Ursa Basin explorers shine new light on shallow water flow
  from: Offshore Engineer
  by: Greg Myers
  Wednesday, April 09, 2008

Targeting the Gulf of Mexico’s Ursa field, academe and industry came together in IODP Expedition 308 to try to improve their understanding of the conditions that cause shallow water flow and the best practices for approaching near-surface, narrow-margin drilling. Greg Myers, who chaired a session dedicated to this initiative at OTC 2007, distills the essence of the team’s findings in this special report for OE.

The unconsolidated soils, excess pore pressures, and deepwater of the Ursa field in the US Gulf of Mexico have long presented challenges to the installation of large structures and wells. Industry and academic partners have been collaborating in a quest to explore the complex geology at Ursa.

The Integrated Ocean Drilling Program (IODP) Expedition 308 focused squarely on Ursa, the overpressured units, and the problem known generically as ‘shallow water flow’ (SWF). However, a more technically appropriate term might be ‘deepwater, near-surface, narrowmargin drilling’, but SWF is the name that has persisted, and will be used in this article.

Geologic setting

The Mars-Ursa salt-withdrawal basin (hereafter referred to as ‘Ursa Basin’) is located 210km south-southeast of New Orleans, Louisiana, on the northeastern Gulf of Mexico continental slope. This region has been studied extensively because of hydrocarbon reservoirs at depth and because of shallow water flow sands near the seafloor. These studies have benefited from the existence of high-resolution seismic data (exploration and hazard) and industry and geotechnical wells.

A southward bulge in the 500m and 1000m bathymetric contours records late Pleistocene deposition from the ancestral Mississippi River. The Mars Ridge, a prominent north-south trending bathymetric area of high relief, bounds the study area to the west. Late Pleistocene shelf, shelf-margin, and turbidite deposits sourced from the Mississippi River are termed the Eastern Depositional Complex. The strata within the Ursa Basin are divided into the sandprone Blue Unit, which is overlain by mud-prone leveed-channel deposition. The Blue Unit is a sand-dominated turbidite unit that was deposited in a broad topographic area of low relief that extended 200km to the east and west and 100km to the north and south. The leveedchannel systems contain a channel and its bounding levees.

The Ursa Canyon channel-levee system immediately overlies the Blue Unit, whereas the Southwest Pass Canyon channel-levee system is younger and lies west of the Ursa Canyon. Multiple masstransport deposits (MTDs) are present in the leveed-channel deposits. The ultimate geometry of the Blue Unit and the overlying leveed channel deposition is that of a wedge: the Blue Unit is approximately horizontal and the leveedchannel deposits thin to the east.

During IODP Expedition 308, three locations were drilled just above the Blue Unit within this sedimentary wedge: sites U1322, U1323, and U1324^1-7.

A seismic cross-section view of the IODP Expedition 308 drill sites as they intersect the Ursa Basin geology is generally characterized by discontinuous, low-amplitude reflectors, which are termed semi-transparent seismic facies. The basal surface cross-cuts stratigraphy. From west-to-east, the base dips at a low angle, then cuts at a high angle down beneath a slide block, then rises up and remains horizontal to site U1322.Within the overall semi-transparent facies, two types of features are observed fault blocks and discontinuous high-amplitude reflectors. At site U1324, triangular fault blocks of undeformed reflectors occur. These are essentially islands of material that remained in place. At site U1322, irregular-shaped, or chaotic, highamplitude reflectors can be seen⁵.

Overpressure

The basic conditions at Ursa are:

�œ a thick accumulation of fine-grained, low-permeability sediment that had been deposited very rapidly and recently in the geologic past;

�œ subsurface drainage inadequate for excess pore pressure to dissipate (even near the seafloor); and

�œ because of first two conditions, pressures nearly at the fracture gradient. This translated not only into a narrow drilling margin, but also very low effective stress and thus low soil strength.

The challenge was, therefore, in soil barely strong enough to support its own weight, the challenge therefore was: �œ to drill closely spaced holes without fracturing or otherwise weakening the formation;

�œ to hold those holes open long enough to run casing and pump cement; and

�œ to enable the resulting structure to support the loads of risers and production hardware for as long as 20 years without intervention⁶.

