Logging Summary

IODP Expedition 311:

Cascadia Margin Gas Hydrates

Introduction

Figure 1. A) Location of IODP Expedition 311 on the Cascadia margin. B) Bathymetry and location of the different sites visited (bathimetry courtesy of D. Kelley, J. Delaney, and D. Glickson, University of Washington, and C. Barnes, C. Katnick, NEPTUNE Canada, University of Victoria; funded by the University of Washington and the W.M. Keck Foundation).

Figure 2. Seismic section (line 89-08) showing the position of the Expedition 311 sites across the accretionary front.

Expedition 311 of the Integrated Ocean Drilling Program (IODP) investigated the occurence and the formation of gas hydrate in the accretionary prism of the northern Cascadia margin. The five sites visited during the expedition defined a SW-NE transect across a ~30km wide Bottom Simulating Reflector (BSR) that runs parallel to the coast along most of the continental slope (see Figure 1).

From Site U1326, at the SW tip of the accretionary prism, to Site U1329, the shallowest site located at the landward limit of the BSR, the transect was designed to sample the complexity of the evolution of a gas hydrate system (see Figure 2).

To constrain the formation of gas hydrates in subduction zones, Expedition 311 had an ambitious drilling program including extensive pressure coring to recover gas hydrate at in situ conditions. Because of the unstability of gas hydrate at surface conditions, and of the strong response of some logging tools such as electrical and acoustic logs to the presence of gas hydrate, logging was a critical component of the operations. The logging program consisted of two phases - the first week of the expedition was dedicated to Logging While Drilling (LWD), in order to identify intervals likely to contain gas hydrate where pressure coring tools should be deployed; the second phase consisted in wireline logging following coring operations in order to complete the geophysical characterization of the sites. A complete overview of the expedition results and preliminary conclusions is available in the Expedition 311 Preliminary Report.

Logging Operations

Logging While Drilling/ Measurements While Drilling (LWD/MWD)

Figure 3a. Logging While Drilling/Measurement While Drilling (LWD/MWD) bottom hole assembly used during Expedition 311.

Following a strategy that was successfuly appliedduring ODP Leg 204 on Hydrate Ridge, offshore Oregon, the detailed planning of the coring operations was determined by the LWD results. The presence and the distribution of gas hydrate should be indicated by high resistivity values in the the resistivity logs and images.

A number of the LWD/MWD tools had been used during ODP Leg 204: Resistivity at Bit (RAB, GeoVISION); the Azimuthal Density Neutron tool (adnVISION); the Nuclear Magnetic Resonance (proVISION); and the Measurement While Drilling (TeleScope, an update of the MWD tool used previously). In addition to these tools, the LWD/MWD tool string used during Expedition 311 included the SonicVISION, which had been used in an earlier version during ODP Leg 196 in the Nankai Trough, and the EcoScope, which had never been used during ODP/IODP. With all the measurements provided by the ADN, the EcoScope also provided several additional resistivity measurements, elemental capture spectroscopy, and the borehole annular pressure while drilling (APWD). The complete bottom hole assembly is shown in Figure 3a.

Wireline logging

Figure 3b. Results Wireline logging tool strings used during Expedition 311. Some of the tools had to be recombined differently because of difficult sea states.

The planned wireline logging program included two logging runs for each site: the triple combo followed by the FMS/Sonic tool string. The first run was to provide data similar to some recorded by the LWD for correlations (resistivity, density, neutron porosity, gamma ray), but was also to provide a caliper log indicating the quality of the hole for the subsequent run. The second run with the FMS/Sonic was to provide high resolution electrical images, and most importantly for the gas hydrate characterization, an acoustic log. Because of difficult sea conditions, the ship heave often exceeded the operating range of the wireline heave compensation system (> 4m), and after damaging a caliper arm early in the expedition, the wireline program was limited in several holes to tool strings devoid of protruding arms. Despite this constraint, it was possible to acquire acoustic logs at all sites, allowing further hydrate characterization and seismic/well integration. In addition, two VSP were acquired as planned in Sites U1327 and U1328. The tool strings including all the tools originally planned are shown in Figure 3b.

The following table summarizes the variuos tool combinations used during Expedtion 311.

