声学指数在城市森林鸟类多样性评估中的应用
边琦, 王成, 程贺, 韩丹, 赵伊琳, 殷鲁秦

Exploring the application of acoustic indices in the assessment of bird diversity in urban forests
Qi Bian, Cheng Wang, He Cheng, Dan Han, Yilin Zhao, Luqin Yin
表1 本研究使用的6个声学指数介绍
Table 1 Six acoustic indices used in this study
声学指数 Acoustic indices 详细信息 Details 参考文献 Reference
声熵指数
Acoustic entropy index (H)
将音频划分为多个频段, 使用Shannon指数来计算时间熵(Ht)和频谱熵(Hf), 两者相乘得到声熵指数, 体现声学信号在时域和频域上的复杂度。The audio is divided into multiple frequency bands and the Shannon index is used to calculate the temporal entropy (Ht) and spectral entropy (Hf), which are multiplied to obtain the acoustic entropy index, reflecting the complexity of the acoustic signal in the time and frequency domains. Sueur et al, 2008
声学复杂度指数
Acoustic complexity index (ACI)
该指数用于测量相邻频率带之间振幅的变化, 反映声强的可变性和不规律性, 特别是鸟鸣声。该指数相对不受恒定强度或持续声音的影响。本研究中使用以下参数计算: min_freq = 2000, max_freq = 11000, j = 5, fft_w = 512。This index is used to measure changes in amplitude between adjacent frequency bands, reflecting the variability and irregularity of sound intensity, especially bird calls. The index is relatively unaffected by a constant intensity of a sustained sound. The following parameters are used in this study: min_freq = 2000, max_freq = 11000, j = 5, fft_w = 512. Pieretti et al, 2011
声学多样性指数
Acoustic diversity index, (ADI)
将频谱图划分为多个频段(默认为10), 并使用Shannon指数计算每个频段中超过阈值(默认为-50 dBFS)的声音所占比例。本研究中使用默认参数计算。The spectrogram is divided into frequency bands (default 10) and the percentage of sounds in each band that exceed the threshold (default -50 dBFS) is calculated using the Shannon index. The default parameters are used for calculation in this study. Villanueva-Rivera et al, 2011
声学均匀度指数
Acoustic evenness index (AEI)
将频谱图划分为多个频段(默认为10), 并使用Gini指数计算每个频段中超过阈值(默认为-50 dBFS)的声音所占比例。本研究中使用默认参数计算。The spectrogram is divided into frequency bands (default 10) and the Gini index is used to calculate the proportion of sounds in each band that exceeded the threshold (default -50 dBFS). The default parameters are used for calculation in this study. Villanueva-Rivera et al, 2011
生物声学指数
Bioacoustic index (BIO)
测量指定频率范围内的声音强度, 频谱中超过阈值部分的面积与大多数鸟类的鸣叫频率区间和声强有关, 本研究中使用以下参数计算: min_freq = 2000, max_freq = 11000, fft_w = 512。The sound intensity in the specified frequency range was measured and the area of the above-threshold portion of the spectrum was related to the frequency range and sound intensity of most birds’ calls. The following parameters were used in this study to calculate: min_freq = 2000, max_freq = 11000, fft_w = 512. Boelman et al, 2007
标准化声景差异指数
Normalized difference sound index (NDSI)
通过计算人工声(1-2 kHz)与生物声(2-11 kHz)的比值来评估人类活动对声景观的影响程度。范围从-1到1, 本研究中使用默认参数计算。NDSI was assessed by calculating the ratio of anthrophony (1-2 kHz) to biophony (2-11 kHz). The range is from -1 to 1 and is calculated using the default parameters in this study. Kasten et al, 2012