**5. Conclusions**

This study is the very first to demonstrate the prediction of the spread of Kiwifruit Decline using remote sensing technologies. Our results indicate that unsupervised clustering can be a reliable algorithm for characterising the disease in its early stages, in order to identify homogenous areas out in the field. Multispectral data can be used to discriminate symptomatic from asymptomatic plants, allowing a quick estimation of disease spread. On the other hand, thermal data have been shown to be

effective in predicting future outbreaks of the disease, by providing an informative tool for directing the sampling activities for aetiological and epidemiological studies.

**Author Contributions:** Conceptualisation, F.S., M.M., P.E., S.P. and A.-K.M.; Data curation, F.S. and S.P.; Formal analysis, F.S. and S.P.; Funding acquisition, M.M., P.E. and A.-K.M.; Methodology, F.S., M.M., P.E., S.P. and A.-K.M.; Supervision, A.-K.M.; Writing: original draft, F.S.; Writing: review and editing, M.M., P.E., S.P. and A.-K.M. All authors have read and agreed to the published version of the manuscript.

**Funding:** SP and AKM were partially funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) under Germany's Excellence Strategy, EXC 2070 – 390732324. FS, PE, MM were partially funded by the Friuli Venezia Giulia region (Italy) CUP: G24I19000810002.

**Acknowledgments:** Simone Saro, from ERSA—Phytosanitary Service of the Friuli Venezia Giulia region (Italy)—for his collaboration and support in field-scouting activities. Luca Zuliani, from Adron Technology srl, for his help in the pre-processing of images and generation of the orthomosaics. Gianni Tacconi from CREA—the Council for Agricultural Research and Agricultural Economics Analysis—for his insights regarding the spread of Kiwifruit Decline in Italy.

**Conflicts of Interest:** The authors declare no conflict of interest.