The basic principles of shallow water flow and its remedies have been covered in a number of earlier papers. Furthermore, near-surface drilling operations and SWF problems at Ursa have been described in great detail in previous works�ö. More recently, the IODP set off on an expedition to drill the overpressured lithologies for the purpose of pure scientific exploration to gain an understanding into how the complex Ursa system works6.

Mass-transport deposits

Mass-transport deposits (MTDs) are sedimentary bodies that have experienced downslope migration and various degrees of internal deformation. Mass transport complexes in the north Gulf of Mexico�fs Ursa region are identified by their low seismic reflectivity, increased resistivity, increased bulk density, high shear strength, and decreased porosity. The seismic signature of MTDs includes high amplitude reflections at their bases and low amplitude, and often non-continuous, internal reflections. Resistivity increases in MTDs (up to 10%) create a contrast with overlying and underlying, nondeformed sediments. Increased resistivity correlates with increased bulk density and visual observations of increased deformation. MTDs are generally defined as the deposits of seismically resolvable mass movement events that include deposits of slumps, slides, and debris flows1, 2.

MTDs are potential drilling hazards and can act as seals for hydrocarbon reservoirs. Several recent studies have focused on the 3D seismic geomorphology of MTDs These studies and others have shown that they are a major component of deepwater environments. A comprehensive study of MTDs in the Amazon Fan showed they have a high degree of consolidation relative to their burial depth. This property of MTDs has been related to the slow penetration time of jetted conductors and suction anchor piles.

IODP Expedition 308 cored, logged and sampled several prominent Ursa region MTDs in the upper 600m (~2000 ft) below seafloor.

It is interpreted that increased density results from consolidation associated with shear deformation. This creates different resistivity-density relations for MTDs and non-failed sediments. Undrained shear strength measured on cores increases by as much as 20% in MTDs where resistivity and density are high. Consolidation during shear deformation also decreased permeability of mudstones in this basin. The combination of rapid deposition and low permeability may have contributed to shallow overpressure in the Ursa region1.

Injectite occurrence

During the last 70,000 years, rapid deposition, directly south of the Mississippi River accumulated a sand and mud sequence, the Blue Unit. This permeable reservoir was rapidly buried by thick mud-rich levees of channel-levee systems accumulating at up to 25m/�f000 years, resulting in substantial overpressures. The Blue Unit is one of the sands that has been the source of SWF and ensuing drilling problems in the Ursa area. Sands penetrated at shallow depths (<1km) in the Gulf of Mexico can flow to the surface and undermine seafloor installations. Here we explore aspects of natural fluid expulsion structures from these �'shallow-water sands�'.

High-resolution seismic data shows no obvious sand injection structures from the Blue Unit, but perhaps nascent dike formation. Also, channel sands stratigraphically above the Blue Unit are overpressured, with a possible pressure source in the latter. In the Ursa area, the maximum principal stress is vertical and constrained by the integrated unit weight of the sediments at IODP site U1324. The stress necessary to form hydrofractures (the minimum principal stress plus negligible tensional strength of the sediment) is about 85% of the maximum principal stress, indicating an approximately isotropic state of stress. High Poisson�fs ratios (0.46-0.49), determined from wireline compressional and shear wave velocities at site U1324, are also consistent with a nearly isotropic state of stress at shallow depths. Estimated maximum fluid pressures are about 80% of the total overburden stress, less than necessary to hydrofracture the sediment. Therefore, the fluid pressure is insufficient to form sand injectites. The formation of sand injectites, as observed in the geologic record, may require very high fluid pressures, approaching lithostatic values. Diverse orientations of ancient injectites support the necessity of very high fluid pressures and a nearly isotropic state of stress.

