Data indicates different geological environment with probable limitations on deepwater hydrocarbon accumulations
- The history of exploration drilling on the shelf and in deep water since 1945. The data, compiled by Petroconsultants, does not include North America.
- Hydrocarbon liquids discovered in deepwater have been found to be denser than those on the continental shelf.
A petroleum system involves all the ingredients necessary for a hydrocarbon accumulation to exist. These include, among other factors, a number of static elements such as reservoir, trap, source rock, cap rock, and their interconnections, as well as the time factor. All these elements must be present and correctly linked in time and space.
Most of the elements of the petroleum system are affected by the context in which they find themselves and there are some particularities in the deepwater environments which affect the eventual nature and volumes of the hydrocarbons trapped. A brief examination of the statistics of discoveries in deep water and a comparison with exploration results on the shelf has been undertaken using data derived from Petroconsultants' Probe Database. This database largely excludes data from North America, therefore, unless otherwise stated, it should be assumed that all the statistics quoted here also exclude North America.
Exploration on the shelf started slowly in the early 1950s, building up gradually with time. By contrast, deepwater exploration started in the early 1980s at a high level of activity. This level of drilling activity has been sustained to the present day. To try to compare the performance in the past and to predict what could happen in the future, the first 14 years of exploration in deep water (1981-1995) are compared with the first 14 years of exploration on the shelf (1951-1965). It is possible to compare the technical successrate in the two environments during similar periods of exploration maturity in the two environments. The technical success rate in a working petroleum system is a measure of the technical complexity of a system and our ability to correctly define the objective.
With similar available technology, the technical success rate is a measure of geological difficulty. It is interesting that 24.8% of exploration wells (254 wells) drilled on the shelf during the first 14 years found hydrocarbons (shelf success rate for exploration is now about 24.5%), whereas the technical success rate of exploration wells (551 wells) in deep water is only 21.9%. This, in spite of the fact that at the onset of offshore exploration, seismology was poorly evolved, limiting our ability to focus exploration objectives, and that high resolution seismic data have been available to us since the beginning of our exploits in deeper waters. Thus, although there is only a small difference in the success rates, this suggests that there could be significant differences in the petroleum systems in the two cases and more complexity in the deep marine environment.
Explorers always attempt to cream (find the largest, most productive fields first) a given play. Significant fields were found during both the 1951-65 and 1981-95 periods but, in general, a higher ratio of significant large fields were found on the shelf than are found today in deepwater. The fact that there have been proportionately fewer very large fields found in deep water may not be significant, but field-size distribution curves do show there is no statistical evidence to support the idea that there should be larger fields in deepwater than on the shelf. This is in spite of the fact that our modern exploration tools are considerably more powerful than were available during the early years of exploration in the offshore environment.
This comparison suggests that the closed reservoir volumes of fields in the deep offshore are no greater and could be smaller than the closed reservoir volumes on the more classic shelf areas. To try to obtain a vision of the future of exploration in deep water, the field size distribution for fields discovered on the shelf between 1951 and 1965 has been compared with the field size distribution for all fields discovered on the shelf. This shows that during the evolution of exploration in this environment, the ratio of large fields has descended with time. Although any statistical projection is dangerous, and there will be very significant exceptions, in general it should be expected that the proportion of very large fields found in the future in deep water will diminish.
An indirect way of obtaining an indication that there are differences in the nature of hydrocarbon fluids in the two environments is to examine the API gravities of the trapped liquids. The gravities of oils discovered on the shelf from 2,111 samples in the Petroconsultants data base were compared. There is a clear modal value in the range of 35-40 API. By contrast, the modal value of the 101 samples recorded for discoveries in the deepwater environment is lower, being at 25-30 API. The statistics are heavily weighted to the low end of the API range by the heavy oils of Brazil, and towards the high end by the discoveries in Norway.
This suggests that there could be differences in the some of the sourcing systems in deep marine environments which could lead to variation in the density of the trapped hydrocarbons. Confirmation of this could be obtained by comparing the ratio of gas-rich to oil-rich fields on the shelf and in deepwater.
More deepwater oil
Global statistics demonstrate that out of 2,762 fields discovered on the shelf, 60% are gas fields and 40% oil fields. In contrast, out of the 212 fields found in deepwater, only 45% are gas fields. This confirms that there is a tendency towards heavier hydrocarbons in deepwater fields and there could be significant differences in hydrocarbon generation/expulsion or in modification of entrapped hydrocarbons after expulsion in petroleum systems in the two environments.
To summarize, the statistics suggest that if the deepwater and shelves are compared, there is a tendency to find that in deepwater:
- Exploration is more risky.
- The largest fields are smaller.
- The hydrocarbons are heavier.
