Introduction

The U.S. Geological Survey (USGS) Coastal and Marine Hazards and Resources Program has actively collected geophysical and sedimentological data in the northern Gulf of Mexico for several decades. Shallow subsurface geologic data collected from this area have been recorded as high-resolution seismic profiles (HRSPs). Prior to the early 1990s, most HRSP data were recorded on paper rolls in analog format as continuous seismic profiles (Bosse and others, 2020b). Over 75 percent of the USGS holdings of shallow geologic seismic reflection data collected from the Outer Continental Shelf in the Gulf of Mexico exist as large-format paper records, which reside in a repository at the USGS St. Petersburg Coastal and Marine Science Center (SPCMSC). The USGS, as part of the National Geological and Geophysical Data Preservation Program (https://www.usgs.gov/programs/national-geological-and-geophysical-data-preservation-program), and in collaboration with the Bureau of Ocean Energy Management (BOEM), has been recovering these data through well-established digitizing and data processing procedures (see Bosse and others [2017a] for methodology). The data are converted to digital format in order to be more easily accessible and useable for end-users. The data are converted to the open-standard Society of Exploration Geophysicists revision Y (SEG-Y) format, and the recovered data are then published online as USGS data releases and data series reports (Sanford and others, 2009a, b; Bosse and others, 2017b, c; Bosse and others, 2018a, b).

Beginning in 2017, the SPCMSC, in collaboration with BOEM, identified and converted a large-scale geophysical investigation that was conducted in Outer Continental Shelf waters offshore of Texas and Louisiana (fig. 1). This dataset was collected in 1980 by the geophysical survey company Intersea Research Corporation under contract to the USGS. The HRSP data were collected using Ocean Research Equipment “boomer” and Edgerton, Germeshausen, and Grier “sparker” HRSP systems. An example of the seismic profile is shown in figure 2. There are 202 seismic lines in the dataset that extend approximately 2,000 linear kilometers, also referred to as line-kilometers (fig. 1), and cover an area of 13,000 square kilometers (km²). Information on equipment, navigation, and survey logs are included in the Intersea (1980) report. The recovered data from this survey are published in a USGS data release (Bosse and others, 2020a), and include the associated navigation, printable and georeferenced imagery, metadata, and original scans of the logbooks. This dataset represents the largest recovery effort by the SPCMSC to date and covers an area where existing near-surface HRSP data are sparse. The dataset (Bosse and others, 2020a) is referred to as “Intersea” throughout this report.

The purpose of this project is to evaluate and interpret the HRSP data in order to characterize the shallow geologic framework of the Gulf of Mexico Outer Continental Shelf. Numerous geophysical studies in the northern Gulf of Mexico have identified specific lithologic and morphologic features that are related to pre-Holocene fluvial and Holocene transgressive marine deposits (Bartek and others, 2004; Flocks and others, 2011; Hollis and others, 2019). Through analysis of the HRSPs, near-surface structures such as buried Pleistocene fluvial channels; relic landforms; and sea floor morphologic features such as shoals can be identified. These kinds of features are important to identify in the near-subsurface because of the potential for uses as sediment resources, and they contain volumes of material suitable for coastal management projects such as beach nourishment (Flocks and others, 2009). From this analysis of the HRSP data, the project developed a sediment-coring strategy that can be implemented in similar investigations of these resources.

Methodology

Data Integration and Static Adjustment

The seismic data visualization and interpretation software OpendTect version 6.0 (OD; https://dgbes.com) was used to analyze and interpret the HRSP data. Within the OD software environment, a project study area was set up to incorporate the HRSP lines (fig. 3). The spatial dimensions for the project are in Universal Transverse Mercator (UTM Zone 15, in meters) for the horizontal axes, and two-way traveltime in milliseconds for the vertical axis. A sound velocity of 1,500 meters per second (m/s) was used to characterize the signal speed within the sediments so that the target depth could be determined, in meters. This velocity is consistent with speed-of-sound measurements within shallow marine sediments (Zheng and others, 2016).

The HRSP data were collected using two different kinds of sound sources that create soundwaves in the water column through displacement. One system uses an electromagnetic plate to mechanically displace the water (boomer system), and the other uses an electric spark to create an explosion (sparker system). The sound receiver is a string of hydrophones (streamer) adjacent to the sound source. The seismic reflection profiles in this study contain characteristics consistent with data acquired in marine environments. The first energy peak (reflector) in an HRSP image, known as the first arrival, is the direct transmission of the acoustic pulse from the sound source to the receiver and does not represent an actual seismic reflector (fig. 2). The profiles then exhibit discontinuous or chaotic reflectors that are attributable to water column noise prior to the first real seismic reflection that represents the seafloor. A seafloor signal can also be repeated within the seismic profile at exactly two times the distance from the seafloor to the sound source. This repeated signal is known as a multiple, which can obscure the seismic reflections at depth. The combination of first arrival, acoustic noise, and multiples degrade data quality in shallower water depths of the study area. Where signal degradation is not an issue, data are of good quality and are suitable for the identification of buried near-surface and seafloor morphological features.

