Back to Geoclutter
R/V Cape Henlopen Cruise 01-17
Science Party:
John Goff (Chief Scientist), University of Texas Institute for Geophysics (UTIG)
Larry Mayer (Co-Chief Scientist), University of New Hampshire (UNH)
Chris Sommerfield (Co-Chief Scientist), University of Delaware (UD)
Robert Burger (UTIG)
Sean Gulick (UTIG)
Sylvia Nordfjord (UTIG)
Eric Jabs (UNH)
Eric Lynskey (UNH)
Barbara Kraft (UNH)
Peter Simpkin (IKB Technologies)
David Walsh(UD)
Andrew Klingbeil (UD)
The primary objective of CH01-17 was to characterize surficial sediment properties within the Office of Naval Research’s Geoclutter natural laboratory on the outer New Jersey shelf. The Geoclutter program is investigating the nature of target-like acoustic returns that are seen on low frequency anti-submarine warfare (ASW) acquisition systems, but which arise from the natural environment. The New Jersey shelf was chosen for Geoclutter because it was previously a natural lab for ONR’s STRATAFORM program, and thus was the focus of considerable prior data collection and interpretive efforts, including high resolution multibeam and sidescan, ultra-high resolution seismic data (e.g., Huntec, CHIRP), and coring efforts. A recent acoustic reconnaissance survey of the Geoclutter natural laboratory using ASW-type equipment discovered an abundance of geoclutter targets. Furthermore, given the very low slopes of most of the bathymetry over the target areas, it evident that such geoclutter is arising from sources in the subbottom. Our investigation of seafloor sediment properties will support efforts to acoustically model geoclutter by providing the boundary characterization for acoustic penetration into the subbottom.
CH01-17 is one of three geological and geophysical field efforts within and in conjunction with Geoclutter designed to support acoustic modeling of reverberation data from potential geoclutter targets. Accurate modeling of the physics of geoclutter will require accurate characterization of the scattering environment. Of primary concern is the structural geometry of potential scatterers (e.g., shallowly buried river channels) and measurements of the contrasting sediment properties that give rise to strong impedance contrasts that are the source of geoclutter targets. Subbottom structure will be determined from data collected recently during a the EN359 cruise aboard the R/V Endeavor, which included densely spaced CHIRP seismic reflection data over primary geoclutter targets identified during the acoustic reconnaissance survey. Subbottom sediment properties will be ascertained from long coring work to be conducted in 2002 using the heave-compensated GLAD800 system aboard the R/V Knorr. This later effort is an outgrowth of the STRATAFORM program.
Another critical characterization is the sediment/water interface. Because ASW sonars project acoustic energy horizontally, interactions with the seafloor are predominantly at very low grazing angles. Small variations in the velocity and impedance contrast at the sediment/water interface can have a large effect in determining how much, or indeed if any, acoustic energy penetrates into the subbottom to interact with potential geoclutter targets. CH01-17 was designed to address this concern, by collecting sediment at the surface for grain size, porosity and density measurements, and by collecting in situ acoustic velocity measurements. Sediment samples were gathered from the seafloor and shallow subsurface using grab sampling and short coring techniques. Velocity measurements were collected using the new Geoclutter Velocity Probe developed by UNH.
In addition to directly supporting Geoclutter acoustic goals, CH01-17 was designed to support geological characterization of the natural laboratory setting, in conjunction with the CHIRP data collected during EN359. These data will help address important questions regarding the geologic evolution of the seafloor during the latest sea level rise, particularly the role of erosive modification to the seafloor in the outer shelf setting. These data will also be used to ground truth the 95 kHz acoustic backscatter map of the region, collected during the STRATAFORM program. This analysis will help us to extrapolate sediment properties determined at the individual station locations to the broader region.
Prior to the cruise, 150 station locations were identified for possible sampling and velocity probe work. Most of these stations were located along 6 transect lines, for later comparison with CHIRP seismic data to be collected on the subsequent EN359 cruise (these lines were subsequently collected successfully). In selecting station locations, particular attention was paid to sampling in the vicinity of high priority geoclutter targets, as identified by Nick Makris’ analysis of the acoustic reconnaissance data collected in April, 2001. More broadly, stations were chosen to sample a wide range of backscatter and bathymetric features as seen in the swath multibeam and sidescan data. Of these 150 stations, it was expected that no more than 125 would be sampled under ideal conditions and circumstances; we anticipated the need for flexibility in decision making at sea regarding selection and ordering of stations. The initial numerical ordering of the stations reflected our preliminary assessment of priority and transit logistics.