By closer consideration in more detail, certain aspects of the petroleum system such as the sedimentology of potential reservoirs and some of the constraints on hydrocarbon generation, expulsion, and migration, possible explanations for these differences can be proposed.
Many reservoirs in present day deepwater were deposited in environments other than that in which they are now found. There are many cases of shallow water carbonates or deltaic deposits which are now in deep water. The geometries and properties of these reservoirs are well studied and have been discussed in many papers and there is no reason to expect that these properties will differ significantly with the depth of water in which they now find themselves.
By contrast, there are some reservoirs which were deposited in deepwater and still find themselves there today. These deepwater sands make some of the very best, and many of the worst, hydrocarbon reservoirs on Earth. Deepwater systems are very variable and the following paragraphs are a brief summary of the processes which govern sedimentation in deep water, and the favorable situations in which good reservoirs can be expected in the deep domain. Sedimentation in deepwater is governed by three main processes:
- Re-sedimentation by mass flows (turbidity currents) along the continental margins.
- Settling of fine particles.
- Reworking by permanent currents.
Only the first process, sedimentation by mass flows (turbidity currants) along the continental margins, can bring sand into the deep basin. The second process, the rain of fine sediment out of the water column, can be considered as the background sedimentation and can only deposit very fine-grained clastic material (i.e. shale particles) or shells from living organisms (mostly calcareous plankton). Permanent ocean currents traverse the oceans and can redistribute sand, but they are not able to bring new sediment into the system. Thus, significant reservoirs in deepwater can only be generated by turbidity currents.
Mass flows encompass a very wide range of mechanical behaviors, from slurry-like, slow moving debris flows to sudden and devastating dense turbidity currents, to sluggish and dilute muddy turbidity currents. The only mechanism which can result in reservoir accumulation in the deep domain is the second type, at least under favorable conditions. It is important to understand roughly what a turbidity current is in order to understand the reservoirs they make.
Turbidity currents can best be visualized as snow avalanches, but snow avalanches with clay and sand instead of snow and ice, and in an aqueous medium rather than in the air. Both turbidity currents and snow avalanches re-sediment almost instantaneously huge amounts of sediment which were previously stored higher up on a slope, where they had piled up for a much longer interval of time.
An average reservoir-making turbidity current would start as a creep of unconsolidated material high on the slope. Progressive acceleration along the slope incorporates water into the mixture until it gets dilute enough to become turbulent. From this point on, the flow is a proper turbidity current in which the coarse particles (sand grains) are maintained in suspension by the turbulence of the flow. A mixed sand/shale turbidity current is differentiated, with a higher concentration of coarse material in the bottom part than at the top. As a result, overflow of a turbidity current over an obstacle, a channel margin for instance, will segregate coarser material upstream of the obstacle from finer material down flow.
Turbidity currents start depositing their load when the velocity decreases. This is naturally achieved when the slope on which the current flow decreases, and particularly when the current reaches a flat area. The coarsest particles are the most difficult to maintain in suspension and are deposited first (i.e. close to the entry point to the deep basin).
The energy of a large scale turbidity current is huge, which explains why some currents can carry sand grade particles as far as several thousands of kilometers away from their source on the shelf. The most famous example is the black shell turbidite, which flowed down from Newfoundland 17,000 years ago and covers a good part of the North Atlantic. Such basin-wide turbidites however also have a shaly top, which means that they produce layer cakes rather than real reservoirs.
Turbidite systems are probably the type of sedimentary systems which show the highest variability in size, morphology, and lithology. The largest deepsea fans associated with major river systems can cover several hundred thousand square kilometres, whereas small systems in tectonic basins may be not exceed a few sq km. The lithology ranges from 100% sandy systems to 100% shaly ones.
The biggest turbidite systems develop directly downslope of large deltas. Deltas normally prograde on the shelf without triggering major gravity flows until they reach the shelf edge. At this point, the incoming sediment cannot be accommodated any more and it is periodically re-sedimented into the basin by turbidity currents. Re-sedimentation is more likely to take place in low sea-level conditions, such as when the coastline gets closer to the shelf edge (as was the case during the Quaternary glaciations, for instance).
Turbidite systems typically comprise a conduit zone on the slope, with canyons and/or channels in which turbidity currents flow very much like rivers on land, and a depositional zone at the bottom of the slope and in the basin where sediments are deposited. In the case of incoming mixed lithologies (sand and shale), the coarsest particles are deposited in proximal position.
Reservoir making processes
- This illustration shows the relationship between a lacustrine source rock and turbidities reservoirs separated by a thickness of shales. The shales could provide a barrier to the migration of hydrocarbons.
