The link between glacier velocity and the drainage of ice‐dammed lakes: Comment on a paper by knight and tweed

Abstract The interesting suggestion by Knight and Tweed (1991) that a link exists between glacier velocity changes and the filling and draining of a marginal ice‐dammed lake at Sólheimajökull, southern Iceland, is examined. The main problem lies not with plausibility, but with a lack of field data t...

Full description

Bibliographic Details
Published in:Hydrological Processes
Main Author: Lawler, D. M.
Format: Article in Journal/Newspaper
Language:English
Published: Wiley 1994
Subjects:
Online Access:http://dx.doi.org/10.1002/hyp.3360080507
https://api.wiley.com/onlinelibrary/tdm/v1/articles/10.1002%2Fhyp.3360080507
https://onlinelibrary.wiley.com/doi/pdf/10.1002/hyp.3360080507
Description
Summary:Abstract The interesting suggestion by Knight and Tweed (1991) that a link exists between glacier velocity changes and the filling and draining of a marginal ice‐dammed lake at Sólheimajökull, southern Iceland, is examined. The main problem lies not with plausibility, but with a lack of field data to substantiate the argument. Some pertinent hydrological, glaciological and meteorological data from the area are presented here, many of which do not entirely support certain inferences and observations made in their paper. It is shown that this very small lake is likely to have limited quantitative significance in the operation of the glacio‐hydrological or solute and sediment transfer systems in this basin. Further, specific, difficulties include the absence of crucial observations on glacier velocity variations and the precise timing and pattern of lake drainage in relation to such velocity changes, the lack of evidence for a significant relationship between precipitation receipt and ice velocity here [a component of the Knight and Tweed (1991) proposition], and the occurrence of the lake drainage event in a rain‐free period (contrary to the authors' suggestion). The opportunity is also taken to correct some incidental errors.