Water depths are measured in meters below rig floor. They were identified during each tool run by a sharp increase in measured natural radioactivity gamma ray when the gamma ray tool crosses the seafloor.

Hole	Water depth (mbrf)	Max. dept (mbsf)	Tools run
1325A	2203	350	GeoVISION/EcoScope/SonicVISION/TeleScope/ProVISION/adnVISION
1325C	2205	259 185	DIT/HNGS DSI/SGT/TAP
U1326A	1838	300	GeoVISION/EcoScope/SonicVISION/TeleScope/ProVISION/adnVISION
U1326D	1838	300	DIT/DSI/SGT
U1327A	1316	300	GeoVISION/EcoScope/SonicVISION/TeleScope/ProVISION/adnVISION
U1327D	1314	295 276	DIT/APS/HLDS/HNGS/TAP WST
U1327E	1313	290	DIT/DSI/SGT
U1328A	1278	300	GeoVISION/EcoScope/SonicVISION/TeleScope/ProVISION/adnVISION
U1328C	1279	292 292 285	DIT/APS/HLDS/HNGS FMS/DSI/GPIT/SGT WST
U1329A	956	220	GeoVISION/EcoScope/SonicVISION/TeleScope/ProVISION
U1329D	956	10 195	DIT/APD/HLDS/HNGS/TAP FMS/DSI/GPIT/SGT

Gas monitoring

Figure 4. Summary of the LWD/MWD data used to monitor for gas. None of the anomalies observed in the pressure and waveform coherence records indicate any significant amount of free gas.

When gas was detected or expected in previous ODP cruises, the lack of any blowout prevention system required a strict monitoring of the gas composition in the cores. The monitoring procedure relied on the analysis of headspace gas samples after each core was recovered to decide whether it was safe to proceed with coring. Because the LWD/MWD tools were deployed prior to any coring, the monitoring for gas was performed by watching carefully several measurements transmitted in real time by the MWD tool: the annular pressure (APWD) and the sonic waveform coherence. For this purpose, the SonicVISION was configured to identify the wave propagating through the borehole fluid. The possible occurence of gas should be indicated by a sharp pressure decrease, possibly preceded by a pressure increase, and by a loss of coherence in the sonic waveforms. The safety protocol designed for expedition 311 required preventive actions for any pressure anomaly exceeding 100 psi. Figure 4 shows a summary of all the APWD records, after subtraction of the best fit linear trends to enhance the pressure anomalies, and of the sonic waveforms coherence records. Overall, it shows that no significant anomaly was ever detected, allowing to drill and core each site to its target.

Data and Results

Overview

We present here a summary of the logging data and some highlights for each site visited. Because of hole instability in the shallower sediments, wireline data are usually not recorded in the upper ~60 m. In addition, the very short time elapsed after the initial formation penetration makes LWD measurements much less affected by borehole degradation than wireline logs. Therefore, we use the LWD measurements preferably for data that were recorded by both sets of tools, such as density and porosity. For still unexplained reasons, the LWD gamma ray readings were generally higher than the wireline data and we display both curves for completeness.

Despite the proximity of the holes in any site, the considerable heterogeneity in lithology and gas hydrate distribution results in apparent discrepancies between the LWD and wireline data. Some of these discrepancies are discussed in greater details for individual sites. Finally, we have added to each figure the compilation of the infrared images (IR) recorded in each site. These images, recorded immediately after core recovery, were the primary means onboard to identify and isolate gas hydrate samples, which were associated with cold anomalies due to the endothermic nature of gas hydrate dissociation.

The data are presented from the most seaward site (U1326), following the transect upward the deformation front to the most landward site (U1329) (see Figures 1 and 2 for locations).

Water saturations

Preliminary estimates of the amounts of gas hydrate are also given in these figures, expressed as water saturation. Since the primary conductor of electric current in the formation is the pore water, its substitution by electrically insulating gas hydrate generates resistivity anomalies that can be used to estimate the fraction of the pore space occupied by water, or water saturation (Sw). Archie (1942) defined a relationship to estimate water saturation from resistivity and porosity logs in traditional hydrocarbon reservoirs, and Collett (1998) has shown that the same relationship can be used in the presence of gas hydrate. The gas hydrate saturation (Sh) can be assumed to be the complement of Sw (i.e. Sh = 1 - Sw) within the gas hydrate stability field. (see the Initial reports of ODP Leg 204 for a complete description). For each site, a value for the cementation exponent (m) in the Archie relationship was estimated from Archiees equation using the resistivity and porosity logs and salinity values measured on core samples, with the Archie coefficient (a) and the saturation exponent (n) were assumed to be 1 and 2, respectively.

