Abstract:

This dataset is used to analyze ambient noise background levels measured in the northern Gulf of Mexico in 2007, 2010, and 2015 near the Deepwater Horizon site. Passive acoustic data were collected using EARS (Environmental Acoustic Recording System) buoys deployed at two sites from 2007-07-07 to 2007-07-16 (south site only, 28°25.23’N, 88°37.07’W; depth 1465 m), 2010-09-11 to 2010-09-23 (south site only, 28°24.61’N, 88°34.26’W; depth 1550 m), and 2015-06-26 to 2015-10-22 (north site only, 28°39.07’N, 88°31.05’W; depth 1570 m). Each file contains one minute’s average power spectral density with 1 Hz resolution between 1-40000 Hz. The data has been high-pass filtered (>200 Hz); the true levels are only for frequencies higher than 200 Hz. The deployment and recovery noise is also present at the beginning and end of the site analysis. This dataset supports the publication: Sidorovskaia, N. A., & Li, K. (2016). Decadal evolution of the northern Gulf of Mexico soundscapes. doi:10.1121/2.0000382.

Suggested Citation:

Sidorovskaia, Natalia. 2020. Ambient noise in the northern Gulf of Mexico determined by passive acoustic monitoring between 2007 and 2015. Distributed by: GRIIDC, Harte Research Institute, Texas A&M University–Corpus Christi. doi:10.7266/n7-jn6j-mg21

Publications:

Sidorovskaia, N. A., & Li, K. (2016). Decadal evolution of the northern Gulf of Mexico soundscapes. doi:10.1121/2.0000382

Purpose:

The dataset was used to study decadal trends of the deepwater northern Gulf of Mexico ambient noise levels measured across broad frequency bands overlapping with soundscapes produced by industrial activities and marine mammals.

Data Parameters and Units:

μPa^2^/Hz. The files are named by mooring site, name, and then sequentially by the minute. For example, S210000.mat contains the first minute’s data at the 2010 south site (S21) and S210001.mat the seconds minute’s data.

Methods:

The data were analyzed in segments of N = 1024000 (5.3 seconds) with a 50% overlap and windowed with a Hamming window before applying the Fast Fourier Transform (FFT) to calculate the power spectral density (PSD); twenty-four PSD outputs were ensemble-averaged to obtain 1-min PSD estimates, and the six spectral bins or frequencies centered around 1-Hz were summed to estimate the 1-Hz band PSDs for each minute of data.

Individual Files: