As with previous paleoceanographic sampling missions, the main operational goal of Leg 202 was to recover complete stratigraphic sections at the preselected sites with as little coring disturbance as possible and as efficiently as possible. Preferred use of the APC coring system over rotary systems, coring of multiple holes at a site to ensure a complete record, and shipboard construction of stratigraphic composite sections and sampling splices became standard strategies over the last several ODP drilling legs that addressed paleoceanographic objectives. During Leg 202, we employed techniques and concepts to further optimize these strategies in order to obtain longer APC sections, to correlate data from multiple holes faster for real-time control on coring offsets and thus minimizing redundant coring, and to correlate and integrate data from cores and downhole logs.
Two coring systems were used during Leg 202: the APC and the XCB. The APC, a "cookie-cutter" type system that cuts cores with minimal coring disturbance, was always the preferred coring system. The drill pipe is pressurized to shear one or two pins that hold the inner barrel attached to the outer barrel. The inner barrel strikes out and cuts the core. The driller can detect a successful cut or "full stroke" on the pressure gauge on the rig floor.
When "APC refusal" occurs in a hole before the target depth is reached, the XCB is generally used to advance the hole. The XCB is a rotary system with a small cutting shoe extending below the large rotary bit. The smaller bit can cut a semi-indurated core with less torque and fluid circulation than the main bit and thus optimizes recovery. XCB coring disturbs the cores, as the torque of the drill rotation shears and breaks the core into segments. In this process, voids fill with drill slurry and short sections of core appear as "biscuits" that range from a few centimeters to decimeters in length. The degree of XCB disturbance depends strongly on the lithology. Disturbance is most severe in partly lithified sediments that are not stiff enough to maintain their integrity under a rotating bit. Given this disturbance, an operational goal during Leg 202 was to continue APC coring as long as possible prior to switching to the XCB.
APC refusal is conventionally defined in two ways: (1) a complete stroke (as determined from the pump pressure reading) is not achieved by the piston because the formation is too hard, and (2) excess force (>60 kilopounds) is required to pull the core barrel out of the formation because the sediment is too cohesive or "sticky." In cases where full stroke can be achieved but excessive force cannot retrieve the barrel, the core barrel can be "drilled over" (i.e., after the inner core barrel is successfully shot into the formation, the rotary bit is advanced to total depth to free the APC barrel).
During Leg 202 we generally accepted only the first APC refusal criterion (incomplete stroke) for the transition to XCB operations. Drilling over stuck APC barrels allowed us to advance many holes significantly deeper with the APC than if the second criterion (excess pullout force >60 kilopounds) had been applied (Table T3). A total of 101 core barrels were drilled over during Leg 202, which is more than during any previous leg. These drill-over operations consumed time (in some cases >1 hr per core in addition to normal operations), and in a few cases resulted in damaged core barrels. Nevertheless, we found that these investments of time and equipment paid substantial dividends in terms of enhanced APC penetration. At the Leg 202 sites dominated by carbonate-rich lithologies and characterized by low to moderate sedimentation rates (Sites 1237–1241), overdrilling increased the APC penetration by 12–126% (averaged 77%) in those 10 holes that were cored to APC refusal. At two sites, APC penetration was pushed to >300 mbsf, and at another two sites to >200 mbsf. In addition, by reserving XCB coring for sediments that were substantially lithified, we maximized the quality of recovery within these intervals, in most cases resulting in high-quality cores suitable for developing composite depth sections through most of the sedimentary sections at each site.
During Leg 202, as during several previous Ocean Drilling Program (ODP) legs with paleoceanographic emphasis, multiple APC holes were drilled at each site to ensure recovery of a complete stratigraphic sequence, despite coring gaps that are present in any one hole. A meters composite depth (mcd) scale for each site was constructed through inter-hole correlation using closely spaced core logging measurements. The mcd scale accommodates core expansion and coring gaps and can be used to define a shipboard splice, a stratigraphically continuous sediment sequence consisting of segments from different holes.
The assembly of composite depth sections with an mcd scale, in principle, can be done with two holes, but in most cases three or more holes are needed because the position of coring gaps is not uniform (and is often unpredictable) as drilling advances downhole. Variations in bit depth may be caused by tides, ship heave (which is not compensated for with APC coring), and limited precision in sensing the depth of the drill bit based on the visual sensing of pipe advances. To optimize the depth offsets between holes, it is desirable to control bit position on a core-by-core basis. This requires knowing the position of a core relative to cores in the previously drilled hole(s) in near real time such that the relative offset of the subsequent core can be estimated and communicated to the driller. Rapid core logging that keeps pace with drilling is an essential requirement of this strategy.
To provide information on core depths for real-time adjustments in bit depth, an automated "fast track" magnetic susceptibility core logger was provided by Oregon State University (OSU) for Leg 202. This track was installed in the JOIDES Resolution core laboratory near the catwalk so that whole-round cores could be analyzed as soon as they entered the laboratory. The OSU Fast Track employs a Bartington MS2 susceptibility meter that was zeroed before each section scan. The usual sampling interval was 5 cm, although intervals were adjusted as needed to keep pace with drilling. Where possible, a 1-s integration time was used. Using this setting, full cores (seven sections) could be analyzed in ~15 min. However, several of the Leg 202 sites had little terrigenous material and low magnetic susceptibility. At these sites, a 10-s integration time was needed and this increased the logging time to ~40 min per core. Even with this slow setting, core flow was significantly faster than the ODP multisensor track (MST) logging time and fast enough to keep up with drilling under most circumstances.
