- Email: [email protected]
Seafloor geological impacts associated with drilling disturbance
ARTICLE IN PRESS Deep-Sea Research II 56 (2009) 4–11
Contents lists available at ScienceDirect
Deep-Sea Research II journal homepage: www.elsevier.com/locate/dsr2
Seafloor geological impacts associated with drilling disturbance I.C.S. Correˆa a,, E.E. Toldo Jr.a, F.A.L. Toledo b a b
ˆnica, Instituto de Geocieˆncias, Universidade Federal do Rio Grande do Sul, CECO-UFRGS, CP. 15001, Porto Alegre, RS 91509.900, Brazil Centro de Estudos de Geologia Costeira e Ocea ´fico, IO-USP, Pc- a do Oceanogra ´fico, Sa ˜o Paulo, SP 05508.900, Brazil Instituto Oceanogra
a r t i c l e in f o
a b s t r a c t
Article history: Accepted 3 August 2008 Available online 30 October 2008
This study describes the sedimentological composition and morphology of the sea bottom in an area located at 902 m depth in the Campos Basin of Brazil, and compares characteristics before and after drilling activity for oil exploration. The results show no significant sedimentological variation in the area affected by drilling. The most noticeable effects were observed during the second (MD2) of three cruises, in terms of change to grain size distribution, total organic carbon and clay mineral composition around the Eagle well. This impact occurred over the seabed in a direction that corresponds to the nearbottom circulation pattern, predominantly northward. In the third cruise (MD3), 12 months after drilling, some recovery was observed. Side-scan sonar imaging was used to explore the extent of the area affected by drilling, and sedimentological samples from it confirmed the effects of drilling. & 2008 Elsevier Ltd. All rights reserved.
Keywords: Drilling impact Oil exploration Campos Basin Brazil
1. Introduction The deposition of sediments on continental margins is a result of complex interactions between tectonic processes, sea-level change, climatic change, and oceanographic processes. Tectonic and eustatic changes control the sediment deposition. Tectonics, eustatics, climate, and oceanographic processes interact to control the supply and distribution of sediments. During periods of intense activity in the marine environment, such as in storms and when strong currents occur, sediments originating from the continental area can be carried beyond the shore zone and deposited on the continental shelf, often reaching the shelf edge and the continental slope, and sometimes moving down slope by landslide or gravitational flow. Interest in how these slope environments may be affected by drilling disturbance comes from the discovery of large, morphologically complex oil fields, for example, in the Gulf of Mexico, in the North Sea, and in the Campos Basin of Brazil. Assessing the response to different parameters controlling the evolution of such deposits—slope gradient, bottom morphology complexity, sediment flow and size, base level changes—is a task requiring detailed geological knowledge. As a first approach to investigating the effects of small-scale sediment disturbance from deep-sea drilling, the MAPEM Project (Environmental Monitoring of Offshore Drilling for Petroleum Exploration, Toldo and Ayup Zouain, 2004) began in 2001 at 902 m depth in the southwestern Atlantic Ocean in the Campos Corresponding author. Tel./fax: +55 51 3316 9855.
E-mail addresses: [email protected] (I.C.S. Correˆa), [email protected] (E.E. Toldo Jr.), [email protected] (F.A.L. Toledo). 0967-0645/$ - see front matter & 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.dsr2.2008.08.014
Basin. This project aims to assess the effect of drilling activity upon oceanic benthic ecosystems, when submitted to the discharge of drill cuttings generated with non-aqueous fluids (NAFs) used in drilling. The present paper describes the texture of the sediments covering a continental slope area situated in the northern part of the Campos Basin. It focuses essentially on characterizing the variation in bottom sediments in this area: prior to drilling activity, 30 days after, and 1 year after the conclusion of a deep-water exploratory well (Correˆa et al., 2004).
