Fungal diseases are a major concern for crop production. Up to now crop protection has mainly relied on fungicide sprays and host genetic resistance. However, their intensive use boosts the adaptation of fungal populations, causing the decrease of pathogen sensivity to fungicides and the breakdown of host resistance. Therefore, novel crop management techniques have to be developed to be used complementary to fungicides and resistant cultivars in sustainable farming. To achieve that, new levers have to be identified. An epidemic causes crop damages if several steps act in sequence: * Initial inoculum survives on soil and stubble during winter, or is spread by wind from remote survival locations, and produces spores that are spread upwards. * A leaf is reached by pathogen spores * This leaf provides a physical and physiological environment that favors epidemiological processes (infection, penetration, lesion growth, sporulation). * The lesions alter the plant functioning, whose consequences are a quantitative and qualitative decrease of yield. Knowing how a plant population influences these steps enables the definition of three types of plant self-defense against airborne fungal diseases: resistance, tolerance, and escape, resistance being subdivided in genetic (the usual one), phylloclimatic and physiological resistance. Regarding the ability to characterize and model the feedback action of disease on canopy architecture as well as on biomass and grain production, the function-structure plant modeling (FSPM) approach seems well suited. FSPM is the combination of a 3D description of the canopy morphogenesis with a functional plant model, where the basic unit is the individual organ. Introducing fungi-leaf interactions in such model also requres coupling it with phylloclimate models. Phylloclimate corresponds to the physical environment actually perceived by each individual aerial organ of a plant population, and is described by physical variables such as spectral irradiance, temperature, on-leaf water and features of around-organ air (wind speed, temperature, humidity, etc.).Characterizing phylloclimate variables, using experimental work or modeling, raises many questions such as the choice of suitable space- and time-scale as well as the ability to individualize plant organs within a canopy. Recent trends and challenging questions in phylloclimate research are discussed, as well as their possible applications in plant epidemiology. Finally, a brief review of emerging needs in terms of collaborations with mathematicians and computer scientists has been adressed.