We are excited to announce our recent membership in the European Meteorological Society (EMS), a non-profit organization dedicated to advancing meteorology and related sciences in Europe. We are proud to be part of this prestigious platform and are eager to contribute to the enhancement of meteorological innovation in Europe through our dedication to scientific progress and professional excellence.
We deeply appreciate the chance to enrich the event's agenda with our presentation on the significance of improving fog prediction methods. Fog, being a severe meteorological phenomenon, has a substantial impact on various transportation modes, including air travel, road navigation, and maritime activities, making it of paramount societal importance. Our study focused on the Poprad-Tatry Airport in Slovakia and utilized a range of machine learning algorithms (such as support vector machines, decision trees, and k-nearest neighbors) to forecast fog with visibility below 300 meters for a lead time of 30 minutes.
This research was conducted within the framework of the SESAR research project PJ.04-29.2. Significantly, our study introduces an innovative approach by incorporating data from remote cameras to enhance fog prediction models. These cameras provide visibility information based on multiple landmarks at different distances and directions from the airport. Our most successful model achieved a score level of 0.89 (0.23) for the probability of detection (false alarm ratio).
One of the key findings from our study highlights the valuable predictive role of the minimum camera visibilities from eight cardinal directions. This significantly enhances the machine learning models' ability to forecast the onset and dissipation of fog, essentially predicting when a fog-free situation transitions into fog and vice versa. This innovative use of camera-based observations addresses limitations associated with automated sensors (typically providing point measurements) and human observers (offering less frequent but more complex observations).
We cordially invite you to join us on Friday, September 8, from 11:30 to 11:45, to delve deeper into our research and its far-reaching implications. Furthermore, you will have the opportunity to meet us in person, as we are one of the few organizations setting up a booth at the University premises during the event. We eagerly anticipate sharing our insights and engaging in discussions on this critical meteorological topic.
