G E O S C I E N C E A U S T R A L I A

Australian Bathymetry
and Topography Grid
June 2009

T.G. Whiteway

Record
2009/21

GeoCat #
67703

A P P LY I N G G E O S C I E N C E TO AU ST R A L I A’ S M O ST I M P O RTA N T C H A L L E N G E S
Australian Bathymetry and
Topography Grid, June 2009

GEOSCIENCE AUSTRALIA
RECORD 2009/21

by

T.G. Whiteway1

1. Petroleum and Marine Division, Geoscience Australia, GPO Box 378, Canberra ACT 2601
Department of Industry, Tourism & Resources
Minister for Resources, Energy & Tourism: The Hon. Martin Ferguson, AM, MP
Secretary: Mr John Pierce

Geoscience Australia
Chief Executive Officer: Dr Neil Williams,PSM

© Commonwealth of Australia, 2009

This work is copyright. Apart from any fair dealings for the purpose of study, research, criticism,
or review, as permitted under the Copyright Act 1968, no part may be reproduced by any process
without written permission. Copyright is the responsibility of the Chief Executive Officer,
Geoscience Australia. Requests and enquiries should be directed to the Chief Executive Officer,
Geoscience Australia, GPO Box 378 Canberra ACT 2601.

Geoscience Australia has tried to make the information in this product as accurate as possible.
However, it does not guarantee that the information is totally accurate or complete. Therefore, you
should not solely rely on this information when making a commercial decision.

ISSN 1448-2177
ISBN 978-1-921498-89-3 web
ISBN 978-1-921498-89-6 DVD
ISBN 978-1-921498-84-8 hardcopy

GeoCat # 67703

Bibliographic reference: Whiteway, T.G., 2009. Australian Bathymetry and Topography Grid, June
2009. Geoscience Australia Record 2009/21. 46pp.
 Australian Bathymetry and Topography Grid – June 2009

Table of Contents
TABLE OF CONTENTS III

1. ABSTRACT 1

IMPORTANT INFORMATION 1

2. INTRODUCTION 2

3. INPUT DATASETS 3

SINGLE BEAM DATA 3

MULTIBEAM (SWATH) BATHYMETRY 4
DIGITAL BATHYMETRY FROM NATMAP 1:250,000 SERIES AND AHS FAIRSHEETS 4
LASER AIRBORNE DEPTH SOUNDER 5
SATELLITE MEASUREMENTS – ETOPOV2G AND ETOPO1 5
AUSTRALIAN TOPOGRAPHY 6
NEW ZEALAND TOPOGRAHY 6
SRTM DEM 6
CHRISMAS, COCOS AND LORD HOWE ISLAND BATHYMETRIC GRIDS 7
GRID FOR THE GULF OF PAPUA AND ADJACENT AREAS 7

4. THE PROCESS 9

STEP 1 – AREA OF INTEREST 12

STEP 2 – DATA PREPARATION 12
STEP 3 – IMPORT DATA INTO INTREPID™ DATABASES BASED ON TILES 13
STEP 4 – CREATE TILE GRIDS, CLIP AND MERGE TILED GRIDS INTO A SINGLE GRID 14
STEP 5 – CLEANING SURVEYS GRID 16
STEP 6 – FILLING ON-SHELF ZONES AND HOLES 16
STEP 7 – MOSAICING ALL GRIDS 17
STEP 8 – CLEANING THE MOSAICED GRID 17
STEP 9 – FINAL ADDITIONS 18
STEP 10 – FINAL CLEANING PROCESS 18
STEP 11 – FINAL PREPARATION OF PRODUCTS 18

5. FINAL PRODUCTS 19

2009 BATHYMETRIC GRID OF AUSTRALIA 19

LINEAGE SHAPEFILE 23

6. 2005 AND 2009 BATHYMETRIC GRIDS – A COMPARISON 24

NEW DATA 24
REVISED ETOPO DATA 28
INTERPOLATION ARTEFACTS 29
METHEDOLOGY ARTEFACTS 32
ERRANEOUS DATA 34

7. CONCLUSION 37

8. ACKNOWLEDGMENTS 38

iii
 Australian Bathymetry and Topography Grid – June 2009

9. REFERENCES 38

10. OTHER DATA SOURCES 39

APPENDIX 1 – MULTIBEAM SURVEYS 40

APPENDIX 2 – SRTM DEM METADATA 46

iv
 Australian Bathymetry and Topography Grid – June 2009

1. Abstract
In 2005 Geoscience Australia and the National Oceans Office undertook a joint project
to produce a consistent, high-quality 9 arc second (0.0025° or ~250m at the equator)
bathymetric grid for Australian waters. In 2009 a number of new datasets were included
in an updated version of the grid.

The 2009 bathymetric grid of Australia has been produced to include recently acquired
datasets, and solutions to issues identified in the previous version. The revised grid has
the same extents as its 2005 counterpart, including the Australian water column
jurisdiction lying between 92˚ E and 172˚ E, and 8˚ S and 60˚ S. The waters adjacent to
the continent of Australia and Tasmania are included, as are areas surrounding Macquarie
Island, and the Australian Territories of Norfolk Island, Christmas Island, and Cocos
(Keeling) Islands. The area selected does not include Australia's marine jurisdiction
offshore from the Territory of Heard and McDonald Islands and the Australian
Antarctic Territory.

This report details the datasets and procedures used to produce the 2009 bathymetric
grid of Australia. As per the 2005 grid, the 0.0025 decimal degree (dd) resolution is only
supported where direct bathymetric observations are sufficiently dense (e.g. where swath
bathymetry data or digitised chart data exist) (Webster and Petkovic, 2005). In areas
where no sounding data are available (in waters off the Australian shelf), the grid is based
on the 2 arc minute ETOPO (Smith and Sandwell, 1997) and 1 arc minute ETOPO
(Amante and Eakins, 2008) satellite derived bathymetry. The topographic data (on shore
data) is based on the revised Australian 0.0025dd topography grid (Geoscience Australia,
2008), the 0.0025dd NZ topography grid (Geographx, 2008) and the 90m SRTM DEM
(Jarvis et al, 2008).

The final dataset has been provided in ESRI grid and ER Mapper (ers) formats. An
associated shapefile has been produced so that the user can identify the input datasets
that were used in the final grid.

IMPORTANT INFORMATION
This grid is not suitable for use as an aid to navigation, or to replace any products
produced by the Australian Hydrographic Service. Geoscience Australia produces the
0.0025dd bathymetric grid of Australia specifically to provide regional and local broad
scale context for scientific and industry projects, and public education.

The 0.0025dd grid size is, in many regions of this grid, far in excess of the optimal grid
size for some of the input data used. On parts of the continental shelf it may be possible
to produce grids at higher resolution, especially where LADS or multibeam surveys exist.
However these surveys typically only cover small areas and hence do not warrant the
production a regional scale grid at less than 0.0025dd.

There are a number of bathymetric datasets that have not been included in this grid for
various reasons. Comments or queries about the data included in the grid (or excluded)
can be directed to: IDEASRequests@ga.gov.au.

1
 Australian Bathymetry and Topography Grid – June 2009

2. Introduction
Geoscience Australia (GA) has been acquiring and collating bathymetric survey, seismic
and sampling data for the Australian jurisdiction since 1963 (Webster et al, 2005). GA is
also the national custodian of bathymetric data for Australia and as such holds data
acquired by a range of other sources including scientific institutions, universities, oil
exploration companies and the Australian Hydrographic Service. The data sources and
quality vary greatly from digitised hydrographic charts (less accurate and often with
digitising errors), to the more accurate multibeam datasets including swath and Laser
Airborne Depth Sounder (LADS).

Buchanan (1998) compiled Geoscience Australia’s first digital bathymetry grid based on
the data sets available at the time. This grid covered 9°S - 45°S, 108°E - 160°E, and was
produced at a 30 arc second (approximately 1km) cell size. In 2000 Petkovic and
Buchanan created a revised version of the 1998 bathymetry grid. In 2005 Webster and
Petkovic released new bathymetric grid, created using a different methodology, with the
addition of new data sets.

While much of the original data and the processes used to create the 2009 grid are similar
to those used in the 2005 grid, the 2009 grid described in this document has been
compiled using a slightly different methodology. It also contains additional multi-beam
data, but still lacks many of the available LADS datasets.

This report documents the procedures used to construct the bathymetric grid to the
extents: 92˚ E – 172˚ E and 8˚ S – 60˚ S (Figure 1). Additionally, in order to provide
users with information on the source data at each grid cell, a lineage shapefile has been
created to provide the source of the data at any given point.