As mentioned above, sand injectites occur in oil fields and in subaerially exposed sections, and constitute important and underappreciated migration paths. Although these features have not developed from the Blue Unit, the state of stress in the latter provides insight on the threshold for injectite occurrence. Sand injection must be at a pressure exceeding the minimum principal stress and the tensional strength of the sediment, and along a plane oriented perpendicular to this stress. In subaerially exposed injectite systems, injections can occur both parallel to the bedding and cross cutting the layering. These geometries have been used to infer a nearly uniform state of stress in outcrop studies of ancient examples. The requirement for the nearly isotropic state of stress is similar to that observed in the Ursa area. The ancient examples apparently require a higher, nearly lithostatic fluid pressure to form injectites. Judging from constraints of the Ursa hydrofracture data, the formation of injectites would require fluid pressure of about 85% of the overburden. This value exceeds previous estimates⁴.

Pore pressure measurements

Pore pressure was measured with two pore pressure penetrometers on IODP Expedition 308 (www.iodp.org) site U1324. Between the seafloor and 300m below the seafloor (mbsf), overpressures reach 80% of the hydrostatic effective stress. In this interval, only low permeability mudstones are present. Beneath 300mbsf, Poisson�fs ratio is approximately 0.2 and the sediments are composed of interbedded mudstone, siltstone, and very fine sandstone. The lower relative pressures beneath 300mbsf are interpreted to be caused by the higher permeability of these sediments. Given the time and expense associated with these measurements, in this environment, in-situ properties must be interpreted from partial dissipation records. If detailed soil properties are available, the in-situ pressure and diffusivity of the sediment can be inferred from modeling of soil behavior for different penetrometer geometries. However, in many cases soil properties are not available, or there are insufficient resources to pursue soil modeling. In these cases, in-situ pressure is inferred from a simple extrapolation approach³.

Drilling at Ursa

All strategies are risking and can lead to flow to the seafloor outside casing and/or loss of structural support for the shallow casing, resulting in casing buckling and premature abandonment. Over and underbalanced drilling has been attempted with varying degrees of success7.

�œ Overbalanced drilling with a riser tends to fracture the formation with loss of drilling mud and cement. In deepwater drilling with near-surface narrow margins, the main problem with this method is that if too large an interval is attempted, the mud weight necessary to control the pressure of the deepest sand encountered will exceed the fracture strength of the casing shoe.

This is the same as with any conventional drilling in overpressures, but in the near-surface, narrow-margin section the acceptable intervals become very small.

Furthermore, dynamic pressures will be even greater due to pump pressure, cuttings load, and surge pressures when running casing. Prudent drilling practices attempt to minimize these dynamic pressure effects. Another important difference between nearsurface drilling and deeper drilling is proximity to the seafloor, and the relative ease with which induced fractures and flow behind pipe can reach the seafloor and become uncontrollable.

�œ Underbalanced riserless drilling with seawater followed by killing with heavy mud; this allows water flow from the sands (ie SWF) during drilling, which tends to wash out the sands and contaminate the cement. Pump and dump drilling, where weighted mud is circulated with returns to the seafloor, can also be employed.

The alternative method is to drill the near-surface section underbalanced, with seawater and no riser, so that cuttings return to the seafloor. In fact, this method is required for at least the first cemented string, which must be in place first to provide a foundation and attachment point for the BOP stack and riser.

The main drawback to underbalanced riserless drilling is that, while drilling, flows from overpressured sands are uncontrolled, which typically results in large washouts. Recognizing this problem, drillers would attempt to drill this section to TD as quickly as possible to minimize washouts.

Once TD is reached, a heavy kill mud is pumped into the open well bore, sufficient to control pressures in the deepest sands, but within the margin dictated by the formation fracture gradient. In practice, kill mud often fails to work as planned, for reasons such as:

�œ the kill mud is diluted as it is being pumped by water flowing from overpressured sands; this can be a runaway effect; �œ if excessively dense, it may flow into the sands, causing the top of the mud column to drop with time;

�œ if the effective fracture gradient is less than expected, for example due to induced fractures or poor cement at the casing shoe, mud may leak off at the shoe, also causing the mud column to drop. A slight modification to the basic procedure is the dynamic kill, in which the static weight of the mud column is underbalanced, but the additional dynamic pressure while circulating is sufficient to balance the formation pressure.