The depth to the seafloor ratio, or the distance between the top of the profile and the seafloor, in the HRSPs is not consistent across each trackline. This is due to the boomer and sparker sensors being flown at different depths within the water column, daily tidal variations, and distance between the sensors and positioning reference (layback). The sound source for the Ocean Research Equipment boomer system is mounted on pontoon sleds and towed at the water surface, whereas the sparker systems are towed within the water column. Because digital HRSPs in this project were scanned and transformed from analog plots of the original data, the vertical scale from plot to plot may differ, as well as the position of the seismic profile on the paper roll. To adjust each seismic profile to a common vertical datum, each HRSP needs to be integrated with a suitable bathymetric surface. In this study, a 3-arc-second digital elevation model (DEM) of the bathymetry in the study area was derived using sounding data available from the National Oceanographic and Atmospheric Administration (NOAA) Bathymetric Data Viewer website (https://www.ncei.noaa.gov/maps/bathymetry/). The sounding data were acquired from approximately the same time period as the Intersea data were acquired (1980s). The DEM was projected to UTM Zone 15 (in meters) and inserted into the OD project for visualization (fig. 4).

To adjust each seismic survey line to the NOAA bathymetry, a two-dimensional (2D) profile of the NOAA bathymetry was projected onto each digital HRSP line. The HRSP line navigation was imported into the geographical information system software QGIS version 3.0 (https://qgis.org/en/site/) along with the NOAA DEM. The Join Attributes by Nearest tool in QGIS was used to affix a bathymetric depth value in meters to each seismic-trace navigation position. The seismic-trace navigation and associated bathymetric depth values were then reimported into OD as 2D horizons. Using the Digitize tool in OD, a 2D horizon representing the seafloor seismic reflector was interpreted for each HRSP (fig. 5A). Once the seafloor for each HRSP was determined, a script was developed that used the delta-resample algorithm in OD to adjust the HRSP by aligning the seafloor seismic reflector with the 2D horizon that represents the NOAA bathymetry (fig. 5B). The delta-resample script extracts the difference (delta) in elevation between the digitized HRSP seafloor reflector and the water depth from the NOAA bathymetry and adjusts each trace in the HRSP by the delta value:

$$Delta=(Z_0\times C_0)-(Z_1\times C_0)$$

,

where

Z₀

= two-way traveltime depth to digitized seafloor in HRSP, in seconds;

Z₁

= depth to NOAA bathymetry at location of Z₀, converted to two-way traveltime, in seconds, using a sound velocity of 1,500 meters per second; and

C₀

= seconds to milliseconds conversion value; constant=1,000.

Because the goal of this project is to identify potential near-surface sediment resources, it was not necessary to include the entire profile (0.5 seconds or about 350 meter [m] depth) of data; therefore, during this step, the HRSPs are also windowed to the top 100 milliseconds of data, or to approximately 75 m below the water surface using a sound velocity of 1,500 m/s (see data window in figure 2 for extent of window). Once the seismic reflectors in each HRSP are consistently depth-referenced across the study area, key reflective horizons can be interpreted.

Interpretation

The Intersea seismic profiles are suitable to resolve seismic reflectors deep within the seismic section. High-angle reflectors are readily visible and likely represent surfaces associated with Pleistocene fluvial systems that extended across the shelf during sea-level lowstands. In the uppermost seismic reflectors, reverberations from the high-power output of each acoustical system create a “ringy” signal seen in seismic reflection patterns that sometimes obscures the near-surface seismic stratigraphy. This ringy signal was especially seen in profiles from the shallowest water depths of the study area. However, not all near-surface features and morphology were obscured in the seismic profile. Various geologic features can be identified in the seismic stratigraphy that may have potential as sediment resources, and based on water depth and amount of overburden, these features may also be considered as targets for direct sediment sampling.

Throughout the study area, a consistent seismic reflector is seen at depth in almost all the HRSPs (fig. 6). This seismic reflector roughly parallels the seafloor and varies from 3 m below sea level in the nearshore portion of the study area to 50 m below sea level in the seaward portion of the study area (fig. 7). The persistence and crosscutting nature of this seismic reflector suggest that it may represent a transgressive surface, which is an erosional feature created by sea level rise. Whether this seismic reflector represents a transgressive surface that resulted from the most recent sea level rise (Marine Isotope Stage 1; Anderson and others, 2014) or an earlier event cannot be determined without direct sampling of the stratigraphy and age determination. This seismic reflector was digitized in OD to create a 2D representation of the surface in each trackline. The 2D horizons were then used to generate a three-dimensional grid. Because of the very short distance between individual traces in the HRSP (less than 1 m), as opposed to the very long distances between tracklines (approximately 5 kilometers), common gridding algorithms, such as kriging, fail to interpolate a representative surface. To produce the grid, a fourth-order trend algorithm was applied to the digitized depth values of the transgressive reflector. The Trend tool in ArcGIS software (v. 10; Esri, Redlands, Calif.) forces a smooth surface through the input sample points to interpolate the regional pattern of the transgressive surface. This output provides an adequate representation of the surface over large areas because the transgressive surface over the 13,000-km² study area is relatively flat.

