DETECTING AND ANALYZING UNDERWATER AMBIENT NOISE OF GLACIERS ON SVALBARD AS INDICATOR OF DYNAMIC PROCESSES IN THE ARCTIC

Abstract: The physical processes accompanying glacier melting generate underwater sound audible in the fjords of marine-terminating glaciers, and are potentially useful in the study of changing glacier dynamics. Measurements were conducted in two Spitsbergen fjords: the Hornsund – surrounded by melt...

Full description

Bibliographic Details
Main Authors: J. Tegowski, A. Lisimenka, P. Blondel
Other Authors: The Pennsylvania State University CiteSeerX Archives
Format: Text
Language:English
Subjects:
Online Access:http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.456.8685
http://promitheas.iacm.forth.gr/uam_proceedings/uam2011/UAM2011 Paper 18.7.pdf
Description
Summary:Abstract: The physical processes accompanying glacier melting generate underwater sound audible in the fjords of marine-terminating glaciers, and are potentially useful in the study of changing glacier dynamics. Measurements were conducted in two Spitsbergen fjords: the Hornsund – surrounded by melting glaciers and the Murchison fjord – devoid of glaciers but full of floating ice floes. Calm weather during the experiment allowed the recording of sounds produced mainly by the processes associated with melting ice without noise coming from wind, rain or breaking waves. Ambient noise recording were made at frequencies from 20 Hz to 24 kHz using an omnidirectional hydrophone deployed at depth of 18 meters. Significant differences in measured noise levels and spectral shapes are observed between the fjords. A parametric signal analysis of the noise has been undertaken to identify distinct geophysical phenomena, such as the noise generated by the bursting of air bubbles trapped in melting growlers and icebergs, waterfalls from glaciers and calving glacier noise. Recognition and quantification of the noise events was done in 1/3-octave frequency bands. In addition, spectral, statistical and wavelet energy parameters were computed as inputs for a neural network clusterization algorithm, resulting in the separation of ambient noise sources. Intensive glaciological research has shown that the glaciers around the Hornsund can be considered typical of the Atlantic sector of the Arctic. This suggests that the results from research carried out in this area will be of much greater than just local importance.