Frequency distribution of samples without detected airgun pulses, as a function of date.

The black bars show the total number of cell-time intervals while the white bars show the number of cell-time intervals without any detected airgun pulses, as a function of date in all four years (2007–2010). The top, nearly flat part of the black bars corresponds to the dates during which all DASAR arrays were deployed in all years. The right edge of the plot is steeper than the left edge because the period of retrieval in October was less variable in timing than the period of deployment in August (see Table 1). The figure shows that cell-time intervals without detected seismic activity are not biased toward certain phases of the migration, and thus any differences between calling rates found between no seismic and seismic cell-time intervals are not contaminated by seasonal effects.