Canadian Forest Service Publications

Classification of annual non-stand replacing boreal forest change in Canada using Landsat time series: A case study in northern Ontario. 2016. Ahmed, O.; Wulder, M.A.; White, J.C.; Hermosilla, T.; Coops, N.C.; Franklin, S.E. Remote Sensing Letters. Vol. 81, No. 1, pp. 29-37.

Year: 2017

Available from: Pacific Forestry Centre

Catalog ID: 37307

Language: English

CFS Availability: PDF (download), PDF (request by e-mail)

Available from the Journal's Web site.
DOI: 10.1080/2150704X.2016.1233371

† This site may require a fee.

Abstract

Standardized protocols for forest change detection and classification based on Landsat time series data are becoming more common for use in characterizing multi-decadal history or trends in forest dynamics. One such protocol, referred to as Composite-2-Change (C2C), is a highly automated process developed in Canada that is applicable across extensive forest regions and includes change detection and typing based on Best-Available-Pixel (BAP) image compositing, spectral trend analysis of breakpoints, object-based segmentation, and Random Forest (RF) classification. The aim of this Letter is to assess the classification accuracy of the Composite-2-Change (C2C) protocol in an eastern Canadian boreal forest environment in northern Ontario. Results demonstrated that the Landsat-derived change detection and attribution approach was approximately 90% accurate for stand-replacing forest change (fire, harvesting, roads), and approximately 75% for four non-stand replacing forest changes caused by spruce budworm and forest tent caterpillar defoliation, wetland and forest flooding caused by localized hydrological variations, and one class of multiple/other disturbances. The C2C protocol approach offers unique independent data layers for modeling that can be used to relate and inform on a range of substantive and subtle changes, which in turn can be labeled and tracked, offering otherwise unavailable information on forest dynamics over large areas.

Plain Language Summary

Standardized protocols for forest change detection and classification based on Landsat time series data are becoming more common for use in characterizing multi-decadal history or trends in forest dynamics. Additional testing was recommended: i) to document the use of the C2C* Landsat time series protocol in other forest environments and ecoregions; and ii) to consider certain change features (e.g., roads) and non-stand replacing change in greater detail to increase the accuracy, comprehensiveness, and overall utility of the protocol. *Composite-2-Change (C2C) is the standardized Landsat time series protocol we have developed and published on previously. The current study was designed to assess annual forest change classification accuracy, with a focus on non-stand replacing change. Stand-replacing forest changes associated with roads, industrial-scale forest harvesting patterns and wildfires were classified with the highest individual class accuracies and with approximately 90% overall classification accuracy. Four classes of annual non-stand replacing change caused by defoliating insect outbreaks and local hydrological changes (e.g., forest and wetland flooding) were approximately 75% accurate. Areas that experienced multiple non-stand replacing disturbances were identified with higher omission and commission error.

Date modified: