Unlocking the time capsule of historic aerial photography to measure changes in Antarctic Peninsula glaciers
Recent studies have reported widespread retreat and acceleration of glaciers on the Antarctic Peninsula, attributed to regional warming. The loss of ice is a contributor to sea-level rise, but its volume and impact on sea level is poorly known. There are few ground measurements of ice thickness chan...
Published in: | The Photogrammetric Record |
---|---|
Main Authors: | , |
Format: | Article in Journal/Newspaper |
Language: | English |
Published: |
Blackwell
2008
|
Subjects: | |
Online Access: | http://nora.nerc.ac.uk/id/eprint/2412/ https://nora.nerc.ac.uk/id/eprint/2412/1/Fox_revised.pdf https://doi.org/10.1111/j.1477-9730.2008.00463.x |
Summary: | Recent studies have reported widespread retreat and acceleration of glaciers on the Antarctic Peninsula, attributed to regional warming. The loss of ice is a contributor to sea-level rise, but its volume and impact on sea level is poorly known. There are few ground measurements of ice thickness change and existing satellite altimeters are ineffective over the mountainous terrain. An accurate assessment of changes in surface height, and hence ice volume, of glaciers on the Antarctic Peninsula over past decades is needed to aid better estimates of their past impact on sea-level rise and predictions of their future contribution. There is an archive of over 30,000 aerial photographs going back to the 1940s for parts of the Antarctic Peninsula and photogrammetry of time-series of these historic photographs is now the only way to reconstruct changes in glacier surface height over the last fifty years. However, the historic aerial photographs are difficult to use for detailed measurements due to inadequate ground control, unfavourable sortie characteristics, incomplete calibration data and use of paper prints. This paper describes a method to provide control for historic photos without ground fieldwork by linking them to a newly-acquired, highly-accurate photogrammetric model adjusted through direct kinematic GPS positioning of the camera. It assesses the achievable accuracy through a worked example using a glacier on the Antarctic Peninsula with typical aerial photography at five dates from 1947 to 2005. Overall measurement accuracy of better than 2 m RMSE in X, Y and Z was achieved for all the photography types, which is precise enough to allow reliable measurement of changes in ice thickness for the glacier over decadal periods. The principal constraints are image quality of the historic photographs and using paper prints. |
---|