DETECTING PATTERNS AND DRIVERS OF ICE ON AND ICE OFF TIMING IN ALASKAN RIVERS WIDER THAN 150 m USING MODIS
Annual river ice freeze-up and breakup have major implications for northern ecosystems and infrastructure and are particularly responsive to climate change. However, a lack of ground-based observations hampers understanding of large-scale patterns in ice timing. Here I detect freeze-up and breakup d...
| Main Author: | |
|---|---|
| Format: | Text |
| Language: | English |
| Published: |
The University of North Carolina at Chapel Hill University Libraries
2019
|
| Subjects: | |
| Online Access: | https://dx.doi.org/10.17615/zmq1-ag93 https://cdr.lib.unc.edu/concern/dissertations/3r0750330 |
| Summary: | Annual river ice freeze-up and breakup have major implications for northern ecosystems and infrastructure and are particularly responsive to climate change. However, a lack of ground-based observations hampers understanding of large-scale patterns in ice timing. Here I detect freeze-up and breakup dates on Alaskan rivers wider than 150 m using MODIS satellite imagery from 2000-2017, the first large-scale detection of ice freeze-up using remote sensing and an expansion of breakup detection to rivers narrower than 500 m. I find statistically significant trends in breakup dates in the North Slope (-0.67 days/year, p<0.05) and West Central regions (-0.63 days/year, p<0.10). I find no long-term regional freeze-up trends. Regional timeseries of ice timing are instead dominated by teleconnections. Pacific Decadal Oscillation and Southern Oscillation Indices in the preceding fall and concurrent spring correlate highly to breakup dates, suggesting regional predictability. Methods described here can detect freeze-up and breakup timing on pan-Arctic rivers. |
|---|