Characterisation of the radiative response of crops covered with tephra

Tephra emissions associated with volcanic eruptions impact human activities, and notably the agricultural sector. To assess the risk of tephra to crops, data on crop vulnerability are needed. However, these are difficult to acquire in the field and as a result they remain limited. Information on crop vulnerability may be retrieved through remote sensing, which provides regular, cost effective and rapid data across time and space data. The collection of the normalized difference vegetation index (NDVI) is a first step into the characterization of crop vulnerability to tephra. In this thesis, the spectral properties of tephra were first assessed. These are dictated by the silica content, the grain size, and the chemical composition of the tephra. Finer grain size of tephra with low contents of Fe and Ti usually led to higher reflectance, whereas the presence of water absorption bands typically produced low reflectance values. Secondly, we conducted an experiment where tephra was applied to wheat plants and the reflectance of the canopy was measured using a multispectral camera attached to a drone. Tephra spraying on the foliage led to a temporary decrease in NDVI when specific tephra deposits were applied. The reduction of NDVI was always more important for finer tephra, with the greater mass load (500 g m-2 compared to 200 g m-2). The effect of tephra on canopy reflectance was also influenced by the phenological stage of the wheat, with a greater NDVI decrease for the early stem extension stage due to the more horizontal orientation of leaves. However, over time the NDVI of the wheat plants re- increased as soon as the tephra deposit was eroded from the leaves. Also, the signal of the leaves which had received tephra at early phenological stage was overshadowed by the appearance of new leaves. The NDVI is a good tool to evaluate the spectral response of wheat plants covered with tephra. Future studies would be needed to further explore the relationship between vegetation index, yield loss and crop vulnerability.