FIGURE 1: Extent of the 2009 Bathymetric grid of Australian (red boundary) and tiles (black lines)

2
 Australian Bathymetry and Topography Grid – June 2009

3. Input Datasets
The following text describes each dataset used in the production of the 2009 grid. The
input datasets used to create the final grid varied greatly in resolution, source and data
type. In this revised version of the Australian bathymetry grid the following datasets were
used:
• Multibeam
• Fairsheets and 1:250,000 Series
• Laser Airborne Depth Sounder (LADS) (Only those LADS datasets that had
previously been converted from LAT were used in this grid)
• Satellite altimetry measurements – ETOPO2v2g
• Australian topography
• New Zealand topography
• SRTM DEM

Single beam datasets were not used in the creation of the 2009 grid (see below). All
datasets once gridded were shifted to ensure cell alignment matched the Australian DEM
prior to mosaicing to ensure the accuracy of the Australian DEM was maintained.
Methodologies used in these processes are outlined in Part 4.

SINGLE BEAM DATA

Measurements of water depth particularly in deep oceans were nearly impossible to
gather prior to the development of echosounder tools. Echosounders are a form of
active sonar that create a sound (or ping) that is bounced off the ocean floor. On the
ping’s return, the transducer (antenna) receives the signal and based on the speed of
return and the known speed of sound through the water column, the water depth is
calculated. Single beam echo sounders collect bathymetry data along a ships path (often
referred to as ship track data).

In previous Australian bathymetry grids single beam has been included particularly for
areas of deep ocean where there are limited other sources of data. In the 2009 grid, the
single beam dataset was considered for inclusion, but was not used due to issues
emerging from its use in previous grids, specifically:
• The inclusion of single beam data in previous grids has resulted in lines/stripes of
high resolution data running through the lower resolution ETOPO data. This
causes artefacts that look like channels and ridges running through the grid
• There are also issues associated with the quality of the raw data. These include:
- levelling and historic navigation issues which means the data is often
inaccurate, and;
- limited time available to clean the data, which means ‘dirty’ areas of data
in the single beam datasets still exist.

Additionally, a majority of the single beam data covers the shelf. In the majority of shelf
areas, there are better quality datasets available including multibeam and LADS that can
be used to produce a more accurate and consistent grid. The compliment of issues
surrounding the use of the single beam data has led to its exclusion from the 2009
version of the grid.

3
 Australian Bathymetry and Topography Grid – June 2009

MULTIBEAM (SWATH) BATHYMETRY

Multibeam echosounders map the seabed using a ‘fan-shaped swath’ of sonar beams.
This allows for large areas of seabed to be mapped in a relatively short time, making
them a more efficient tool than single beam echosounders.

Multibeam or swath datasets are an extremely valuable source of bathymetric data, providing
highly accurate broad coverage for both on and off shelf areas. As the national custodian for
bathymetric data, GA holds data from government organisations, academic and foreign
institution multi-beam survey data. Appendix 1 lists the swath surveys used in this grid
(Figure 5).

All GA swath datasets are initially cleaned in CARIS HIPS/SIPS software and are
exported as xyz files for use in gridding packages. The swath datasets used in this grid
have not been decimated or averaged prior to gridding.

DIGITAL BATHYMETRY FROM NATMAP 1:250,000 SERIES AND AHS FAIRSHEETS

The following section has been extracted from Webster and Petkovic (2005).

“In 1994, GA received a set of charts from the Australian Hydrographic Service, Royal
Australian Navy (AHS), sourced from the NATMAP 1:250,000 bathymetric series. These
charts included contours on the face of the map and soundings as spot depths on the
back. The spot depths for a subset of the charts, mainly for the NW Shelf and Gulf of
Carpentaria, were captured digitally by the AHS, and these data were provided to GA.
The digitisation of the remainder of the charts was continued by GA and completed in
2001.

At the request of GA, the AHS provided an additional set of charts as scanned images
for the shelf in selected regions. These 'fairsheets' displayed hand-written spot depths for
surveys completed over various periods of time. To save costs the fairsheets were
digitised selectively, leaving out near coastal data points, and, as a general rule, digitising
lines to a minimum spacing of 2 nautical miles (~3.7 km). Some charts of high interest to
clients were digitised in entirety. There are 1049 charts in the collection.”

These raw datasets have been collected in a range of different Datums, and due to time
constraints many of these have not been converted back to a single datum (MSL). The
raw data points are stored in GA’s Oracle database and can be extracted by location as
required.

4
 Australian Bathymetry and Topography Grid – June 2009

LASER AIRBORNE DEPTH SOUNDER

The following section has been extracted from Webster and Petkovic (2005).

“The AHS has been operating its Laser Airborne Depth Sounder (LADS) systems since
1993. The system uses red and green laser light emitted from a stabilised platform inside
a fixed wing aircraft. The light pulses reflected from the sea surface (red) and sea floor
(green) are separately detected, and the time difference between the sea surface and sea
floor returns allows calculation of the water depth.

Data are typically acquired at a rate of 900 soundings in a 240 m wide x 5 m long swath
per second. The system works best in clean water of the pristine coral reefs of the Coral
Sea down to 70m depth. Shallower maximum depths are achieved in other areas, such as
Great Barrier Reef, Bass Strait, Timor Sea (40-50m), South Australia (30-35m), Torres
Strait (15-25m), and anywhere waters are not clear.”

The LADS data is of particular importance to the large scale 0.0025° bathymetry grid
because;
• it fills the area between the coastline (Australian DEM) and the offshore datasets
such as swath, and
• it provides excellent edge boundaries for reefs, coastlines and islands.

“AHS provided GA with LADS data around Australia, including 1773 LADS datasets. In
order to be included in the gridding process, each dataset needed to be converted from
HTF format into XYZ, be datum converted from LAT to MSL and, each dataset needed
to be imported into Intrepid™ as an Intrepid™ database. Due to time constraints only
the LADS data used previously in the 2005 Australian bathymetry grid was used for this
grid. The LADS datasets used included data from Queensland waters particularly around
the Great Barrier Reef.”

Only limited LADS datasets were used in the production of the 2009 grid due to the
limited coverage they provide, and lack of conversion time and storage available to
convert these datasets from the LAT to MSL Datum.

SATELLITE MEASUREMENTS – ETOPOV2G AND ETOPO1

The 2 arc minute ETOPO2v2g (U.S. Department of Commerce, 2006) data based on
predicted bathymetry from satellite altimetry (Smith and Sandwell, 1997) was used as
base data for the grid where more accurate data sources were not available (i.e. for much
of the deep ocean).

In 2008 during development of the revised Australian bathymetric grid, a 1 arc minute
version of ETOPO, ETOPO1 (Amante and Eakins, 2008) was released. In the
Australian region, this dataset is based primarily on GEBCO estimated Seafloor
bathymetry. It was used as a late inclusion to replace data in the top left of the Australian
bathymetry grid where both he Australian, and ETOPO2v2g data were inaccurate.

The ETOPO grids were resampled to 0.0025dd using a bilinear conversion as the nearest
neighbour method created grid artefacts including striations. In the 2009 version of the
grid the ETOPO data appears to be less smooth and less detailed than the ETOPO data
in the 2005 bathymetry grid. In the 2005 bathymetry grid the ETOPO data was re-

5
 Australian Bathymetry and Topography Grid – June 2009

interpolated in Intrepid, in the 2009 version there was no interpolation in order to keep
the values consistent with those produced in the original ETOPO data.

This predicted bathymetry (Figure 2) was included in areas where the high-resolution
data (swath and digitised charts) were not available, and was not used on the Australian
shelf. All other datasets were mosaiced on top of the ETOPO data. The predicted
bathymetry contributes the characteristic artefact of rippled noise which has a wavelength
of about 25km and amplitude of about 200m in the deep water.

AUSTRALIAN TOPOGRAPHY
The Australian 9" DEM (Geoscience Australia, 2008) (Figure 3) was used as the
topographic dataset for the Australian Continent and surrounding islands. This is GA’s
4th edition DEM and has been calculated using ANUDEM v 5.2.2.

FIGURE 2: ETOPO2v2g (U.S. Department of Commerce, 2006)

NEW ZEALAND TOPOGRAHY

The 250m New Zealand DEM was sourced from Geographx (NZ) Ltd (2008) in Ascii
format, non-projected, NZ Mercator datum (Figure 3). The projection of this dataset was
modified to WGS84 and, as with the Australian DEM, was then directly mosaiced over
the other datasets without feathering or smoothing.

SRTM DEM
The 90m Shuttle Radar Topography Mission (SRTM) DEM (Figure 3) was sourced for
the southern areas of Papua New Guinea and Indonesia from the CGIAR Consortium

6
 Australian Bathymetry and Topography Grid – June 2009

for Spatial Information (CIGAR-CSI) in the WGS84 datum in the ASCII format. The
individual datasets were mosaiced and then resampled to 0.0025dd using ArcGIS nearest
neighbour method, and values below 0m were removed so that only land based values
were used. The SRTM data was then directly mosaiced over the other datasets without
feathering or smoothing.