2. Location The Campos Basin lies to the southwest of the South Atlantic Ocean, in an area off the Brazilian continental margin between latitudes 201300 S (the region of the Vito´ria high in the State of Espı´rito Santo) and 241S (the region of the Cabo Frio high in the State of Rio de Janeiro). The basin covers over 100,000 km2 (Fig. 1). Over 70% is in water deeper than 200 m. The continental shelf has an average width of 100 km, and the depth of the shelf break ranges from 80 m deep in the northern part of the basin to 130 m in the southern part, with a mean depth of 110 m (Viana et al., 1998). The slope base is shallower in the north of the basin than in the south, with average depths ranging from 1550 to 2000 m, respectively. The transition from continental slope to a continental rise is marked by an intermediate zone, called the Sao Paulo Plateau, a deep ocean area with low gradients controlled by the presence of salt diapers. The Eagle well, around which the environment was monitored, was operated by the company UNOCAL.
ARTICLE IN PRESS ˆa et al. / Deep-Sea Research II 56 (2009) 4–11 I.C.S. Corre
5
Miocene, the most recent having occurred between 85,000 and 53,000 years ago. Coralline deposits have been mapped in waters between 570 and 800 m. The area described in this report is located on the continental slope. It shows a bottom morphology cut by numerous valleys and canyons, mostly oriented east–west (Fig. 1). Water depth ranges from 800 to 1050 m, taking only into account the area included in the 500 m radius from the well, which, as stated above, is at a depth of 902 m. The morphology of the continental slope is regular, with homogeneously distributed isovalue curves. The average declivity of the area is on the order of 4%, typical of many continental slopes.
4. Current patterns on the continental shelf edge
Fig. 1. Location of the study area (white dot) on the SE continental margin of Brazil and 3D view of continental slope morphology in the adjacent area with sampling design for the 54 stations.
In this study, water masses were vertically grouped into: surface and deep water. Surface waters are mainly influenced by mid-latitude atmospheric circulation. They consist of the Superficial Tropical Water (STW) and South Atlantic Central Water (SACW) (Fig. 2). The contour current, also called the Brazil Current (BC), flows along the east, southeast and south coasts as a relatively shallow surface current following the line of the shelf break at a maximum depth of 700 m. This current carries the STW to the southeast between the surface and 200 m water depth. The processes of mass and energy exchange with the atmosphere through intense radiation and excessive evaporation make the
3. Sedimentology and geomorphology Siliciclastic and bioclastic sands occupy the inner to middle part of the continental shelf of the Campos Basin. Predominately siliciclastic sands dominate the outer shelf, ranging from moderately rounded to well-rounded. At the southern area of the shelf, sands are finer than in the central and northern portions. They have quartzose, locally glauconitic, composition, with only a few types of heavy minerals. Bioclastic constituents are primarily molluscs, bryozoans, algae fragments, and foraminifers. In the central and northern portions of the shelf, between its middle and outer parts, sands are well sorted, covered with a thin layer of iron oxide form extensive sub-aqueous dunes that drift to the NE. Those dunes have an average height of 0.5–1 m and are dozens of meters long. From the edge of the shelf seaward, fine sands occur in the southern part of the basin, whereas coarser sands are found to the north and mixed with bioclastics. Muddy sands cover extensive areas and are bioturbated. In the mid-slope, at depths ranging from 550 to 1200 m, these deposits change gradually to fine sands, sandy silts, ferricrusts and deep-water corals. Antarctic Intermediate Water (AAIW), with its high dissolved oxygen content, is responsible for the generation of crusts. The continental shelf is relatively flat, characterized by bedforms from ripples to sand dunes, locally with lithified sediments up to 100 m long and 5 m high, as well as long carbonate bars over 5 km in length and heights up to 10 m. Sets of parallel canyons can be seen beyond the edge of the continental shelf, and, according to Viana et al. (1998), seismic data indicate that these systems were active on the shelf edge during former periods of low sea level. The upper slope is flat, gradually changing to a convex outer form in its intermediate part, at depths ranging from 600 to 1200 m. Its convex shape is attributed to an accumulation of deposits from gravitational flows developing irregular bottom topography. Such deposits occupy an area of approximately 1660 km2, with six flow events identified in high-resolution multi-channel seismic profiles, all after the Upper
Fig. 2. Stratification of water masses near the Brazilian southeastern continental margin, latitude 221S. STW ¼ Superficial Tropical Water, SACW ¼ South Atlantic Central Water, AAIW ¼ Antarctic Intermediate Water, NADW ¼ North Atlantic Deep Water, and AABW ¼ Antarctic Bottom Water (modified from Viana et al., 1998).