The stratigraphy above this seismic reflector represents the most recent Holocene marine deposits. Subtracting the grid of the possible transgressive surface from the DEM of the seafloor provides thicknesses (isopachs) for these Holocene deposits (fig. 8), which range from 1 to 8 m in thickness. Because the possible transgressive surface is relatively flat with respect to the seafloor, the isopach map reveals positive morphologies on the surface, such as shoals, which are commonly targeted in the northern Gulf of Mexico for sediment resources (Flocks and others, 2009). The area of greatest sediment thickness, which is located in the northeastern part of the study area, is related to deposits described later in this section but may also include spurious gridding artifacts because the shallow water depth in this area restricted access during the Intersea survey (see figures 1 and 7 for trackline locations).

The shoaling in the northeast portion of the study area (fig. 8) is associated with a wedge deposit that is apparent in the HRSP from this area. Figure 9 shows a relatively level pattern of subsurface reflectors from HRSP Intersea 336b trackline, with the seafloor sediment thickness above this reflector increasing from approximately 7.5 m at the seaward end of the profile to 13 m nearshore, thus creating a wedge-shaped deposit. The isopach map of this deposit also shows how the deposit thickens landward (fig. 10). There is a gap in the HRSP coverage east of this feature because the water was too shallow for operation of the survey vessel and equipment used to collect the Intersea data. The decrease in water depth in this area that created the gap in HRSP coverage is a result of the shoaling stratigraphy represented by the pattern of reflectors shown in figure 9. Existing HRSPs farther to the east of this shoaling feature are of poor quality, so it is not possible to track the eastern extent of the feature; thus, the complete spatial extent of the shoaling feature cannot be determined with existing data. Studies have identified large shoals in this area, in particular Trinity Shoal (Edrington and others, 2010). The extent of Trinity Shoal is shown in figure 8 and correlates with the shoaling reflectors from the HRSP.

In the northwest portion of the study area, there is significant shoaling (fig. 8). These shoals are derived from sediments reworked during a transgression associated with the now-infilled Pleistocene fluvial valley of the Sabine River (Dellapenna and others, 2011; Anderson and others, 2014). The largest shoal, Sabine Bank (fig. 11), has been extensively studied for potential sediment resources (Dellapenna and others, 2011; Rodriguez and others, 1999; Forde and others, 2010). Other shoals apparent in the Intersea dataset are not as well known. Identifying and digitizing the depth to the base of these shoals in the HRSP (fig. 11) provides estimates of shoal thickness (fig. 12). The shoals around Sabine Bank vary in surface expression and thickness (fig. 13). Although not as large as Sabine Bank, the other shoals in the area have potential as sediment resources.

Coring Strategy

In order to groundtruth the seismic reflection interpretations and apply sediment type to the seismic stratigraphy, direct sampling of the sediments is necessary. To identify shallow geologic and seafloor morphologic features in marine environments, 9-m-long sediment cores are commonly collected using a vibratory system (vibracore). Analysis of vibracores can include visual interpretation, sediment grain size, and other analytic procedures that provide data and insight into the sedimentology and stratigraphy. Seismic reflectors represent changes in density between lithologic units (for example, a transition from sand to silt) and are correlated with stratigraphic horizons. Once the seismic reflectors have been correlated to the sedimentology and stratigraphy in a vibracore or vibracores, the stratigraphic units can then be extrapolated across the study area using the HRSPs. The distribution of deposits suitable for sediment resources (for example, sand-filled units) can be identified, and the sand content and volume of these units can be quantified.

The following criteria had to be met for sediments within the study area to be considered suitable for further exploration and sampling by vibracore to determine if they are appropriate as a sediment resource: the target sediments must be at the seafloor or have minimal (less than 3 m) overburden of nonsuitable sediments (for example, silts and clays), and the maximum water depth, or combined water depth and overburden thickness, must be less than 33 m. These criteria were then applied to the selection of potential vibracore sites, although for exploratory purposes, some sites were selected that do not fit all the criteria. Also, areas where previous investigations used vibracore sampling (in particular, Sabine Bank) were avoided. The proposed coring strategy is broad and at a reconnaissance level, so that areas will be marked by one or two cores that are centered on the HRSP to provide ground-truth to the seismic reflectors. The purpose of the exploratory coring attempt is not to provide the high-resolution stratigraphic characterization necessary for engineering and design of borrow sites, but rather to investigate large-scale features that can be delineated by using HRSPs. Figure 14 shows potential core site selections based on the criteria needed for selection as discussed above, and targets positive morphologic features (for example, shoals) and transgressive sediment deposits. Sediment thickness values for each relevant sediment deposit were extracted from the isopach maps at each potential vibracore site, and these values are shown in table 1.