FIGURE 3: Australian (GA, 2008), NZ (Geographx, 2008) and SRTM (Jervis et al., 2008) Digital Elevation
Models (DEM’s) used in the production of the 2009 bathymetric grid of Australia

CHRISMAS, COCOS AND LORD HOWE ISLAND BATHYMETRIC GRIDS

For the Christmas, Cocos and Lord Howe Islands, bathymetric grids had been produced
with hand edited data. Fledermaus with the 3D editor were used to compare input data
points, and points not consistent with the higher accuracy datasets (swath and LADS)
were deleted prior to gridding. The island datasets have been gridded in Intrepid and
have been incorporated into the revised bathymetry grid to improve accuracy in these
regions.

GRID FOR THE GULF OF PAPUA AND ADJACENT AREAS

The grid in the region of the Gulf of Papua (Figure 4) was created using a combination
of processes from GMT (Generic Mapping Tools) to average the input data points and
ANUDEM to grid the final product (Daniell, 2008). The Gulf of Papua grid comprises
the following input datasets:
• Multibeam sonar
• Australian Hydrographic Service (AHS) Fair sheet and survey data
• Ship Track and Miscellaneous data (including external surveys)
• SRTM
• Landsat derived Bathymetry

7
 Australian Bathymetry and Topography Grid – June 2009

FIGURE 4: Gulf of Papua and adjacent areas grid (Daniell, 2008) used in the production of the 2009
bathymetric grid of Australia

8
 Australian Bathymetry and Topography Grid – June 2009

4. The Process
A number of programs were used to create the 2009 revised Australian bathymetry grid,
in particular:
• ArcGIS for review of gridded data, mosaicing without feathering and the
creation of the lineage shapefile.
• ER Mapper for gridding replacement data in problem areas, and for data review.
• Intrepid™ for initial raw data editing for Fairsheets, gridding and mosaicing with
feathering.

The process used to develop the grid is summarised in Table 1, which defines an eleven
step sequence.

Table 1 – Summary of steps and processes used to create the 2009 bathymetric grid

STEP / INTERIM GRID ANALYSIS

INPUT PARAMETERS
PROCESSES TYPE

STEP 1 – AREA OF INTEREST

STEP 2 – DATA PREPARATION

STEP 3 – IMPORT DATA INTO INTREPID™ DATABASES

STEP 4 – CREATE TILE GRIDS, CLIP AND MERGE TILED GRIDS INTO A SINGLE
GRID

 Cell Size = 0.0025

Surveys grid  Minimum Curvature tension =
 LADS Gridded 0.5
 Swath together  Maximum Iterations = 100
 Fairsheets with AHD Datum (Intrepid™)  Maximum Residual = 1.0 m
 Fairsheets with Non-AHD Tiles  Extrapolation = 10 cells
datums clipped and  Additional Smoothing = No
 Fairsheets with Unknown mosaiced  Initial Cell Assignment =
datums in ArcGIS Nearest Neighbour
 Reef s lines  Output Precision =
IEEE4ByteReal

STEP 5 – CLEANING SURVEY GRIDS

Cleaned in  Erroneous data set to null

Surveys grid cleaned
ER Mapper using region editing

STEP 6 – FILLING ON-SHELF ZONES AND HOLES

Sourced
Australian DEM
Internally

9
 Australian Bathymetry and Topography Grid – June 2009

 Resampled to 0.0025dd cell

Sourced
size
NZ DEM from
 Shifted to match the
Geographix
Australian DEM
 Resampled to 0.0025dd cell
Sourced
size
SRTM DEM from
 Shifted to match the
CGIAR
Australian DEM

Surveys with Land

 Fairs heets/swath/LADS
Mosaic
mosaiced dataset
with no
 Australian DEM (values set to N/A
feathering
0m)
(ArcGIS)
 NZ DEM (values set to 0m)
 SRTM DEM (values set to 0m)

 Holes in on-shelf zones filled

using ER Mapper’s gridding
tool parameters:
 Boundary Tension = 0%
On-shelf filled grid Mosaic  Interior Tension = 0%
 Fairs heets/swath/LADS with no  Convergence speed = 40%
mosaiced dataset feathering  Convergence Limit = 0.1
 Filled on-shelf areas (ArcGIS)  Minimum No Iterations = 255
 Aspect Ratio = 1
 Filled areas mosaiced under
Fairsheets/swath/LADS
dataset

STEP 7 – MOSAICING ALL GRIDS

Sourced
Internally  Shifted to match the
Gulf of Papua grid
(Daniell, Australian DEM
2008)
 Shifted to match the
Sourced Australian DEM
Internally  Converted to point
Christmas Island Grid
and data  Re-gridded in Topogrid
added (ArcMap) with additional data
(GA2476)

Sourced  Shifted to match the

Cocos Island Grid
Internally Australian DEM

Sourced  Shifted to match the

Lord Howe Island Grid
Internally Australian DEM

Sourced
 Resampled to 0.0025dd cell
from US
size
ETOPO2v2g Grid Departmen
 Shifted to match the
t of
Australian DEM
Commerce

10
 Australian Bathymetry and Topography Grid – June 2009

Higher accuracy grid

Mosaic
 Gulf of Papua grid
with no
 Christmas, Cocos and Lord N/A
feathering
Howe Island grids
(ArcGIS)
 On-shelf filled

Mosaic
Higher accuracy and ETOPO2v2g
with no
 Higher accuracy grid N/A
feathering
 ETOPO2v2g
(ArcGIS)

High resolution/ETOPO2v2g and

Land Mosaic
with no
 Australi an DEM N/A
feathering
 NZ DEM
(ArcGIS)
 SRTM

STEP 8 – CLEANING MOSAICED GRID

 Erroneous data set to null

using region editing
 Holes filled using ER
Mapper’s gridding tool
Cleaned grid parameters:
Cleaned in
 High resolution/ETOPO2v2g and  Boundary Tension = 0%
ER Mapper
Land  Interior Tension = 0%
 Convergence speed = 40%
 Convergence Limit = 0.1
 Minimum No Iterations = 255
 Aspect Ratio = 1

Cleaned mosaiced grid Mosaic

 High resolution/ETOPO2v2g and with no
N/A
Land cleaned feathering
 Fill areas (ArcGIS)

STEP 9 – FINAL ADDITIONS

Final Additions
Mosaic
 Cleaned mosaiced grid
with no
 GA2476 leg 3 N/A
feathering
 GA2447
(ArcGIS)
 GA2413

STEP 10 – CLEANING FINAL ADDIITIONS GRID

Sourced
 Resampled to 0.0025dd cell
from US
size
ETOPO1 Grid Departmen
 Shifted to match the
t of
Australian DEM
Commerce

11
 Australian Bathymetry and Topography Grid – June 2009

Mosaic
Addition of ETOPO1 data with no
N/A
 Final Additions grid feathering
(ArcGIS)

 Erroneous data set to null

using region editing
 Holes filled using ER
Mapper’s gridding tool
parameters:
Cleaned grid Cleaned in
 Boundary Tension = 0%
 Final Additions grid ER Mapper
 Interior Tension = 0%
 Convergence speed = 40%
 Convergence Limit = 0.1
 Minimum No Iterations = 255
 Aspect Ratio = 1

Mosaic
Cleaned mosaiced final grid
with no
 Final Additions Grid N/A
feathering
 Fill areas
(ArcGIS)

STEP 11 – FINAL PREPARATION OF PRODUCTS

 Grid values set to integer, as

Integer the cell size and data quality
Integer grid
(ArcGIS) does not support sub-metre
accuracy

Clip  Grid clipped to final extents

Clipped grid
(ArcGIS) (9°S - 45°S, 108°E - 160°E)

STEP 1 – AREA OF INTEREST

The Area of Interest was defined to match the 2005 Bathymetry grid. Two degrees were
added to each edge of the grid and removed at the end of processing to reduce edge
effects. In figure 1 the outer red boundary represents the final outer boundary of the
grid. The initial grids were interpolated to the outer black boundary prior to clipping.

AOI extents: 92, 172, -8, -60

Extended AOI extents: 90, 174, -6, -62

STEP 2 – DATA PREPARATION

Raw Data
Raw data points used in this grid included swath, fairsheets and LADS datasets. Due to
the large extent of raw data and limited computing capacity, the area of interest was
divided into 32 tiles, 14 degrees wide by 11 degrees long, with 4 degrees of overlap
(trimmed before merging). All raw datasets were converted to XYZ files from the

12
 Australian Bathymetry and Topography Grid – June 2009

database (in tiles as represented in figure 1), and each input dataset was exported to XYZ
format on a tile by tile basis.

ETOPO2v2g
ETOPO2v2g has a cell size of 0.033dd. In order to incorporate the data into the
bathymetry grid the data was resampled to a 0.0025dd cell size using bilinear
interpolation in ArcGIS. The bilinear interpolation process interpolates the new values of
each smaller cell using a bilinear function. Original values of the ETOPO2v2g data have
not been preserved, as a result.

