Dr Luca Rossini

• Modelling in Crop Protection

Theoretical development of models to describe the interaction between environmental factors and different trophic levels in an agroecosystem: host plant, insect pests, plant diseases, natural enemies, and control strategies (human actions). To pursue this objective, I focus on Ordinary (ODE) and Delay (DDE) Differential Equations and on computational methods for parameter estimation from experimental data. I am used to work with state-estimation techniques (e.g., Kalman Filtering) as well, as they are the key to improve the model estimations based on field data. Although field and lab observations are the starting point for the theoretical developments, my main interest is to obtain general models that can be possibly applied to the majority of terrestrial arthropods and plant pathogens, provided the biological traits of the specific case study.

• Metrology applied to insect science and plant protection at large

This line of research involves both laboratory and field environment. In laboratory, I am interested in the development of standard methods to collect, analyse, and share the information from experiments (e.g., insects’ life tables). I mainly focus on analytical methods to extract quantitative information on the response of the species to external factors. Based on that, I develop methods for model parameter estimation. On the field, I am interested in the development of models describing the measurement process, to further improve the information that measurement can provide with the associated uncertainty, as well.

• Optimal control applied to agroecology and crop protection

This topic came as an output of my Marie Curie Postdoctoral Fellowship project (MSCA-PF-2022 Project PestFinder n. 101102281 – 2023-2025). This line of research is based on the intuition that integrated pest management (or any decision-making process in ecology) can be seen as an optimisation problem composed of a dynamic system, control actions, constraints, and decision variables. Besides the formulation of the problem according to the optimal control principia, I focus on the development of methods to find fast solutions, based on both parallel computing and mathematical approximations.

• Resource allocation and sensing policies

This line aims to optimise the data collection process in agroecology. The goal is to use state-estimation techniques to set optimisation problems which allows model-based data collection campaigns.

• Alternative control methods and agroecological multitrophic interactions

 Study of insect-to-plant, insect-to-insect, and insect-to-enemy interaction and how they are affected by environmental parameters and human actions. This topic involves bioecological studies based on traditional and molecular approaches, the evaluation of insect predators and parasitoids, and the evaluation of the effect of entomopathogenic organisms (mainly fungi). The biological information coming from this line of research actively supports the model development, as it is the main source providing hints and assumptions to model the interaction between trophic levels and of the control actions.