As explained previously, most turbidites are deposited from mixed sand-shale flows, and comprise a bottom sand-rich part and a shale layer at the top. As a result, turbidites would usually make layer cakes rather that proper reservoirs. However, some of the very best reservoirs in the world have been deposited from turbidity currents in the deep domain. This section will examine the conditions under which good reservoirs can be built by turbidity currents in the deep domain. There are three main ways of making reservoirs from turbidites: amalgamation. sieving, and selective sourcing of sand.
- Amalgamation: This occurs where the energy of an incoming turbidity current is sufficient to erode the shaly top of the previously deposited turbidite. It is likely to happen in proximal settings (i.e. within channels where the turbulence is very high, or at the outlet of the channels), not in the distal pans of the system.
- Sieving: Sieving or selective trapping of the sand fraction, can occur in specific settings. One such setting is the flat area on a slope or local lows between growing salt diapirs in the Gulf of Mexico, in which sand is deposited while the shaly upper part of the current can overflow to the deep basin. A somewhat similar process is obtained where turbidites are deposited in deep areas swept by bottom currents. Bottom currents typically flow along the base of slope (contour currents sensu strictu.) or perpendicular to the incoming turbidity currents which flow down slope.
The permanent current can then winnow the shale out of the incoming flow, thus producing good reservoirs. In addition to that, bottom currents can locally prevent deposition, ensuring that the reservoir is not connected with sand upslope.
- Selective sourcing: This process is perhaps the most obvious way of making massive sand bodies in a deep basin by re-depositing sand only. This very favorable condition can happen wherever there is a sorting agent between the primary delivery of sand to the basin (river) and the feeder canyon/channel of the system. Specific climatic conditions and hinterland geology can also result in an exclusive supply of sand to the rivers, which gives the same type of sand bodies.
Little is known at present of the mechanical behavior of sand flows. It seems that they do not travel as far into the basin as mixed flows. Sand flows tend to make sand heaps. with a mounded geometry, not far from the entry points to the basin, while mixed flows spread their sand content over much wider areas.
In addition to those processes which ensure vertical communication between successive beds, longer-term sea level variations on the shelf (due either to eustatic sea-level changes or to tectonic activity) also contribute to alternating episodes during which sedimentation switches from relatively sand-prone (low relative sea level stand) to relatively shale-prone (high sea level stands, during which most of the sand is trapped on the proximal part of the shelf).
One problem with turbidite reservoirs is their effectiveness as a reservoir. The sandy part of turbidites deposited by mixed flows dies out progressively, leaving a reservoir with a high clay content. This means that the distal part of the turbidite system can be a waste zone rather than a reservoir or seal. By contrast, the lateral onlap onto the margin of the basin normally provides a good stratigraphic trap with a weakness - the entry point to the system (feeder channel). This regions is often infilled in the late stages of the system by turbidites which can make a drain towards the shelf.
A very characteristic feature of all outcropping turbidites is their flat geometry. Outcropping turbidites obviously were deposited either as sheet-like sands, continuous over tens of kilometers or infill pre-existing topographies. In both cases, the tops of the sand bodies lie flat and palaeo-horizontal.
At the same time, published models frequently describe or illustrate mounded sands, also interpreted as turbidites. It seems that the apparent conflict results from the different types of basin settings. Most turbidites at the outcrop represent the infill of former foredeep basins by mixed sand and shale flows. On the other hand, the mounded sand-rich turbidites are usually observed in subsurface in cratonic basins, and are made of a stacking of sand flows rather than mixed flows.
Shaly mounds are common, in particular on continental slopes where huge collapses of shaly material result in thick mounded sedimentary bodies which do not contain any sand. At present, there is no unequivocal way to distinguish between sandy and shaly features from their geometry only.
For a good reservoir to be deposited, three elements are necessary:
- An identified source of sand (a river channel with a siliciclastic hinterland).
- A conduit to lead the sand from the sandy shelf to the prospect (a canyon system).
- A sand trap (thalweg, low in a slope, or at least a decrease in slope).
Most fan deposits do not fulfill these criteria and are unlikely to be originators of reservoirs. Fan systems are essentially dispersive and transport distances of sand-rich turbidity currents are limited. Thus, the quality and thickness of reservoirs will degrade rapidly from the end of the conduit.
As in the case of reservoirs, there are two possible ranges of origins of source rocks currently found in deepwater environments:
- Deposited in a deepwater environment, possibly with a deepwater reservoir in the same context as the source rock
- Deposited in a less deep environment (lacustrine, deltaic, or shallow marine). It is not uncommon to find such a source in close juxtaposition with (overlain by) sandstones and shales which themselves were deposited in deeper water associated with deposits in the deep water.
There are three principal factors which control the formation of a marine source rock: the organic productivity at the surface of the water body, the presence of a reducing environment in the sediments on the sea floor, and a low rate of inorganic sedimentation. Studies of recent sediments have shown that the deposits at great depth of water are poor in organic matter, suggesting that during its passage down a large water column the organic matter is destroyed - probably eaten or dissolved.
