These data sets consists of approximately flowline-parallel bathymetric profiles in the North and South Atlantic Oceans, along the northern and southern East Pacific Rise, and along the Pacific-Antarctic Ridge. These data are used in a paper submitted to Geophysical Research Letters entitled "A spreading-rate dependence for periodic signals embedded in otherwise aperiodic abyssal hills".

The first zip file, AHspectral_profiles.zip, contains unprocessed profiles, each containing the following columns:  (1) longitude, (2) latitude, (3) depth in m, (4) crustal age in my, and (5) half spreading rate in 1/1000^th of a cm. The second zip file, AHspectral_processed.zip, the profiles have been processed to be a function of crustal age, resampled at constant interval, and detrended with an order 6 polynomial. Header gives number and data points and sampling interval in my.

The profile naming convention consists of the cruise ID of the data collection survey (except for NEPR data, which come from too many bathymetric sources, and so are just identified as “nepr”), and a 3-digit code indicating (a) whether data profile was to the east (e)or the west (w) of the ridge axis, (b) a number indicating which of one or more profiles where utilized for that cruise ID, and then (c) a letter indicating which segment of the profile was used. If the data were sampled from multibeam data, that is indicated by an “mb” suffix; otherwise the data were gathered from mgd77 trackline data files.

All profiles were processed as follows:

1. Profile points were converted from [latitude, longitude, depth] values to [age, depth], utilizing the global crustal age map of Müller et al. (2016), and the “grdtrack” module in Generic Mapping Tools (GMT; https://www.generic-mapping-tools.org/). Paleo-half spreading rates along each profile are also determined using the same procedure.
2. The [age, depth] profiles were resampled at a regular spacing using a linear interpolation to facilitate discrete fast Fourier transformation (FFT). NATL, SATL and PAR profiles were resampled at 0.01 my spacing, and NEPR profiles at 0.005 my.
3. Resample profiles were detrended using a best-fitting, 6^th-order polynomial, employing the “trend1d” module in GMT.