Site U1326

Figure 5. Summary of the logging data recorded at Site U1326. In the resistivity column, the Deep Induction and SFLU (Spherically Focussed Log Unfiltered) curves were recorded with the wireline tools, the others with LWD. The last column on the right is a compilation of the Infra Red (IR) images recorded on the core liner of the recovered sections to detect gas hydrate.

Located at the SW end of the Expedition 311 transect, Site U1326 was drilled Êon an uplifted ridge of accreted sediments. This site was the last one visited for coring and wireline logging operations. Because of time constraints and concerns about an upcoming storm, wireline operations were limited to a single run with a tool string composed of the Gamma Ray/Resistivity/Sonic tools (SGT/DIT/DSI). A summary of the data recorded in this site is shown in Figure 5. The most significant feature in the LWD data is a ~20m interval above 100 mbsf, characterized by bright RAB images and high resistivity values (Phase shift 16 in and Button deep average). The wireline resistivity curves (deep induction and SFLU) and the compressional velocity log (Vp) almost parallel each other in this interval, following a pattern identical to the LWD resistivity, but ~10 m shallower. The high resistivity and Vp values, while density or porosity do not change, suggest significant amounts of gas hydrate, but the offset between the two holes indicates significant lateral heterogeneity. Indeed, the water saturation is possibly as low as 40% between 80 and 90 mbsf in Hole U1326A, but steeply dipping features identified in the RAB images show that gas hydrate is present in beds dipping by as much as 85¡, indicating the sediment deformation in this ridge at the tip of the accretionary prism.
Below ~100 mbsf, the uniformly low resistivity values suggest that only little, if any, gas hydrate is present, except for Êa ~2m interval above 260 mbsf in Hole U1326A. This interval coincides with a loss in sonic waveform coherence in the LWD monitoring data (see Figure 4), that could be associated with free gas.

Site U1325

Figure 6. ummary of the logging data recorded at Site U1325. In the resistivity column, the Deep Induction and SFLU (Spherically Focussed Log Unfiltered) curves were recorded with the wireline tools, the others with LWD. The last column on the right is a compilation of the Infra Red (IR) images recorded on the core liner of the recovered sections to detect gas hydrate.

Figure 7. Summary of the logging data recorded at Site U1327. In the resistivity column, the Deep Induction and SFLU (Spherically Focussed Log Unfiltered) curves were recorded with the wireline tools, the others with LWD. The comparison of the Infra Red (IR) images from Hole U1327C and U1327D with the resistivity logs acquired in Holes U1327A, U1327D and U1327E illustrates the strong lateral heterogeneity of gas hydrate distribution at this site.

Site U1325 was drilled in a slope basin located seaward of of the deformation front, behind the ridge of accreted sediments drilled in U1326 (see Figure 1). This site was the first one visited by Expedition 311, and a strict application of the safety protocol guidelines resulted in excessive pumping rate and poor data in the first 20 m of Hole U1325A. Pumping rates were subsequently reduced at the time of spud-in for the following sites, while maintaining the capability to monitor the safety of the drilling operations. Because of difficult sea state, wireline operations were limited to two strings without arms (HNGS/DIT and SGT/DSI). Rapidly deteriorating hole conditions prevented the recording of acoustic data below 175 mbsf. A summary of the logging data is shown in Figure 6.

The highly variable resistivity log and the succession of dark and bright layers in the RAB image show that gas hydrate may be present in a series of thin layers alternating with gas-hydrate-free sediments. The most distinct occurrences, between 175 and 240 mbsf, result in water saturation values as low as 40 %. Unfortunately, no acoustic data could be recorded in this interval. However, because of the location of this site in an undisturbed basin, the wireline and LWD logs agree very well over the entire interval logged, and tend to confirm the finely layered distribution of gas hydrate.

