Examining class boundaries variations of arctic vegetation in the past 17 years using MODIS satellite images
The concern about rising arctic temperature and its consequences has grown recently. There have been some researches using NDVI (normalized difference vegetation index) to assess the arctic vegetation’s respond to the global warming. This paper uses both NDVI and trained supervised classification to...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Master Thesis |
| Language: | English |
| Published: |
The University of Edinburgh
2018
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/1842/35476 |
| Summary: | The concern about rising arctic temperature and its consequences has grown recently. There have been some researches using NDVI (normalized difference vegetation index) to assess the arctic vegetation’s respond to the global warming. This paper uses both NDVI and trained supervised classification to provide a general understanding of the vegetation changes in productivity and species distributions in Arctic tundra and boreal forest ecosystems over the past 17 years (2000 - 2016) based on MODIS (Moderate Resolution Imaging Spectrometer) satellite data and CAVM (circumpolar arctic vegetation map). Some specific findings of arctic plants species in the ABoVE (Arctic-Boreal Vulnerability Experiment) area were reported. Some increasing NDVI was found in several regions in the circumpolar area but partly browning trends were also shown in the result maps. This study also reveals a contrast in the respond of different plant species to the climate change. Due to the classification uncertainties, it is difficult to summarise any change patterns but improvements have been suggested for future research. The emerging trends and possible reasons were discussed. This study demonstrates the use of supervised classification on medium resolution remote sensing data for studying the spatio-temporal dynamics of plants in the Arctic. |
|---|