Seascape classification layer from the Darwin Harbour 2011 Marine Survey (GA0333)  

Authors / CoAuthors

Siwabessy, J. | Tran, M. | Huang, Z. | Nichol, S.L. | Atkinson, I.

Abstract

This dataset contains seascape classification layer derived from bathymetry and backscatter, and their derivative from seabed mapping surveys in Darwin Harbour. The survey was undertaken during the period 24 June to 20 August 2011 by iXSurvey Australia Pty Ltd for the Department of Natural Resources, Environment, The Arts and Sport (NRETAS) in collaboration with Geoscience Australia (GA), the Darwin Port Corporation (DPC) and the Australian Institute of Marine Science (AIMS) using GA's Kongsberg EM3002D multibeam sonar system and DPC's vessel Matthew Flinders. The survey obtained detailed bathymetric map of Darwin Harbour. Refer to the GA record ' Mapping and Classification of Darwin Harbour Seabed' for further information on processing techniques applied (GeoCat: 79212; GA Record: 2015/xx)

Product Type

dataset

eCat Id

83951

Contact for the resource

Keywords

• Marine Data

( Theme )

• multibeam

( Theme )

• marine survey

( Theme )

• bathymetry

( Theme )

• backscatter

( Theme )

• marine environmental baselines

• AU-NT

Australian and New Zealand Standard Research Classification (ANZSRC)

• Marine Geoscience

• Published_External

Publication Date

2015-01-01T00:00:00

Creation Date

Security Constraints

Australian Government Security ClassificationSystem    

Classification - unclassified

Legal Constraints

Creative Commons Attribution 4.0 International Licence    

Access - license

Use - license

Status

Purpose

Maintenance Information

asNeeded

Topic Category

environment

Series Information

Lineage

Multiple spatial layers of physical data were classified using the Iterative Self Organising (ISO) Unsupervised Classification methodology available in ArcGIS (v.10). This methodology performs unsupervised classification based on a series of input raster bands using the ISO Cluster and Maximum Likelihood Classification tools. Datasets used in the derivation of the seabed habitat classification included bathymetry, slope, rugosity, backscatter and p rock. Statistically, there are an optimum number of classes into which the data can be partitioned that minimises uncertainty. The method we used is called the Distance Ratio method. Classifications were carried out for 2 to 9 classes, with the distance ratio estimated for each class. The distance ratio is the ratio of the average of the mean distance of each class member from its class mean to the overall average distance of each member from the overall mean. This value provides an indication of how well the data matches the assigned classification. The optimal number of classes occurs where the distance ratio has a local minimum, indicating that the addition of more classes will not improve the classification accuracy as much as the addition of previous classes. Choosing fewer classes will not explain the variation in the classes as thoroughly.

Parent Information

Extents

[-12.59, -12.32, 130.69, 130.94]

Reference System

Spatial Resolution

Service Information

Associations

Downloads and Links

File download  

Source Information

Legacy product, source data not available. 130.69 130.94 -12.59 -12.32