ARTICLE IN PRESS 6
ˆa et al. / Deep-Sea Research II 56 (2009) 4–11 I.C.S. Corre
STW a warm water mass, with temperatures above 20 1C, salinities above 36, and low concentrations of dissolved nutrients (Castro, 2003). Besides the STW water mass, the South Atlantic Central Water (SACW) is found along the whole Brazilian continental margin, formed by sinking of surface water in the region of subtropical convergence (Fig. 2). Starting at the region of subtropical convergence, the SACW flows to the east and northeast. Along the southeast and southern coasts of Brazil, it is carried immediately below at depths ranging from 200 to 500 m, with a maximum of 700 m. The SACW is a cold-water mass with temperatures below 20 1C, salinities between 34.6 and 36, and high nutrient concentrations. Deep-water masses found below the thermocline are Antarctic Intermediate Water (AAIW), North Atlantic Deep Water (NADW), and Antarctic Bottom Water (AABW). The AAIW, which occupies the mid-slope between 550 and 1200 m, has temperatures ranging from 6 to 2 1C, high dissolved oxygen contents, and a minimum salinity of 34.2. This current flows to the north from Cabo Frio. The NADW corresponds to a 2000-m thick layer that flows to the south, with high salinity, low nutrient levels, but rich in oxygen. The dense and cold AABW develops below the NADW, in waters deeper than 4000 m (Fig. 2). Data obtained from the study area show a complex hydrodynamic context between the continental shelf edge and the upper slope, influenced by different strata of water masses, tidal currents, storm waves, vortices, and meanders generated by BC. These processes cause erosion, transportation, and deposition of sediments along the area, as well as towards the deep sea (Viana et al., 1998). Bottom currents measured at the edge of the Campos Basin shelf indicate the presence of bidirectional currents crossing the continental shelf with mean values of 25 cm s1 and peaks of 50 cm s1. A spectral analysis of the data reveals the predominance of cycles with periods longer than 31 h, which can be influenced by the predominant wind regime from NE and prevailing winds from SW, associated with the passage of coldfront storms. Cycles of astronomical tides, with minimum and maximum amplitudes of 0.5 and 1.5 m, respectively, also can generate a pattern of bidirectional currents. Statistical analyses show that the pattern of currents running parallel to the beach, or essentially northeastward, is mainly related to the passage of cold fronts. Satellite images showing ocean surface temperature indicate the presence of vortices with diameters greater than 50 km, associated with the propagation of BC meanders. Those vortices are displaced from their original place of formation in the central area of the shelf to its border, at an average speed of 4–35 cm s1. The frequency of such occurrences, though not well established, varies from 15 to 45 days, with residence time ranging from a few days to less than a month. Preliminary analyses of current records obtained close to the bottom at a water depth of 400 m indicate that the currents associated to those vortices may reach speeds above 35 cm s1. During the entire discharge period the maximum velocity measured near the bottom, 28 m above the seabed, varied between 15 and 20 cm s1 and did not present a defined pattern, although the currents had a predominantly northward direction. For a detailed description of the currents for different depths along the water column for the period of discharged cuttings, see Pivel et al. (2008).