ETOPO1
ETOPO1 has a cell size of 0.0167dd. In order to incorporate the data into the
bathymetry grid the data was resampled to a 0.0025dd cell size using bilinear
interpolation in ArcGIS. The bilinear interpolation process interpolates the new values of
each smaller cell using a bilinear function. Original values of the ETOPO2v2g data have
not been preserved, as a result.

Australian Topography
The Australian topography grid required no conversions and was used as the base grid to
which all other grid were shifted to maintain a single cell configuration throughout the
processing.

New Zealand Topography

The projection of this dataset was modified from NZ Mercator datum to WGS84. Values
below 0m were removed so that only land based values were used.

SRTM DEM
The SRTM data was downloaded as individual blocks of data. These datasets were
mosaiced and then resampled from a cell size of 0.00083dd to 0.0025dd using ArcGIS
nearest neighbour method. Values below 0m were removed so that only land based
values were used.

Christmas, Cocos and Lord Howe Island Bathymetric grids

The Christmas, Cocos and Lord Howe Island Bathymetric grids required no conversions.

Gulf of Papua and Adjacent Areas Bathymetric grid

The Gulf of Papua and Adjacent Areas Bathymetric grid required no conversions.

STEP 3 – IMPORT DATA INTO INTREPID™ DATABASES BASED ON TILES

Each raw XYZ file was imported into an Intrepid™ database using WGS84 datum with
geodetic projection. Intrepid was chosen because it is widely used within GA as a
geophysical processing tool, and the gridding tool allows considerable flexibility of
parameter control, grid masking, and auxiliary operations such as stitching, merging and
editing (Billings and Fitzgerald, 1998).

13
 Australian Bathymetry and Topography Grid – June 2009

STEP 4 – CREATE TILE GRIDS, CLIP AND MERGE TILED GRIDS INTO A SINGLE GRID

Gridding
The datasets gridded in this process included:
• Multibeam – swath
• Fairsheets with AHD Datum
• Fairsheets with Non-AHD datums
• Fairsheets with Unknown datums
• LADS

The multibeam, fairsheet and LADS datasets were gridded together to minimise the
propagation of edge effects where the different datasets meet. The grids were computed
using Intrepid software, which requires data to be present in an Intrepid database or set
of databases.

“Intrepid was chosen because of its ease of use with large datasets, its combination of
GUI and batch operation, fine control over parameter setting, large user base within the
organisation, associated grid manipulation utilities, and mature version history. To create
the present grid Intrepid was used in batch mode accessing several databases sequentially.
These same databases can be updated as the need requires, and grids of any specification
can be created from them.

Intrepid was originally designed to deal with potential field data, and uses the minimum
curvature algorithm which is not well suited to topography. Other software such as that
of Hutchinson (1989) deals effectively with stream-lines and roughness, for example, but
doesn’t allow masking, which is essential to the present task.

In order to simulate the roughness of topographic character in Intrepid, the minimum

curvature tension parameter was set to mid-way between pure minimum curvature and
pure minimum potential. This avoids some of the over-shoot that results from minimum
curvature surface fitting in areas of poor control, while at the same time doesn’t produce
severe cusps at data points. Inspection of test grids indicates that this compromise setting
gives sufficiently smooth results yet retains some of the characteristic roughness of
topographic data.

Billings and Fitzgerald (1998) discuss some of Intrepid’s gridding options and its
limitations. The main gridding parameters are given in the table below.” (Webster and
Petkovic, 2005)

14
 Australian Bathymetry and Topography Grid – June 2009

Table 2 - Gridding parameter settings for Intrepid

Parameter Value
Latitude range 8° S – 60° S
Longitude range 92° E – 172° E
Cell size 0.0025° (9”)
Minimum curvature tension 0.5
Maximum iterations 100
Maximum residual 1.0 m
Extrapolation 10 cells (~1.5 km)
Setting no-data areas to null Yes
Additional Smoothing No
Initial cell assignment nearest neighbour
Output precision IEEE4ByteReal

Clip and Mosaic

Each grid was clipped in ArcGIS to remove the 2 degrees of excess (or 4 degrees of
overlap) per tile. The final clipped grids were then mosaiced together using the ‘mosaic to
new raster’ function in ArcGIS (Table 1). This is a direct overlay process with no
feathering or smoothing used to produce a grid of the entire area (figure 5). The final grid
mosaiced in this process is from here on called the ‘surveys grid’.

FIGURE 5: Surveys grid

15
 Australian Bathymetry and Topography Grid – June 2009

STEP 5 – CLEANING SURVEYS GRID

All grids created from raw data had data errors that required editing or removal. In the
case of Fairsheet data there were a number of digitising errors (in the raw data) that were
identified after the first gridding process. These included extreme low and high points in
areas of gentle slope. The erroneous points identified were checked against the original
Fairsheets (which have previously been scanned and rectified). Where a digitising error
was confirmed, the point was corrected (through editing in Intrepid™) prior to gridding,
and the Australian Hydrographic Office was sent the x, y and z coordinates of any
digitising errors for review.

In all other datasets the raw data was not modified due to the dataset size. The size of the
datasets meant that editing program (Intrepid™) could not handle the datasets without
regular crashing. Therefore, inaccurate data areas identified in any dataset (including
some unresolved areas in the Fairsheets) were removed from the grided datasets using a
clip process.

The removal of inaccurate data was undertaken in two steps:

1. For each grid the areas where data values were less than -9999m depth or greater
than 1000m were set to null.
2. For each dataset a vector file (ER Mapper) of data errors to be removed was
created, and used to clip out errors using the region set to null.

STEP 6 – FILLING ON-SHELF ZONES AND HOLES

In the on-shelf zone of Australia, and in some off-shelf areas, the data was dense enough
that there were only small holes with no data coverage. In initial versions of the
bathymetric grid, the ETOPO data was mosaiced behind the greater accuracy datasets to
fill these gaps. However, in many cases the ETOPO data was vastly different to the
higher quality data and caused spurious holes or ridges. Consequently in this version of
the grid, areas where on-shelf gaps occurred were filled with data interpolated in ER
Mapper™.

In previous versions there have also been issues caused by filling holes through
interpolation that are connected to land areas. When ER Mapper interpolates between
high land values and low ocean values the interpolation can create areas where ocean
values are higher than the land.

In order to minimise these issues in this version of the grid, all values of the land grids
(including Australia, SRTM and New Zealand) were set to 0m. These grids were then
mosaiced (with no feathering) into the swath/LADS/fairsheets combination to set a
value of 0m on the coast. An interpolation process was then run to fill the gaps between
the sea values and land values.

The use of this process creates a secondary artefact in that there will always be a smooth
interpolation from ocean values to the land (i.e.) deep water drop offs may be removed.
However, this process does remove the occurrence of land values (i.e. > 0m) in the
ocean area. The collection of data along the coast (both land and ocean) is vitally
important to reduce errors in this part of the grid for future products.

16
 Australian Bathymetry and Topography Grid – June 2009

A number of interpolation methods and input parameters were trialled including tin to
raster and topo to raster in ArcGIS, Intrepid’s nearest neighbour interpolation and ER
Mapper’s tin and minimum curvature interpolation techniques. The process selected was
minimum curvature interpolation in ER Mapper. This process was able to grid large
datasets effectively and created fewer artefacts than many other interpolation options.

A number of parameters for the minimum curvature technique were also trialled. The
final parameters chosen minimised the tension requirements while ensuring that the
tension was great enough to prevent the data from fluctuating wildly, as per the ER
Mapper default values.
• Boundary Tension = 0%
• Interior Tension = 0%
• Convergence speed = 40%
• Convergence Limit = 0.1
• Minimum No Iterations = 255
• Aspect Ration = 1

The Minimum Curvature interpolation process re-interpolates all input data.

Consequently the data from the interpolated grids was used to fill the holes only.

STEP 7 – MOSAICING ALL GRIDS

Grids were mosaiced with the highest accuracy grids on top and the lowest accuracy on
the bottom. All grids were shifted prior to mosaicing to match the Australian DEM to
ensure that both major products released from Geoscience Australia have common
topography. Note that as a result the NZ DEM and SRTM DEM datasets will have been
shifted slightly and will not line up perfectly with their original location.

In this grid the Gulf of Papua, Lord Howe, Christmas Island, ETOPO2v2g and Cocos
Island grids were mosaiced with the surveys grid with a feathering process to reduce edge
effects. The feathering process smooths the edges so that one grid mosaics with another
grid without the large steps seen when no feathering is applied. The feathering process
has been minimised so that only cells adjacent to the joins are feathered. The parameters
used were:
• Overlap limit = 20m
• Filter length = 2500m
• Smoothing iterations = 100
• Smoothing residual = 1

All other grids (New Zealand Digital Elevation Models, the SRTM DEM) were mosaiced
without feathering. These grids were directly overlaid with no smoothing to retain cliff
lines and their original data values. If smoothing were undertaken on the DEM, cliff
edges would be smoothed as part of the process.

