Fungi cause more than $100 billion in crop damage worldwide annually. Fungal infections are spread by the distribution of fungal spores via water flow, wind currents, and insects. Detecting and monitoring spores will facilitate disease management strategies to reduce economic loss. In this project the detection of pathogens causing fusarium head blight (FHB) was used as a model system. The detection of other fungal species will follow as the technology matures.

Among the top 10 fungal pathogens in molecular plant pathology, Puccinia spp. are ranked third and Fusarium spp. are ranked fifth. Rust and fusarium head blight are devastating plant diseases and several pathogen species have a large economic impact on wheat production worldwide. Three highly specialized species of fungi are responsible for the rust diseases of wheat: Puccinia graminis, cause of stem rust; Puccinia recondite, cause of leaf rust; and Puccinia striiformis, cause of stripe rust.

The southeastern part of the prairies in Western Canada is the main stem and leaf rust hazard region. Under normal conditions in this region, there is usually a race between rust epidemic development and wheat maturity. In the case of winter wheat, crop maturity usually wins the race and yield losses due to rust are minimal. The later maturing spring wheat cannot be successfully grown in the rust area of the Canadian prairies without a high level of cultivar resistance or other effective methods of rust control.

Currently, the development of rust resistant wheat cultivars provides the most cost effective means of controlling rust. However, as the rust races are constantly changing the new races of rust can quickly increase to epidemic levels. The major mechanism for rust fungi dispersal is atmospheric transport and deposition. Trap nurseries and other currently used molecular methods like real time PCR assay to follow movements of rusts, are both labor intensive and expensive. Therefore, an early warning system for the arrival of rust pathogen spores is highly desired. There is no such commercial product currently available. Developing such a system is a technology breakthrough in the agriculture sector.

The project had two main objectives:

1) To design and build a sensor within a spore trap to detect the presence of specific spores. The sensor within a spore trap specifically detects the presence of FHB and rust spores without the need for the transportation, incubation and visual inspection of a traditional spore trap filter.

2) To test the sensor and build and test a wireless sensor network in the field to detect and monitor the presence of the spore, its populations, and its proliferation. Besides the detection of the spore, other environmental conditions for the growing of the spore were also collected by the sensor network, including temperature, wind speed, humidity, and total sunlight. A software program will be developed based on the model to assist the prediction of the spore.

The development of the fungus sensor stated in the first objective was successful as the device was installed and tested in the field to collect data.

The second objective of the project was achieved ahead of time as the investigators and students had more experience in building wireless sensor network and Internet of Things. Each sensor network provides weather and environment data to the users and can be placed anywhere in Saskatchewan using Bell wireless internet to store the data on the “cloud”. The prediction algorithms have been developed and will be improved as new data are collected in the three systems installed in the field around Saskatchewan.

The results from this project can also be used in other applications, in particular the new biosensor we developed. Numerous applications in agriculture can be built around the sensor and IoT system with our expertise. The work and our expertise are getting attention from the other researchers in Saskatchewan, particular in plant sciences and computer sciences. Internationally, researchers in Vietnam, Australia, and Columbia are interested in working with us to use the system and algorithms in real applications. In additional to IoT sensing system with machine learning techniques for modeling and prediction, spectroscopy can be used in agriculture and plant sciences, security, food, and other areas. We are currently using the built sensor to monitor the water flow in plants to study their drought tolerant characteristics. The system is to be used in the near future to monitor the efficacy of the pathogen cleaning system.

Funding from Agriculture Research Branch, Saskatchewan Ministry of Agriculture and Western Grains Research Foundation.