Ocean Bottom Pressure Records


With a recent grant from the U.K. Natural Environment Research Council, the PSMSL is working to greatly improve its provision of data from in-situ ocean bottom pressure recorders (OBPRs) from all possible sources, a remit given to the PSMSL by IAPSO in 1999. In order to provide a valuable product to the community, we would like to gather views on what would be most useful. To simplify this process as much as possible, we have put together a straw-man initial proposal for consultation, followed by a few questions to elicit proposed improvements. Please take a little time to read through the proposal, and then answer the following questions.

Currently, we supply OBPR data through the following sites: Global Undersea Pressure (GLOUP; an early attempt to do something similar), ACCLAIM bottom pressure measurements from NTSLF, historical OBPR data from BODC, OBPR data from BODC, or search BODC data holdings (data download requires registration) using the "Data Type" refine to restrict the Instrument to "Bottom pressure sensor tide gauge" and the "Originator" refine to restrict the Organisation to "National Oceanography Centre, Liverpool" and "National Oceanography Centre, Southampton".


We will provide 3 levels of data.

  1. Raw pressures (all calibrations applied), as originally sampled, simply given as a list of times (centre of averaging period), durations (length of averaging period), and pressures, in a standard format.
  2. Hourly averaged values.
  3. Daily averaged values.

Apart from (1), for which each deployment will be in a separate file, data from a single “site” will be concatenated into single files, but different deployments at the same “site” will be distinguished by a flag labelling the deployment number. Deployments are considered to be at the same “site” if they are intended to form a single time series; they will generally be within a few km and a few hundred metres of depth of each other. The different positions and depths of each deployment will be given as auxiliary information, wherever known.

To compute hourly values from datasets with disparate sampling, we will first perform a least-squares tidal fit (tides at diurnal and shorter period only), and subtract the tide, before interpolating and averaging the residuals as necessary, then restore the tide reconstructed at the hourly sampling points. Values will be hourly averages, centred on 30 minutes past the hour. A detrending for each deployment will be performed following Watts and Kontoyiannis (exponential plus linear trend, reference below). Each file will contain 1) the hourly data, including tides and drift, 2) the residual after removing tide and drift, 3) the tide that was removed, 4) the drift that was removed, plus a flag to indicate whether the residual pressure for that hour is produced by interpolation (no good data in that hour), and a time stamp.

Daily values will be constructed from the time series after removal of diurnal and shorter period tides, and will include data before and after drift removal. They will be the average of 24 hourly values, from 00:30 to 23:30 on each day. While this choice of a simple arithmetic average of the residuals differs from the normal usage of a “daily” value in tidal studies, where some tidal filter would be applied, we believe that this will make it simpler to compare model results that have not been similarly filtered.


There follow a few specific questions, followed by a more general one to capture any issues not addressed so far.

  1. The tidal analysis does not include fortnightly, monthly, or longer period tides. We prefer to treat gravitational tides at periods longer than 2 months separately, and will look at providing auxiliary information for these. Between the diurnal period and 2 months, tides can be calculated by least-squares fitting, but the value of the fit varies dramatically depending on the length of the record and the amount of other oceanographic variability at these frequencies. Would you like the files provided to also contain a) the fitted tide (regardless of how good the fit is), b) a predicted model tide?
  2. What file format do you prefer a) ascii, b) NetCDF, c) IEEE unformatted binary, d) other (please specify).
  3. Would you like a dedrifted and detided version of the data at the original sampling rate?
  4. What is the shortest record that would be of interest to you? a) 1 day, b) 1 week, c) 1 month, d) 6 months or longer.
  5. What is your main interest? a) tides, b) ocean circulation, c) other (please specify).

And finally, please tell us if there is something missing from our proposal that you would like to see, if there is any technical point we have not mentioned that you would like to flag as important (ideally with guidance as to what we should do about it), or any other general comments that you would like to make. Please send responses to Mark Tamisiea (mtam at noc.ac.uk).

Watts, D. R. & Kontoyiannis, H. Deep-ocean bottom pressure measurement: Drift removal and performance. J. Atm. Oceanic Tech. 7, 296-306 (1990).