Classification of broadband echoes from prey of a foraging Blainville’s beaked whale
Echolocation signal (FM click) of M. densirostris, observed during this study. The whale recorded was a conspecific foraging with the whale discussed in this study. (a) Time series (arbitrary vertical scale), (b) frequency spectrum.
Two representative examples of echograms displaying scattered FM clicks of a foraging beaked whale which are used to identify echo trains: (a) low density of scatterers, (b) high density of scatterers. Click number is shown on the vertical axis and the horizontal axis shows time since last emitted click. Echo strength is indicated by color with red corresponding to higher sound pressure and blue corresponding to lower sound pressure. Plot (a) shows a single echo train as the whale approaches a target. The horizontal black line represents a period of greater than between outgoing clicks that meet a preset threshold level which, in this case, indicates the whale’s switch to the lower amplitude buzz.
Example of automated structure analysis results. Plus signs indicate nulls and horizontal dashed lines indicate mean relative target strength of frequency band displayed. An example of the two criteria for a null are illustrated in the plot (b): (1) The dip near does not meet the criterion of being more than lower than both adjacent peaks; (2) the dip at does not meet the criterion of being more than lower than the average of the two adjacent peaks.
Broadband acoustic signatures of three prey selected by the whale. Each plot is comprised of frequency spectra of a series of echoes that make up one echo train. Only echoes meeting the ENR criterion are included, hence not all echoes in each echo train are displayed. The echo number of displayed echoes are indicated by the presence of a tick mark. Echo values are interpolated across the gaps associated with echoes not chosen such as between echoes 15 and 17 in plot (a). Plots (a) and (b) show examples of high target strength prey observed at deep depths (below ) in environments with low scatterer density. Plot (c) shows an example of lower target strength prey observed in shallower water (above ) in environments with high scatterer density.
Normalized distributions of various characteristics of echoes from whale-selected prey and randomly chosen scatterers not selected by the whale (black and gray bars, respectively). The characteristics are: (a) number of nulls, (b) location of first null, (c) spacing between the first two nulls, and (d) relative target strength. Each characteristic of an individual echo is weighted by the fraction of the total number of echoes within a train that the echo represents. Distributions are normalized by the total number of echo trains (whale selected or nonselected). Vertical dashed line in target strength plot (d) represents absolute target strength value of based on an arbitrary source level of .
Depth distribution of prey categorized by scatterer density. Depth is truncated above as no predation events were observed at shallower depths.
Normalized distributions of number of nulls in frequency responses of selected and nonselected scatterers (black and gray bars, respectively) in (a) shallow, high density aggregations and (b) deep, low density aggregations. Weighting and normalization are identical to Fig. 5.
Normalized distributions of relative target strengths of whale-selected prey and nonselected scatterers (black and gray bars; respectively) in: (a) shallow, high density aggregations, and (b) deep, low density aggregations. Vertical dashed line represents absolute target strength value of based on an arbitrary source level of . Weighting and normalization are identical to Fig. 5.
Distribution of echo trains (bold) and echoes (parenthesis) from whale-selected and nonselected scatterers observed in three dives examined. All echoes selected for analysis (bottom half of table) had an echo-to-noise ratio, ENR, of at least .
Article metrics loading...
Full text loading...