| IODP Proceedings Volume contents Search | ||
 | ||
| Expedition reports Research results Supplementary material Drilling maps Expedition bibliography | ||
|
doi:10.2204/iodp.proc.319.103.2010 OperationsShingu, Japan, port callLoading for IODP Expedition 319 began on 8 May 2009 at Shingu, Japan. A "prespud" meeting was held aboard the D/V Chikyu on 9 May 2009. The Chikyu departed Shingu at 0930 h (all times given are local) on 10 May 2009 for Site C0009 a few days later than originally planned because of bad weather. Transit to Site C0009Upon arrival at Site C0009, the ship was set in dynamic positioning mode, the Big Head transducers were lowered, and a detailed seabed survey began at 1700 h on 12 May, finishing on 15 May. Official water depth at the site is 2082.3 m drilling depth below rig floor (DRF) (2054 m mud depth below sea level [MSL]). Two arrays of 10 transponders were deployed during 16 and 17 May and were immediately calibrated. The 36 inch conductor pipe, gamma ray attenuation (GRA)/Mudmat, and drilling ahead tool (DAT) were rigged up and the vessel moved 5 nmi upstream to lower the 36 inch conductor through the GRA and Mudmat on 17 May. Hole C0009AOperations at Site C0009 consisted of several drilling phases (summarized in Table T1), as well as casing and hole completion. Initial operations began with a run in the hole and jet-in after the remotely operated vehicle (ROV) seafloor survey. Hole C0009A (33°27.4704′N, 136°32.1489′E) was spudded at 0930 h on 19 May, followed by jetting-in the 36 inch conductor. The 36 inch conductor was successfully installed to 2136.8 m DRF (54.5 m drilling depth below seafloor [DSF]) by 1730 h on 19 May. However, the top connection of the 36 inch DAT sheared because of excessive slackoff and bending of the pipe above the DAT. A successful fishing run was completed on 23 May and the 26 inch drilling bottom-hole assembly (BHA) reentered the 36 inch wellhead at 0700 h on 24 May. Drilling with the 26 inch BHA, including MWD tools (Table T2), continued to 2795 m DRF (712.7 m DSF) and successfully reached the target depth on 25 May. A dynamic positioning system (DPS) malfunction occurred on 26 May, the investigation of which continued until 30 May, when the DPS was fixed and carefully observed in order to evaluate repairs. During this time, the 20 inch casing was prepared for installation. A final decision regarding the DPS failure was made on 31 May and operations resumed. Preparations for running the 20 inch casing began on 1 June and the casing was installed to 2786.2 m DRF (703.9 m DSF) on 2 June, with cement operations concluding on 3 June. Rigup for running riser pipe and blowout preventer (BOP) installation continued until 7 June. Meanwhile, a thruster failure occurred on 4 June and investigations into the cause continued until 8 June. Running down the BOP with riser pipe began on 8 June, including attachment of fairings to the uppermost 11 sections of the riser pipe. Problems with the riser tensioner Riser Antirecoil System (RARS) valves were discovered on 13 June and troubleshooting continued until 17 June. The rotating telescopic joint load ring became stuck on 19 June; however, the BOP was successfully connected to the wellhead on 21 June. Following a BOP pressure test, the 17 inch drill-out cement assembly was run into the hole on 24 June and the 3 m cement plug was drilled out to 2798 m DRF (715.7 m DSF) by 25 June. After this, two leak-off tests (LOTs) were conducted below the casing shoe on 26 June (see "Leak-off test"). Following the LOTs, the hole was drilled with a 12¼ inch drilling BHA with MWD tools, including Power-V and MWD (Table T2), beginning on 28 June. Cuttings and drilling mud gas were continuously collected during drilling. The 12¼ inch drilling reached TD at 3686 m DRF (1603.7 m DSF) on 2 July, including several short trips and circulation from bottoms up. After reaming, circulation, and cleaning the hole, coring with the 10⅝ inch rotary core barrel (RCB) bit started on 5 July (Table T3). A total of nine cores were collected (Table T4) with RCB coring operations finishing on 8 July after reaching a TD of 3676.2 m DRF (1593.9 m DSF). The first two cores had very poor recovery because of problems with equipment. After coring, the cored section was reamed with the 12¼ inch drilling assembly to open the hole and collect additional cuttings. Opening the hole with the 12¼ inch drilling assembly continued to a TD of 3686 m DRF (1603.7 m DSF). During this time, lost circulation material was added to drilling mud to remediate minor lost circulation in the hole (see "Mud program"). On 12 July, two wireline logging runs were conducted: a first run with Environmental Measurement Sonde (EMS)–High-Resolution Laterolog Array (HRLA)–Platform Express (PEX)–gamma ray and a second run with Formation MicroImager (FMI)-Hostile Natural Gamma Ray Spectrometry Cartridge (HNGC)-Sonic Scanner-EMS-Power Positioning Device and Caliper Tool (PPC) (see "Logging" in the "Methods" chapter for details). Prior to a third wireline logging run with the MDT, a 12¼ inch wiper trip assembly was run to clean the hole. After the wiper trip, MDT logging was conducted on 14 and 15 July (see "Logging" in the "Methods" chapter).
