False-colour spectrograms: Gympie and Woondum National Park

This dataset contains 796 false-colour spectrogram images, one for each of the 398 days from the two audio recordings sites Gympie and Woondum National Park. These spectrograms were produced using the software analysis programs see Towsey, M., Truskinger, A., & Roe, P. (2017) Audio Analysis Software (Version 16.06.3430.0) [Computer software]. Brisbane: QUT Ecoacoustics Research Group.

These 24-hour false colour spectrograms (Towsey, M., Zhang, L., Cottman-Fields, M., Wimmer, J., Zhang, J., & Roe, P. (2014). Visualization of long-duration acoustic recordings of the environment. Procedia Computer Science, 29, 703-712. doi:10.1016/j.procs.2014.05.063) are representations of the the acoustic events occurring across each 24-hour period (midnight to midnight) in the continuous audio recordings from the 22 June 2015 and 23 July 2016 at the two recording sites. Each image is 1440 pixels wide and 668 pixels high. The 1440 horizontal pixels represent the number of minutes in one day. There is one vertical pixel for each frequency bin. Each vertical pixel represents 43.07 Hz, there are 256 frequency bins to cover the frequency range of 0 to 11.025 kHz. Each image has three spectral acoustic indices mapped to the red, green and blue channels of the image.

The three spectral acoustic indices mapped to red, green and blue in the top image is the Acoustic Complexity Index (ACI), Temporal Entropy (ENT) and Events per second (EVN). The three spectral acoustic indices mapped to the red, green and blue in the bottom image is Background Noise (BGN), Power (POW) and Spectral Peak Tracks (SPT).

The time scale at the base of the top and bottom images has ticks marked at every 30 minutes, however these images can also be read to the nearest minute. There are alternating horizontal black and white dots at a spacing of 3 minutes apart which makes the determination of the exact minute possible.

本数据集包含796幅假彩色声谱图，每份对应于两个录音地点——Gympie和Woondum国家公园中的398天。这些声谱图系采用Towsey, M., Truskinger, A., & Roe, P. (2017)所开发的软件分析程序[计算机软件]（版本16.06.3430.0）生成[Computer software]，由布里斯班昆士理工大学生态声学研究中心编制。这些24小时的假彩色声谱图（Towsey, M., Zhang, L., Cottman-Fields, M., Wimmer, J., Zhang, J., & Roe, P. (2014). 环境长期声学记录的可视化. 计算机科学进展，29，703-712. doi:10.1016/j.procs.2014.05.063）是对从2015年6月22日和2016年7月23日连续音频记录中每个24小时（午夜至午夜）发生的声学事件的表征。每张图像宽度为1440像素，高度为668像素。1440个水平像素代表一天中的分钟数。每个频率分箱对应一个垂直像素。每个垂直像素代表43.07 Hz，共有256个频率分箱以覆盖0至11.025 kHz的频率范围。每张图像都有三个声谱声学指数映射到图像的红色、绿色和蓝色通道。顶部图像中映射到红色、绿色和蓝色的三个声谱声学指数分别是声学复杂性指数（ACI）、时间熵（ENT）和每秒事件数（EVN）。底部图像中映射到红色、绿色和蓝色的三个声谱声学指数分别是背景噪声（BGN）、功率（POW）和声谱峰值跟踪（SPT）。顶部和底部图像底部的时间尺度上，每隔30分钟标记一个刻度，但这些图像也可以精确到分钟。每隔3分钟交替出现水平黑点和白点，使得精确确定分钟成为可能。