Map of the study area. The black square on the map of China indicates the oxbow lake where biologging experiments were conducted. This lake is part of the Tian-e-Zhou Baiji National Natural Reserve of the Yangtze River, Shishou, Hubei Province. The acoustic monitoring station was located near the confluence of the Yangtze River and Poyang Lake, which is located in the middle reaches of the Yangtze River. The densities were estimated at Station 0-3. The dash lines of the lower map indicated the water direction.
Example of four click trains of Yangtze finless porpoise recorded with a stereo acoustic data logger, A-tag. The vertical axes show the received SPL (dB, upper) and inter-click interval (ms, lower).
ROC curve for the present acoustic filter for detecting click trains of finless porpoises with optimum CV value. We adopted a value of 0.40 for the CV.
Histogram of inter train intervals obtained from 15 tagged porpoises.
Estimated detection probability of click trains by distance. Probabilities were calculated from source level values recorded from porpoises in a wild habitat ( Li et al., 2009 ). (c) was obtained by (a) with a weight of 4.4% of on-axis beam and (b) with weight of 95.6% of on-axis beam. The solid line of (a) and (b) shows , which was calculated as a fitted curve .
Correlation between the average number of porpoises and the number of click trains detected in 1 min at Stations 0–2 (left) and during 1 day at Station 3 (right). The number of porpoises was averaged over either every hour (left) or day (right).
Changes in the estimated density of finless porpoises . Gray bars indicate periods that were not monitored.
Parameters used in the detection filters for sonar signals. The ROC curve for each parameter was examined. Appropriate values for each parameter were chosen according to the highest correct-detection and lowest false-alarm rates. ICI and CT were abbreviations of inter-click interval and click train, respectively.
Detailed information of tagged porpoises. Both data logger worked during the time of effective data. The data were excluded when the animal’s depth was shallower than 0.3 m and its swimming speed from the beginning of respiration was slower than 0.2 m/s; these data are shown as % of the data excluded.
Article metrics loading...
Full text loading...