Discussion and Conclusion

Approximately 2,000 line-kilometers of HRSPs were collected in the northern Gulf of Mexico offshore of Texas and Louisiana in 1980 for use in geologic framework studies. Results from the shallow geologic framework interpretations presented in this report show that these data are suitable for identifying sediment resources and for sediment resource investigations, particularly for features such as shoals, which are commonly associated with sandy deposits that can be used in shoreline restoration and management efforts (Flocks and others, 2009).

Because of the ringy nature of the seismic data collected in this area and other noise inherent in this HRSP dataset, the shallowest buried fluvial features typically are obscured and difficult to identify. Higher-resolution technology, such as broadband swept-frequency systems (for example, chirp), is better suited to detect these features in the shallow subsurface. However, most chirp systems do not provide an equal amount of acoustic energy to the substrate, and deeper features, such as the thalweg of a fluvial system, may not be adequately imaged.

To develop an investigative coring strategy, this study focuses on identifying features seen in the HRSPs that are at the seafloor or in the shallow subsurface seismic stratigraphy that may represent shoals and other transgressive deposits in less than 30 m of water. Because numerous scientific investigations of Sabine Bank have been conducted, that area of interest was excluded from our analyses and is not included in this report.

The transgressive surface identified in HRSPs in this study may have suitable sand content in areas where the sediments have accumulated into thicker deposits relative to the surrounding environment. Sediment accumulation occurred mainly in the central portion of the study area, which is where the site selections for potential vibracoring target the thickest deposits (sites 1–8 in fig. 14; table 1). The shoreward-thickening deposits (wedge of sediments) in the northeast study area (sites 28–31 in fig. 14, table 1) are of unknown composition. The positive expression of the deposits on the seafloor, up to 13 m (fig. 9), is commonly consistent with sand deposits such as remnants of former barrier islands or sandy spits (Twichell and others, 2013). However, in former fluvial environments, these features can also be associated with fine-grained deltaic stratigraphy, such as prodelta muds; therefore, sediment sampling is necessary to evaluate these deposits across the study area. The seismic reflectors that represent this deposit thin and pinch out at the southern end of the deposit (fig. 15). This thinning of the deposit is shown by the contours in figure 10 that decrease from north to south. Below this deposit are additional seismic reflectors that may represent beds that are fluvial deposits (fig. 15). An opportunity to sample the underlying stratigraphy in an area where the deposit is not covered by the wedge sediments is located at sites 25–27 (fig. 14; table 1).

Throughout the study area, there are deep seismic reflector patterns consistent with buried Pleistocene fluvial systems (fig. 15). These seismic reflector patterns have been identified in numerous studies across the Gulf of Mexico (Kindinger and others, 1994; Bartek and others, 2004), and when these areas have been sampled to determine the associated sedimentology, the sediments were composed of sandy material. Sediment-resource studies target these fluvial systems where they are near the seafloor (Flocks and others, 2009). In the Intersea HRSP dataset, there is commonly a near-surface artifact that is due to the ringy nature of the boomer and sparker acoustic sources used in the 1980 survey. These kinds of artifacts can obscure the geologic structure, and because this artifact is prevalent throughout the Intersea data, it makes it difficult to identify any fluvial systems within vibracore sampling depth. However, a potential subsurface target that may represent fluvial deposits has been identified in the HRSP at sites 23 and 24 (fig. 14; table 1). The HRSP where a possible fluvial channel appears in the reflectors is shown in figure 16.

Numerous shoal features of various sizes and orientation are distributed across the northwest part of the study area (fig. 12) and have the highest potential for sediment resources. The largest shoal (Sabine Bank) has been extensively studied for this purpose (Dellapenna and others, 2011) and was not assessed in this study. Many smaller shoals reflected in the HRSP data (fig. 11) can be targeted with direct sampling (sites 9–22 in fig. 14; table 1). From the isopach measurements derived from the HRSP interpretations, there is an area of shoaling of up to 9 m above the basal transgressive surface, and these features are specifically targeted by sites 9–16 (fig. 14; table 1). Relative to the other sites reviewed in this study, this area of shoaling is closest to shore, within 10 m of water depth or less, and is an optimal location for sediment sampling. Smaller shoal features are targeted farther offshore in sites 17–22 in 10–12-m water depth, and sites 21–22 in 18-m water depth (fig. 14; table 1).