Cold arctic and antarctic currents traverse the oceans floors establishing oxygenic conditions at the sediment water interface and further diminishing the possibility of establishing conditions favorable to source rock deposition. Thus, it is unlikely that in the true deep open oceanic waters a good thick source rock is likely to be deposited and preserved. On the other hand, it is possible to establish suitable conditions for the deposition of source rocks in fairly deep waters along the passive margins (continental slopes) such as in association with either an oxygen minimum layer or with upwelling currents.
Upwelling currents develop on the continental slope and are caused by the movement of cold currents across the continental slope in conjunction with an opposing offshore dominant wind direction. These upwelling currents are rich in organic matter. Modern examples of such a current can be found offshore Chile (famous for rich fisheries) and offshore Namibia. This current upwelling offshore Namibia generates deposits rich in organic matter at water depths between 200 meters and 1,000 meters. Thick intervals of shale with dispersed organic matter of terrestrial detritic origin (plant matter) are found associated with deltas such as the Niger Delta and Offshore Angola and these are probably effective source rocks. These shales likely extend into deepwater and are associated with sands derived from the delta complex.
The case of source rocks found currently in a deepwater context but deposited initially in shallow water is well known and is found in several proven petroleum systems (Brazil, Angola, and probably the West Shetlands). These examples are all associated with oceanic opening, and the source rocks are either of lacustrine origin deposited in the grabens or of marine origin deposited in a post rift sequence. Thermal subsidence linked to the end of rifting initiated an increase of water depth which continued during the period of drift and ocean formation. One of the problems challenging explorers in this circumstance is the correct prediction of the extension of the source facies towards the offshore. It is necessary to judge whether the most distal rifts contain a lacustrine facies or whether the shales in the post rift marine sequence retain their source qualities far offshore.
Source rock maturity
To generate hydrocarbons, a source rock needs to attain a threshold temperature dependent on the nature of the source rock and temporal and physical conditions of that source rock. Although the nature of the chemical process is the same in the deep and shallow offshore, there are additional difficulties in attaining a sufficient level of maturity to generate hydrocarbons in deep water.
First, it is well known that on the passive margin when one goes from the continent to the offshore, the thickness of sediment increases, passes through a maximum, and then diminishes again. As a result, for a source rock of a given age, it will be less buried and therefore cooler than for the equivalent source rock on the shelf. This can mean that a source rock which is effective in the shallow water may not be mature enough for the generation/expulsion of hydrocarbons. Furthermore, in deep water, the temperature of the sea water just above the sediment-water interface at water depths greater than 1,000 meters is never greater than 4-5C. This compares with possible temperatures onshore of up to 30C. This also has the effect of lowering the temperature attained in a given depth of sediment.
Modeling shows that the levels of maturity (beginning of the oil window) can be deepened by about 500 metres. This difference appears to us to be excessive and our model probably does not completely correctly simulate the cooling phenomenon. However, it demonstrates that lower maturities for a given depth of sediment could result from the lower temperature at the sediment-water interface.
Global statistics show that the oils discovered in deepwater are in general relatively heavy (20-30 API). Although the effect of biodegradation cannot be ignored, the relative lack of maturity of the source rocks could explain these high densities.
Migration from sources
The "background" sediment in deepwater is shale. Thus, significant thicknesses of shale should be predicted and potential reservoirs, even though of good quality, will often be isolated from a potential source rock by shale. Furthermore, the presence of a structure or laterally continuous feeder beds to focus the migration paths of generated hydrocarbons is unlikely. For a petroleum system to work in deepwater, the connection between the source and the reservoir needs to be efficient. In a petroleum system in which source rocks formed in association with rifling are coupled with later turbiditic reservoirs, there could be migration problems.
In fact, the turbiditic unit develops in association with an argillaceous series which can be quite thick. If these shales are too thick, migration can be inefficient and will necessitate the presence of faults. The most favorable situation would be where the source rocks are in close proximity to the reservoirs and likely to be efficient traps of hydrocarbons, especially heavy oil products generated and expelled early. In the absence of later lighter products due to the cool thermal regime, the resulting reservoirs are likely to contain heavier than normal oils.
Seventy percent of the earth's surface is covered by sea, of which a large part can be described as deepwater, but the constraints of the petroleum system dictate that only the relatively restricted surface area underlain by continental plates can be considered as prospective for hydrocarbons. Although analogous fields to those found in shallow water can be found in deepwater, the influence of the deepwater environment on the petroleum system suggests that there is little reason to expect better quality reservoirs or larger volume accumulations in deepwater than in shallow water. Despite this, there remain considerable areas of deepwater with potential to contain entrapped hydrocarbons that have not been explored.
Copyright 1996 Offshore. All Rights Reserved.