Wind stress forcing of the North Sea 'pole tide'

We conduct numerical simulations of the wind forcing of sea level variations in the North Sea using a barotropic ocean model with realistic geography and bathymetry to examine the forcing of the 14 month 'pole tide', which is known to be anomalously large along the Denmark–Netherlands coast. The simulation input is the monthly mean surface wind stress field from the National Centers for Environmental Prediction (NCEP) reanalysis for the 40 year period 1958–1997. The ocean model output sea level response is then compared with 10 coastal tide gauge records from the Permanent Service for Mean Sea Level (PSMSL) over the same period of time. Besides the strong seasonal variations, several prominent quasi-periodicities exist near 7 years, 3 years, 14 months, 9 months and 6.5 months. Correlations and spectral analyses show remarkable agreement between the model output and the observations, particularly in the 14 month, or Chandler, period band. The latter indicates that the enhanced pole tide found in the North Sea along the Denmark–Netherlands coast is actually the coastal set-up response to wind stress forcing with a periodicity of around 14 months. We find no need to invoke a geophysical explanation involving resonance enhancement of the pole tide in the North Sea to explain the observations.     

Joint traveltime inversion of wide-angle seismic data and a deep reflection profile from the central North Sea

We report results from the Seismic Wide-Angle and Broadband Survey carried out over the Mid North Sea High. This paper focuses on integrating the information from a conventional deep multichannel reflection profile and a coincident wide-angle profile obtained by recording the same shots on a set of ocean bottom hydrophones (OBH). To achieve this integration, a new traveltime inversion scheme was developed (reported elsewhere) that was used to invert traveltime information from both the wide-angle OBH records and the reflection profile simultaneously. Results from the inversion were evaluated by producing synthetic seismograms from the final inversion model and comparing them with the observed wide-angle data, and an excellent match was obtained. It was possible to fine-tune velocities in less well-resolved parts of the model by considering the critical distance for the Moho reflection. The seismic velocity model was checked for compatibility with the gravity field, and used to migrate and depth-convert the reflection profile. The unreflective upper crust is characterized by a high velocity gradient, whilst the highly reflective lower crust is associated with a low velocity gradient. At the base of the crust there are several subhorizontal reflectors, a few kilometres apart in depth, and correlatable laterally for several tens of kilometres. These reflectors are interpreted as representing a strike section through northward-dipping reflectors at the base of the crust, identified on orthogonal profiles by Freeman et al. (1988) as being slivers of subducted and imbricated oceanic crust, relics of the mid-Palaeozoic Iapetus Ocean.     

Rift kinematics preserved in deep-time erosional landscape below the northern North Sea

Our understanding of continental rifting is, in large parts, derived from the stratigraphic record. This record is, however, incomplete as it does not often capture the geomorphic and erosional signal of rifting. New 3D seismic reflection data reveal a Late Permian-Early Triassic landscape incised into the pre-rift basement of the northern North Sea. This landscape, which covers at least 542 km², preserves a drainage system bound by two major tectonic faults. A quantitative geomorphic analysis of the drainage system reveals 68 catchments, with channel steepness and knickpoint analysis of catchment-hosted palaeo-rivers showing that the landscape preserved a >2 Myr long period of transient tectonics. We interpret that this landscape records a punctuated uplift of the footwall of a major rift-related normal fault (Vette Fault) at the onset of rifting. The landscape was preserved by a combination of relatively rapid subsidence in the hangingwall of a younger fault (Øygarden Fault) and burial by post-incision sediments. As such, we show how and why erosional landscapes are preserved in the stratigraphic record, and how they can help us understand the tectono-stratigraphic evolution of ancient continental rifts.     

On the nature of regional seismic phases–II. On the influence of structural barriers

The blockage of the L _g wave by crustal barriers such as continental margins and graben structures has long been recognized as providing a very useful tool for mapping large-scale lateral crustal variations along the propagation path. Numerical simulation of L _g -wave propagation in complex anelastic media using the pseudospectral method provides insight into the nature of the propagation process using both snapshots of the wavefield and synthetic seismograms. A variety of 2-D structures have been investigated, including the influence of sediments, crustal thickness and attenuation.
Thick sedimentary basins covering a graben structure can have a major influence, since they remove L _g energy by generating P conversion and scattering–the principal mechanisms for strong L _g attenuation across a graben. The reduction of the L _g energy is reinforced by anelastic attenuation in the sediments as well as the influence of the gradually thinning crustal waveguide associated with an elevated Moho.
The extinction of L _g in a sequence of explosions fired across the central graben of the North Sea can be simulated by numerical calculations for the structure derived from refraction experiments.     

A Hybrid Collocation Method For Calculating Complete Theoretical Seismograms In Vertically Varying Media

A numerical method is presented for calculating complete theoretical seismograms, under the assumption that the earth models have velocity, density and attenuation profiles which are arbitrary piece-wise continuous functions of depth only. Solutions for the stress-displacement vectors in the medium are expanded in terms of orthogonal cylindrical functions. Our method for solving the resulting two-point boundary value problems differs from that of other investigators in three ways. First, collocation is used in traditionally troublesome situations, e.g. for highly evanescent waves, at turning points, and in regions having large gradient in material properties. Second, in some situations (high frequencies and small gradients) P and S -waves decouple and we use a different solution method for each wave type, instead of trying to force a single method to find all solutions. For example, above the P - and S -waves turning points an approximate fundamental matrix may be used for each wave type. At the P -wave turning point, the fundamental matrix may be used for the S -wave components but collocation is used for the P -wave. Between the P - and S -wave turning points collocation is used for the evanescent P -wave and the fundamental matrix is used for the S -wave. At the S -wave turning point and below, collocation is used for both. Third, the computational algorithm chooses the appropriate solution method and depth domain upon which it is employed based upon a specified error tolerance and the known inaccuracies of the various approximations employed. Once solutions of the boundary value problems are obtained, a Fourier—Bessel transform is then applied to get back into the space-time domain.