Geo-referenced digital photographs and videos with associated GPS waypoints and tracks for terrestrial ecosystem types from Victoria Island, King William Island and continental Nunavut, 2014

• Main Authors: McLennan, Donald, Wagner, Johann
• Format: Dataset
• Language: English
• Published: Canadian Cryospheric Information Network 2015
• Subjects: Arctic; Cambridge Bay; GIS; Mapping; Satellite imagery; Vegetation; Victoria Island; Polar Knowledge Canada/Canadian High Arctic Research Station; Nunavut; Canada; King William Island; William Island; victoria island
• Online Access: https://dx.doi.org/10.21963/12526; https://www.polardata.ca/pdcsearch/?doi_id=12526

Geo-referenced digital photographs and videos with the associated GPS waypoints and tracks of various terrestrial ecosystem types have been taken during the summer of 2014 on Victoria Island, King William Island and continental Nunavut. The photographs and videos, in combination with additional GPS waypoints and tracks, characterize site, soil and vegetation characteristics of the arctic landscape. The geo-referenced photographs and videos, as well as the GPS data will be entered in a GIS software and overlaid on satellite imagery. By matching pixel colours of the satellite images with different ecosystem types in the field, a model will be developed that will allow the assessment of ecosystem types over large areas, based solely on satellite imagery. The digital photographs were taken from ground with GPS-enabled cameras (geographical coordinates written in the image file¿s exif information). Digital videos of the landscape below were shot from helicopter and floatplane. The geographical coordinates are stored in a separate text file associated with the video file. Additionally, waypoints and tracks associated with the photographs and videos were recorded with a handheld GPS device. : Purpose: Geo-referenced digital photographs and videos with associated GPS waypoints and tracks of various terrestrial ecosystems have been taken during the summer of 2014 on Victoria Island, King William Island and continental Nunavut for the purpose of training a model using Random Forest software aimed at predicting and delineating terrestrial ecosystems over large areas. The model predictors will include data in processed optical satellite imagery (R, G, B, ifR), predictors developed from associated digital elevation models (e.g., slope, elevation, slope position, soil drainage models), as well as predictors developed from archived optical data such as spring snow bank persistence . : Summary: Pictures and videos and the associated geographical coordinates have been taken of the arctic landscape on Victoria Island, King William Island and continental Nunavut. These information will aid in development of a model which will allow determination of vegetation and habitat types from satellite images.