obtained from the bottom samples collected in three sampling operations in the studied area of the continental slope. Three oceanographic cruises were made, denoted by MD1, April 19–24, 2001, prior to drilling; MD2, July 23–27, 2001, 30 days after drilling; and MD3, June 22–26, 2002, 12 months after drilling. The field program included bottom-sampling, onboard laboratory analyses, and storage of collected samples, ending with the subsequent transfer of samples and data for the other analyses in biological, geological and chemical laboratories. Field operations related to the management of the vessel Satro 25, as well as recording and interpreting the sonar data, were carried out by the Petroleum and Environment Geo-Service (PEG). The vessel was equipped with a global positioning system and the software Hydropro, which were used for navigation and positioning for the duration of the fieldwork, the precision of the system being 71 m. As described by Toledo et al. (2004), the positioning of the sample stations was recorded as UTM coordinates, whose reference datum was WGS 84. The positioning of the box-cores and the side-scan sonar during the MD2 and MD3 cruises was carried out by a hydroacoustic positioning system (USBL) manufactured by SonarDynes. This system consists of a transceiver attached to the hull of the vessel and a beacon attached to a cable close to the sensor. This hydro-acoustic system provides the positioning of the towed equipment in relation to the previous position of the vessel on the surface. The sampling plan around the Eagle well involved 54 stations (Fig. 1), with 54 box-core samples planned for each sampling cruise. A concentric radial sampling grid was adopted, with six sediment samples collected in circles situated at 50 and 100 m from the well, and with 12 other samples collected at circles situated at 150, 300 and 500 m from it (Fig. 3). A distance scale was thus established between the sampled spots and the density of the sampling grid so as to adequately cover the processes affecting the ocean bottom. Six additional reference samples were collected at 2500 m from the drilling area. Each sample collected during each cruise was analyzed for grain size, composition of the 2-mm fraction and total organic
5. Methodology The methodology used in the MAPEM project has been described by Toledo et al. (2004). The present report gives results
Fig. 3. Plan view of the morphology on the continental slope area and location of 48 sample stations.
ARTICLE IN PRESS ˆa et al. / Deep-Sea Research II 56 (2009) 4–11 I.C.S. Corre
carbon (TOC) determination. Samples were collected using a 50 50 cm box-corer, with a maximum sampling height of 70 cm, resulting in a maximum total sample volume of 1.75 m3. Samples collected for geological purposes were sieved and pipetted, and their grain size assessed and classified according to the Wentworth scale (1922). A part of each sample collected was used to identify the clay mineral components of the o2-mm fraction using X-ray diffractometry. With grain size data obtained from the sediment analysis, the statistical parameters of the samples were calculated using the Folk and Ward formulas (1957). The same spots, or others just a few meters distant, were sampled to obtain an integrated statistical model based on the concept of repeated measurements. The sonar images were obtained through the EdgeTech DF 1000 digital sonar, based on three survey routes, on both the MD2 and the MD3 cruises. The equipment was mounted on an EdgeTech 560 processing unit. The whole system was connected and towed by means of a 3500 m-long electromechanical cable specifically suited for towing such a system. The sonar acquisition software was ISIS, 5.0 version, by Triton Elics. During the survey, the towfish sonar was towed above the seabed at a height of approximately 10–20% of the slant range used to collect the best possible information. Near the Eagle well, lines with a 100 m ground range were obtained. The positioning of the towfish on the sea bottom was carried out by means of USBL described above.
7
Table 1 shows minimum silt percentages of 26% on MD1 and 29% on MD3, with maximum values of 48% on MD1 and 51% on MD3. MD2 shows greater variation between the minimum and maximum silt values, the minimum being of the order of 17%, and the maximum 51%. The mean value among all three sampling operations was consistent, ranging from 38% on MD1 to 42% on MD2. Clay presents significant variations when comparing the MD2 cruise to the others (Table 1). On MD1, the minimum content was 32%, while the maximum was 61%; on MD2, those values became more extreme, the minimum content being 10% and the maximum 56%. The clay content observed on MD3 was very similar to that seen on MD1, the minimum being 23% and the maximum 60%. The mean clay content was 51% on MD1, 45% on MD2 and 50% on MD3. On MD1, median grain size (Md) ranged from medium silt (0.0180 mm) to coarse clay (0.0023 mm); on MD2, the median ranged from fine sand (0.1450 mm) to coarse clay (0.0028 mm). On MD3, it ranged from coarse silt (0.0580 mm) to coarse clay (0.0022 mm). Average median grain size was very fine silt (0.008 and 0.004 mm), reflecting different sand contents in MD1 and MD2. Thus changes in sediment texture were minimal through the time between MD1 and MD2, for sand, silt and clay contents, except for the site within the submarine canyon. The variations in grain size distribution, at the six reference stations (49, 50, 51, 52, 53 and 54), were not significant enough to
6. Results 6.1. Grain size distribution Samples described here are from stations located within a 500 m radius from the well. The amount of gravel in the surface sediments from MD1, MD2 and MD3 showed very little variation, as seen in Table 1. The small amounts of gravel were predominantly bioclastic and not cuttings (Fig. 4). In fact, cuttings generally break down during sample preparation, particularly those composed of fine material. The sand content in the bottom-surface samples from the three sampling cruises shows a significant variation in texture. A considerable variation can be seen between the first and second samplings (Table 1), with values increasing from 29% to 73%, then decreasing to 48% on MD3. Sand content is higher in the southern areas, along an east–west axis cut by the canyon. This increase in sand, more specifically at stations 29, 31, 33 and 41 (Figs. 3 and 5), is probably due to bottom currents (gravitational flows) in the canyon area and winnowing of finer sediments washed through the valley system.
