Outline of project

Stratigraphic and magmatic characteristics of transient asthenospheric upwelling

Project outline

1. Background:

(Dixon, Fitton et al. 1981) documented the alkaline, rift-related magmatism of Jurassic age in the North Sea and discussed possible stretching and melt generation parameters.

(Latin, Dixon et al. 1990) showed that although the magmatism occurred where β-factors were largest, existing theoretical models of melt generation related to stretching (McKenzie and Bickle 1988) could not satisfactorily account for melt production at the low β-factors inferred (c.1.5 to 2 at most) - the “misfit problem” - nor was there an obvious explanation for the occurrence of magmatism before any discrete rift-bounding faults had developed.

Using sequence stratigraphic evidence, (Underhill and Partington 1993) showed that the principal episode of magmatic activity occurred on the crest of a transient broad dome at the point in time (Bathonian/Bajocian) when it reached its highest elevation relative to sea-level, shortly before it became the location of the Upper Jurassic rift triple junction and then subsided. The estimated uplift is of the order of 500 metres, though other workers suggest it may be more (Perrot, van der Poel et al. 1987). It occurred on a dome approximately 1,000 km in diameter defined by the limit of the erosion surface, marked by correlative conformities.

(Fitton, Foulger et al. 2007) make it clear that enriched basalts indistinguishable from OIB, that also have identical geochemical signatures to the Icelandic basalts with positive ∆Nb, occur too widely in the oceans, as well as in long-lived continental rift zones, and on too great a range of scales for all to be the product of conventional asthenospheric upwellings or plumes, whatever characteristics are attributed to such upwellings. An origin in the selective melting of fusible streaks in the mantle is a better explanation for many while the existing concept of a plume as an upwelling of mantle with higher potential temperature, accepted by most workers, can be retained for clear large-scale phenomena such as Hawaii, Iceland, Galapagos, the Cape Verdes and the Kenya dome.

Hawaii and the Cape Verdes are distant from rifts, demonstrating that melt can be generated without additional lithospheric thinning in such settings.

In Iceland (Fitton, Saunders et al. 2003) and Kenya (Latin and White 1993) is it is also apparent that melt generation is too great to be attributed purely to a combination of elevated potential temperature and lithospheric stretching: a component of upward asthenospheric flow must be involved.

It has also been shown in both Hawaii (Watson and McKenzie 1991) and more recently in the Cape Verdes (Pim, Peirce et al. 2008) that the long-wavelength swell in both must be in part due to dynamic support from this upward flow and not simply due to the presence of a low density cushion, whether hot or wet.

Points arising:

In the North Sea, therefore, there is an unanswered question as to which of the potential end-member models best accounts for the transient uplift in the Jurassic given that associated magmatism preceded faulting and the stretching maximum.

The mid-Jurassic North Sea Dome was clearly a transient phenomenon. However, the spatial and temporal dimensions would appear to be on a scale too large to be anything other than asthenosphere “anomaly” driven, for which the term “mega-drip” will be used for convenience.

The fusible streaks hypothesis for alkali basalt production when applied as a modification to the McKenzie and Bickle model removes the misfit problem by challenging the inherent assumption that the asthenosphere was homogeneous, dry, fertile garnet/spinel/plagioclase lherzolite, as used in the experiments on which the parameterisation was based. The other possible origin of the misfit was that the temperature of the onset of melting for dry peridotite had been incorrectly inferred by extrapolation from the 15% minimum observable in actual experiments prior to 1988. More recent work using more refined melt detection techniques suggests that the extrapolation was correct (Robinson, Wood et al. 1998) but if the source is different this becomes irrelevant. That volatile-, iron- and alkali-rich mantle will melt at a lower temperature at a given depth is well known.

The fusible streaks model can explain melt generation either by low β-factors (or simply asthenospheric upwelling into thinner regions of the lithosphere even without active stretching), or by elevated potential temperatures alone, or both in combination.

If extension is absent, having not yet occurred, then melt generation must be due to localised enrichment in fusible streaks, without any temperature excess, or a localised temperature excess operating on “normal” fusible streaks, or both, or an alternative localised compositional anomaly such as enrichment in water (a “wet spot”), with or without the fusible streak component. The critical additional factor to be explained is the uplift, with the further complication that it was transient. A model that is characterised only by enrichment in fusible streaks cannot account for the uplift, as the fusible streaks will have a higher density than normal mantle. The uplift must be due to lower mantle density, the result either of higher than ambient temperature or the influx of water. A dynamic component of uplift due to flow may have been present but it can only conceivably have been sustained by the more fundamental density anomaly. The implications of the transient character of the uplift are considered further below.

