When it comes to Earth observations in AI for disaster risk reduction, is it feast or famine? A topical review

Earth observations (EO) have successfully been used to train artificial intelligence (AI)-based models in the field of disaster risk reduction (DRR). Given the number of in situ and remote (e.g., radiosonde or satellite) monitoring devices, there is a common perception that there are no limits to the availability of EO for immediate use in such AI-based models. However, a mere fraction of EO are actually being used in this way. We explore reasons for this discrepancy. First, we look at the types of EO needed to train AI-based models for DRR applications. Second, we consider the two main types of EO (in situ vs remotely sensed) and, for each, we explore: main characteristics, possible challenges, and innovative solutions, which are drawn from selected use cases. Finally, we suggest steps that can be taken to make EO more user ready and to facilitate its uptake in AI for DRR and beyond.

KUGLITSCH Monique;  ALBAYRAK Arif;  LUTERBACHER Juerg;  CRADDOCK Allison;  TORETI Andrea;  MA Jackie;  PADRINO VILELA Paula;  XOPLAKI Elena;  KOTANI Rui;  BEROD Dominique;  COX Jon;  PELIVAN Ivanka; 

2023-09-27

IOP PUBLISHING LTD

JRC132091

1748-9326 (online),   

https://dx.doi.org/10.1088/1748-9326/acf601,    https://publications.jrc.ec.europa.eu/repository/handle/JRC132091,   

10.1088/1748-9326/acf601 (online),