In a third method, weighted mud is pumped continuously while drilling without a riser, to achieve overbalance and thus prevent caving, ideally resulting in an in-gauge hole. This technique was utilized successfully on IODP Expedition 308. The main drawbacks to this method are:

�œ the high cost of the large quantities of mud necessary while drilling; �œ logistics of providing such a large quantity of mud, typically in excess of what most drilling vessels can hold; and

�œ discharge of drilling mud to the seafloor environment. Emerging, commercially available mud recovery systems are likely to reduce or eliminate the need to utilize the pump and dump method⁷.

Recommendations

Lessons learned at Ursa have been invaluable in understanding the conditions that cause SWF and the best practices for approaching near-surface, narrow-margin drilling. One measure of success is that today drilling into SWF zones is a routine part of deepwater Gulf of Mexico operations⁷.

1. The requirements of a properly constructed deepwater well can be divided into excavation, foundation, and plumbing. SWF and related phenomena, associated with narrow-margin drilling through near-surface overpressured sands can degrade the well�fs integrity in all three areas.

2. The three main strategies for addressing SWF are:

�œ overbalanced drilling with a riser and weighted mud;

�œ underbalanced, riserless drilling with seawater, followed by a weighted mud kill; and

�œ more recently, pump-and-dump drilling, where weighted mud is pumped while drilling, with returns to the seafloor.

Primarily the first two approaches were used at Ursa.

3. The history of drilling at Ursa reflects the difficult choice between conventional overbalanced and underbalanced riserless drilling, each with its own drawbacks (eg formation damage by fracturing vs sand washouts). Riserless drilling through the first two cemented strings ultimately succeeded in the second batch set.

4. The number of cemented casing strings used to drill through the

�eBlue Unit�f overpressured sands increased incrementally from one in the discovery well to four in the second batch set.

5. More favorable geological conditions (thinner Blue Unit and thicker post-Blue overburden) also contributed to the ultimate success of the second batch set.

6. Technological developments that contributed to the SWF control effort included:

�œ development and dissemination of best practices, such as controlled drilling and gel sweeps to minimize cuttings loading; �œ new casing systems;

�œ low-temperature foam cement;

�œ use of LWD and PWD tools during all stages of drilling; ]

�œ measurement of in-situ pore pressure in geotechnical boreholes; and

�œ high resolution 3D seismic surveying to better define the subsurface geology.

7. Deepwater mud recovery systems that establish a dual gradient drilling regime while recovering drilling muds, without a riser.

8. SWF-related foundation failure that leads to buckled casing can result from:

�œ poor cement across large sand washouts, causing inadequate lateral support;

�œ induced fractures propagating to the seafloor and reducing soil strength, thus causing inadequate vertical support; and

�œ soil creep around severely caved sands, causing the overburden to subside⁷. OE

References

1 B Dugan et al. Physical properties of mass transport complexes in the Ursa region, northern Gulf of Mexico, IODP Expedition 308, determined from log, core and seismic data. OTC 2007.

2 GJ Iturrino et al. Interpretation of downole measurements, deformation analyses, and lithologic characterization in the Ursa Basin, Gulf of Mexico. OTC 2007.

3 H Long et al. In situ pore pressure at IODP site U1324, Ursa Basin, Gulf of Mexico. OTC 2007.

4 JC Moore. Fluid migration and state of stress above the Blue Unit, Ursa Basin: relationship to the geometry of injectites. OTC 2007.

5 DE Sawyer et al. Lateral varations in core, log, and seismic attributes of a mass transport complex in the Ursa Region, IODP Expedition 308. OTC 2007.

6 CD Winker et al. 2007 geology of shallow water flow at Ursa: 1. Setting and causes. OTC 2007.

7 CD Winker et al. 2007 geology of shallow water flow at Ursa: 2. Drilling principles and practice. OTC 2007.

This article represents a synthesis of the special session entitled 'Shallow-Water Flow in the Ursa Basin: an IODP/industry cooperation' at the 2007 Offshore Technology Conference (OTC) in Houston, co-chaired by Greg Myers and Kelly Kryc of IODP-MI. Special thanks to Craig Shipp of Shell.

Click here to register to receive your own copy of Offshore Engineer each month.