What are the possibilities in the North Sea and how might one distinguish between them?

Because the dome subsided with no residual excess elevation to counteract the extension-related subsidence in the thinned area and apparently spent no significant period at maximum elevation, there was essentially no steady state where a low density column existed. This suggests that a transient low density body was unconnected at its base, in contrast to the conventionally envisaged continual column, or plume, and must have dissipated by spreading sideways once it reached the base of the lithosphere.

The dome subsided too quickly for this static hot body (“mega-drip”) to have dissipated by conduction as the thermal time constant was too long. It is also difficult to see how there could have been a significant dynamic support component to the uplift, except in the final stage of upwelling when the displaced mantle attempted to escape, and then as the mega-drip shortened upward and spread outward. This simply clarifies the rationale for the mega-drip model to explain the uplift.

Possible models:

There are then two possible end-member explanations based on whether or not the absence of evidence of extension at the surface prior to magmatism also applies at depth. Model One is that magmatism is the result of elevated temperature in a mega-drip causing the solidus in a fusible streak to be intersected. The occurrence of magmatism before faulting and crustal stretching is therefore a true reflection of the total state of extension in the lithosphere, i.e. that it had not yet occurred. The magmatism is thus by inference asthenosphere-derived (and may carry identifiable isotope characteristics) there being no significant time for the mega-drip to transfer heat to the lithospheric mantle, which was not yet being stretched. As the mega-drip spreads and dissipates and stretching proceeds, localised by the uplift itself, there is no additional magmatism as the fusible streaks have already been melted out by the elevated temperatures and no fresh asthenosphere is being fed into the thinned zone as it develops. This last proviso assumes that the lithosphere is not moving significantly relative to the underlying asthenosphere over the time interval concerned and normal cooling and subsequent thickening occurs.

Model Two invokes depth- and time-distributed stretching to account for the magmatism by postulating that the stretching envelope initially widens upwards, giving very diffuse extension in the upper crust and sharply focussed thinning at the base of the lithosphere, before inverting to the opposite configuration. While there is evidence that extension does begin diffusely in the crust (Cowie, Underhill et al. 2005) before narrowing down onto master rift faults that step inwards as rifting proceeds, there is no clear evidence of a narrow base-lithosphere thin zone, which would be counter-intuitive in the ductile realm. This model nevertheless links the magmatism firmly to the extension even though it apparently pre-dates it.

2. Project

The discussion thus far leads directly on to research questions and programme of work for a PhD project that could help discriminate between these two models:

Research questions:

• Can the characteristics of transient asthenospheric upwelling be described and identified from the geological record?

• Do the hitherto separately documented characteristics of stratigraphic uplift and alkaline magmatism associated with Central North Sea doming in the Jurassic combine to identify transient asthenospheric upwelling?

• Can a sequence of similar occurrences be identified and linked spatially and temporally to support a dripping tap model to explain transient asthenospheric upwelling? i.e. are there other examples of transient uplifts that generate magmatism at the crest and at the time of maximum elevation, before any significant extensional faulting or even in the absence of such faulting, that are linked together, for example connecting the mid-Jurassic North Sea dome to other later examples in adjacent parts of NW Europe?

• Is melt generation associated with transient upwelling dependent on lithospheric stretching or can it be due entirely to a transient temperature anomaly initiating selective melting, perhaps of fusible streaks or blobs in the asthenospheric mantle?

Approach:

By updating and refining studies of uplift, extension and magmatism previously focussed on the Jurassic activity in the North Sea, these hypotheses might be tested and the work extended to other locations and time periods.

Localities:

(Dixon, Fitton et al. 1981) assembled the scattered data then available on magmatic activity in the wider North Sea area during the Mesozoic and pointed to an apparent southward migration of alkaline activity from the Middle Jurassic triple-junction centre through the Central Graben to the Early Cretaceous undersaturated volcanism at Zuidwal near the Dutch coast. Further afield are the alkaline centres of the Wolf Rock (Cretaceous) (Harrison, Snelling et al. 1977), the Aptian basanites of Skåne in South Sweden (Tappe 2004), and the extensive alkaline Tertiary magmatism in the Bohemian graben (Ulrych, Dostal et al. 2008).

North Sea:

It will be useful to re-visit, update and draw together the various published contributions on the actual uplift, subsidence and stretching history of the triple-junction area, as well as examine the possible behaviour of rising bodies, whether from numerical or analogue modelling, and to re-examine the question of whether the stratigraphic data support or contradict modelling results.

Zuidwal:

The two key properties of this location in the Dutch area of the southern North Sea are:

• The availability of seismic and well data and published regional geology (Ziegler 1990; Sissingh 2004; Wong, Batjes et al. 2007) that should allow the regional stratigraphy and palaeogeographic context up to, during and after the magmatic activity to be reconstructed.

