Comment on ’A spectral analysis of geoid undulation and gravity-anomaly data computed with Pratt's isostasy theory applied to Moho depth variations in Fennoscandia’, by T. Nord and L. E. Sjöberg

In Fennoscandia the lithospheric structure is unusual compared to Phanerozoic Europe. All available data show that even though the topography is low, the crustal depth is in some places well below 50km and the lithospheric thickness has been estimated to be about 200 km. This structure leads to a density distribution that can be expected to have an impact on the potential field anomalies. The key parameter, however, is the compensating mass balance inside the lithosphere. Seismic profiling generally indicates high lower crustal density, but does not reveal high-density mass distributions inside the crust or in the uppermost part of the lithosphere needed to locally balance the Moho depth undulations. It can be expected that small-balancing density contrasts are distributed to rather great depth. From a geophysical point of view, taking a compensation depth of 30 or 50km (inside the crust) to be responsible for all compensation is not realistic. In general, for a better understanding of the compensating mass distribution, one has to take into account the topography and the internal structure of the crust and the lithosphere. The important raw data set for the study by Nord & Sjoberg is a Moho map for Fennoscandia. The problem here is that any Moho map is based on a very limited number of seismic reflection or refraction profiles. The exact Moho depth for each profile can only be estimated with an accuracy of about 2 to 5 km in limited parts along the profile. Thus it is easy to imagine that any Moho depth map can only give a rough estimate on the real variations of the crust-mantle boundary and is critically dependent on the imagination of the person who draws the isolines or on the algorithm of the applied computer program. In Fennoscandia in particular, areas exist where no or very few data are available. This is true for most of the continent-ocean transition for SW Norway, for the southern Baltic Sea, and for the entire SE part of the area of investigation. Nord & Sjoberg used a Moho depth map from Luosto (1990), which differs in some way from maps published