STEP 8 – CLEANING THE MOSAICED GRID

Following the completion of a mosaiced version of the bathymetry grid, a number of
edge effects and flaws in theof data required editing. These areas of data were located
and set to no-data using region identification in ER Mapper. The holes created were then

17
 Australian Bathymetry and Topography Grid – June 2009

interpolated using the ER Mapper Gridding tool (with parameters as set in Step 7) and
these fill areas were mosaiced with no feathering.

STEP 9 – FINAL ADDITIONS

On completion of the grid three additional datasets were identified that needed to be
added. These datasets (GA2413, GA2477 and GA2476 – Leg3) were mosaiced into the
final dataset without feathering in ArcGIS.

STEP 10 – FINAL CLEANING PROCESS

Following addition of new datasets a final version of cleaning was undertaken to ensure
that there were no remaining edge effects or errors. The cleaning process was undertaken
as per those outlined in Step 7.

Additionally in this step the top left section of the grid was removed and replaced with
ETOPO1. ETOPO1 is a newly released version of ETOPO data that has higher
accuracy, including data in regions plagued by poor data in ETOPOv2g. This data was
used to fill areas where both swath and ETOPO2v2g datasets had inaccuracies.

STEP 11 – FINAL PREPARATION OF PRODUCTS

The final version of the bathymetry grid was converted to an integer grid as the 0.0025dd
resolution and source accuracy of the data does not support higher accuracy. This
version was then clipped to the required extent and projected to WGS84. The final 9”
grid was then re-produced in ER Mapper, ESRI grid and ASCII XYZ formats.

18
 Australian Bathymetry and Topography Grid – June 2009

5. Final Products
2009 BATHYMETRIC GRID OF AUSTRALIA
The 2009 bathymetric grid of Australia has been produced for modelling, visualisation
and map presentation (Figure 6 through 11). Each dataset was gridded at 0.0025dd
without smoothing, and overlayed with the best accuracy grids (swath) on top, and the
least accurate (ETOPO) grid at the bottom (without feathering).The terrain models for
Australia, New Zealand and Indonesia / Papua New Guinea have been directly overlaid
with no smoothing to retain the raw data for each grid.

FIGURE 6: 2009 bathymetric grid of Australia

19
 Australian Bathymetry and Topography Grid – June 2009

FIGURE 7: 2009 bathymetric grid of Australia - Queensland

FIGURE 8: 2009 bathymetric grid of Australia – north-west Australia

20
 Australian Bathymetry and Topography Grid – June 2009

FIGURE 9: 3D view of the 2009 bathymetric grid of Australia – Western Australia

21
 Australian Bathymetry and Topography Grid – June 2009

FIGURE 10: 3D view of the 2009 bathymetric grid of Australia – Eastern Australia

FIGURE 11: 3D view of the 2009 bathymetric grid of Australia – Western Australia

22
 Australian Bathymetry and Topography Grid – June 2009

LINEAGE SHAPEFILE
An associated dataset, the lineage shapefile, was created to show the source input
datasets for each part of the final grid (Figure 12).

The shapefile was created by converting each grid to an integer grid, and mosaicing these
in the same order as the grids were mosaiced to produce the 2009 bathymetry grid. The
mosaiced grid was then exported to shapefile. Note that there has been a smoothing
process applied to the edges of the swath, fairsheets, LADS and around the island grids
to allow for streamlined gradient between each data type. As such the areas where edges
of each different dataset meet will not exactly replicate the original data sources in these
locations.

FIGURE 12: Lineage shapefile for the 2009 bathymetric grid of Australia

23
 Australian Bathymetry and Topography Grid – June 2009

6. 2005 and 2009 Bathymetric Grids – A

Comparison
In creating a revised bathymetric grid, a number of requirements were met. The
following outlines the differences between the 2005 and 2009 bathymetric grids of
Australia, particularly focussing on the required outcomes for the 2009 grid.

NEW DATA
The 2009 grid incorporates a number of new datasets where available. In particular a
number of new Swath datasets were included as illustrated in figures 13 and 14. However
no new LADS datasets were included in this grid due to time constraints.
a.

b.

FIGURE 13: Addition of Swath data off the East Australian shelf in the 2009 grid (b) (2005 grid (a))

24
 Australian Bathymetry and Topography Grid – June 2009

a.

b.

FIGURE 14: Addition of Swath data from Survey GA 2476 in the 2009 grid (b) (2005 grid (a))

25
 Australian Bathymetry and Topography Grid – June 2009

A range of improved DEMS were also used in the creation of this grid. The latest DEMs
including the new 2008 Australian DEM, the SRTM DEM (figure 15b) and the 250m
New Zealand DEM were resampled to match the cell size of the Australian DEM, and
were mosaiced directly over the final dataset to give continuity and improved land data
accuracy.
a.

b.

FIGURE 15: Inclusion of high quality DEMs increased the overall quality as seen with the inclusion of
the SRTM DEM in the Indonesian area in the 2009 grid (b) (2005 grid (a))

26
 Australian Bathymetry and Topography Grid – June 2009

Additionally, a number of higher accuracy bathymetric datasets were also incorporated.

Grids produced in separate processes for the Lord Howe Island (figure 16), Gulf of
Papua and Christmas Island were a vast improvement on previous grids of these areas.
The new datasets were directly overlaid to provide higher quality data for these areas.
a.

b.

FIGURE 16: Addition of a higher accuracy grids in the 2009 grid (b), including the Lord Howe Island
grid in the image above, increased the overall quality (2005 grid (a))

27
 Australian Bathymetry and Topography Grid – June 2009

REVISED ETOPO DATA

ETOPO2v2g and ETOPO1 datasets were used as background datasets in the 2009
version of the bathymetric grid of Australia. In the 2005 grid the ETOPO data was
resampled to a 0.0025dd cell size using an interpolation process that smoothed the data
and added apparent detail (figure 17a). In the 2009 grid the ETOPO data was resampled
to 0.0025dd using a bilinear process to minimise interpolation. This means that the
ETOPO data in the 2009 grid is coarser and seems to have less detail. However it is
more true to its original source (figure 17).
a.

b.

FIGURE 17: The ETOPO data in the 2005 (a) grid is much smoother and appears to show more detail
than the 2009 (b) grid as it has been interpolated using a different method.

28
 Australian Bathymetry and Topography Grid – June 2009

INTERPOLATION ARTEFACTS
In the 2005 grid a number of coastline interpolation artefacts appeared particularly along
the Queensland coastline (figures 18 and 19). In the 2009 grid, coastal bathymetry data
was interpolated to meet a DEM with a flat height of 0m. This forced the interpolation
to maintain an even rise, minimising interpolation artefacts around the coastline.
a.

b.

FIGURE 18: Removal of interpolation artefacts in the 2009 grid (b) (2005 grid (a))

29
 Australian Bathymetry and Topography Grid – June 2009

a.

b.

FIGURE 19: Removal of interpolation artefacts in the 2009 grid (b) (2005 grid (a))

30
 Australian Bathymetry and Topography Grid – June 2009

There were also areas in the 2005 grid where interpolation processes had been used on
the Australian DEM, causing smoothing of the final terrain data (figure 20). These
interpolation artefacts were removed by directly overlaying the DEM data for each
terrain over the bathymetry data.
a.

b.

FIGURE 20: Removal of interpolation errors over the DEM from the 2005 grid (a) (2009 grid (b))

31
 Australian Bathymetry and Topography Grid – June 2009

METHEDOLOGY ARTEFACTS
When two different quality datasets are gridded together and one has higher values than
the other (effects of positional accuracy, depth accuracy, datum issues etc) the
interpolation package will interpolate to make the edges of the grids meet. Therefore
highs will drop off quickly to meet low values and vice versa (figure 21a).
a.

b.

FIGURE 21: The interpolation process for the 2005 grid (a) in many cases produced a longed (less
sharp) boundary between the different data types compared to mosaicing and feathering method used
in the 2009 grid (b).

In the 2009 grid a different process was applied to allow the production of a shapefile
showing the general input dataset for each of the grid cells. The process involved

32
 Australian Bathymetry and Topography Grid – June 2009

gridding like accuracy datasets in the same process, and feather mosaicing them with
other datasets that had been gridded together. This differed from previous grids where all
datasets were gridded together. For this reason, there are smoothing artefacts around
sections of data that were mosaiced, similar to those seen when all datasets are
interpolated together (figure 21b). However, in the 2009 grid, in some areas they will
appear sharper, as the boundary change distance has been minimised to avoid smoothing
through other higher accuracy datasets.

