Setting up of alert systems for the early detection of fungal diseases on Rosa spp.

Plants belonging to the Rose genus are among the main products of national and international floricultural sector, both as landscaping shrubs and for cut flower production. Depending on the type of production, rose plants are usually grown in soilless systems, e.g., pot, in open field conditions or into greenhouses. Both types of cultivation are very susceptible to fungal pathogen attacks that can strongly affect plant growth and aesthetic appearance constituting a serious concern for producers. The decreasingly availability of effective fungicides, for both environmental issues and the increasing pathogen resistance, necessitates the development of alternative methods for pathogen control. In particular, an early detection of pathogens is critical to face the ambitious challenge of a sustainable disease management. In this regard, the development and validation of models based on pathogen cycle stage in relation with the micrometeorological conditions seem a promising strategy for the pathogen spread prediction and prevention. Thus, the aim of this work was the development of alert systems for fungal disease outbreaks for rose plants based on new models on micrometeorological parameters or the improvement of existing ones. The models were validated with the help of imaging tools (i.e., multispectral camera) and molecular biology methods. In particular, two experimental sites were set up: i) landscaping floribunda roses in a nursery, 384 plants organized in 24 blocks of 16 plants belonging to three cultivars with different sensitivity to fungal pathogens, monitored with a weather station, five intra-canopy air humidity and temperature sensors, two substrate humidity and temperature sensors; ii) cut flower roses in a protected environment, 288 plants organized in 18 blocks of 16 plants belonging to three cultivars with different sensitivity to fungal pathogens, monitored with 5 extra- and intra-canopy air temperature and humidity sensors and four substrate humidity and temperature sensors. The spread of fungal pathogens, i.e., Botrytis cinerea and Sphaerotheca pannosa, two of the main fungi affecting aerial parts of rose plants, was monitored for one year in combination with multispectral image elaborations and molecular detection of pathogen DNA as well as a weekly record of plant symptoms. Main findings and major problems regarding the early detection of pathogens with these integrated techniques will be discussed focusing on the future perspectives of the applications of these precision agriculture techniques in the ornamental sector.