The availability of the Fast Track magnetic susceptibility data, in many cases, allowed us to verify complete recovery of the APC-cored interval with two holes or with a third hole that was spot cored to cover specific gaps. By minimizing redundant drilling, the Fast Track strategy effectively preserved time for deeper penetration with the APC by overdrilling and for occupying three alternate sites that added greatly to the overall scientific results of Leg 202. Use of the OSU Fast Track for most composite section development and drilling offset decisions had an additional scientific benefit, as it allowed the ODP MST to be run more slowly to optimize data quality. This strategy allowed us to increase the use of the slower MST sensors, such as the natural gamma counters, which provided a useful data set for core-log integration. Effectiveness of stratigraphic correlation and core logging would have been seriously compromised without the extra core logging track on Leg 202. Based on our experience, we recommend that future operations requiring rapid verification of complete recovery employ a similar Fast Track strategy.
A revised meters composite depth (rmcd) scale was generated to account for the differential stretching and squeezing observed between correlated cores from each hole drilled at a particular site of Leg 202. The stretching or squeezing between cores from different holes may reflect small-scale differences in sedimentation, but more commonly is denoted by distortion caused by the drilling process.
To map cores to the splice, we developed an rmcd scale at some sites using the inverse correlation method of Martinson et al. (1982). A MATLAB program (N.G. Pisias, pers. comm., 2002) was used to correlate each core from a site to the splice record for that site. For cores already included in the splice, no correlation is necessary. For cores not in the splice, the inverse technique calculates a mapping function that allows for small changes in the mcd depth scale of the core and maximizes the correlation coefficient between the core and splice records. The program allows for exclusion of portions of the core that are not well correlated to the splice or are from intervals that contain coring distortion, such as flow-in. The generation of an rmcd scale for a given core results in a table that contains meters below seafloor (mbsf), mcd, and rmcd values for any sample from that core at a resolution equivalent to the MST data used for the mapping.
Composite depth scales typically are expanded by 10%–20% when compared to drilled intervals because of the expansion of cores upon recovery as a result of elastic rebound, expansion of volatile hydrocarbons (mostly biogenic methane), and mechanical stretching during the coring process. This expansion is assumed to occur without significant uptake or loss of interstitial water per unit mass of dry sediment. If this assumption is correct, sediment densities determined by wet and dry mass and volume measurements are most compatible with those measured in situ. Calculations of mass accumulation rate (MAR) require information on both sediment density and linear sedimentation rate (LSR). LSR is typically based on the mcd scale, which suffers from core expansion and must be corrected to be compatible with density measurements.
To facilitate the calculation of MAR, during Leg 202 we established the corrected meters composite depth (cmcd) scale, which adjusts the mcd scale for empirically observed expansion. A cmcd datum is produced by dividing the mcd value by the average expansion of the mcd scale relative to the mbsf scale over a sufficiently long interval so that random variations in drill pipe advance are negligible. The cmcd scale provides a complete stratigraphic sequence that is the same length as the total depth cored. The cmcd scale is a close approximation of actual drilling depths, and unless further corrected by logging data, the cmcd scale should be used when calculating LSR or MAR.
At three sites (1238, 1239, and 1241), logging operations produced data sets that were of sufficient quality to allow for core-log integration. Core-log integration produces yet another depth scale, equivalent log depth (eld). This depth scale has the advantage that it corrects for stretching and squeezing within cores. The disadvantage is that it is rarely a complete data set for correlation of the entire hole. The eld scale typically begins at ~100 mbsf, which is where the drill pipe is positioned during logging operations. Where available and where logging data are of sufficient quality, the eld is the best estimate of in situ depth and is ideal for calculating MAR.
To determine eld, logging and whole-core MST data were imported into the Sagan software package (version 1.2) and culled as necessary to remove extraneous errors associated with voids in cores or with poor sensor contact with the borehole. Because core logging data generally have a higher resolution than downhole logging data, it is necessary to smooth the core logs before comparing them with downhole logs. Sagan allows the correlation of individual cores within different holes with the data series recovered from logging. We found that Formation MicroScanner (FMS) data integrated around the borehole using 64 buttons was particularly useful because its high depth resolution compared well with that of core gamma ray attenuation (GRA) density measurements (Fig. F7). We also found good agreement between some cored intervals, natural gamma data (Fig. F8), and gamma densities measured both in the cores and in the borehole (Fig. F9).
Because we had near real-time Fast Track magnetic susceptibility data to assess coring depth offsets between holes, we attempted to make adjustments in drilling depth as coring progressed. As we scrutinized the data to make these decisions, we noticed that the differential offsets appeared to vary systematically through a daily cycle. We hypothesized that these variations were due to variations in water depth resulting from the tides.