Prior to the cruise, we anticipated collecting both a short core and a velocity probe measurement at each station visited, using the Smith-Macintire grab sampler as a backup. However, the short core proved problematic under adverse sea state and when sampling sediment with significant shell content (see below). The grab sampler was therefore used as our primary sediment collection tool. A resistivity probe was used on both grab samples (prior to sediment collection when sea water remained in the sampler) and short cores, to be converted both to porosity and density properties. We were able to situate the corer and velocity probe on the Henlopen’s deck such that the corer could be deployed from the stern A-frame and the velocity probe from the port J-frame. Initially, the grab sampler was deployed from the A-frame as well, but performance and safety were greatly enhanced when it was moved to the starboard CTD winch. We thus stretched the Henlopen’s logistics to their limit, but where able to efficiently deploy all three devices at each station.
Although delayed initially by weather, and then by difficulties with both sampling devices, we visited 99 of the pre-selected stations, collecting velocity measurements at all 99, grab samples at 98, successful cores at 25 and core-derived hand samples at four others (Figure 1). All high priority geoclutter targets were sampled, as were primary targets for investigating backscatter and morphologic variability. The nearly perfect performance of the In situ Sound Speed and Attenuation Probe (ISSAP) is particularly noteworthy, as this was the first use of this recently developed instrument. We anticipate that data collected by the probe will represent a fundamental advance in the accuracy and reliability of in situ velocity measurements of seafloor sediments.
Figure 1.
Locations of sample stations visited.
Background image is STRATAFORM sidescan (lighter shades = higher
backscatter) and bathymetry (contours in meters), from Goff et al. (1999).
Although the location of all deployments were recorded by watch standers, the bridge also kept track of position by marking on-bottom locations with the ship’s differential GPS system, and then later recording them in the ship’s log. We consider these marks to be more accurate than the watch stander log because they more precisely match the moment that each instrument hit the bottom, and so these locations are given in the table of deployment results given below.
In support of the Geoclutter program, we have developed, built, and deployed a relatively inexpensive, robust, ship-deployable device - ISSAP (Figs. 2, 3) – for measuring sound speed and attenuation in near-surface sediments.. The system uses four 2.54 cm (diameter) by 20 cm long probes that are inserted into the seafloor by 180 kg of reaction weight deployed on a coaxial cable free-swinging within a protective tripod that allows the probes to be inserted vertically on slopes up to 20 degrees. The probes operate at frequencies of either 50 or 100 kHz and are arranged in a diamond shaped pattern with nominal path separations of 20 or 30 cm. An onboard computer and topside electronics control the paths selected and the number of measurements per path. A typical deployment involves measurements across five paths including both long (30 cm) and short (20 cm) paths. Along with the acoustic probes, the ISSAP also has a color video camera that provides imagery of the seafloor and the probes as they penetrate, a 50 kHz altimeter to independently monitor height off the bottom, and temperature, pressure, pitch, roll, and heading sensors to monitor the stability and orientation of the platform. Finally, a bottom sense switch provides yet another indication of the platforms height above the bottom.
Figure
2. The ISSAP (In situ Sound Speed and
Attenuation Probe) during recovery
Fig. 3. Block diagram of major components of ISSAP – insert on the bottom left is detail of probe.
The transmit and receive pulse for each measurement is sent up the coax and digitized at 2 MHz on the topside acquisition computer and sent to a processing computer. An entire measurement cycle (150 measurements) takes less than one minute and results in approximately 75 Mbytes of data. A typical station (2 bottom water and two sediment cycle) produces about 300 Mbytes of data. The fundamental measurement is that of the travel time (time-of-flight) between the transmit and received pulse. Travel times are determined by several methods (threshold, zero-crossing and correlation techniques) and converted to sound-speed through a calibration process. There are two levels of calibration available. The most precise involves collecting data in distilled water at a known temperature and using the well-established variation in sound speed with temperature to precisely determine the separation of each pair of transducers. This is done at the beginning, end and several times during a cruise. We also carry out an ongoing calibration by measuring the speed of sound in seawater (at known temperature) before and after each penetration into the seafloor. These bottom water calibrations also allow us to determine if the insertion of the probes into the bottom resulted in a change in their relative path length.