Soon after MDT operations finished and the tools were pulled out of the hole, the hole was opened to 17 inches (on 16 July). Drilling mud gas was collected during this hole opening phase. Hole opening operations continued until the hole was reamed down to 3650 m DRF (1567.7 m DSF) on 19 July. Several wiper trips and circulation bottoms ups were conducted to clean the hole prior to installing 13⅜ inch casing. Preparations for running casing began on 21 July, and the casing was successfully installed on 22 July. The 13⅜ inch casing shoe was set at 3624.3 m DRF (1542 m DSF) (Fig. F1). Cementing the 13⅜ inch casing took place on 23 July. During cementing operations, landing of the second cement plug could not be confirmed, and cement loss was observed (84% efficiency for the cement job). Upon pulling the cementing assembly to the surface, cement was found on the ports of the tool, indicating the possibility of a cement leak from the tool at the casing hanger. A junk basket with a 12 Five VSI sensor arrays were rigged up on 24 July, and a check shot test was performed. There was a problem found in the VSP cables, so the sensor arrays were retrieved and checked. Four arrays of VSP tools (a total of 16 shuttles) were put back into the hole after finding a communication problem between the Number 4 and Number 5 arrays. The walkaway VSP experiment with the Kairei was started at ~2000 h on 24 July. One line and one circle shooting transect for the walkaway VSP experiment were completed on 25 July, followed by a zero-offset VSP on the same day (see "VSP operations"). Preparations to retrieve the BOP started on 26 July, and all riser pipes and the BOP were safely retrieved on 30 July. Retrieving transponders via ROV was completed early on 31 July, and the ROV set the corrosion cap into the wellhead on the same day. All operations for Hole C0009A were completed at 2359 h on 31 July. The Chikyu started moving to IODP Site C0010 (proposed Site NT2-01J) at 0001 h on 1 August. Leak-off testAfter 20 inch casing was run to 703.9 m DSF and cemented to 708.6 m DSF, the hole was deepened to 715.7 m DSF. The hole was circulated and cleaned. Preparations for the LOT started at 2315 h on 25 June. From ~0001 to 0200 h on 26 June, there were various problems with a valve (the upper inline blowout preventer [IBOP]). The lower IBOP kill line was flushed and tested at 1000 pounds per square inch (psi). The first LOT began at 0215 h, with 1.08 g/cm3 of KNPP mud, at a pump rate of 0.25 bbl/m. Pumping stopped when the pump pressure reached 116 psi, with a total volume pumped of 2.2 bbl. The leak-off pressure was 105 psi, or an equivalent mud weight (EMW) of 1.106 g/cm3 at the casing shoe. The maximum pressure was recorded as 116 psi (EMW = 1.109 g/cm3). After pumping stopped there was a flowback volume of 0.5 bbl. Preparation for the second LOT started at 0306 h. Both mud tanks were filled for a line pressure test. Pressure was held at 272 psi. At 0322 h, the LOT proceeded with 1.08 g/cm3 KNPP mud at a pump rate of 0.25 bbl/m. The pump was stopped when pressure reached 135 psi, for a total pumped volume of 3.6 bbl. The leak-off pressure was 105 psi (EMW = 1.106 g/cm3). The maximum pump pressure was 133 psi (EMW = 1.113 g/cm3). After pumping stopped there was a flowback volume of 0.5 bbl. VSP operations
The VSP experiment was divided into two parts: a walkaway VSP in conjunction with the research vessel Kairei, and a zero-offset VSP conducted solely by the Chikyu (Table T5). Details regarding VSP operations can be found in "Downhole measurements" in the "Methods" chapter. Operations for the walkaway VSP were adjusted from the precruise plans because of scheduling uncertainties and delays with expedition operations. One line of air gun shooting, combined with one circuit around the Chikyu, was achievable in the time available. Before ocean-bottom seismometer (OBS) deployment, synchronization of radio signals between the Kairei and the Chikyu was checked and confirmed. The Kairei began operations by dropping eight OBSs to augment an array of four broadband ocean-bottom seismometers (BBOBSs) that had been deployed in 2008. After cementing operations were completed, a 12 Preparations for the