This is the first time I'm posting a question on here so I hope I'm doing it right. I've tried searching on here for an answer but haven't found anything particularly relevant.
I have some data collected from a line array of 32 hydrophones. I'm trying to estimate the power spectral density of the underwater noise at different tow speeds of the array using Welch's method. I think the noise should be approximately Gaussian, however on examination of the time series from each hydrophone, there are loud "bangs" occurring at random times on random hydrophones. The bangs are about 10 times the amplitude of most of the time-series values lasting for only fractions of a second.
Am I correct in assuming that the "bangs" will distort the estimate of the spectrum? Will the averaging used in Welch's method help to overcome this and if so is there any way to quantify the effects? Or is there any way of removing the "bangs" without distorting the PSD too much?
I don't have much formal background in signal processing so I apologise if the question is basic or lacking in information.