Tracking forest attributes across Canada between 2001 and 2011 using a k nearest neighbors mapping approach applied to MODIS imagery

Abstract

Mapping Canada’s forests is a significant challenge given their extent and the interprovincial differences in forest inventories. We created new sets of nationally consistent forest attribute maps for the years 2001 and 2011 by building upon previously published work with the objective to determine if sequential maps of forest attributes could be used to quantify changes over time. We first refined our previously published methodology of using the k nearest neighbors (kNN) prediction method and MODIS spectral reflectance data as predictive variables. The maps were generated using an improved reference dataset and a new analytical kNN workflow. We then evaluated 2001 to 2011 changes in two key attributes, aboveground biomass and percent tree cover, on pixels identified from published sources as having undergone fire, harvest, or postdisturbance regrowth during that period. For all three change types, average changes in both aboveground biomass and percent tree cover between 2001 and 2011 matched expectations relative to the dynamics of Canadian forests. Our results support the use of sequential national maps of forest attributes for evaluating regionally aggregated disturbance-related changes in forest properties. The new forest attribute maps are available from Beaudoin et al. (2017 ; doi: 10.23687/ec9e2659-1c29-4ddb-87a2-6aced147a990) at http://ouvert.canada.ca/data/fr/dataset/ec9e2659-1c29-4ddb-87a2-6aced147a990 .

Sapwood biomass carbon in northern boreal and temperate forests

“…Notwithstanding the remaining uncertainties, which are unavoidable with the currently available data, the present study involves several innovations that go far beyond state of the art in the synergy of satellite and forest inventory data for biomass estimation: Theoretical considerations on the sapwood distribution in the tree stem make it possible to apply relatively more frequent measurements of sapwood area for estimation of sapwood biomass. The application of a global stand density product (Crowther et al, ) enables the gap to be bridged between inventory measurements at the tree level and satellite observations at the forest stand scale. The application of a dominant tree genera map covering the boreal and temperate forests in Canada (Beaudoin et al, ), the USA (Wilson et al, ), Europe (Brus et al, ) and Russia (Schepaschenko et al, ) allows genera‐specific differences in biomass allometry to be taken into account. …”

“…In order to address the above‐mentioned research objectives, here we: (a) develop a theoretical framework to estimate sapwood biomass for a given stem biomass; (b) collect measurements of sapwood proportions from the BAAD, an additional extensive literature review and our own studies; and (c) derive a sapwood biomass map based on remote sensing products (accompanied by an uncertainty estimate), inferred allometric relationships and new products on the distribution of stand density and tree genera (Figure ). A global stand density product (Crowther et al, ) allows scaling from measurements at tree level to areal estimates, and this is supplemented with information on tree species distribution (Beaudoin, Bernier, Villemaire, Guindon, & Guo, ; Brus et al, ; Schepaschenko et al, ; Wilson, Lister, & Riemann, ; Wilson, Lister, Riemann, & Griffith, ) that enable the application of allometric relationships for tree genera in addition to the leaf type level.…”

“…Measurements of and are aggregated into classes at the genus level ( Betula , other broadleaf diffuse‐porous, Quercus , other broadleaf ring‐porous, Larix , Abies , Picea , Pinus , Pseudotsuga and Tsuga ) in order to account for the variation in their relationship. A corresponding map of the dominant tree genera in northern boreal and temperate forests is derived at 0.01° resolution from available up‐scaled forest inventory data on tree species distribution for Canada (Beaudoin et al, ), the USA (Wilson et al, ), Europe (Brus et al, ) and Russia (Schepaschenko et al, ). The original tree species distribution maps are reclassified based on the tree genera classes identified above, adding other needleleaf evergreen and other (unidentified) and resampled or reprojected (nearest neighbour) to 0.01° resolution and geographical coordinates.…”

“…We use the BAAD to establish the relationships in Equations (1) and (2) (Beaudoin et al, 2017(Beaudoin et al, , 2018, the USA (Wilson et al, 2012(Wilson et al, , 2013,…”

“…3. The application of a dominant tree genera map covering the boreal and temperate forests in Canada (Beaudoin et al, 2017(Beaudoin et al, , 2018, the USA (Wilson et al, 2012(Wilson et al, , 2013, Europe (Brus et al, 2012) and Russia (Schepaschenko et al, 2011) allows genera-specific differences in biomass allometry to be taken into account.…”