33
 Australian Bathymetry and Topography Grid – June 2009

ERRANEOUS DATA
A range of errors including irregular ship track lines, Swath spikes, ETOPO gravity
anomaly highs and Fairsheets digitising errors were identified in the 2005 grid. Generally
these errors are products of the raw data, causing high and low anomalies (figure 22),
lines of erroneous data (figure 23) and large scale interpolation artefacts (figure 24).
Where identified these errors were removed through editing the output grid. However,
the en mass errors seen in the single beam data meant that this entire dataset was
removed prior to interpolation.
a.

b.

FIGURE 22: Removal of spurious highs and lows seen in the 2005 grid (a) (2009 grid (b))

34
 Australian Bathymetry and Topography Grid – June 2009

a.

b.

FIGURE 23: Removal of ship track errors seen in the 2005 grid (a) (2009 grid (b))

35
 Australian Bathymetry and Topography Grid – June 2009

a.

b.

FIGURE 24: Removal of data processing errors seen in the 2005 grid (a) (2009 grid (b))

36
 Australian Bathymetry and Topography Grid – June 2009

7. Conclusion
The 2009 iteration of the bathymetric grid has addressed a range of issues seen in the
2005 version, and has included a number of new datasets. The production of an
associated lineage file provides users with an indication of the source of the raw data.

The 2009 bathymetry grid provides excellent regional context, and the greater volume of
higher accuracy data on the shelf provides an improving local context. However, the
production of an Australian region bathymetric grid should be seen as an iterative
process of review and re-compilation to fully exploit the latest methods for managing
and gridding new data.

37
 Australian Bathymetry and Topography Grid – June 2009

8. Acknowledgments
The following people provided advice, data and information during the process:

Mike Sexton, Cameron Mitchell, Cameron Buchanan, Peter Milligan, Murray Woods,
Mark Webster, Andrea Cortese, James Daniell, Michael Morse, Peter Petkovic, George
Bernadel and Richard Mileczko.

9. References
Amante, C. and B. W. Eakins, ETOPO1 1 Arc-Minute Global Relief Model:
Procedures, Data Sources and Analysis, National Geophysical Data Center, NESDIS,
NOAA, U.S. Department of Commerce, Boulder, CO, August 2008.

Billings, S.D. and D.J. FitzGerald, 1998. An integrated framework for interpolating
airborne geophysical data with special reference to radiometrics. Exploration
Geophysics, 29, 284-289.

Buchanan, C., 1998. 30 arc second gridded bathymetry. GA - Geoscience Australia, CD-
ROM.

Carroll, D., (ed), 1996. Geodata 9 second DEM user guide. Australian Surveying & Land
Information Group, Canberra.

Daniell, J. J., 2008. Development of a bathymetric grid for the Gulf of Papua and
adjacent areas: A note describing its development, J. Geophys. Res., 113, F01S15,
doi:10.1029/2006JF000673.

Geographx, 2008. New Zealand 250m Digital Elevation Model, available from
http://www.geographx.co.nz/downloads.html

Geoscience Australia, 2008. GEODATA 9 Second DEM and D8, Version 3 and Flow
Direction Grid 2008, Data User Guide, Canberra, Australia, available from
http://www.ga.gov.au/nmd/products/digidat/dem_9s.jsp

Jarvis A., H.I. Reuter, A. Nelson, E. Guevara, 2008. Hole-filled seamless SRTM data V4,
International Centre for Tropical Agriculture (CIAT), available from
http://srtm.csi.cgiar.org.

Petkovic, P. and C. Buchanan, 2002. Australian bathymetry and topography grid (January
2002). [CDROM]. Canberra: Geoscience Australia.

Smith, W. H. F., and D. T. Sandwell, 1997. Global seafloor topography from satellite
altimetry and ship depth soundings, Science, v. 277, p. 1957-1962, 26 Sept., 1997.

Webster, M.A., and P. Petkovic, 2005. Australian Bathymetry and Topography Grid,
June 2005. Record 2005/12.

38
 Australian Bathymetry and Topography Grid – June 2009

10. Other Data Sources

ETOPOv2g
U.S. Department of Commerce
National Oceanic and Atmospheric Administration
National Environmental Satellite, Data, and Information Service
National Geophysical Data Centre
http://www.ngdc.noaa.gov/mgg/fliers/06mgg01.html

ETOPO1
U.S. Department of Commerce
National Oceanic and Atmospheric Administration
National Environmental Satellite, Data, and Information Service
National Geophysical Data Centre
http://www.ngdc.noaa.gov/mgg/global/global.html

NZ 250m DEM
Geographx
The Dominican Observatory
34 Slamanca Rd
Kelburn
Wellington
New Zealand
http://www.geographx.co.nz/downloads.html

SRTM 90m DEM

CGIAR Consortium for Spatial Information
http://srtm.csi.cgiar.org/

COCOS, CHRISTMAS AND LORD HOWE ISLAND BATHYMETRIC GRIDS

R. Mleczko
Risk and Impact Analysis Group
Geoscience Australia
GPO Box 378, Canberra, ACT, 2601

GULF OF PAPUA AND ADJACENT AREAS GRID

J. Daniell
Seabed Mapping and Characterisation Group
Marine Geoscience and Environment Group
Geoscience Australia
GPO Box 378, Canberra, ACT, 2601

39
 Australian Bathymetry and Topography Grid – June 2009

Appendix 1 – Multibeam Surveys

SURVEY CODE SURVEY NAME GA ID VESSEL DATA OWNER SYSTEM
SST01/2008 Hobart-Sydney Transit 2467 R/V Southern Surveyor CSIRO Simrad EM300 (30 kHz)
SS05/2008 PULSE test, Tasmania 2466 R/V Southern Surveyor CMAR Simrad EM300 (30 kHz)
SS04/2008 Tsunami Detection Buoy 2465 R/V Southern Surveyor BoM Simrad EM300 (30 kHz)
SS03/2008 Eastern GAB 2464 R/V Southern Surveyor CMAR Simrad EM300 (30 kHz)
Bonney Coast, Kangaroo
SS02/2008 2463 R/V Southern Surveyor SARDI Simrad EM300 (30 kHz)
Island
SS01/2008 SE Tasmania 2462 R/V Southern Surveyor CMAR Simrad EM300 (30 kHz)
SS06/2008 Lord Howe Island 2461 R/V Southern Surveyor CSIRO/GA Simrad EM300 (30 kHz)
SS07/2007 GBR Reefs 2447 R/V Southern Surveyor JCU Simrad EM300 (30 kHz)
SO189-2 Sumatra 2460 R/V Sonne BGRH Simrad EM120 (12 kHz)
SO189-1 Sumatra 2459 R/V Sonne BGRH Simrad EM120 (12 kHz)
BGR/IFM-
SO186-1 SeaCause I 2458 R/V Sonne Simrad EM120 (12 kHz)
GEOMAR
SO186-2 SeaCause II 2457 R/V Sonne BGR Simrad EM120 (12 kHz)
WA Margins reconnaissance
GA-2476 2476 R/V Sonne GA Simrad EM120 (12 kHz)
(Legs 1-3)
NBP0408 NBP0408 2456 Nathaniel B. Palmer NSF Simrad EK500
East Cay 2455 James Kirby JCU
Sumatra subduction zone 2454 HMS Scott NOCS/RAN/UKHO
SS04/2007 WA Pelagic 2434 R/V Southern Surveyor CMAR Simrad EM300 (30 kHz)
SST03/2007 Hobart-Fremantle Transit 2433 R/V Southern Surveyor CMAR Simrad EM300 (30 kHz)
ANT-XXIII/9 ANT-XXIII/9 2432 R/V Polarstern AWI Atlas Hydrosweep DS2
BoM tsunami buoy
SS03/2007 2431 R/V Southern Surveyor BoM Simrad EM300 (30 kHz)
deployment
SS02/2007 TASMOUNTS2 2430 R/V Southern Surveyor CSIRO Simrad EM300 (30 kHz)
Sea Trials
Hobart Seatrials 02/2007 2429 R/V Southern Surveyor CSIRO Simrad EM300 (30 kHz)
SS02/2007

