What about the asthenosphere viscosity? Comment on ‘Sea-level change, glacial rebound and mantle viscosity for northern Europe’ by K. Lambeck, C. Smither and P. Johnston

Abstract

Lambeck et al. (1998a; hereafter LSJ) present a thorough In my own studies (Fjeldskaar 1994, 1997) I have shown that the best-fitting models have a low-viscosity asthenosphere. study of the sea-level change connected to the deglaciation of Fennoscandia. They establish a comprehensive observational Based on Fjeldskaar (1994), the thickness and viscosity of the asthenosphere are not precisely determined, but are within database on crustal rebound for the whole of northern Europe. This database is used in a high-resolution modelling of the the range shown in Fig. 1. It is clearly demonstrated that the asthenosphere has a thickness of less than 150 km; a thicker rebound, based on the observed deglaciation pattern. A key result is an estimate of the effective elastic thickness of the asthenosphere shows significant discrepancies from the observed uplift data. lithosphere and the viscosity structure of the mantle. LSJ give a brief overview of previous results on the mantle Models similar to the one favoured by LSJ were tested in Fjeldskaar (1994, model no. 14). It was pointed out that models rheology models established by a similar approach. LSJ’s comment to models with a low-viscosity asthenosphere, and with a lower mantle viscosity of 2.0×1021 Pa s or higher, without a low-viscosity asthenosphere, are not viable options. The various other previously published viscosity models, is that the ‘differences are attributable to a number of factors’. These reason is that the short-wavelength components decay too slowly, giving significant deviations in central parts of the uplift dome. factors include the fact that the other studies used only a subset of the data, and that some of them assumed the ice sheet to be This is clearly demonstrated in the theoretical present uplift pattern, calculated with the rheology parameters of LSJ (given perfectly known. A third factor is that none of the studies ‘has examined the entire range of possible earth-model parameters’. in Lambeck et al. 1998b). The predicted present-day uplift centre lies further to the west than the observed centre. This would This immediately gives the reader the impression that LSJ do not have similar shortcomings, and that the final resulting not be the case if a low-viscosity asthenosphere was used (Fjeldskaar 1994, Fig. 6). Furthermore, the rheology model of earth rheology model is better constrained than the previous studies. However, this is not the case. LSJ have ignored models LSJ cannot produce the observed shoreline tilts (Fjeldskaar 1994). This is also demonstrated in LSJ, Figs 28 and 29. with a low-viscosity asthenosphere. On this basis it is rather peculiar that they (1) criticize others for not having used the In conclusion, LSJ have not been able to come up with a model for the earth rheology that gives optimum fit to entire range of possible earth models and (2) give the reader the impression that models with a low-viscosity asthenosphere the observed rebound because models with a low-viscosity asthenosphere have not been considered. The LSJ rheology is are unrealistic.