Along with measurement of time-of-flight (and thus sound speed) we can also compare the digitized pulses to attempt to measure sediment attenuation. Several approaches will be used to measure attenuation. The relative amplitude of the received wave forms over the different path-lengths is one indication of attenuation as is the spectral ratio (or difference) between the sea water received waveform and the sediment received waveform. Finally we will use the filter correlation technique of Courtney and Mayer (1993) that was developed especially for short time series of the type we are measuring.
During Henlopen Cruise 01-17 the system
performed flawlessly recovering water column and sediment data at 99 stations
selected to represent a range of seafloor backscatter types. More than 40 gigabytes of digital data was
collected as well as more than 20 hours of video. While the detailed analyses of these data have only just begun,
initial observations indicate that the data are of excellent quality . Calibration runs resulted in a standard
deviation of .354 m/sec. Overall
variation of the bottom water measurements over the 99 stations was less than 4
m/sec with standard deviations being less than one meter per second at any
given site (well within the expected change due to small variations in bottom
water temperature) and indicating that
the system geometry remained constant and timing precise. Real and substantial variations in seafloor
sound speed were measured (station values ranged from a low of approximately
1570 m/sec to highs of over 2000 m/sec) with an apparent (an not unsurprising)
correlation between increased sound speed and higher backscatter. The system
was deployed in sediments ranging from muddy, silty sands to gravels and shell
hash deposits with a video record of each deployment site also recorded.
While we were uncertain that we would be able to also measure attenuation when we proposed the development of the ISSAP, the clean recovery of well developed waveforms gives us confidence that we will also be able to extract attenuation values at most if not all of the stations. Most importantly, the tremendous redundancy of our measurements at each station (approx. 300 measurements) will allow us to put well grounded confidence limits on our measurements and thus understand the true local variability of sound speed and attenuation in the Geoclutter area
The Smith-Macintyre grab sampler (Fig. 4) consists of a matched pair of jaws, spring loaded in the open position and triggered to close upon bottom contact. Initial attempts to deploy it from the side extension of the A-frame proved logistically difficult and somewhat unsafe due to poor line of sight between the winch operator and the technician deploying and recovering the device. In addition, our success rate was initially very poor, and the logistics of moving the trawl wire from the corer to the grab sampler were cumbersome. After a day and a half of difficult use, the grab sampler was moved to the CTD winch, primarily for safety reasons and also because the A-frame sheave was obnoxiously squeaky. We were pleasantly surprised to have 100% success rate after the move. We believe this was largely a result of the less jerky nature of the CTD winch and the much thinner size of the CTD wire, which allowed clear detection of bottom hits. New confidence in use of the grab sampler provided us with impetus to revisit a number of early and high priority stations that we had failed to collect samples at.
Figure 4. The Smith-Macintyre grab sampler during
deployment
The Smith-Macintyre grab sampler typically collected between 8 and 14 cm in depth of seafloor sediment, usually with a well-preserved water/sediment interface and with sea water still contained above the sampled sediments. Upon retrieval of the grab sampler, a resistivity probe was inserted into the sample, after first collecting a calibration value in the retained seawater, and then recording resistivity values approximately every 2 cm to the bottom of the center of the grab sample. Sediments were then extracted and bagged for later analysis using a syringe-style plug corer. In well preserved cores, subsamples were bagged in ~4-5 cm depth intervals. On occasions where there was an abundance of shell hash or gravel, an extra hand sample was collected, primarily to increase the total mass used to estimate weight percent of such coarse material. After bagging samples, the grab sample was examined by hand and a description was logged. In a number of outer shelf samples, a small number of well-rounded stones where found among otherwise very fine grained sediments. We presume that these may represent drop stones, and collected these by hand, where found, and bagged them separately.
Bagged samples have been transferred to UTIG, where coarse and fine fractions will be separated, and the coarse fraction analyzed for grain size distribution using a settling tube. A representative sample of the fine fraction will be preserved and sent to UD for eventual completion of grain size analysis using a sedigraph.
A hydraulically damped gravity corer (a.k.a. the Slow Corer; Figure 5) was used to collect undisturbed cores of the sediment-water interface and surficial strata for aboard-ship porosity determinations and subsequent sedimentological analyses in the laboratory.
Figure 5. Slow corer recovered with nearly full core barrel.