• The possible continuation of discrete episodes of uplift and magmatism southward from the Jurassic focus (Forties) that could be a reflection of activity in a single deep-seated asthenospheric source - the dripping tap model of transient uplift generation. A consideration of probable absolute motion paths for the Eurasian plate at that time suggests that the southward track, if it exists, is in about the right direction. Two or more drips are clearly better than one in connecting uplift and magmatism to a deep source.

For the second point, the dating, both stratigraphic and isotopic, needs to be re-visited and assessed.

Models:

The question of the mantle source for the magmatism in any of these settings is clearly important. If the lithospheric mantle is clearly involved then a mechanism is needed, either thinning or heat transfer by conduction or by melt transport, in order to link the asthenospheric upwelling to the melt production. A significant result would be to find that magmatism associated with transient drip-type uplifts but not rifting had asthenospheric signatures.

3. References:

Cowie, P. A., J. R. Underhill, et al. (2005). "Spatio-temporal evolution of strain accumulation derived from multi-scale observations of Late Jurassic rifting in the northern North Sea: A critical test of models for lithospheric extension." Earth and Planetary Science Letters 234(3-4): 401-419.

Dixon, J. E., J. G. Fitton, et al. (1981). The tectonic significance of post-Carboniferous igneous activity in the North-Sea Basin. Petroleum geology of the continental shelf of North-West Europe; Proceedings of the second conference. L. V. Illing and G. D. Hobson, Publisher Heyden and Son, London, United Kingdom: 121-137.

Fitton, J. G., G. R. Foulger, et al., Eds. (2007). The OIB paradox, Geological Society of America (GSA), Boulder, CO.

Fitton, J. G., A. D. Saunders, et al. (2003). "Does depleted mantle form an intrinsic part of the Iceland plume?" Geochemistry Geophysics Geosystems 4.

Harrison, R. K., N. J. Snelling, et al. (1977). "The Wolf Rock, Cornwall: new chemical, isotopic age and palaeomagnetic data." Geological Magazine 114(04): 249-264.

Latin, D. and N. White (1993). "Magmatism in extensional sedimentary basins." Workshop on Modes of deformation; from the brittle upper crust through detachments to the ductile lower crust Annali di Geofisica 36(2): 123-138.

Latin, D. M., J. E. Dixon, et al., Eds. (1990). Mesozoic magmatic activity in the North Sea basin; implications for stretching history, Geological Society of London, London, United Kingdom.

McKenzie, D. and M. J. Bickle (1988). "The Volume and Composition of Melt Generated by Extension of the Lithosphere." Journal of Petrology 29(3): 625-679.

Perrot, J., A. van der Poel, et al., Eds. (1987). Zuidwal; a Neocomian gas field, Graham & Trotman, London, United Kingdom.

Pim, J., C. Peirce, et al. (2008). "Crustal structure and origin of the Cape Verde Rise." Earth and Planetary Science Letters 272(1-2): 422-428.

Robinson, J. A. C., B. J. Wood, et al. (1998). "The beginning of melting of fertile and depleted peridotite at 1.5 GPa." Earth and Planetary Science Letters 155(1-2): 97-111.

Sissingh, W. (2004). "Palaeozoic and Mesozoic igneous activity in the Netherlands; a tectonomagmatic review." Geologie en Mijnbouw. Netherlands Journal of Geosciences 83(2): 113-134.

Tappe, S. (2004). "Mesozoic mafic alkaline magmatism of southern Scandinavia." Contributions to Mineralogy and Petrology 148(3): 312-334.

Ulrych, J., J. Dostal, et al. (2008). "Late Cretaceous to Paleocene melilitic rocks of the Ohre/Eger Rift in northern Bohemia, Czech Republic: Insights into the initial stages of continental rifting." Lithos 101(1-2): 141-161.

Underhill, J. R. and M. A. Partington (1993). "Jurassic thermal doming and deflation in the North Sea: implications of the sequence stratigraphic evidence." Geological Society, London, Petroleum Geology Conference series 4: 337-345.

Watson, S. and D. McKenzie (1991). "Melt Generation by Plumes - a Study of Hawaiian Volcanism." Journal of Petrology 32(3): 501-537.

Wong, T. E., D. A. J. Batjes, et al., Eds. (2007). Jurassic, Royal Netherlands Academy of Arts and Sciences, Amsterdam, Netherlands.

Ziegler, P. A. (1990). Geological atlas of western and central Europe. Maatschappij B.V. :, Shell Internationale Petroleum.