Fig. 4. Cuttings from sample MD2-05, taken from a box-core.
Table 1 Descriptive analysis of textural properties.
MD1 Minimum Maximum Mean MD2 Minimum Maximum Mean MD3 Minimum Maximum Mean
Gravel (%)
Sand (%)
Silt (%)
Clay (%)
Md (mm)
Mz (mm)
s
SkI
TOC (%)
0 1 0
3 29 11
26 48 38
32 61 51
0.0180 0.0023 0.0039
0.0096 0.0028 00043
2.50 3.45 2.85
0.25 0.35 0.09
0.95 1.60 1.27
0 1 0
5 73 14
17 51 42
10 56 45
0.1450 0.0028 0.0058
0.0666 0.0028 0.0055
2.50 3.53 2.87
0.22 0.61 0.01
0.39 2.00 1.19
0 1 0
3 48 10
29 51 40
23 60 50
0.0580 0.0022 0.0043
0.0272 0.0022 0.0045
2.20 3.47 2.78
0.27 0.55 0.05
0.68 2.10 1.25
Md ¼ median (mm), Mz ¼ mean size (mm), standard deviation (s), SkI ¼ asymmetry, TOC ¼ total organic carbon (%).
ARTICLE IN PRESS 8
ˆa et al. / Deep-Sea Research II 56 (2009) 4–11 I.C.S. Corre
6.3. Description of clay minerals
Fig. 5. Sand concentration variability (%), between MD1 (upper) and MD2 (lower).
characterize an increase or decrease of the sediments texture in this area, except at station 49, where sand increased from 11% on MD1 to 23% on MD2. This may be related to winnowing of finer sediments by bottom currents as this station was located it edge of a submarine canyon. 6.2. Total organic carbon (TOC) The TOC observed in Table 1 shows variations over the three sampling cruises. On MD1, TOC varied from 0.95% to 1.60%; on MD2, it varied from 0.39% to 2.00%, and on MD3, TOC ranged from 0.68% to 2.10%. There was a drop in total organic carbon on MD2, while those of MD1 and MD3 were similar. The areas with a drop in organic carbon relative to MD1 were the same areas as those in which greater variations in contents of sand, silt and clay were seen.
The clay minerals reported below are related to a semiquantitative analysis of surface samples in the fraction smaller than 2 mm. Assessment of proportions for different phases of the identified clay minerals (and gibbsite) was by integration of the areas of peaks 001 normalized to 100. It was then possible to make relative internal parameterizations, thus obtaining a semiquantitative approximation for the identified phases. Features of the program EVA-DIFFPLUSs were used to obtain the area values. Table 2 shows the variation in kaolinite content. There was a reduction in its minimum content between MD1and MD3, while the maximum values were the same. On MD1, the areas close to the well presented slightly higher kaolinite contents, reaching values close to 70%, and decreasing to around 60% near the edges of the area. After the well was drilled, an inversion in values seems to have occurred. In the central part of the area, close to the well, the contents are lower, around 60%, while at the edges, they increased to around 75%. Comparison of results from the cruises shows that some material produced by the drilling of the well may have entered the central part of the area, leading to a reduction in the kaolinite content there. The increase observed at the edges of the area may be related to an influx of material through natural processes, such as gravitational flows or bottom currents. Table 2 shows a decrease in minimum values of smectite from MD1 to MD3 and an increase of a similar order in maximum values. On MD1, the areas closer to the well present lower smectite contents, reaching values below 25%, and increasing to values above 35% near the edges of the area. The distribution is the opposite of the distribution seen for kaolinite. After the well was drilled, values became inverted: in the central part of the area, close to the well, the contents increased a little to around 30–35%, while at the edges to the west they decreased to values lower than 15%. To the east, they varied and reached up to 30%. Twelve months after drilling the smectite values increased considerably in the central part of the area, reaching values higher than 50%. Near the western edge, these values decrease to around 15%, remaining around 30% near the eastern edge. Comparing all sampling results shows a general decrease in smectite contents from the central area to the edges, the opposite of the kaolinite distribution. The increase seen in the central part of the area close to the well may be related to the deposit by decantation of clay material coming from the well or even from the drilling fluid, rich in bentonite (Fig. 6). Regarding the illite content (Table 2) the minimum values remained constant between MD1 and MD3, whilst there was an increase in maximum values. This clay mineral, like smectite, comes from material originating from the well as cuttings and
Table 2 Descriptive analysis of clay minerals.