Site 1240 illustrates variations in drilling depths well because we APC cored the entire recovered section and were able to construct a splice that spans a 2.5-day-long time interval of operations. We obtained a deep-ocean tide prediction for the drill site courtesy of Dr. Gary Egbert at OSU. A comparison of the two (Fig. F10) shows that local tides and differential drilling offsets share a 12-hr period and have a similar magnitude (~2 m) over the period of operations. This similarity suggests that deep-sea tides produce up to a 2-m change in the offset between cores that varies within a 12-hr period. As coring proceeds at a site, tides can either add to or subtract from the depth offsets, depending on the time at which two holes are begun relative to the tidal cycle (i.e., on a rising or falling tide). We imagine that in the future tidal predictions could be used to adjust drill pipe advances as needed to maintain coring overlap in multiple holes. Effective implementation of this strategy, along with the provision for heave compensation during APC coring, would likely provide for complete recovery in two holes at a site, rather than the three to five holes per site that are typically needed in current operations.
Physical properties are the expression of the lithologic, textural, and structural variations in sediments, and, in many cases, they can be used to empirically infer these primary properties over long intervals at much higher resolution than can be practically measured using traditional laboratory-based direct measurements on discrete samples. During Leg 202, reflectance spectroscopy data were calibrated with direct geochemical measurements to provide rapid optical estimates of carbonate and total organic carbon (TOC) percentages in sediments as well as relative abundance of oxides, such as hematite and goethite, and organic pigments known as chlorins.
The reflectance spectroscopy data included 31 reflectance values over the visual spectrum (400 to 700 nm) averaged over 10-nm intervals. First derivatives of these data with respect to wavelength are commonly used to emphasize variations in spectral shape. Empirical equations were calibrated based on least-squares multiple regressions fit to discrete chemical data. Regression terms were selected with a stepwise procedure and were retained in the equations only if significant above a 95% level. The advantages of using this technique in real-time during the cruise are that the technique provides us with (1) a better understanding of sediment composition and physical properties and 2) a detection of scales of variability in carbonate and TOC that could not be attained rapidly at sea based on direct measurements with low depth resolution. Variations detected by the detailed optical data set can be targeted rapidly for manual sampling to make sure that extreme events are verified with precise chemical techniques. One example of the successful use of reflectance spectroscopy estimates of lithology during Leg 202 is the detection at Site 1237 of rhythmic variations in estimated TOC at scales of 5 to 10 m (Fig. F11). Based on preliminary shipboard age models, such variations may be associated with known ~400-k.y. cycles of Earth's orbit and suggest a climatic and biogeochemical response to this external forcing.
A new digital imaging system (DIS) implemented by ODP in 2002 and installed on the JOIDES Resolution during Leg 200 was used routinely for the first time during Leg 202. Essentially, all cores were imaged with this tool. As would be expected with a new, custom-made measurement system, we encountered some problems and came up with some recommendations for the operation of the system and use of the data, as well as with some recommendations for future improvements that would maximize the benefit of such measurements at sea. Although we found the digital images extremely useful for purposes of basic visual description (especially for postdescription review when the cores were not easily available for viewing), we also found that the digital image calibration was not sufficiently stable or reliable when using the digital data as a measure of quantitative brightness or color.
We found systematic millimeter- and centimeter-scale stripes in the digital images that result from the system's attempt to adjust image exposure based on calibration pixels at the site of the digital array. These pixels were inadvertently aligned with millimeter and centimeter marks on reference rulers along the sides of each core section. This problem can be mitigated by careful alignment of the camera or by using low-contrast markings on reference rulers that will not trigger automatic adjustments in camera calibration.
Significant variations in the brightness of digital images from the top to the bottom of core sections occurred because of variations in ambient lighting in the laboratory. Such ambient lighting effects may be corrected in one of three ways: (1) by enclosing the imaging system in a sealed box that shades the core sections from ambient light, (2) by increasing the brightness of the lamps that illuminate the core sections so that ambient light is overpowered by a constant source, or (3) by modifying the system to move core sections under a fixed camera so that ambient light is effectively constant rather than having the system move the camera over fixed core sections as the system is now designed.
By analyzing standard materials, we also found that small (millimeter scale) variations in camera-to-core distance (which inevitably result from irregularities in cut core surfaces) result in variations in the brightness of core images. Such an effect could be mitigated by increasing the distance between the light source and the camera system relative to the core surfaces so that small variations in the core surface are negligible relative to the total system geometry. Increases in the core-to-lamp distance, however, work against the goal of increasing light intensity to overpower ambient light, and this effect must also be considered.
Significant temporal drift in the calibration of the digital camera system relative to white standards dictates frequent recalibration of the camera system. Ideally, such calibrations should be done automatically between each run. Calibration standards could include multiple gray levels that would compensate for nonlinear responses of the camera system across its full range of brightness.
Finally, we found it awkward to set the camera's aperture and focus manually in sediment sequences that vary significantly in their reflectance or geometry. Although manual settings are fine for occasional measurements, a greater degree of automation in camera settings and calibrations would be beneficial for ODP operations in which the camera is used continuously for long periods of time by a variety of operators who are working under stressful conditions at sea.