Observation bias correction reveals more rapidly draining lakes on the Greenland Ice Sheet
Rapid drainage of supraglacial lakes on the Greenland Ice Sheet enables the establishment of surface-to-bed hydrologic connections and subsequent basal water delivery. Estimates of the number and spatial distribution of rapidly draining lakes vary widely, and no study has quantified the impact of ob...
| Main Authors: | , |
|---|---|
| Format: | Article in Journal/Newspaper |
| Language: | English |
| Published: |
Wiley
2017
|
| Subjects: | |
| Online Access: | https://www.repository.cam.ac.uk/handle/1810/269630 https://doi.org/10.17863/CAM.15840 |
| Summary: | Rapid drainage of supraglacial lakes on the Greenland Ice Sheet enables the establishment of surface-to-bed hydrologic connections and subsequent basal water delivery. Estimates of the number and spatial distribution of rapidly draining lakes vary widely, and no study has quantified the impact of observation bias due to cloud cover in satellite imagery on reported frequency of rapid lake drainage. To better understand the rapid drainage mechanism, we map and track an average of 515 supraglacial lakes per year in central West Greenland from 2000-2015. We test four previously published definitions of rapid lake drainage and find the proportion of rapidly draining lakes to vary from 3% to 38% and to be strongly dependent on observation frequency. We then apply an observation bias correction and test three new drainage criteria, which reveal a bias corrected rapid drainage probability of 36-45%. When observation bias is addressed, we can also show that lakes above 1600m are as likely to drain rapidly as lakes located at lower elevations. We conclude that inconsistent detection methodologies and observation bias have obscured the true frequency of rapidly draining lakes, and that the rapid lake drainage mechanism will establish surface-to-bed hydrologic connections at increasing distance from the margin as supraglacial lakes expand inland under climate warming. S.W.C. acknowledges financial support from the Gates Cambridge Trust at the University of Cambridge. PC acknowledges funding from the European Research Council under the European Union's Horizon 2020 research and innovation programme (grant agreement no. 683043). |
|---|