Southern Ocean GLOBEC Bathymetry: Progress Report - February, 2002

by Tom Bolmer (WHOI), Bob Beardsley (WHOI), Carol Pudsey (BAS), Peter Morris (BAS), Peter Wiebe (WHOI), Eileen Hofmann (ODU), and John Anderson (Rice).

Go to the recent NBP0202 Cruise Track

Go to the original Marguerite Bay Digital bathymetry for the Southern Ocean GLOBEC study area

Introduction

One objective of the SO GLOBEC program is to produce a better knowledge of the sea floor bathymetry in the program study area. Much of Marguerite Bay and the adjacent shelf was poorly charted when the program was started in 2000, so we first created a local area improved version (ETOPO8.2A) of the Sandwell and Smith ETOPO2 2-min digital gridded bathymetry for the SO GLOBEC study area. The first SO GLOBEC mooring cruise on the R/V Lawrence M. Gould in March 2001 quickly showed that the 2-min resolution of ETOPO8.2A does not resolve many of the canyons and abrupt changes in topography which characterize Marguerite Bay and the inner- to-mid shelf region, nor is it particularly accurate in even the more uniform terrain regions.

ETOPO8.2 merged with Digitized Navigation Charts and GTOPO30.

Data Collection 2001:

To begin to improve this situation, high-quality swath bathymetry data were collected during both SO GLOBEC broadscale surveys using the SeaBeam system aboard the R/VIB Nathaniel B. Palmer in 2001. In addition, an extensive multibeam survey was conducted along the deep trough leading into Marguerite Bay aboard the R/VIB James Clark Ross in 2001. While the 2001 Palmer SeaBeam data may need some adjustment for sound speed corrections, an initial attempt has been made to merge these new data sets with all existing center trackline and multibeam data available at the U.S. Geophysical Data Center (NGDC) or provided by other investigators.

All of the Multi-Beam Data Assembled by Hawaii AGU 2002.

All of the NGDC Track lines Assembled by Hawaii AGU 2002 plotted over the Multibeam data.

New Bathymetry:

Figure 4 shows the initial composite bathymetry based on these sources. Additional multibeam data will be collected in 2002 by the Palmer and Ross, and we hope to obtain additional data collected by other investigators. Thus, we hope to provide an improved digital bathymetry by spring 2003. Please contact Tom Bolmer (tbolmer@whoi.edu) or Bob Beardsley (rbeardsley@whoi.edu) with any questions about this collaborative project. We encourage any investigator with high-quality digital bathymetry data collected in the SO GLOBEC study area to contact us and consider adding their data into the combined data base.

All of the Data Assembled by Hawaii AGU 2002 gridded.

All of the Data Assembled by Hawaii AGU 2002 gridded and merged with the ETOPO8.2 and GTOPO30 Data Sets.

PLOT as of April 1, 2002

Below is a plot of the data as of April 4, 2002. This is created from the gridded data to date. The gridded data were turned into an ASCII XYZ file in which the NaNs were removed. The data were then merged with data from the GTOPO30 land data base. These data were then run through the GMT blockmean and surface programs to create a new grid entirely derived from the assembled bathymetry data and the land data. The below plot shows the result of this work. The original grid had 3 cells of padded data and was gridded at 125 meters. If no data occured in a grid cell, a NaN was used in that cell.

All of the Data Assembled by April 3, 2002, were merged with the GTOPO30 dataset and then run through the GMT Surface program. The figure on the left uses the rainbow coloring scheme used in the figures above. The figure on the right uses the Haxby coloring scheme for a more "tradional" coloring of the data. Please note that the data set in the two figures shows the status of the data to date. It is not a final product. The data still need to be checked for bad values and the sound velocity profiles need to adjusted.

PLOT as of May 1, 2002

Below is a reworking of the data using the Antarctic Digital Database CD-ROM published in 1993 by the ICSU Scientific Committee on Antarctic Research (SCAR) and compiled jointly by the British Antarctic Survey, the Scott Polar Research Institute and the World Conservation Monitoring Centre; Cambridge, UK. The data from the Multibeam, NGDC MGD77, and underwiegh JGOFS data that werea gridded using the mbsystem mbgird program were made into ASCII data with no NaN's. This data was then merged with the SCAR coastline data and the GTOPO30 land 30second data. The new data were then surfaced using the GMT surface program. This new plots has the coastline comfoming much better to the assembled data than has prefiously been available.

Recent NBP0202 Cruise track

Status as of May 31, 2002

Below is a plot of the data as of May 31, 2002. This includes the data from NBP0202 and selected sections of NBP0201. This is created from the gridded data assembled to date. The gridded data were turned into an ASCII XYZ file in which the NaNs were removed. The data were then merged with data from the GTOPO30 land data base. These data were then run through the GMT blockmean and surface programs to create a new grid entirely derived from the assembled bathymetry data and the land data. The below plots show the result of this merging. The original grid had 3 cells of padded data and was gridded at 125 meters. This version includes the coastline from the SCAR 1:1 million antarctic coastline. The blue line is the edge of the ice shelf as of the SCAR data base creation.
There are some problems with the coastline data in the upper Right corner of the plot. Also, in the South in the ice shelf off of Alexander Island there are some contouring artifacts that must still be addressed.

Data Assembled by May 31, 2002.
The figure on the left uses the rainbow coloring scheme used in the figures above. The figure on the right uses the Haxby coloring scheme for a more "tradional" coloring of the data. Please note that the data set in the two figures shows the status of the assembled data to date (May 31, 2002). It is NOT a final product. The data still need to be checked for bad values and the sound velocity profiles need to adjusted. On NBP0202 an attempt was made to adjust the velocity of sound in the NBP0103 data set, but I believe work on this leg is still needed. I am still trying to fine tune a coloring scheme to best show the seafloor features.

The figure on the left is from the etopo 8.2 data that was merged with the Didigitized charts as referenced in The Sothern Ocean Bathymetry page. This figure was colored with the sample color file as the plot to the upper right. None of the new cruise data or NGDC data searchs have been added here. This clearly shows the problems with the Gravity derived bathymetry and the improvements we have made in this area in understanding the bathymetry.

Acknowledgements:

We want to thank all those who helped in collecting and processing multibeam data on the Palmer and Ross during this last year. Funding for the post-cruise processing of the Palmer data and merging the Palmer, Ross and other data is covered by NSF Office of Polar Programs grant OPP810092.04.

Return to the TOP

Go to the original Marguerite Bay Digital bathymetry for the Southern Ocean GLOBEC study area

This page was created by Tom Bolmer.
send comments and questions to:

Tom Bolmer.
Last Modified 04/04/02