CEFASe40207b8-8739-46b4-9177-03bc430dcd13
English
dataset
Centre for Environment, Fisheries & Aquaculture Science
Data Manager
+44 (0)1502 562244
Cefas Lowestoft Laboratory
Pakefield Road
Lowestoft
Suffolk
NR33 0HT
UK
data.manager@cefas.co.uk
pointOfContact
2020-12-09T09:07:17
MEDIN Discovery Metadata Standard
Version 2.3.7
urn:ogc:def:crs:EPSG::32600
OGP
2015 - 2015 Centre for Environment, Fisheries & Aquaculture Science (Cefas) DPLUS026 British Virgin Islands Seabed Classification Map
Darwin BVI mapping
2015-12-31
publication
CEFASe40207b8-8739-46b4-9177-03bc430dcd13
http://www.cefas.co.uk/
Predicted seabed classification map for part of Sir Francis Drake Channel
south of Tortola, British Virgin Islands.
Centre for Environment, Fisheries & Aquaculture Science
Data Manager
+44 (0)1502 562244
Cefas Lowestoft Laboratory
Pakefield Road
Lowestoft
Suffolk
NR33 0HT
UK
data.manager@cefas.co.uk
originator
Centre for Environment, Fisheries & Aquaculture Science
Data Manager
+44 (0)1502 562244
Cefas Lowestoft Laboratory
Pakefield Road
Lowestoft
Suffolk
NR33 0HT
UK
data.manager@cefas.co.uk
custodian
notPlanned
Geographic Information System
NDGO0005
Habitat characterisation
Habitat extent
SeaDataNet P021 parameter discovery vocabulary
2011-03-25
revision
Sea bed
GEMET, version 1.0
2008-06-01
publication
Hydrography
GEMET - INSPIRE themes, version 1.0
2008-06-01
publication
Public data (Crown Copyright) - Open Government Licence Terms and Conditions apply
otherRestrictions
Public data (Crown Copyright) - Open Government Licence Terms and Conditions apply
English
inlandWaters
SeaVoX Vertical Co-ordinate Coverages
2010-05-18
revision
Unknown
-70
-60
10
20
2015-04-01T00:00:00.000Z
2015-09-30T00:00:00.000Z
http://data.cefas.co.uk/#/View/18174/order
dataset
A new habitat map for the site was produced by analysing and interpreting the
available acoustic data and the ground truth data collected by the dedicated
survey of North St George's Channel rMCZ. The process is a combination of two
approaches, statistical modelling and image analysis.
To map substrata and assemblage types across the study site, object-based
image analysis (OBIA; Blaschke, 2010) was utilised. The technique was
implemented in the software package eCognition® v8.8 combined with a
predictive modelling approach using the Random Forest2 algorithm (Breiman,
2001) application within eCognition®. This consists of a classification model
aimed at predicting a target variable (in this case, sediment composition)
based on exhaustively sampled auxiliary variables (in this case, the acoustic
data). The technique has been used in previous studies to predict sediment
type (Li et al., 2011a). Li et al. (2011b) showed that the Random Forest
algorithm outperformed a range of other modelling techniques for predicting
substrate type (Liaw and Wiener, 2002).