Most petroleum geologists will have a good knowledge of sequence stratigraphy – an important discipline that has revolutionised our work over the last 30 years. Sequence stratigraphy examines the stratigraphic geometries and associated patterns of sedimentary facies that are generated by relative sea level change. In doing so, it is a valuable tool for predicting the occurrence of, for example, reservoir and source rock facies and understanding the architecture of reservoirs.
Equally importantly, it provides a catalyst for the integration of seismic and well and outcrop data as well as detailed sedimentological, biostratigraphic and geochemical studies. Put simply, many of the deepwater plays being explored for today are lowstand fans, predicted from sequence stratigraphic principles and identified on high-resolution seismic data.
Eustatic Sea Level Change
Sequence stratigraphic methodology first came to prominence with the publication of the seminal papers by Peter Vail and his colleagues from Exxon in 1977. Since then, sequence stratigraphic studies have become commonplace and the science has developed its own particular jargon to account for the countless ways in which sediments respond to sea level change.
But Vail and his colleagues did not just bring sequence stratigraphy to the petroleum geologists’ tool box; they reactivated an old idea that some sea level changes are both synchronous and global in nature. Such changes are termed eustatic, a term first introduced by the Swiss geologist Eduard Suess in 1888.
Following the publications by Vail and his colleagues, eustasy once again began to grow in the minds of geologists, cemented in 1987 by another paper from the Exxon school, led by Bilal Haq, which offered a clear statement of a high-resolution sea level curve for the Mesozoic and Cenozoic. Many geologists accepted this and set out to find the sea level events in their particular area, but sceptics like Andrew Miall doubted that a eustatic record could be discernible given tectonic and sediment supply influences, or that biostratigraphy was of sufficient resolution to say that sea level changes were truly synchronous. They also suggested that no viable mechanism existed to explain rapid eustatic sea level changes in the geological past. Miall’s publications remain a valuable source of commentary on the validity of eustatic cycles.
So debate has raged over eustatic sea level changes and their expression for about 30 years. But why does this matter to the petroleum geologist?
Value of Sequence Stratigraphy and Eustasy
Sequence stratigraphy has now become established as a primary interpretation methodology for petroleum geologists. It helps explain the true geometric relationship of sediment packages and, because the technique is based around an understanding of the temporal and spatial relationship of sediments, we can recognise if, for example, they are physically connected – useful for estimating reservoir volume, and more effective than simple, potentially misleading, lithostratigraphic correlations. It is an effective means of combining datasets such as biostratigraphy, sedimentology, logs and seismic within one integrated framework, and is a powerful means of predicting away from known data points. The geometries of sequence stratigraphy predict what facies may be expected up, down and laterally in the depositional system, so the occurrence of reservoirs, source rocks and seals may be predicted regionally from relatively sparse datasets.
All these are valid benefits, regardless of whether a eustatic model underpins one’s view of sequence stratigraphy or not. However, if a eustatic model can be applied then other powerful advantages accrue, including the development of a framework for detailed and precise correlation, mapping and isopaching, and the use of analogues and generic play concepts. Outside the direct economic benefits, an understanding of eustasy provides insight into the sedimentary evolution of the Earth and links to changes in palaeoclimate, both important for our understanding of future climate change.
But first we need to address the question of whether eustatic sea level change can be recognised in the Earth’s sedimentary record, and if so, what mechanisms drive it?
Synchronous Sedimentary Sequences
There is growing evidence that within the resolution of the stratigraphic calibration tools available (typically biostratigraphy, but also isotope stratigraphy), it is possible to demonstrate that there are synchronous global rises and falls of sea level throughout the Phanerozoic. The precision of stratigraphic calibration tools has much improved recently, so that biozonal/isotopic resolution can typically be in the order of a few hundred thousand years or less (as calibrated by orbital forcing, or “Milancovitch” cyclicity). Whilst not perfect, it suggests that a sea level event as represented by changing facies, occurring in the same or equivalent biozone in multiple locations around the world, is likely to be the same eustatic event, although expressed differently depending on the local tectonic and sedimentary setting.
Of course, to build up a eustatic model requires the examination of thousands of sedimentary sections around the globe, analysed in a consistent manner using calibrated biostratigraphy/isotopes. Such work is usually carried out in industry and has major competitive commercial advantages, hence results are rarely published in detail (in Neftex we have in the last ten years built a model which currently has over 125 global sequences in the Phanerozoic, for use in correlation and mapping).
Nonetheless, some results of sequence stratigraphic synthesis have been published. Volumes such as SEPM Special Publication 60 or our Arabian Plate Sequence Stratigraphy are helpful for demonstrating the data that lies behind eustatic models. In 2008 Haq and Schutter published their model of Palaeozoic eustasy indicating the reference and ancillary sections from which the model was derived, although not their interpretation strategy or biostratigraphic calibration.
