A volcano can be defined as a complex system, not least for the hidden clues related to its internal nature. Innovative models grounded in the Artificial Sciences, have been proposed for a novel pattern recognition an...A volcano can be defined as a complex system, not least for the hidden clues related to its internal nature. Innovative models grounded in the Artificial Sciences, have been proposed for a novel pattern recognition analysis at Mt. Etna volcano. The reference monitoring dataset dealt with real data of 28 parameters collected between January 2001 and April 2005, during which the volcano underwent the July-August 2001, October 2002-January 2003 and September 2004-April 2005 flank eruptions. There were 301 eruptive days out of an overall number of 1581 investigated days. The analysis involved successive steps. First, the TWIST algorithm was used to select the most predictive attributes associated with the flank eruption target. During his work, the algorithm TWIST selected 11 characteristics of the input vector: among them SO<sub>2</sub> and CO<sub>2</sub> emissions, and also many other attributes whose linear correlation with the target was very low. A 5 × 2 Cross Validation protocol estimated the sensitivity and specificity of pattern recognition algorithms. Finally, different classification algorithms have been compared to understand if this pattern recognition task may have suitable results and which algorithm performs best. Best results (higher than 97% accuracy) have been obtained after performing advanced Artificial Neural Networks, with a sensitivity and specificity estimates over 97% and 98%, respectively. The present analysis highlights that a suitable monitoring dataset inferred hidden information about volcanic phenomena, whose highly non-linear processes are enhanced.展开更多
High-resolution digital topography is essential for land management and planning in any type of territory as well as the reproduction of the Earth surface in a geocoded digital format that allows several Digital Earth...High-resolution digital topography is essential for land management and planning in any type of territory as well as the reproduction of the Earth surface in a geocoded digital format that allows several Digital Earth applications.In a volcanic environment,Digital Elevation Models are a valid reference for multi-temporal analyses aimed to observe frequent changes of a volcano edifice and for the relative detailed morphological and structural analyses.For the first time,a DTM(Digital Terrain Model)and a DSM(Digital Surface Model)covering the entire Mt.Etna volcano(Italy)derived from the same airborne Light Detection and Ranging(LiDAR)are here presented.More than 250 million 3D LiDAR points have been processed to distinguish ground elements from natural and anthropic features.The end product is the highly accurate representation of Mt.Etna landscape(DSM)and ground topography(DTM)dated 2005.Both models have a high spatial resolution of 2 m and cover an area of 620 km2.The DTM has been validated by GPS ground control points.The vertical accuracy has been evaluated,resulting in a root-mean-square-error of±0.24 m.The DTM is available as electronic supplement and represents a valid support for various scientific studies.展开更多
文摘A volcano can be defined as a complex system, not least for the hidden clues related to its internal nature. Innovative models grounded in the Artificial Sciences, have been proposed for a novel pattern recognition analysis at Mt. Etna volcano. The reference monitoring dataset dealt with real data of 28 parameters collected between January 2001 and April 2005, during which the volcano underwent the July-August 2001, October 2002-January 2003 and September 2004-April 2005 flank eruptions. There were 301 eruptive days out of an overall number of 1581 investigated days. The analysis involved successive steps. First, the TWIST algorithm was used to select the most predictive attributes associated with the flank eruption target. During his work, the algorithm TWIST selected 11 characteristics of the input vector: among them SO<sub>2</sub> and CO<sub>2</sub> emissions, and also many other attributes whose linear correlation with the target was very low. A 5 × 2 Cross Validation protocol estimated the sensitivity and specificity of pattern recognition algorithms. Finally, different classification algorithms have been compared to understand if this pattern recognition task may have suitable results and which algorithm performs best. Best results (higher than 97% accuracy) have been obtained after performing advanced Artificial Neural Networks, with a sensitivity and specificity estimates over 97% and 98%, respectively. The present analysis highlights that a suitable monitoring dataset inferred hidden information about volcanic phenomena, whose highly non-linear processes are enhanced.
基金This work was partially supported by the Ministero dell’Istruzione,Universitàe Ricerca through the Italian Project FIRB FUMO‘Sviluppo Nuove Tecnologie per la Protezione e Difesa del Territorio dai Rischi Naturali’.
文摘High-resolution digital topography is essential for land management and planning in any type of territory as well as the reproduction of the Earth surface in a geocoded digital format that allows several Digital Earth applications.In a volcanic environment,Digital Elevation Models are a valid reference for multi-temporal analyses aimed to observe frequent changes of a volcano edifice and for the relative detailed morphological and structural analyses.For the first time,a DTM(Digital Terrain Model)and a DSM(Digital Surface Model)covering the entire Mt.Etna volcano(Italy)derived from the same airborne Light Detection and Ranging(LiDAR)are here presented.More than 250 million 3D LiDAR points have been processed to distinguish ground elements from natural and anthropic features.The end product is the highly accurate representation of Mt.Etna landscape(DSM)and ground topography(DTM)dated 2005.Both models have a high spatial resolution of 2 m and cover an area of 620 km2.The DTM has been validated by GPS ground control points.The vertical accuracy has been evaluated,resulting in a root-mean-square-error of±0.24 m.The DTM is available as electronic supplement and represents a valid support for various scientific studies.