Site U1327

Site U1327 was drilled near ODP Site 889, where the largest amounts of gas hydrate had been identified during ODP Leg 146. While LWD operations proceeded without any trouble, the wireline logging program encountered several drawbacks due to bad sea conditions. At the end of the first triple combo run in Hole U1327D, during which heave was measured at ~3.5 m, the HLDS caliper arm was damaged, and the subsequent Vertical Seismic Profile (VSP) also resulted in damage to the WST tool. Despite these difficulties, the data recorded were of good quality, but it was necessary to drill a new hole, U1327E, in order to acquire an acoustic log, crucial for the complete characterization of the gas hydrate distribution. The arm-free tool string used in Hole U1327E did not meet any further obstacles. A summary of the data collected in Site U1327 is shown in Figure 7.

As in Hole U1326A, the most striking feature in the data recorded in Hole U1327A is a high resistivity 20-meter thick interval, showing as a bright lawyer between 120 and 140 mbsf in the RAB image, which was interpreted as a significant occurence of gas hydrate. The water saturation estimate suggest that gas hydrate occupies up to 60% of the pore space in this interval. However, the different pressure coring tools deployed in the same interval in adjacent Hole U1327C failed to recover any gas hydrate, and the IR images recorded on the cores from Hole U1327C suggested that the main occurrence of gas hydrate in this hole was ~20 m deeper than in Hole U1327A. The subsequent IR images in Hole U1327D and wireline resistivity logs in Holes U1327D and U1327E showed more heterogeneity in gas hydrate distribution, the only agreement being between the log data and IR images recorded in the same hole (U1327D). Overall, the combination of the logs and IR images suggest that some amounts of gas hydrate is present in all the holes between 80 and 240 mbsf, but this distribution seems to be different in every hole. VSP The original plan for the VSP was to acquire a high resolution survey with only 5 meters between each station over the entire hole. Difficult sea conditions and an enlarged hole did not allow us to record more than 16 stations in Hole U1327D, none shallower than 182 mbsf. The results of the VSP in Hole U1327D are shown in Figure 8.

Figure 8. VSP results in Hole U1327D. A) Stacked traces with original automated travel time picks. B) Corrected picks. The two lines indicate least square linear fit to the data above and below the BSR. C) Comparison of the VSP inversion results with the sonic log recorded in Hole U1327E and the inversion performed on the VSP data from Hole 889B.

Figure 9. Summary of the logging data recorded at Site U1328. In the resistivity column, the Deep Induction and SFLU (Spherically Focussed Log Unfiltered) curves were recorded with the wireline tools, the others with LWD. The last column on the right is a compilation of the Infra Red (IR) images recorded on the core liner of the recovered sections to detect gas hydrate.

The primary purpose of the VSP was to define an accurate time vs. depth relationship to tie precisely the well data with the seismic line crossing the site. The arrival times picked in the stacked traces in Figure 8a, after correction for firing delays and tool position, define two distinct velocity trends above and below ~250 mbsf in Figure 8b. The contrast between the higher velocity above and lower velocity below, presumably corresponding to the occurrence of gas hydrate and free gas, respectively, causes the BSR observed in this site. Figure 8c shows the results of a Bayesian inversion of these data (see Malinverno and Briggs, 2004), giving a more complete picture of the velocity changes at the origin of the BSR. The comparison with logging and VSP data in nearby Holes U1327E and 889B shows that the velocity transition zone responsible for the BSR occurs at different depths despite the proximity of the holes (U1327D and U1327E are ~15 meters apart. Hole 889B is ~500m to the west).

Site U1328

Figure 1

Figure 9.

The LWD data recorded in Hole U1328A show the highest concentrations of gas hydrate encountered during Expedition 311. The water saturation derived from the LWD resistivity drops below 20% in intervals between 5 and 20 mbsf, indicating that gas hydrate could occupy more than 80% of the pore space. The RAB images show that some of the highest concentrations occur within steeply dipping fractures that act as conduits feeding the near surface gas hydrate accumulation. Gas hydrate concentrations then decrease with depth and gas hydrate is only sparsely present below 50 mbsf, in particular within a thin hydrate-filled steep fracture at ~95 mbsf. VSP A clement weather and good hole conditions allowed to record stations every 5 meters over most of the open interval in Hole U1328C, with 35 stations recorded successfully between 286 and 106 mbsf. Apparent interference from the pipe and/or the wireline prevented from recording any meaningful signal at shallower depths. The results are shown in Figure 10.

