View in EDS

Quantifying and mapping tree cover in Alberta and Manitoba's prairie landscapes.

Saved in:
Bibliographic Details
Title: Quantifying and mapping tree cover in Alberta and Manitoba's prairie landscapes.
Authors: Shafeian, Elham1 (AUTHOR) elham.shafeian@usask.ca, Mood, Bryan J.1 (AUTHOR), Belcher, Kenneth W.2 (AUTHOR), Laroque, Colin P.1 (AUTHOR)
Source: International Journal of Remote Sensing. Mar2026, Vol. 47 Issue 6, p2555-2580. 26p.
Subjects: Prairies, K-means clustering, Prairie ecology, Remote-sensing images, Ecological regions, Forest density, Random forest algorithms
Geographic Terms: Alberta, Manitoba
Abstract: The study aimed to map and measure non-forest tree cover across the prairie regions of Alberta and Manitoba. These non-forest trees are often overlooked in literature, even though they provide important ecosystem services, such as soil protection and carbon storage. We used Sentinel-2 (S2) and Sentinel-1 (S1) data as predictors and very high-resolution (VHR) Google images as reference data. Random Forest (RF) classification, in the prairies of Alberta and Manitoba, produced accurate tree cover maps with overall accuracy and kappa of the best-performing model (S2 imagery for September–October 2023–2024) of 0.98 and 0.97, respectively. For ecological stratification and quantification, we applied K-means clustering to group ecodistricts based on environmental variables such as soil texture and land cover that resulted in 19 and 14 clusters in the prairies of Alberta and Manitoba, respectively. The cluster analysis assisted in quantifying the distribution of tree cover within each ecodistrict. Our results indicated that there are 28,498 km2 and 13,265 km2 of tree cover in the prairies of Alberta and Manitoba, respectively. The maximum tree cover was found in Cluster 5 of the prairies of Alberta (32% tree coverage) and Cluster 13 of the prairies of Manitoba (82.5% tree coverage). Cluster 5 of the prairies of Alberta is scattered but concentrated primarily in the north and northeast, while Cluster 13 of the prairies of Manitoba is concentrated primarily in the southeastern part. This study provides the first systematic evaluation of non-forest tree cover and demonstrates the effectiveness of using RF for accurate tree cover mapping and K-means clustering for ecodistrict stratification that allows scalable quantification of prairie landscapes. [ABSTRACT FROM AUTHOR]
Copyright of International Journal of Remote Sensing is the property of Taylor & Francis Ltd and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database: Engineering Source
Description
Abstract:The study aimed to map and measure non-forest tree cover across the prairie regions of Alberta and Manitoba. These non-forest trees are often overlooked in literature, even though they provide important ecosystem services, such as soil protection and carbon storage. We used Sentinel-2 (S2) and Sentinel-1 (S1) data as predictors and very high-resolution (VHR) Google images as reference data. Random Forest (RF) classification, in the prairies of Alberta and Manitoba, produced accurate tree cover maps with overall accuracy and kappa of the best-performing model (S2 imagery for September–October 2023–2024) of 0.98 and 0.97, respectively. For ecological stratification and quantification, we applied K-means clustering to group ecodistricts based on environmental variables such as soil texture and land cover that resulted in 19 and 14 clusters in the prairies of Alberta and Manitoba, respectively. The cluster analysis assisted in quantifying the distribution of tree cover within each ecodistrict. Our results indicated that there are 28,498 km2 and 13,265 km2 of tree cover in the prairies of Alberta and Manitoba, respectively. The maximum tree cover was found in Cluster 5 of the prairies of Alberta (32% tree coverage) and Cluster 13 of the prairies of Manitoba (82.5% tree coverage). Cluster 5 of the prairies of Alberta is scattered but concentrated primarily in the north and northeast, while Cluster 13 of the prairies of Manitoba is concentrated primarily in the southeastern part. This study provides the first systematic evaluation of non-forest tree cover and demonstrates the effectiveness of using RF for accurate tree cover mapping and K-means clustering for ecodistrict stratification that allows scalable quantification of prairie landscapes. [ABSTRACT FROM AUTHOR]
ISSN:01431161
DOI:10.1080/01431161.2026.2618106