zero-offset VSP experiment began immediately following the walkaway VSP; the Chikyu deployed a set of air guns suspended from a crane off the port side of the ship. These were shifted to a point farther from the Chikyu to avoid damage to the azimuth thrusters. A delay of ~4 h occurred because of repeated failures of the shear pin on the compensation line. An additional delay was caused by the use of faulty shackles on the air gun assembly—the crane operator noticed that condemned shackles had been used and called for their replacement before deploying the air gun at 1145 h. The zero-offset VSP began with the VSI array in the same position as the walkaway VSP, but between shootings the array was moved upward in the casing at 122 m intervals to provide data throughout the borehole. Shooting and recording began at 1620 h. It was soon discovered that Seismometers 9 and 10 were malfunctioning, so the zero-offset VSP was run using only eight seismometers. At 1916 h, pulling out of the hole for the VSI tool string began, finishing rigdown at 0100 h on 26 July. Mud programOne of the major differences in riser drilling compared with riserless drilling is the use of weighted mud to control or balance formation pressures and sweep cuttings from the hole. Riser drilling is commonplace in the petroleum industry, and weighted mud has been used for some riserless drilling projects in IODP (e.g., Flemings et al., 2005). However, riser drilling with weighted mud is a new element in IODP drilling that carries important considerations for merging scientific and operational objectives. Continuous monitoring of mud weight, annular pressure, mud losses, and other circulation data during riser drilling provides important scientific data to constrain formation pore fluid pressure and state of stress (e.g., Zoback, 2007). On the other hand, problems related to mud weight during drilling or cementing may jeopardize successful drilling of the borehole itself, as well as the ability to achieve postdrilling scientific objectives including observatory installations and active source seismic experiments (e.g., VSP studies) that require hydraulic sealing or coupling of casing to the formation. Unsuccessful hole completion (cementing operations) can also impair deepening of the borehole by later drilling. For these reasons, we describe some key observations and operations related to the mud program and hole conditions during drilling of Hole C0009A. In general, mud weight is selected such that the pressure of the mud column in the borehole is sufficient to balance formation fluid pressure but remains below the fracture pressure (approximately equal to the least principal stress plus the tensile strength of the formation at a given depth). If mud weight is too low (underbalanced condition), fluid may enter the borehole, leading to partial or complete borehole collapse or, in the case of significant gas, possible blowout. If mud weight is too high (overbalanced condition), the pressure of the mud in the borehole can cause fracturing of the formation and/or mud losses can occur with varying degrees of severity. Fracturing and loss of circulation because of overbalancing can impact the ability to circulate cuttings and to control pressure in the borehole, potentially leading to packoffs or hole collapse. In addition to the risks of borehole failure, if the downhole pressure during cementing leads to overbalance (i.e., exceeds the fracture pressure) such that the cement column cannot be supported by the formation, this can lead to insufficient cement fill outside the casing and/or a final cement level lower than planned because of cement loss under the column weight (e.g., Abbas et al., 2004; Nelson and Guillot, 1990). In shallow riser holes drilled in deep water, hole balance is especially challenging because (1) small differences in mud weight are integrated down the entire length of the riser and thus translate to large differences in absolute pressure in the borehole, and (2) at shallow depths below seafloor the difference between formation pore pressure and the least principal stress (effective stress) is generally small in magnitude (Fig. F2). For example, in 2000 m water depth, the pressure in a