However, some eustatic models differ from each other for the same time period – not surprising as interpretation strategies vary between workers, and views on biostratigraphic calibration may differ. Therefore, as a fundamental note of caution, before considering how a succession relates to a published eustatic sea level curve or model, one should ask the question “is the sequence stratigraphic interpretation strategy used to construct this model the same as mine?”
The first step to understanding the different driving mechanism for eustasy is to determine the rate and magnitude of eustatic sea level change.
Earlier estimates of eustatic sea level suggested single cycle changes of up to 400m, but more precise estimates using back-stripping (see papers by Ken Miller and co-workers) show that short term eustatic changes are typically 20–80m in magnitude.
The pace of eustatic sea level change is more difficult to measure, but there are successions in which orbital forcing cycles offer a ‘clock’ by which to estimate pace and duration. One such is the Late Jurassic Kimmeridge Clay as exposed in North West Europe, where key sea level changes seem to relate to a major 405,000-year cycle. Orbital forcing cycles in Cenomanian chalk sequences calibrate sea level rises as taking place between 80,000–180,000 years. Thus in both the Late Jurassic and the mid-Cretaceous we can see evidence for eustatic sea level rises and falls of tens of metres over less than 500,000 years. This high magnitude and rapid pace perhaps somewhat surprisingly points towards glacio-eustasy (expansion and contraction of land-grounded ice sheets) as the main driving mechanism during this and other periods. There is certainly a strong link between sequence stratigraphy and orbital forcing in the Kimmeridge Clay and Cenomanian Chalk examples, indirectly suggesting the potential for glacio-eustasy, as it is hard to imagine other mechanisms creating climate-linked changes in sea level. Can this really be valid for Mesozoic rocks said to be deposited in “greenhouse” conditions?
There is little dispute that changes in land-grounded ice volumes have been the primary control on eustatic sea level changes over the last 30 million years. During this time there is abundant evidence for significant cycles of ice growth and destruction at the poles, with cyclicity linked to orbital forcing. Tentatively, the Cenozoic record of polar ice continues to be pushed back further into the Palaeogene with evidence for ice-rafted debris and substantial shifts in the isotopic palaeotemperature proxies. There is indisputable evidence for major polar ice in the Late Carboniferous – Early Permian, the latest Ordovician and within the Neoproterozoic. There is also growing evidence for significant glaciation affecting Gondwana in parts of the Ordovician, Silurian, Devonian and Early Carboniferous.
Rapid Climate Change
What then of the rest of the Phanerozoic? Earth history is typically divided into “icehouse” with pronounced polar glaciations and “greenhouse”, when polar ice has been absent or negligible. How reasonable is it to suppose that Earth history has been dominated by these extreme states over long periods of time? It is of course undeniable that the Earth has experienced periods of extreme warming – as evidenced by crocodiles and ferns near the poles during the Late Cretaceous – but do these records prove a continuous state of “greenhouse” conditions for large episodes of geological time? Or is palaeoclimate more variable than has been suspected and there have been episodic ‘cold-snaps’ in greenhouse times, leading to polar ice-sheet expansion and reduction, giving resultant changes in sea level? Recent compilations of palaeotemperature proxies would suggest that the latter is indeed the case.
There is a growing body of evidence to support the presence of volumetrically significant polar ice during what is commonly regarded as greenhouse times. This includes direct physical evidence within sediments, like dropstones and tillites, and proxy evidence, such as isotope records and glendonites. What is remarkable is that short-term (3rd order) eustatic sea level changes seem to show a correlation with isotope proxy records of palaeotemperature – it is hard to imagine such a coincidence without recourse to glacio-eustasy. Longer term sea level cycles can in part be ascribed to tectonic events, whilst, as argued by Bryan Lovell in a recent Presidential Address to the Geological Society, magmatic underplating of the crust could cause relatively rapid and high frequency regional sea level changes. These need to be differentiated from genuine eustatic signals.
A Consistent Framework
The concepts of eustatic sea level change are old ones, but improvements in the resolution of stratigraphic calibration, together with more rigorous sequence stratigraphic analysis of numerous successions worldwide, means that a consensus on a eustatic sea level model throughout the Phanerozoic is possible. Furthermore, the pace and magnitude of eustatic sea level change and its coincidence with shifts in palaeoclimatic proxy data suggest that glacio-eustasy may well be a key driving mechanism, even in supposed “greenhouse” times.
A eustatic sea level model has powerful benefits for the industry by providing a consistent framework for correlation, facies mapping and the prediction of petroleum systems components. It also provides better selection of play analogues and aids the pursuit of global plays.