University of Leicester
ggy140.pdf (12.85 MB)

Crustal properties of the northern Scandinavian mountains and Fennoscandian shield from analysis of teleseismic receiver functions

Download (12.85 MB)
journal contribution
posted on 2019-07-24, 14:16 authored by Walid Ben Mansour, Richard W. England, Stewart Fishwick, Max Moorkamp
The presence of high mountains along passive margins is not unusual, as shown by their presence in several regions (Scandinavia, Greenland, East US, SW Africa, Brazil, West India and SE Australia). However, the origin of this topography is notwell understood. Themountain range between the Scandinavian passive margin and the Fennoscandian shield is a good example. A simple Airy isostatic model would predict a compensating root beneath the mountains but existing seismic measurements of variations in crustal thickness do not provide evidence of a root of sufficient size to produce the necessary compensation. In order to better constrain the physical properties of the crust in northern Scandinavia, two broad-band seismic networks were deployed between 2007 and 2009 and between 2013 and 2014. A new map of crustal thickness has been produced from P-receiver function analysis of teleseismic data recorded at 31 seismic stations. The map shows an increase in crustal thickness from the Atlantic coast (38.7 ± 1.8 km) to the Gulf of Bothnia (43.5 ± 2.4 km). This gradient in thickness demonstrates that the Moho topography does not mirror the variation in surface topography in this region. Thus, classical Airy isostatic models cannot explain how the surface topography is supported. New maps showing variation in Poisson's ratio and Moho sharpness together with forward and inverse modelling provide new information about the contrasting properties of the Fennoscandian shield and crust reworked by the Caledonian orogeny. A sharp Moho transition (R > 1) and low value of Vs(3.5 ± 0.2 km s-1) are observed beneath the orogen. The shield is characterized by a gradual transition across the Moho (R < 1) and Vsof 3.8 ± 0.1 km s-1which is more typical of average continental crust. These observations are explained by a Fennoscandian shield underplated with a thick layer of high velocity, high density material. It is proposed that this layer has been removed or reworked beneath the orogen.


We gratefully acknowledge support from the NERC geophysical equipment facility; loans 833 and 959, and the assistance of Jomar Gellein and Hasse Palm during the fieldwork. Financial support was provided by the Geological Survey of Norway. Parts of the data processing were undertaken using the SAC software (Goldstein & Snoke 2005) and figures were produced using GMT software (Wessel et al.2013).



Geophysical Journal International, 2018, 214 (1), pp. 386-401

Author affiliation

/Organisation/COLLEGE OF SCIENCE AND ENGINEERING/School of Geography, Geology and the Environment


  • VoR (Version of Record)

Published in

Geophysical Journal International


Oxford University Press (OUP), Royal Astronomical Society





Acceptance date


Copyright date


Available date


Publisher version


Supplementary data are available at GJ I online. Figure S1. (a) Teleseismic events (Mw > 5.8) recorded by SCANLIPS2 between July 2007 and September 2009 for P-receiver function study. (b) Teleseismic events (Mw > 5.8) recorded by SCANLIPS3D between July 2013 and September 2014 for P-receiver function study. In both cases the blue star corresponds to the centre of the seismic array. Figure S2. (a) Synthetic P-receiver functions for a Moho depth of 43 km and Vp/Vs of 1.73 from three crustal models. In red, a model with two layers (upper and lower crust) for the crust and step discontinues. In green, a model with three layers (upper, lower crust and transitional Moho with highVp) and step discontinuities. In blue, a gradual model between upper crust and Moho. (b) Sensibility of amplitude of direct P arrival and Ps conversion for different slowness and for two crustal model (on the left a step Moho model and on the right a gradual Moho model). Table S1. Informations about seismic stations and number of events used in this study for each instruments (70XX: SCANLIPS2 -13XX: SCANLIPS3D). Table S2. Compilation of P-RFs analysis and previous works (Ottemoller & Midzi ¨ 2003; Olsson et al. 2008; Silvennoinen et al. 2014).



Usage metrics

    University of Leicester Publications


    No categories selected


    Ref. manager