column of 1.03 g/cm3 mud weight (assumed seawater density) and that of a 1.07 g/cm3 mud weight is ~0.8 MPa (20.2 MPa versus 21 MPa) (Fig. F2). For typical marine sediments, this excess pressure generated by the mud column in the riser itself would exceed the lithostatic stress to ~160 mbsf. In contrast, for riserless drilling the pressure generated by the mud column would be integrated only from the seafloor downward because during drilling the borehole annulus is subject to a seawater hydrostatic boundary condition at the wellhead. Borehole pressures are typically reported as both absolute pressures and an equivalent circulating density (ECD) (units of g/cm3), which is computed as the total pressure divided by the product of depth below a reference datum and gravitational acceleration. Mud program in Hole C0009AWe conducted drilling and casing of the upper ~704 m DSF in riserless mode but used a light mud (1.04 g/cm3; for comparison, the average seawater density between mean sea level and the wellhead is ~1.03 g/cm3) to sweep the hole at intervals (Table T3). During these operations, we collected MWD data, including annular pressure while drilling (APWD), which provided a direct measurement of mud pressure at the bit. This pressure may differ from that computed from mud weight measured on the drillship because of the additional pressure that must be applied to overcome frictional resistance as the mud flows up the annulus or because of cuttings from the formation suspended in the mud that can increase its density. Mud was sampled and tested twice daily by a service company (see C0009MUD.PDF in OPERATIONS in "Supplementary material") to confirm the mud weight and chemical composition, so we believe that a significant difference between actual and intended mud weight is unlikely. For riserless operations in the upper 700 m DSF, the ECD computed from APWD measurements generally ranged from 1.13 to 1.16 g/cm3 (Fig. F3; Table T3). The pumping rate during drilling was 1000 gallons per minute (gpm). The measured borehole pressures are smaller than the lithostatic stress (i.e., the overburden) below the 54.5 m DSF surface casing (Fig. F3B). However, if the minimum horizontal stress is smaller than the vertical stress, as is likely to be the case in a basinal setting, the fracture gradient could be smaller than the lithostatic gradient. For riser drilling, the planned mud weight was 1.08–1.10 g/cm3, or greater if deemed necessary (Kobayashi et al., 2009). Drilling proceeded from 27 June to 6 July using a nominal mud weight of 1.08 g/cm3. After some difficulty with coring operations (see C0009MUD.PDF and DAILYRPT.PDF in OPERATIONS in "Supplementary material"), mud weight was increased to 1.09 g/cm3 on 6 and 7 July, and again to 1.10 g/cm3 from 8 July until the end of riser drilling operations (Table T3; see also C0009MUD.PDF and DAILYRPT.PDF in OPERATIONS in "Supplementary material"). During drilling, a pumping rate of ~700 gpm was used. In the first period of riser drilling from 27 June to 8 July, mud losses of ~10–20 m3/day were observed (Fig. F4; see also DAILYRPT.PDF in OPERATIONS in "Supplementary material"). After the mud weight was increased to 1.10 g/cm3 on 8 July, mud losses increased to ~40–50 m3/day and were sustained at this rate for the remainder of riser drilling operations. The mud losses prior to 8 July are characterized as "seepage" (losses < 1.5 m3/h), whereas those after 8 July qualify as "partial lost returns" but remain at the lower range of values for this categorization (Abbas et al., 2004). The APWD tool was not run during any phase of riser drilling. However, MDT measurement provided a direct measurement of the static mud column pressure during the wireline logging operation (Fig. F5A). The measured pressure gradient corresponds to an ECD of 1.13 g/cm3 (Fig. F5B, F5C), which exceeds the shipboard mud weight measurements of 1.10 g/cm3 (see "Downhole measurements" and C0009MUD.PDF in OPERATIONS in "Supplementary material"). This difference could be caused by cuttings load in the annulus. Because the MDT measurements were obtained in a static column, the circulating ECD is likely to have been higher still (cf. Fig. F3). The measured mud column