Bubbles trapped in arctic lake ice : Potential implications for methane emissions
The amount of methane (CH(4)) emitted from northern lakes to the atmosphere is uncertain but is expected to increase as a result of arctic warming. A majority of CH4 is thought to be released through ebullition (bubbling), a pathway with extreme spatial variability that limits the accuracy of measur...
| Published in: | Journal of Geophysical Research |
|---|---|
| Main Authors: | , , , , |
| Format: | Article in Journal/Newspaper |
| Language: | English |
| Published: |
Institutionen för geologiska vetenskaper
2011
|
| Subjects: | |
| Online Access: | http://urn.kb.se/resolve?urn=urn:nbn:se:su:diva-66529 https://doi.org/10.1029/2011JG001761 |
| Summary: | The amount of methane (CH(4)) emitted from northern lakes to the atmosphere is uncertain but is expected to increase as a result of arctic warming. A majority of CH4 is thought to be released through ebullition (bubbling), a pathway with extreme spatial variability that limits the accuracy of measurements. We assessed ebullition during early and late winter by quantifying bubbles trapped in the ice cover of two lakes in a landscape with degrading permafrost in arctic Sweden using random transect sampling and a digital image processing technique. Bubbles covered up to similar to 8% of the lake area and were largely dominated by point source emissions with spatial variabilities of up to 1056%. Bubble occurrence differed significantly between early and late season ice, between the two lakes and among different zones within each lake (p < 0.001). Using a common method, we calculated winter fluxes of up to 129 +/- 486 mg CH(4) m(-2) d(-1). These calculations are, on average, two times higher than estimates from North Siberian and Alaskan lakes and four times higher than emissions measured from the same lakes during summer. Therefore, the calculations are likely overestimates and point to the likelihood that estimating CH(4) fluxes from ice bubble distributions may be more difficult than believed. This study also shows that bubbles quantified using few transects will most likely be unsuitable in making large-scale flux estimates. At least 19 transects covering similar to 1% of the lake area were required to examine ebullition with high precision in our studied lakes. authorCount :5 |
|---|