Figure 10. VSP results in Hole U1328C.
A) Stacked traces with original automated picks. B) Corrected picks. The red line indicates the least square linear fit for the entire data set. C) Comparison of the Bayesian inversion results with the two passes of the sonic log recorded in the same hole.

Figure 11. Summary of the logging data recorded at Site U1329. In the resistivity column, the Deep Induction and SFLU (Spherically Focussed Log Unfiltered) curves were recorded with the wireline tools, the others with LWD. The last column on the right is a compilation of the Infra Red (IR) images recorded on the core liner of the recovered sections to detect gas hydrate.

The 5 meters spacing provided a high resolution image of the velocity structure in this seismically 'blank' area, and both the time vs. depth relationship (Figure 10b) and the results of the Bayesian inversion (Fig 10c) show low velocity values and a very low variability that are consistent with the low seismic reflectivity. The comparison with the Vp logs recorded during two passes in the same hole confirm these results.

Site U1329

At the shallowest water depth encountered during our operations, Site U1329 marks the landward end of the Expedition 311 transect, at the NE edge of the regional gas hydrate occurrence (see Figure 1). While it was the last site drilled with the LWD/MWD tools, it was also the first one for coring and wireline logging operations. All scheduled logging operations proceeded without incidents, but the wireline logs in Hole U1329D revealed an enlarged hole over most of the open interval, impairing the quality of the data. A summary of the data collected in Site U1329 is shown in Figure 11.

As expected from its location at the limit of the regional area of gas hydrate occurrence, the logging data recorded in Site U1329 reveal only little, if any, gas hydrate in the sediments penetrated at this site. The data are characterized by a gradual increase in density and resistivity with depth, with a marked sharpening of the trend at 165 mbsf and then at 185 mbsf. The high resistivity below this depth, illustrated by the bright RAB images, and combined with high density and low porosity, is the indication of increasingly consolidated sediments, possibly composed of thick debris flows. Unfortunately, the low core recovery in this extremely indurated formation did not allow us to fully characterize the nature of this interval.

Summary

These results are still preliminary, and much remains to be learned from the data recorded during Expedition 311, but a few conclusions can be already drawn at this stage:

(1) As part of an adequate protocol, the LWD/MWD tools provide a safe and reliable way to monitor the occurrence of free gas, flows, or other drilling hazards.

(2) The use of LWD tools to help identify targets for pressure coring tools was again a valuable strategy to optimize the recovery of gas hydrate. However, this strategy has to be considered carefully in highly heterogenous settings such as accretionary complexes.

(3) The combined interpretation of the LWD and wireline data along the Expedition 311 transect, particularly of the resistivity and acoustic data, provides a rich picture of the complex distribution of gas hydrate across the Cascadia margin. These preliminary results suggest in particular that, except for short 'anomalous' intervals, gas hydrate concentrations are generally lower than previously estimated from ODP Leg 146. One of the primary objectives of the upcoming work will be to calibrate the different methods to quantify accurately the amounts of gas hydrate identified from the logs.

References

Collett, T.S., Well log evaluation of gas hydrate saturations, Trans. SPWLA 39th Logging Symposium, paper MM 1998.Malinverno, A. and Briggs, V.A., 2004. Expanded uncertainty quantification in inverse problems: Hierarchical Bayes and empirical Bayes. Geophysics, 69: 1005-1016.

Gilles Guerin: Logging Staff Scientist, Borehole Research Group, Lamont-Doherty Earth Observatory of Columbia University, PO Box 1000, 61 Route 9W, Palisades, NY 10964, USA

Alberto Malinverno: Logging Staff Scientist, Borehole Research Group, Lamont-Doherty Earth Observatory of Columbia University, PO Box 1000, 61 Route 9W, Palisades NY 10964, USA

Greg Myers: LWD LoggingScientist, Borehole Research Group, Lamont-Doherty Earth Observatory of Columbia University, PO Box 1000, 61 Route 9W, Palisades, NY 10964, USA

Peter Jackson : Logging Scientist, Geophysics and Marine, British Geological Survey, Kingsley Dunham Centre, Keyworth, Nottingham NG12 5GG, UK