GEOSCAN Search Results: Fastlink


TitleRemote predictive mapping of surficial materials on northern Baffin Island: developing and testing techniques using Landsat TM and digital elevation data
AuthorBrown, O; Harris, J R; Utting, D; Little, E C
SourceGeological Survey of Canada, Current Research (Online) 2007-B1, 2007, 12 pages,
PublisherNatural Resources Canada
Mediaon-line; digital
File formatpdf (Adobe Acrobat Reader)
NTS37E/02; 37E/03; 37E/04; 37E/05; 37E/06; 37E/07; 37E/10; 37E/11; 37E/12; 37E/13; 37E/14; 37E/15
Areanothern Baffin Island; Baffin Uplands; Conn Lake; Barnes Ice Cap
Lat/Long WENS-76.0000 -73.0000 71.0000 70.0000
Subjectssurficial geology/geomorphology; mapping techniques; remote sensing; satellite imagery; LANDSAT imagery; digital terrain modelling; statistical analyses; glacial deposits; tills; glaciolacustrine deposits; fluvial deposits; bedrock geology; Archean; Rae Domain; Western Churchill Province; classification; Phanerozoic; Cenozoic; Quaternary; Precambrian
Illustrationssketch maps; satellite images; digital elevation models; photographs; tables; 3-D diagrams; Box-and-whisker diagrams
ProgramNorthern Resources Development Program
LinksLANDSAT TM data available free on GeoGratis - Données du LANDSAT-TM disponibles gratuitement au GéoGratis
Released2007 03 01
AbstractConsidering the vastness of Nunavut, the paucity of regional-scale surficial geology maps for the territory, the significant expense of working in a remote region, and the increasing availability of affordable, remotely sensed data, it is timely to develop and test remote predictive mapping techniques for producing surficial geology maps. The goal of this remote predictive mapping project is to produce a surficial materials map, which will be used to expedite subsequent ground-based mapping and sampling.
This paper describes techniques used to produce a surficial materials map for an area in northern Baffin Island using remote predictive mapping techniques with LandsatTMand digital elevation data. The predictive maps produced in advance of the field work (i.e. "ground truthing") were found to be approximately 50% accurate. To improve remote predictive mapping accuracy to at least 80%, high-resolution imagery may need to be included in the remote predictive mapping protocol.