Long-Term Change Detection Wins Esri Canada's App of the Month for December 2021

A Long-Term Change Detection project was initiated by the NWT Centre for Geomatics in response to client demand to monitor various changes on the NWT Landscape.  This included anthropogenic change such as seismic lines and other industry related activities, and climate driven or natural changes like forest fires, permafrost thaw slumps, coastal erosion and variable water levels.  To do this we used a type of remote sensing analysis that was first attempted in national parks of the NWT by Natural Resources Canada (NRCan), called tasseled cap analysis. 

When this type of analysis was initially completed, it was completed on smaller portions of the territory.  This was done to limit the volume of data to process by conventional workstations, and to process discrete zones of the territory to reduce spectral variability.  Spectral variability can occur when processing larger areas, where some portions may have more snow and ice for instance, while other areas may have green vegetation.  Other factors affecting spectral variability include high incidence of shadows, as seen in mountainous regions.

To assess long-term change across the entire NWT required a great deal of computing power and specialized modeling to manage spectral variability across the territory.  A large volume of data was required to complete the analysis, including satellite imagery (Landsat 1984-2019) over the peak growing season for each year of available data in the NWT.  This was facilitated using Google Earth Engine, a cloud-based platform that can leverage super-computing capabilities and direct access to the satellite imagery.

The analysis resulted in image data showing patterns of different colours across the landscape.  These patterns can be validated with ground-based observation and attributed to changes such as old forest fires, erosion, new infrastructure, and coastal erosion to name a few. 

The results of this analysis is provided through a story map.  The story map provides some high-level descriptions of the analysis and the outputs.  It also provides some examples of the types of change seen on the landscape and describes some of the work ahead needed to help classify other types of changes that were detected in the results of the analysis.

Why does this matter?  For many years scientists have been trying to quantify change in the NWT and this has been difficult to do because of the large area of the territory and the temporal considerations when analyzing change over time (including availability of consistent data).  Quantifying change may help us better understand impacts to wildlife, cultural resources, source water protection, and food security.  With this, land managers will be better able to make information-based decisions on how to adaptively manage resources in response to a changing landscape.

A user guide was created to highlight various types of landscape change and how these changes appear in the data.  This guide is intended for all audiences and gives a broad overview of the data.  We intend to share this data with researchers conducting work in the Northwest Territories.  If you are interested in this data please contact the NWTCG and we will be happy to discuss this with you.

Link:

https://experience.arcgis.com/experience/2effc9c8150a4abebdc9ef587865ab8e

Short Link

https://arcg.is/0Tez0z

Link to Esri Site:

https://resources.esri.ca/news-and-updates/app-of-the-month-long-term-la...

Link to CBC's the Weekender Radio Show Interview with Steve Schwarz:

https://www.cbc.ca/listen/live-radio/1-197-the-weekender