pressure also slightly exceeds the fracture pressure equivalent to 1.10–1.11 g/cm3 measured by an LOT at the 20 inch casing shoe (703.9 m DSF) (Fig. F5) but is lower than the fracture pressure determined from an MDT test at 874.3 m wireline log matched depth below seafloor (WMSF), which yielded a value of the least principal stress equivalent to an ECD of ~1.20 g/cm3. These values are similar to the predrilling estimate of fracture gradient of 1.15 g/cm3 at 700 mbsf but lower than the predrilling estimate of 1.34 g/cm3 at 1500 mbsf (Kobayashi et al., 2009). DiscussionIn the riserless borehole drilled to 700 m DSF, measured borehole pressures are ~10% greater than those computed for a static column of the planned mud weight (~1.15 versus 1.04 g/cm3). This difference may be due to circulation pressures during drilling. Alternatively, it could be related to cuttings load in the annulus or to a difference between actual and intended mud weight. During riser drilling, the measured pressure in the borehole exceeded the pressure expected for a static column of 1.10 g/cm3 mud. This discrepancy illustrates the importance of APWD data to enable real-time monitoring of downhole pressure and ECD, especially to account for the additional downhole pressure generated by circulation. APWD can also be used as an indirect indicator of overbalanced or underbalanced hole conditions. The observed static mud weight in the riser hole during the MDT wireline run exceeded the fracture pressure determined by the LOT but lies below that measured by MDT hydraulic fracture testing. A mud pressure at or near the fracture gradient could partly explain the observed mud losses from seepage into the formation or hydraulic fracturing. Additional pressure generated during mud circulation would further increase any overbalance present. The offset between planned and actual static mud weight was only a few percent, but because of the large water depth (and thus riser length), the difference in pressure at 700 m DSF between a column of 1.10 and 1.13 g/cm3 is ~1 MPa, equivalent to ~20%–25% of the vertical effective stress at this depth. Although this is highly speculative, possible damage or fracturing of the formation may also provide a partial explanation of difficulty encountered during cementing operations for the 13⅜ inch casing, including lack of resistance, low efficiency, and apparently incomplete filling of the annulus (based on zero-offset VSP results; see "Downhole measurements") (see DAILYRPT.PDF in OPERATIONS in "Supplementary material") and could also be related to the apparent washout of the borehole below the casing shoe at 703.9 m DRF as observed in caliper and other logs (see "Logging results"). |
||
inch gauge ring was run into the hole to determine if any cement was left inside the 13⅜ inch casing. The junk basket run encountered resistance at 2075–2080 m DRF (near the wellhead). After several attempts, the junk basket was rerun with an 8¼ inch gauge ring, and the hole was found to be sufficiently clear to run the Versatile Seismic Imager (VSI) tool string. These modifications to the operations plan (running the junk basket rather than a cement scraper, as well as the cancellation of the cement-bond log) were made in part to adjust the VSP operation time to match the availability of the shooting vessel, the R/V Kairei.
inch gauge junk basket was run through the casing; however, after three failed attempts at trying to pass 2080 m DRF, the gauge ring was changed to 8¼ inch, which was able to reach 3300 m DRF. After pulling out of the hole following the junk basket run, the VSI tool string, consisting of 20 seismometers, was lowered into the hole to 2085.5 m DRF at 0730 h on 24 July. Checks of the calipers on each seismometer, which are used to couple the tools to the formation, indicated a short circuit somewhere in the string. Therefore, the VSI string was pulled from the hole at 1030 h and 4 seismometers were removed to reduce risks of further cable problems, leaving 16 seismometers remaining on the VSI tool for the walkaway VSP. The VSI tool was again lowered into the hole at 1600 h and locked in the casing from 2999.2 to 3227.8 m DRF. The Kairei deployed four air gun strings (see