mbvelocitytool
Section: MB-System 5.0 (1)
Updated: 10 November 2014
Index
NAME
mbvelocitytool - Interactive water sound velocity profile editor.VERSION
Version 5.0SYNOPSIS
mbvelocitytool [-Byr/mo/da/hr/mn/sc -Eyr/mo/da/hr/mn/sc -Fformat -Ifile -Ssvpfile -Wsvpfile -V -H]DESCRIPTION
MBvelocitytool is an interactive water sound velocity profile (SVP) editor used to examine multiple SVPs, to create new SVPs, and to model the impact of SVP modification on swath bathymetry data. SVPs created using MBvelocitytool can be used by the program mbprocess to recalculate swath bathymetry from raw travel time and angle data.In general, MBvelocitytool is used to examine SVPs obtained from swath data files (see mbsvplist manual page), XBTs, CTDs, or databases, and to construct new profiles consistent with these various sources of information. The SVPs are represented by a set of paired depth and velocity values which are connected by linear interpolation. Users may load a number of SVPs for display. Users may load or create a single editable SVP and then interactively modify this profile.
When users load swath bathymetry data containing raw travel time and angle data (many but not all swath data format include this information), MBvelocitytool recalculates the bathymetry by raytracing through the current SVP model. If the bathymetry in the input swath file has been edited (e.g. with mbedit), then the associated edit save file (*.esf file) will be loaded as well, and the beams flagged as bad will not be used by MBvelocitytool. The bathymetry of each ping is fit with a line, and bathymetry residuals are calculated for each good beam relative to the linear fit. The average of the bathymetry residuals is displayed along with "error bars" indicating the standard deviations of the residuals. Anomalously shallow bathymetry maps into negative residuals and deep bathymetry into positive residuals; the residuals are displayed so that shallow is up and deep is down. If the seafloor is reasonably smooth so that a linear fit is appropriate, then the residuals will accurately reflect any problems with the water velocity profile. If the water velocity profile is correct, then the residual plot will be roughly flat. If the water velocity profile is significantly in error, then the outer beam depths may anomalously shallow (edge curl up) or deep (edge curl down). In practice, the editable velocity profile is altered interactively until a reasonably residual pattern is achieved.
In order to calculate bathymetry values from travel time observations, geometrical raypaths are traced through the SVP for each beam. Because the sound velocity gradients are uniform between the depth-velocity nodes (linear interpolation), the raypaths are be calculated analytically as pieces of circular arcs. This raytracing algorithm is the same used in the program mbprocess.
One important aspect of the raytracing is the handling of the initial takeoff angles associated with each beam or sounding. In general, the raytracing will begin at a point in the sound speed model that has a sound velocity different than the surface sound velocity (SSV) used by the mapping sonar for the original beamforming. The usual approach is to use Snell's law to adjust the starting angle for this change in sound velocity. This amounts to an assumption that the original SSV was correct and that the rays pass through an insignificantly thick layer in which the sound speed equals the SSV before transitioning to the sound speed implied by the SVP. This is the default setting for raytracing in MB-System. Alternatively, one can proceed with raytracing using the original angle but this is rarely useful or correct. Finally, if the SSV used by the sonar is judged to have been incorrect, then the takeoff angle must be corrected for the erroneous beamforming as well as for the difference between the SSV and the initial raytracing sound velocity. This correction must take the sonar geometry into account because the impact of changing the SSV on a beam angle from a flat receive array is very different from a V-shaped or curved array. All three of these angle correction modes are available in mbvelocitytool.
MBvelocitytool can be used in conjunction with mbprocess. If the user uses the Save swath svp file option to save an SVP model developed through the analysis of a particular swath data file, MBvelocitytool also sets the associated mbprocess parameter file so that mbprocess recalculates the bathymetry using the new SVP model. The program mbset may be used to set the SVP file in the parameter file for any swath data file. Users may also save SVP models without setting any mbprocess parameters by using the Save editable profile option.
Sometimes the bathymetry residuals show structure indicative of persistent artifacts in the bathymetry (e.g. certain parts of the swath may be persistently shallower or deeper than the rest of the swath). In this situation, it is possible to export the residuals and to then apply them in mbprocess as static corrections to the bathymetry. This is accomplished by using the Save residuals as offsets option under the File menu.
If a user attempts to read in swath bathymetry that does not contain the travel time and beam angle data required for bathymetry recalculation, MBvelocitytool will estimate the travel times and angles from the bathymetry by assuming a 1500 m/s half-space (and then post a warning dialog). Although the user can proceed to model bathymetry recalculation by modifying the active SVP just as with proper data, the travel times and angles are not in general correct and so the modeling and any results it gives are, well, bogus. A more useful approach is to leave the SVP alone and simply export the residuals to be applied as static corrections in mbprocess. This approach allows users a practical means of correcting older multibeam bathymetry that was originally calculated with an incorrect SVP but which contains no travel time or angle data.
MB-SYSTEM AUTHORSHIP
David W. Caress
Monterey Bay Aquarium Research Institute
Dale N. Chayes
Center for Coastal and Ocean Mapping
University of New Hampshire
Christian do Santos Ferreira
MARUM - Center for Marine Environmental Sciences
University of BremenOPTIONS
Last Updated: 10 November 2014