Improved <i>k</i>-NN Mapping of Forest Attributes in Northern Canada Using Spaceborne L-Band SAR, Multispectral and LiDAR Data
Satellite forest inventories are the only feasible way to map Canada’s vast, remote forest regions, such as those in the Northwest Territories (NWT). A method used to create such inventories is the <i>k</i>-nearest neighbour (<i>k</i>-NN) algorithm, which spatially extends in...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-02-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/14/5/1181 |
| _version_ | 1797473905302044672 |
|---|---|
| author | André Beaudoin Ronald J. Hall Guillermo Castilla Michelle Filiatrault Philippe Villemaire Rob Skakun Luc Guindon |
| author_facet | André Beaudoin Ronald J. Hall Guillermo Castilla Michelle Filiatrault Philippe Villemaire Rob Skakun Luc Guindon |
| author_sort | André Beaudoin |
| collection | DOAJ |
| description | Satellite forest inventories are the only feasible way to map Canada’s vast, remote forest regions, such as those in the Northwest Territories (NWT). A method used to create such inventories is the <i>k</i>-nearest neighbour (<i>k</i>-NN) algorithm, which spatially extends information from forest inventory (FI) plots to the entire forest land base using wall-to-wall features typically derived from Landsat data. However, the benefits of integrating L-band synthetic aperture radar (SAR) data, strongly correlated to forest biomass, have not been assessed for Canadian northern boreal forests. Here we describe an optimized multivariate <i>k</i>-NN implementation of a 151,700 km<sup>2</sup> area in southern NWT that included ca. 2007 Landsat and dual-polarized Phased Array type L-band SAR (PALSAR) data on board the Advanced Land Observing Satellite (ALOS). Five forest attributes were mapped at 30 m cells: stand height, crown closure, stand/total volume and aboveground biomass (AGB). We assessed accuracy gains compared to Landsat-based maps. To circumvent the scarcity of FI plots, we used 3600 footprints from the Geoscience Laser Altimeter System (GLAS) as surrogate FI plots, where forest attributes were estimated using Light Detection and Ranging (LiDAR) metrics as predictors. After optimization, <i>k</i>-NN predicted forest attribute values for each pixel as the average of the 4 nearest (<i>k</i> = 4) surrogate FI plots within the Euclidian space of 9 best features (selected among 6 PALSAR, 10 Landsat, and 6 environmental features). Accuracy comparisons were based on 31 National Forest Inventory ground plots and over 1 million airborne LiDAR plots. Maps that included PALSAR HV backscatter resulted in forest attribute predictions with higher goodness of fit (adj. R<sup>2</sup>), lower percent mean error (ME%), and percent root mean square error (RMSE%), and lower underestimation for larger attribute values. Predictions were most accurate for conifer stand height (RMSE% = 32.1%, adj. R<sup>2</sup> = 0.58) and AGB (RMSE% = 47.8%, adj. R<sup>2</sup> = 0.74), which is much more abundant in the area than mixedwood or broadleaf. Our study demonstrates that optimizing <i>k</i>-NN parameters and feature space, including PALSAR, Landsat, and environmental variables, is a viable approach for inventory mapping of the northern boreal forest regions of Canada. |
| first_indexed | 2024-03-09T20:23:33Z |
| format | Article |
| id | doaj.art-ade140ff62bc4accb9c2e833ee9ba1ad |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-09T20:23:33Z |
| publishDate | 2022-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-ade140ff62bc4accb9c2e833ee9ba1ad2023-11-23T23:42:42ZengMDPI AGRemote Sensing2072-42922022-02-01145118110.3390/rs14051181Improved <i>k</i>-NN Mapping of Forest Attributes in Northern Canada Using Spaceborne L-Band SAR, Multispectral and LiDAR DataAndré Beaudoin0Ronald J. Hall1Guillermo Castilla2Michelle Filiatrault3Philippe Villemaire4Rob Skakun5Luc Guindon6Laurentian Forestry Centre, Canadian Forest Service, Natural Resources Canada, 1055 du P.E.P.S., P.O. Box 10380, Station Sainte-Foy, Québec City, QC G1V 4C7, CanadaNorthern Forestry Centre, Canadian Forest Service, Natural Resources Canada, 5320-122 Street NW, Edmonton, AB T6H 3S5, CanadaNorthern Forestry Centre, Canadian Forest Service, Natural Resources Canada, 5320-122 Street NW, Edmonton, AB T6H 3S5, CanadaNorthern Forestry Centre, Canadian Forest Service, Natural Resources Canada, 5320-122 Street NW, Edmonton, AB T6H 3S5, CanadaLaurentian Forestry Centre, Canadian Forest Service, Natural Resources Canada, 1055 du P.E.P.S., P.O. Box 10380, Station Sainte-Foy, Québec City, QC G1V 4C7, CanadaNorthern Forestry Centre, Canadian Forest Service, Natural Resources Canada, 5320-122 Street NW, Edmonton, AB T6H 3S5, CanadaLaurentian Forestry Centre, Canadian Forest Service, Natural