Estimating whale density from their whistling activity: Example with St. Lawrence beluga

Abstract

A passive acoustic method is developed to estimate whale density from their calling activity in a monitored area. The algorithm is applied to a loquacious species, the white whale (Delphinapterus leucas), in Saguenay fjord mouth near Tadoussac, Canada, which is severely affected by shipping noise. Beluga calls were recorded from cabled coastal hydrophones deployed in the basin while the animal density was estimated visually from systematic observations from a fixed-point on the shore. Beluga calling activity was estimated from an algorithm extracting the call events in time–frequency space, while simultaneously tracking the masking intensity resulting from local shipping noise. The activity index was summarized in 15- and 30-min bins using four different metrics. For bins containing more than 40% of valid data, the metrics were compared to the corresponding visual observations. The estimated mean acoustic detection range generally exceeded the fjord width, and extended to the whole ∼3-km long monitored area under low-noise conditions. The significant linear relations of the visual estimates with the calling activity metrics allowed assessing expected number of visually detected belugas in the basin from a weighted regression model, with a mean standard error of 7.1%.

Introduction

The mouth of Saguenay Fjord at Tadoussac, Québec, Canada (Fig. 1) is frequently visited by belugas from the St. Lawrence population. This area forms a small cusp-like basin about 3.5 km in diameter and 150 m deep, which is isolated from the adjacent St. Lawrence estuary by a shallow (<20-m deep) sill. Belugas can be visually localized over the whole basin from elevated points on both shores. Because of the small size and the simple topography of the basin, belugas calling in the area have a high probability of being detected from the recordings of a hydrophone deployed in the basin. The shallow sill tends to isolate the basin from possible distant calling sources from the St. Lawrence. Therefore, detected calls must come from belugas that are present in the fjord. Systematic observations carried out by Parks Canada during summer months since 2003 (Saguenay St. Lawrence Marine Park, Parks Canada, unpublished data) show that the most occupied zone corresponds to the basin as delimited in Fig. 1 for the visual scanning area. With these particular conditions, the probability that beluga calls recorded in the basin come from the same beluga groups seen by coastal observers is very high. These characteristics make this area a quasi ideal location for comparing visual and acoustic observations, and exploring the possibility of using passive acoustics to monitor local beluga density.

However, local shipping noise from the car ferry line connecting the main road on both sides of the fjord (Fig. 1) and from the whale watching fleet operating in this area complicates the situation. Two ferries cross the fjord from both sides simultaneously every 20 min during daytime (8:00–21:00). During night, one ferry serves both sides and crosses every 20 or 30 min. A third ferry is added during the busiest traffic season, from June 9th to September 7th, and operates from 10:30 to 16:30, which increases the crossing pace to one ferry per 13 min. During that period, one ferry is always transiting when the other two are loading and unloading at the two docks. Besides the usual shipping noise from the transits and docking manoeuvres, impulsive noise is generated when the ferries are docked from load transfers over the gangway by transiting trucks and cars. Noise from boats and ships from the whale watching fleet, yachting, and merchant ships adds to the whole picture. This traffic peaks in mid-July during the tourist season (June to October) and varies daily from 6:00 to 21:00 with a maximum after lunch. Besides the ferries, the number of boats present in the study area during the visual observations in September 2008 averaged 1.4 (SD = 1.5) for a maximum of 8, with 3 or more cruising boats 20% of the time (Saguenay St. Lawrence Marine Park, Parks Canada, unpublished data).

The first attempt to measure whale sounds in their natural environment was carried out on belugas frequenting the same study area at the mouth of the Saguenay fjord by Schevill and Lawrence [1]. In this pioneering bioacoustics work, the authors highlight the loquacious behavior of belugas and underline the good correspondence between the visual and acoustic occurrences by reporting “…we never sighted them without hearing them and rarely heard them without seeing them, except after dark”. In this paper, we want to see if this acoustical behavior of belugas can be quantitatively exploited to count them. The objectives of the present study were to determine if beluga communication activity measured by coastal hydrophones deployed in this regularly frequented but often noisy habitat of the mouth of the Saguenay fjord was (1) correlated with the number of visually estimated individuals, and (2) if so, whether an estimator of beluga density based on a statistical model of the observed visual counts vs passive acoustic metrics could be developed.