Table 2. Acoustic parameters used in Classification and Regression Tree (CART) analysis.

Duration (90%) (s)	90% of the duration of the annotated call
Bout	Number of repetitions of the same call type
Low frequency (Hz)*	Lowest frequency component of the call
High frequency (Hz)*	Highest frequency component of the call
Bandwidth (90%) (Hz)	90% of the difference in frequency between high and low frequency
Start frequency (Hz)*	Starting frequency of fundamental
End frequency (Hz)*	Ending frequency of fundamental
Peak frequency (Hz)*	Frequency of the spectral peak
Center frequency (Hz)*	The frequency that divides the sound equally into two intervals of equal energy
Frequency trend*	Start F0/End F0
Aggregate entropy (bits)	A measure of total disorder in the call (RavenPro, 1.5)

Notes.

Log transformed parameters are indicated with an asterisk (*).