S06 - Session P6 - Technologies and post harvest methods for successful horticultural production - Setting up of fungal risk model for the early detection of powdery mildew on Rosa spp

Authors: Silvia Traversari, Sonia Cacini *, Piero Battista, Beatrice Nesi, Daniele Massa, Catello Pane, Bernardo Rapi, Maurizio Romani, Francesco Sabatini

Rose plantsare among the main products of the national and international floricultural sector for cut flower production in greenhouses soilless systems. A serious concern for producers is constituted by fungal diseases that can strongly affect plant growth and aesthetic appearance compromising flower marketability. The always lower availability of efficient fungicides, for both ecological problems and pathogen resistance issues, draws attention to the clear need for alternative methods for pathogen control. In this context, early detection of fungal pathogens is a key aspect to face the ambitious challenge of sustainable disease management. The development and validation of epidemiological models constitute a promising strategy for pathogen spread prediction and prevention. In particular, the aim of this work was the tuning of models based on the pathogen cycle stage in relation with the microclimatic condition within the plant canopy. The model was validated through the detection and quantification of pathogen DNA by RT-PCR on three cultivars of cut flower roses with different sensitivity to fungi, grown in a greenhouse organised in 18 randomly distributed blocks of 16 plants and monitored with 5 extra- and intra-canopy temperature and humidity sensors, spatially distributed in the cultivation area. The spread of powdery mildew ( Sphaerotheca pannosa var. rosae ) was monitored on every block for two years, through weekly reports of pathogen diffusion and symptom severity on plant canopy. The weekly observations were compared with the output of a risk prediction model based on the micrometeorological conditions measured in each block allowing its tuning. The DNA quantification enabled the detection of pathogen presence also during the incubation stage and the observed data suggested the possibility to obtain an early forecast of fungal cycle stage. These results will be used for the implementation of future operational alert systems for fungal outbreaks on rose plants.