40
 Australian Bathymetry and Topography Grid – June 2009

SO-94 SO-94 2428 R/V Sonne IFM GEOMAR Atlas Hydrosweep DS

SO-113_2 SO-113 L2 2427 R/V Sonne IFM GEOMAR Atlas Hydrosweep DS
KR9813 KR9813 2426 R/V Kairei JAMSTEC
SST02/2007 Lacepede Shelf 2425 R/V Southern Surveyor GA Simrad EM300 (30 kHz)
SS01/2007 Bight Basin 2424 R/V Southern Surveyor GA Simrad EM300 (30 kHz)
SST01/2007 SSTransit 01/2007 2423 R/V Southern Surveyor GA/CSIRO Simrad EM300 (30 kHz)
YK01-15 YK01-15 2422 S/V Yokosuka JAMSTEC Seabeam 2112 (12 kHz)
YK01-14 YK01-14 2421 S/V Yokosuka JAMSTEC Seabeam 2112 (12 kHz)
YK02-07 YK02-07 2420 S/V Yokosuka JAMSTEC Seabeam 2112 (12 kHz)
SO-190Transit SO-190T 2418 R/V Sonne IFM GEOMAR Simrad EM-120 (12 kHz)
SS09/2006 LHRIDGE 2416 R/V Southern Surveyor UNSW Simrad EM300 (30 kHz)
SS11/2006 Tasmounts1 2415 R/V Southern Surveyor CMAR/DEH Simrad EM300 (30 kHz)
SST03/2006 SSTransit03/2006 2414 R/V Southern Surveyor GA Simrad EM300 (30 kHz)
SS10/2006 NSW Slopes 2413 R/V Southern Surveyor GA Simrad EM300 (30 kHz)
MR04-01 MR04-01 2412 R/V Mirai JAMSTEC Seabeam 2112 (12 kHz)
SST02/2006 SSTransit 02/2006 2411 R/V Southern Surveyor UNSW Simrad EM300 (30 kHz)
SST01/2006 SSTransit 01/2006 2409 R/V Southern Surveyor GA Simrad EM300 (30 kHz)
SS06/2006 Rowley Shoals 2408 R/V Southern Surveyor GA Simrad EM300 (30 kHz)
SS05/2006 Meso_Eddies 2406 R/V Southern Surveyor UWA Simrad EM300 (30 kHz)
MR03K04L5 MR03K04_Leg5 2405 R/V Mirai JAMSTEC Seabeam 2112 (12 kHz)
NBP0501 NBP0501 2404 R/V Nathaniel B Palmer NSF Simrad EM-120 (12 kHz)
SS03/2006 TRULL06 2402 R/V Southern Surveyor UTAS Simrad EM300 (30 kHz)
Hobart Trials 2401 R/V Southern Surveyor CSIRO Simrad EM300 (30 kHz)
SS04/2006 Leeuwin Currents 2400 R/V Southern Surveyor UWA Simrad EM300 (30 kHz)
J.K_11/04 2399 James Kirby JCU Reson Seabat 8101
MOGAM06 2398 R/V OGS Explora OGS Reson Seabat 8150 (12 kHz)
SS02/2006 Palaeo Murrays 2397 R/V Southern Surveyor ANU Simrad EM300 (30 kHz)
VT82 (MD153L2?) AUSFAIR_GAB Transit 2396 R/V Marion Dufresne IPEV Thompson Marconi TSM 5265
MD153 AUSFAIR_ZoNeco12 2395 R/V Marion Dufresne IPEV Thompson Marconi TSM 5265
MD152 TAS-NZ Transit 2394 R/V Marion Dufresne IPEV Thompson Marconi TSM 5265
MR04K03L2 MR04K03_Leg2 2393 R/V Mirai JAMSTEC Seabeam 2112 (12 kHz)

41
 Australian Bathymetry and Topography Grid – June 2009

EW9510 EW9510 2392 R/V Maurice Ewing LDEO Thompson Marconi TSM 5265
EW0002 EW0002 2391 R/V Maurice Ewing LDEO Thompson Marconi TSM 5265
SS10/2005 Wabio_2 2390 R/V Southern Surveyor CSIRO Simrad EM300 (30 kHz)
US Antarctic
NBP0406 NBP0406 2389 R/V Nathaniel B Palmer Simrad EM-120 (12 kHz)
Program
MR00K07L2 MR00K07_Leg2 2388 R/V Mirai JAMSTEC Seabeam 2112 (12 kHz)
MR01K05L2 MR01K05_Leg2 2387 R/V Mirai JAMSTEC Seabeam 2112 (12 kHz)
MR02K04L2 MR02K04_Leg2 2386 R/V Mirai JAMSTEC Seabeam 2112 (12 kHz)
MR03K03L2 MR03K03_Leg2 2385 R/V Mirai JAMSTEC Seabeam 2112 (12 kHz)
MR03K04L1 MR03K04_Leg1 2384 R/V Mirai JAMSTEC Seabeam 2112 (12 kHz)
SS09/2005 Naturaliste Rocks 2383 R/V Southern Surveyor UWA Simrad EM300 (30 kHz)
MD148 PECTEN 2382 R/V Marion Dufresne SydneyUni/IPEV Thompson Marconi TSM 5265
SWS2003 2381 Conference Dataset Various
SS06/2005 Cornea-Cartier 2375 R/V Southern Surveyor AIMS/ANU Simrad EM300 (30 kHz)
MANAUTE MANAUTE 2374 N.O. L'Atalante IFREMER Simrad EM-12D (12 kHz)
Sydney Woolongong Transit 2373 HMAS Melville RAN Atlas-Fansweep (20 -100)
VANC29MV VANC29 2372 R/V Melville SIO Seabeam 2000 (12 kHz)
VANC25MV VANC25 2371 R/V Melville SIO Seabeam 2000 (12 kHz)
VANC22MV VANC22 2370 R/V Melville SIO Seabeam 2000 (12 kHz)
VANC21MV VANC21 2369 R/V Melville SIO Seabeam 2000 (12 kHz)
VANC15MV VANC15 2368 R/V Melville SIO Seabeam 2000 (12 kHz)
EW0112 EW0112 2367 R/V Ewing LDEO Atlas Hydrosweep
EW9910 EW9910 2364 R/V Ewing LDEO Atlas Hydrosweep
SS09/2004 SS200409 2363 R/V Southern Surveyor Uni of Sydney Simrad EM300 (30 kHz)
Topas Trials 2361 R/V Southern Surveyor GA Simrad EM300 (30 kHz)
SS11/2004 Notove 2360 R/V Southern Surveyor ANU Simrad EM300 (30 kHz)
SS10/2004 N_Fiji_1 2359 R/V Southern Surveyor UTAS Simrad EM300 (30 kHz)
SS07/2004 WinterEnv 2358 R/V Southern Surveyor CSIRO Simrad EM300 (30 kHz)
SS06/2004 Cotrove 2357 R/V Southern Surveyor ANU Simrad EM300 (30 kHz)
Sydney Hobart Transit 2356 R/V Southern Surveyor CSIRO Simrad EM300 (30 kHz)
SS04/2004 S E Ecosystems 2355 R/V Southern Surveyor CSIRO Simrad EM300 (30 kHz)

42
 Australian Bathymetry and Topography Grid – June 2009

EM300 Trials 2354 R/V Southern Surveyor CSIRO Simrad EM300 (30 kHz)
BHP_HIRES 2353 BHPP
Rottnest 2352 HMAS Cook RAN-JCU Seabeam (12 kHz)
Albany 2351 HMAS Cook RAN-JCU Seabeam (12 kHz)
Adelaide 2350 HMAS Cook RAN-JCU Seabeam (12 kHz)
Eaxa 2349 HMAS Cook RAN-JCU Seabeam (12 kHz)
GBR 2348 HMAS Cook RAN-JCU Seabeam (12 kHz)
ASHMORE 2347 HMAS Leeuwin RAN Atlas-Fansweep (20 -100)
AUSTREA3 AUSTREA3 2346 N.O. L'Atalante LINZ/IFREMER Simrad EM-12D (12 kHz)
VANC30MV VANC30 2343 R/V Melville SIO Seabeam 2000 (12 kHz)
VANC28MV VANC28 2342 R/V Melville SIO Seabeam 2000 (12 kHz)
VANC27MV VANC27 2341 R/V Melville SIO Seabeam 2000 (12 kHz)
VANC26MV VANC26 2340 R/V Melville SIO Seabeam 2000 (12 kHz)
VANC24MV VANC24 2339 R/V Melville SIO Seabeam 2000 (12 kHz)
VANC23MV VANC23 2338 R/V Melville SIO Seabeam 2000 (12 kHz)
VANC20MV VANC20 2337 R/V Melville SIO Seabeam 2000 (12 kHz)
VANC19MV VANC19 2336 R/V Melville SIO Seabeam 2000 (12 kHz)
VANC16MV VANC16 2333 R/V Melville SIO Seabeam 2000 (12 kHz)
VANC14MV VANC14 2332 R/V Melville SIO Seabeam 2000 (12 kHz)
VANC12MV VANC12 2331 R/V Melville SIO Seabeam 2000 (12 kHz)
VANC11MV VANC11 2330 R/V Melville SIO Seabeam 2000 (12 kHz)
VANC10MV VANC10 2329 R/V Melville SIO Seabeam 2000 (12 kHz)
NBP0209 NBP0209 2328 R/V Nathaniel B Palmer NSF Simrad EM-120 (12 kHz)
NBP0008 NBP0008 2327 R/V Nathaniel B Palmer NSF Seabeam 2112 (12kHz)
NBP9603a NBP9603a 2326 R/V Nathaniel B Palmer NSF Seabeam 2112 (12kHz)
EW0003 EW0003 2325 R/V Ewing LDEO Atlas Hydrosweep
EW9914 EW9914 2324 R/V Ewing LDEO Atlas Hydrosweep
EW9912 EW9912 2323 R/V Ewing LDEO Atlas Hydrosweep
EW9911 EW9911 2322 R/V Ewing LDEO Atlas Hydrosweep
EW9512 EW9512 2321 R/V Ewing LDEO Atlas Hydrosweep
EW9511 EW9511 2320 R/V Ewing LDEO Atlas Hydrosweep