Resources Canada, 1055 du P.E.P.S., P.O. Box 10380, Station Sainte-Foy, Québec City, QC G1V 4C7, CanadaSatellite forest inventories are the only feasible way to map Canada’s vast, remote forest regions, such as those in the Northwest Territories (NWT). A method used to create such inventories is the <i>k</i>-nearest neighbour (<i>k</i>-NN) algorithm, which spatially extends information from forest inventory (FI) plots to the entire forest land base using wall-to-wall features typically derived from Landsat data. However, the benefits of integrating L-band synthetic aperture radar (SAR) data, strongly correlated to forest biomass, have not been assessed for Canadian northern boreal forests. Here we describe an optimized multivariate <i>k</i>-NN implementation of a 151,700 km<sup>2</sup> area in southern NWT that included ca. 2007 Landsat and dual-polarized Phased Array type L-band SAR (PALSAR) data on board the Advanced Land Observing Satellite (ALOS). Five forest attributes were mapped at 30 m cells: stand height, crown closure, stand/total volume and aboveground biomass (AGB). We assessed accuracy gains compared to Landsat-based maps. To circumvent the scarcity of FI plots, we used 3600 footprints from the Geoscience Laser Altimeter System (GLAS) as surrogate FI plots, where forest attributes were estimated using Light Detection and Ranging (LiDAR) metrics as predictors. After optimization, <i>k</i>-NN predicted forest attribute values for each pixel as the average of the 4 nearest (<i>k</i> = 4) surrogate FI plots within the Euclidian space of 9 best features (selected among 6 PALSAR, 10 Landsat, and 6 environmental features). Accuracy comparisons were based on 31 National Forest Inventory ground plots and over 1 million airborne LiDAR plots. Maps that included PALSAR HV backscatter resulted in forest attribute predictions with higher goodness of fit (adj. R<sup>2</sup>), lower percent mean error (ME%), and percent root mean square error (RMSE%), and lower underestimation for larger attribute values. Predictions were most accurate for conifer stand height (RMSE% = 32.1%, adj. R<sup>2</sup> = 0.58) and AGB (RMSE% = 47.8%, adj. R<sup>2</sup> = 0.74), which is much more abundant in the area than mixedwood or broadleaf. Our study demonstrates that optimizing <i>k</i>-NN parameters and feature space, including PALSAR, Landsat, and environmental variables, is a viable approach for inventory mapping of the northern boreal forest regions of Canada.https://www.mdpi.com/2072-4292/14/5/1181forest vegetation inventoryPALSARLandsatLiDARGLAS<i>k</i>-NN |
| spellingShingle | André Beaudoin Ronald J. Hall Guillermo Castilla Michelle Filiatrault Philippe Villemaire Rob Skakun Luc Guindon Improved <i>k</i>-NN Mapping of Forest Attributes in Northern Canada Using Spaceborne L-Band SAR, Multispectral and LiDAR Data Remote Sensing forest vegetation inventory PALSAR Landsat LiDAR GLAS <i>k</i>-NN |
| title | Improved <i>k</i>-NN Mapping of Forest Attributes in Northern Canada Using Spaceborne L-Band SAR, Multispectral and LiDAR Data |
| title_full | Improved <i>k</i>-NN Mapping of Forest Attributes in Northern Canada Using Spaceborne L-Band SAR, Multispectral and LiDAR Data |
| title_fullStr | Improved <i>k</i>-NN Mapping of Forest Attributes in Northern Canada Using Spaceborne L-Band SAR, Multispectral and LiDAR Data |
| title_full_unstemmed | Improved <i>k</i>-NN Mapping of Forest Attributes in Northern Canada Using Spaceborne L-Band SAR, Multispectral and LiDAR Data |
| title_short | Improved <i>k</i>-NN Mapping of Forest Attributes in Northern Canada Using Spaceborne L-Band SAR, Multispectral and LiDAR Data |
| title_sort | improved i k i nn mapping of forest attributes in northern canada using spaceborne l band sar multispectral and lidar data |
| topic | forest vegetation inventory PALSAR Landsat LiDAR GLAS <i>k</i>-NN |
| url | https://www.mdpi.com/2072-4292/14/5/1181 |
| work_keys_str_mv | AT andrebeaudoin improvedikinnmappingofforestattributesinnortherncanadausingspacebornelbandsarmultispectralandlidardata AT ronaldjhall improvedikinnmappingofforestattributesinnortherncanadausingspacebornelbandsarmultispectralandlidardata AT guillermocastilla improvedikinnmappingofforestattributesinnortherncanadausingspacebornelbandsarmultispectralandlidardata AT michellefiliatrault improvedikinnmappingofforestattributesinnortherncanadausingspacebornelbandsarmultispectralandlidardata AT philippevillemaire improvedikinnmappingofforestattributesinnortherncanadausingspacebornelbandsarmultispectralandlidardata AT robskakun improvedikinnmappingofforestattributesinnortherncanadausingspacebornelbandsarmultispectralandlidardata AT lucguindon improvedikinnmappingofforestattributesinnortherncanadausingspacebornelbandsarmultispectralandlidardata |