Figure 1.
The sampling frequency was adjusted for each call type.
Table 1.
Sound pressure level (dB) of broadcasted calls measured at 1.5 m distance from the loudspeaker in a sound proof chamber.
Table 2.
Description of the acoustic parameters used in the analyses.
Figure 2.
Amplitude attenuation in relation to habitat type, distances and calling height.
The call types are shown in the different panels: H = harsh bark, S = scream, W = wahoo, G = grunt, C = clear call, B = clear bark. The maximal amplitude of the amplitude envelope (MaxAmpl) is calculated in mV. Mean values and standard errors are indicated.
Table 3.
Reliability in relation to call type and acoustic parameters measured as coefficient of variation (CV).
Figure 3.
Variability of the fundamental frequency (F0) by using two different calculation methods.
The x-axis shows the different call types, G: grunt, C: clear call and B: clear bark. The y-axis shows the relative differences of the acoustic parameters for each condition. Panel rows represent different heights, while panel columns represent different distances. (A) The variability under different conditions calculated via the automatic macro. (B) The variability calculated manually. The plots indicate mean values from ten repetitions at one locality; error bars indicate the confidence interval of 95 %. The horizontal lines denote a variation of 0 %.
Table 4.
Validity of the fundamental frequency in tonal calls in relation to distance, height and habitat.
Figure 4.
Validity of four different parameters in relation to habitat type, transmission height, distances and call type.
Four different acoustic parameters are shown in the graphs; (A) Duration, (B) DFA1 mean, (C) PF max, (D) PF mean. The x-axis shows the different call types, H: harsh bark, S: scream, W: wahoo, G: Grunt, C: clear call and B: clear bark. The y-axis shows the relative differences of the acoustic parameters for each condition. Panel rows represent different heights, while panel columns represent different distances. The plots indicate mean values across five different localities; error bars indicate the confidence interval of 95 %. The horizontal lines denote a variation of 0 %.
Figure 5.
Comparison of peak frequency (PF mean) from two different call types.
The Peak frequency (PF mean) of a tonal grunt (A) and a scream (B) recorded in the soundproofed chamber compared to the same grunt recorded in a dense habitat at 0.5 m height and 6.25 m distance and the same scream recorded in a dense habitat at 0.5 m height and 50 m distance. PF mean is indicated.
Table 5.
Validity of characteristic sound parameters in relation to distance, height and habitat.
Table 6.
Validity of characteristic sound parameters in relation to distance, height and habitat for calls (screams, grunts, and juvenile clear calls) types with an original recording distance below five meter.