43
 Australian Bathymetry and Topography Grid – June 2009

TAN0308 NORFANZ 2312 R/V R/V Tangaroa NIWA Simrad EM-300 (30 kHz)
MR03K04L6 MR03K04_Leg6 2309 R/V Mirai JAMSTEC Seabeam 2112 (12 kHz)
SO 168 ZEALANDIA 2308 R/V Sonne IFM GEOMAR Simrad EM-120 (12 kHz)
EW0201 EW0201 2306 R/V Ewing LDEO Atlas Hydrosweep
EW0114 EW0114 2305 R/V Ewing LDEO Atlas Hydrosweep
KIWI10RR KIWI10RR 2304 R/V Roger Revelle SIO Seabeam 2112 (12 kHz)
NBP9801 NBP9801 2101 R/V Nathaniel B Palmer NSF Seabeam 2112 (12kHz)
BMRG08MV BMRG08MV 2007 R/V Melville SIO Seabeam 2000 (12 kHz)
MW9304 MW9304 1969 R/V Moana Wave SOEST MR1
NBP9702 NBP9702 1824 R/V Nathaniel B Palmer NSF Seabeam 2112 (12kHz)
EW9513 EW9513 1705 R/V Ewing LDEO Atlas Hydrosweep
SOJN06MV SOJN06MV 1435 R/V Melville SIO Seabeam 2000 (12 kHz)
SOJN04MV SOJN04MV 1431 R/V Melville SIO Seabeam 2000 (12 kHz)
WEST10MV WEST10MV 1274 R/V Melville SIO Seabeam 2000 (12 kHz)
BMRG07MV BMRG07MV 1273 R/V Melville SIO Seabeam 2000 (12 kHz)
WEST12MV WEST12MV 1266 R/V Melville SIO Seabeam 2000 (12 kHz)
WEST11MV WEST11MV 1265 R/V Melville SIO Seabeam 2000 (12 kHz)
EW9203 EW9203 1212 R/V Ewing LDEO Atlas Hydrosweep
EW9202 EW9202 1211 R/V Ewing LDEO Atlas Hydrosweep
KN145L12 Wombat Plateau 1180 R/V Knorr WHOI Seabeam 2112 (12 kHz)
KN145L4 Perth Canyon 1179 R/V Knorr WHOI Seabeam 2112 (12 kHz)
SOJN08MV SOJN08MV 1178 R/V Melville SIO Seabeam 2000 (12 kHz)
SOJN05MV SOJN05MV 1175 R/V Melville SIO Seabeam 2000 (12 kHz)
BMRG05MV BMRG05MV 1173 R/V Melville SIO Seabeam 2000 (12 kHz)
BMRG04MV BMRG04MV 1172 R/V Melville SIO Seabeam 2000 (12 kHz)
WEST09MV WEST09MV 1166 R/V Melville SIO Seabeam 2000 (12 kHz)
BMRG06MV BMRG06MV 1154 R/V Melville SIO Seabeam 2000 (12 kHz)
WEST08MV WEST08MV 1153 R/V Melville SIO Seabeam 2000 (12 kHz)
WEST06MV WEST06MV 1152 R/V Melville SIO Seabeam 2000 (12 kHz)
WEST05MV WEST05MV 1149 R/V Melville SIO Seabeam 2000 (12 kHz)
SS07/2005 Wabio 296 R/V Southern Surveyor CSIRO Simrad EM300 (30 kHz)

44
 Australian Bathymetry and Topography Grid – June 2009

SS01/2005 Fraser2 295 R/V Southern Surveyor Uni of Newcastle Simrad EM300 (30 kHz)
SS08/2005 Mentelle 293 R/V Southern Surveyor GA Simrad EM300 (30 kHz)
SS03/2005 Carpbio 283 R/V Southern Surveyor CSIRO/GA Simrad EM300 (30 kHz)
SS05/2005 Arafura 282 R/V Southern Surveyor GA/NOO Simrad EM300 (30 kHz)
SS04/2005 Carpreefs2 276 R/V Southern Surveyor GA Simrad EM300 (30 kHz)
SS02/2005 Mellish 274 R/V Southern Surveyor GA Simrad EM300 (30 kHz)
SS05/2004 Kenn Plateau 270 R/V Southern Surveyor GA Simrad EM300 (30 kHz)
SS03/2004 Bremer 265 R/V Southern Surveyor GA Simrad EM300 (30 kHz)
VANC13MV VANC13 247 R/V Melville SIO Seabeam 2000 (12 kHz)
MD131 AUSCAN2 245 R/V Marion Dufresne AGSO Thompson Marconi TSM 5265
SS04/2003 Carpreefs1 238 R/V Southern Surveyor GA Reson 8101 (240 kHz)
MD130 AUSCAN1 236 R/V Marion Dufresne AGSO Thompson Marconi TSM 5265
EW0113 EW0113 235 R/V Ewing LDEO Atlas Hydrosweep
FR01/02 Torres Strait 234 R/V Franklin GA Reson 8101 (240 kHz)
HI339G New Zealand Star Banks 233 HMAS Melville RAN Atlas Fansweep 20 (100kHz)
SS01/2000 Bioview 224 R/V Southern Surveyor CSIRO Simrad EM1002 (100kHz)
AUSTREA2 AUSTREA2 223 N.O. L'Atalante GA/NOO Simrad EM-12D (12 kHz)
AUSTREA1 AUSTREA1 222 N.O. L'Atalante AGSO Simrad EM-12D (12 kHz)
FAUST2 FAUST2 221 N.O. L'Atalante GA/Uni of Paris VI Simrad EM-12D (12 kHz)
ZoNeco5 ZoNeco5 220 N.O. L'Atalante AGSO Simrad EM-12D (12 kHz)
SO 136 TASQWA 212 R/V Sonne AGSO Atlas Hydrosweep (15.5 kHz)
SOJN07MV SOJN07MV 210 R/V Melville SIO Seabeam 2000 (12 kHz)
MD110 Margau 208 R/V Marion Dufresne IPEV/GA Thompson Marconi TSM 5265
ADEDAV ADEDAV 157 N.O. L'Atalante GA/IFREMER Simrad EM-12D (12 kHz)
TASMANTE TASMANTE 125 N.O. L'Atalante GA/IFREMER Simrad EM-12D (12 kHz)
RS9401 Macquarie Ridge 124 R/V Rig Seismic AGSO MR1
TRANSNOR TRANSNOR 123 N.O. L'Atalante IFREMER Simrad EM-12D (12 kHz)

45
 Australian Bathymetry and Topography Grid – June 2009

Appendix 2 – SRTM DEM Metadata

PROCESSED SRTM DATA VERSION 4
The data distributed here are in ARC GRID, ARC ASCII and Geotiff format, in decimal
degrees and datum WGS84. They are derived from the USGS/NASA SRTM data. CIAT
have processed this data to provide seamless continuous topography surfaces. Areas with
regions of no data in the original SRTM data have been filled using interpolation methods
described by Reuter et al. (2007).

DISTRIBUTION
Users are prohibited from any commercial, non-free resale, or redistribution without explicit
written permission from CIA. Users should acknowledge CIAT as the source used in the
creation of any reports, publications, new data sets, derived products, or services resulting
from the use of this data set. CIAT also request reprints of any publications and notification
of any redistributing efforts. For commercial access to the data, send requests to Andy
Jarvis (a.jarvis@cgiar.org).

NO WARRANTY OR LIABILITY
CIAT provides these data without any warranty of any kind whatsoever, either express or
implied, including warranties of merchantability and fitness for a particular purpose. CIAT
shall not be liable for incidental, consequential, or special damages arising out of the use of
any data.

ACKNOWLEDGMENT AND CITATION

We kindly ask any users to cite this data in any published material produced using this data,
and if possible link web pages to the CIAT-CSI SRTM website (http://srtm.csi.cgiar.org).

Citations should be made as follows:

Jarvis A., H.I. Reuter, A. Nelson, E. Guevara, 2008, Hole-filled seamless SRTM data V4,
International Centre for Tropical Agriculture (CIAT), available from
http://srtm.csi.cgiar.org.

REFERENCES
Reuter H.I, A. Nelson, A. Jarvis, 2007, An evaluation of void filling interpolation methods
for SRTM data, International Journal of Geographic Information Science, 21:9, 983-1008.

46