plant diseases through their reflectance and emittance, allowing an early diagnosis in a quick, easy, non-invasive, and specific way. 

Considering this context, I am currently developing my research and PhD thesis on “Early detection and identification of plant diseases caused by bacteria based on proximal sensing from a precision agriculture perspective” at the Faculty of Sciences of the University of Porto (FCUP) and INESCTEC – more specifically at the Centre for Robotics in Industry and Intelligent Systems (CRIIS). This project aims to develop predictive methods, based on the spectral properties of plants, for the detection and early identification of bacteria responsible for diseases in different agricultural crops. It comprehends the development of reliable and fast diagnosis methodologies through the combination of fundamental science (e.g., plant physiology and biochemistry), diverse optical sensors and artificial intelligence techniques. The tests are being carried out under laboratory and field conditions, using as a case study different pathovars of the genus Xanthomonas spp. and of the species Pseudomonas syringae, as well as different crops such as kiwi, walnut, and tomato. Its validation will justify expanding this type of study to other crops and other pathogens, such as fungi that cause damage and lead to losses in terms of agricultural crops. 

The goal is to improve the diagnosis (detection of the disease) at an early stage of the infection process, thus enabling early interventions. By preventing and controlling the spread of the infection and pathogen, and by modifying cultural practices, it is possible to act and prevent certain crops from being totally affected and compromised. This research also explores the possibility of using predictive methods to map diseases, allowing a targeted and precise application of plant protection products. This can lead to a reduction in the use of pesticides and herbicides, which translates into a beneficial impact on the protection of the environment and ecosystem services, the producers’ profits and the quality of the products that reach the final consumer. This project is aligned with the challenges that European agriculture is currently facing and with the scope of the European Green Deal, namely the operationalisation mechanisms established in the “Farm to Fork” to reduce the use of plant protection products and fertilisers by 50% by 2030. The development of more automated, objective, and sensitive diagnostic methods is, for all these reasons, crucial to boost disease detection in crops of agronomic interest. 

Working at INESC TEC has been extremely rewarding and allows me to collaborate with amazing multidisciplinary teams (CRIIS and the Centre for Applied Photonics – CAP). This collective work has provided me with tools and knowledge in various fields (including engineering, physics, biology, agronomy, statistics, modelling, data science, machine learning, etc.). I also had the opportunity to collaborate with INESC TEC’s Commission for Diversity and Inclusion, which allowed me to better understand the value and variety of different employees and researchers. It has been a challenging journey, but it gives me a huge sense of professional and personal fulfilment.  

PHP Code Snippets Powered By :