Arctic Ecological Classifications Derived from Vegetation Community and Satellite Spectral Data
Saved in:
| Title: | Arctic Ecological Classifications Derived from Vegetation Community and Satellite Spectral Data |
|---|---|
| Authors: | Atkinson, David M. (Author), Treitz, Paul (Author) |
| Source: | Remote Sensing, Vol 4, Iss 12, Pp 3948-3971 (2012) Remote Sensing; Volume 4; Issue 12; Pages: 3948-3971 |
| Publisher Information: | MDPI AG, 2012. |
| Publication Year: | 2012 |
| Subject Terms: | IKONOS, vegetation mapping, Science, 0211 other engineering and technologies, Vegetation -- Remote sensing, arctic, tundra vegetation, correspondence analysis, cluster analysis, remote sensing, 02 engineering and technology, 15. Life on land, 01 natural sciences, Vegetation mapping -- Arctic regions, 13. Climate action, Vegetation classification -- Arctic regions, 0105 earth and related environmental sciences |
| Description: | As a result of the warming observed at high latitudes, there is significant potential for the balance of ecosystem processes to change, i.e., the balance between carbon sequestration and respiration may be altered, giving rise to the release of soil carbon through elevated ecosystem respiration. Gross ecosystem productivity and ecosystem respiration vary in relation to the pattern of vegetation community type and associated biophysical traits (e.g., percent cover, biomass, chlorophyll concentration, etc.). In an arctic environment where vegetation is highly variable across the landscape, the use of high spatial resolution imagery can assist in discerning complex patterns of vegetation and biophysical variables. The research presented here examines the relationship between ecological and spectral variables in order to generate an ecologically meaningful vegetation classification from high spatial resolution remote sensing data. Our methodology integrates ordination and image classifications techniques for two non-overlapping Arctic sites across a 5° latitudinal gradient (approximately 70° to 75°N). Ordination techniques were applied to determine the arrangement of sample sites, in relation to environmental variables, followed by cluster analysis to create ecological classes. The derived classes were then used to classify high spatial resolution IKONOS multispectral data. The results demonstrate moderate levels of success. Classifications had overall accuracies between 69%–79% and Kappa values of 0.54–0.69. Vegetation classes were generally distinct at each site with the exception of sedge wetlands. Based on the results presented here, the combination of ecological and remote sensing techniques can produce classifications that have ecological meaning and are spectrally separable in an arctic environment. These classification schemes are critical for modeling ecosystem processes. |
| Document Type: | Article Other literature type |
| File Description: | application/pdf |
| Language: | English |
| ISSN: | 2072-4292 |
| DOI: | 10.3390/rs4123948 |
| DOI: | 10.32920/14640015.v1 |
| DOI: | 10.32920/14640015 |
| Access URL: | http://www.mdpi.com/2072-4292/4/12/3948/pdf https://doaj.org/article/da12fb9b87c54b878b2052f5e2309076 https://www.mdpi.com/2072-4292/4/12/3948 https://doi.org/10.3390/rs4123948 http://dblp.uni-trier.de/db/journals/remotesensing/remotesensing4.html#AtkinsonT12 https://dblp.uni-trier.de/db/journals/remotesensing/remotesensing4.html#AtkinsonT12 http://ui.adsabs.harvard.edu/abs/2012RemS....4.3948A/abstract |
| Rights: | CC BY |
| Accession Number: | edsair.doi.dedup.....9ed82ec14d3da9f9e55bfdadd530c245 |
| Database: | OpenAIRE |
Full Text Finder 
Nájsť tento článok vo Web of Science | Abstract: | As a result of the warming observed at high latitudes, there is significant potential for the balance of ecosystem processes to change, i.e., the balance between carbon sequestration and respiration may be altered, giving rise to the release of soil carbon through elevated ecosystem respiration. Gross ecosystem productivity and ecosystem respiration vary in relation to the pattern of vegetation community type and associated biophysical traits (e.g., percent cover, biomass, chlorophyll concentration, etc.). In an arctic environment where vegetation is highly variable across the landscape, the use of high spatial resolution imagery can assist in discerning complex patterns of vegetation and biophysical variables. The research presented here examines the relationship between ecological and spectral variables in order to generate an ecologically meaningful vegetation classification from high spatial resolution remote sensing data. Our methodology integrates ordination and image classifications techniques for two non-overlapping Arctic sites across a 5° latitudinal gradient (approximately 70° to 75°N). Ordination techniques were applied to determine the arrangement of sample sites, in relation to environmental variables, followed by cluster analysis to create ecological classes. The derived classes were then used to classify high spatial resolution IKONOS multispectral data. The results demonstrate moderate levels of success. Classifications had overall accuracies between 69%–79% and Kappa values of 0.54–0.69. Vegetation classes were generally distinct at each site with the exception of sedge wetlands. Based on the results presented here, the combination of ecological and remote sensing techniques can produce classifications that have ecological meaning and are spectrally separable in an arctic environment. These classification schemes are critical for modeling ecosystem processes. |
|---|---|
| ISSN: | 20724292 |
| DOI: | 10.3390/rs4123948 |