The overall success of the coring effort was mixed; sample recovery was highly dependent on bottom type, which varied considerably throughout the study area. Attempts to core the medium-grained, shell- and gravel-rich sands that characterize the mid-shelf (50-70 m water depths) failed, because large shell fragments became caught between the cutting edge of the core barrel and the sealing mechanism and caused the sample to wash out during instrument recovery. On the other hand, the corer performed flawlessly in the fine-grained sandy and muddy bottoms that characterize the outer-shelf areas (70-130 m water depths). It is worth noting that the extremely dense relict clays of the northeastern portion of the study area were aptly sampled by the corer. This material is so consolidated that is resists deformation when squeezed by hand. Core lengths ranged from 20 cm to 88 cm with an overall average of about 30 cm.
The sediment-water interface was well-preserved in all of the cores (Figure 6). Numerous small (5 mm long) polychaete worm tubes and a 0.1-1 cm thick low-density fluff layer typified the interface. Burrowing macrobenthos were not apparent in any of the samples.
Continuous electrical resistivity profiles were measured at 1 mm vertical resolution aboard ship for 19 of the 25 cores recovered; shell beds precluded measurements in six cores. Measurements were made from the core top within the barrel, bottom water in place. Sediment resistivity was converted to porosity following the well-known relationship Rs/Rw = f -n, where f is the porosity of the sample (expressed as the ratio of the volume of void space to the total volume of the sample), Rs is the resistivity of the saturated bulk sediment, Rw is the resistivity of the porewater (i.e., bottom water retained in the core), and n is a constant (1.5-2.0 for sands and dense clays). Accordingly, maximum and minimum sediment porosities in cores were 70% and 35%, respectively, with downcore profiles indicative of compaction (net decrease in porosity) and grain size changes (variable porosity). Importantly, intra-station core top and grab-sample porosities were observed to converge, implying that the grab sampling had a negligible (if any) effect on seafloor porosity properties. Following the porosity measurements, cores with overlying water were stored upright at room temperature prior to transferal to the marine sedimentology laboratory at the University of Delaware.
Figure 6. Slow core from Station 33
In the laboratory, all cores were split lengthwise with a circular saw and piano wire for physical description, photography, and X-radiography. Core half-rounds were then sealed in plastic bags, labeled, and placed in a refrigerator set to 4°C. As planned, one core halve will be analyzed for sediment grain size with the other retained as an archive.
The instrumentation developed for, and samples and measurements collected during CH01-17 provide the basis for a number of research projects and corresponding publications.
UNH will take the lead on:
- The development of the velocity probe.
- Investigation of acoustic velocity as a function of sediment type.
- Quantitative ground truth of 95 kHz backscatter map.
UTIG will take the lead on:
- Mapping of seafloor sedimentary properties within the Geoclutter region, and providing such information to Geoclutter acousticians.
- Geological characterization of the seafloor, in conjunction with analysis of EN359 CHIRP data.
- Modeling of CHIRP seismic reflection using synthetics based on in situ velocity, grab sample and physical properties data.
- Characterization of sediment variability – a collaborative effort with the ONR “Characterization of Uncertainty in the Natural Environment”
UD will take the lead on:
- Sediment grain-size analysis of cores
- Analysis of porosity data with respect to sediment type and physical diagenesis.
There will be also an opportunity to compare three different collocated measurements of the acoustic velocity, impedance and attenuation of seafloor sediments. Steve Schock (Florida Atlantic University), using a newly developed refractive CHIRP methodology, collected such data over our central (stations 4-33) and northern (stations 129-141) transects during EN359. Acoustic frequencies used during this experiment ranged from 1 to 24 kHz. In addition, Charles Holland (Pennsylvania State Univ.), during the Geoclutter acoustic reconnaissance survey, used a moored vertical array (DUSS) for a wide angle reflection and bottom loss experiment in the vicinity of our station 21. The useable frequency band for this experiment was ~400 - 6000 Hz. The principals in these measurements have all expressed an interest in comparing their measurements derived from such disparate methods and frequencies, and it is anticipated that they will collaborate in doing so.
References
Courtney,
R.C. and Mayer, L.A., 1993. Calculation
of acoustic parameters by a filter correlation method, J. Acous. Soc. Amer, v.
93, pp. 1145-1154.
Goff, J. A., Swift, D.
J. P, Duncan, C. S., Mayer, L. A., and J. Hughes-Clarke, J., 1999. High resolution swath sonar investigation of
sand ridge, dune and ribbon morphology in the offshore environment of the New
Jersey Margin. Mar. Geol., v. 161, pp.
309-339.