ISHS VI International Congress on Hazelnut BIO-ETHOLOGY OF ANISANDRUS DISPAR F. AND ITS POSSIBLE INVOLVEMENT IN DIEBACK (MORIA) DISEASES OF HAZELNUT (CORYLUS AVELLANA L.) PLANTS IN CENTRAL ITALY

ISHS VII International Congress on Hazelnut NEW OBSERVATION ON BIOLOGY OF EUROPEAN SHOT-HOLE BORER [XYLEBORUS DISPAR (F.)] ON HAZEL IN NORTHERN LATIUM (CENTRAL ITALY)

In this work, inspired by the needs of the H2020 European project PANTHEON for the precision farming of hazelnut orchards, we propose a data-driven pest detection system. Indeed, the early detection of pests represents an essential step towards the design of effective crop defense strategies in Precision Agriculture (PA) settings. Among the possible pests, we focus on true bugs as they can heavily compromise hazelnut production. To this aim, we collect a custom dataset in a realistic outdoor environment and train a You Only Look Once (YOLO)-based Convolutional Neural Network (CNN), achieving ≈ 94.5% average precision on a holdout dataset. We extensively evaluate the detector performance by also analyzing the influence of data augmentation techniques and of depth information. We finally deploy it on a NVIDIA Jetson Xavier on which it reaches ≈ 50 fps, enabling online processing on-board of any robotic platform.

A large proportion of the insects which have invaded new regions and countries are emerging species, being found for the first time outside their native range. Being able to detect such species upon arrival at ports of entry before they establish in non-native countries is an urgent challenge. The deployment of traps baited with broad-spectrum semiochemical lures at ports-of-entry and other high-risk sites could be one such early detection tool. Rapid progress in the identification of semiochemicals for cerambycid beetles during the last 15 years has revealed that aggregation-sex pheromones and sex pheromones are often conserved at global levels for genera, tribes or subfamilies of the Cerambycidae. This possibly allows the development of generic attractants which attract multiple species simultaneously, especially when such pheromones are combined into blends. Here, we present the results of a worldwide field trial programme conducted during 2018–2021, using traps baited with a standardised 8-pheromone blend, usually complemented with plant volatiles. A total of 1308 traps were deployed at 302 sites covering simultaneously or sequentially 13 European countries, 10 Chinese provinces and some regions of the USA, Canada, Australia, Russia (Siberia) and the Caribbean (Martinique). We intended to test the following hypotheses: 1) if a species is regularly trapped in significant numbers by the blend on a continent, it increases the probability that it can be detected when it arrives in other countries/continents and 2) if the blend exerts an effective, generic attraction to multiple species, it is likely that previously unknown and unexpected species can be captured due to the high degree of conservation of pheromone structures within related taxa. A total of 78,321 longhorned beetles were trapped, representing 376 species from eight subfamilies, with 84 species captured in numbers greater than 50 individuals. Captures comprised 60 tribes, with 10 tribes including more than nine species trapped on different continents. Some invasive species were captured in both the native and invaded continents. This demonstrates the potential of multipheromone lures as effective tools for the detection of ‘unexpected’ cerambycid invaders, accidentally translocated outside their native ranges. Adding new pheromones with analogous well-conserved motifs is discussed, as well as the limitations of using such blends, especially for some cerambycid taxa which may be more attracted by the trap colour or other characteristics rather than to the chemical blend.