MD1 Minimum Maximum Mean MD2 Minimum Maximum Mean MD3 Minimum Maximum Mean
Kaolinite (%)
Smectite (%)
Ilite (%)
Gibbsite (%)
50 80 65
10 50 30
1 8 3
0 3 1
40 80 70
5 60 20
2 20 5
1 5 2
20 80 60
2 70 30
2 40 6
0 8 3
ARTICLE IN PRESS ˆa et al. / Deep-Sea Research II 56 (2009) 4–11 I.C.S. Corre
drilling fluid. Table 2 shows that gibbsite content varied in the same way as illite, the mean generally remaining between 1% and 3%, and increasing from MD1 to MD3. It should be noted that the behavior of gibbsite mirrored that of kaolinite. Where there was an increase in gibbsite content, an increase in kaolinite was also seen. The origin of gibbsite is the same as that for kaolinite. The material entering the area under study was probably related to gravitational flows or bottom currents. Overall, there was an increase in gibbsite and kaolinite between MD2 and MD3, from the well to the edges of the area. This relative increase in kaolinite and gibbsite was related to the natural contribution made by clay minerals, probably brought in by the currents. The increases observed in smectite and illite, in inverse proportion to those of kaolinite and gibbsite, were linked to cuttings and material from the well.
Fig. 6. Variability of smectite concentrations in the clay fraction (%), between MD1 (upper) and MD2 (lower).
9
At the six reference stations, the prevailing clay mineral is kaolinite, followed by smectite, with smaller but similar values for illite and gibbsite. There was a general increase in kaolinite and gibbsite on MD2 and MD3 relative to MD1. They were probably carried by currents running through the study area, and are likely to be of continental origin. 6.4. Side-scan sonar images The side-scan sonar images were obtained during the second and third cruise, and additional underwater video footage was produced targeting part of the studied area. This material was processed and interpreted to generate supplementary information for biological, geological and chemical analysis in order to establish relationships between the features recorded by the side-scan sonar images and those depicted in video and photographs. The use of images generated by side-scan sonar allows a visualization of the bottom in the area studied. Using the different types of underwater reflection patterns, it was possible to identify the type of sea bottom and the deposition of the material produced by drilling activity. The reflection patterns traced on the side-scan sonar image were subsequently compared with the sediments collected through the box-corer, as well as with photographs obtained by digital video. Based on the data around the Eagle well (Fig. 7), sonographic patterns were seen that indicate the presence of different materials, on MD2 and MD3. These patterns were grouped into four types: Weak reflection sonographic pattern: this pattern occurred around the Eagle well, and was extended in the ENE direction. Its clear aspect, indicates a low return of the acoustic signal (Fig. 8). The extent of the area of this sonographic pattern on MD3 was noticeably smaller than that from MD2 (Fig. 9). Medium-to-strong reflection sonographic pattern: this pattern occurred in the quadrants situated W and NW from the well, characterized by a reflection of medium to strong intensity (Fig. 7). The main spreading direction of cuttings in this direction is in agreement to the near-bottom circulation pattern, predominantly to northward. In the MD3 cruise, the area with this sonographic pattern was reduced (Fig. 9). Strong reflection sonographic pattern: this pattern occurred within the limits of the medium-to-strong reflection sonographic pattern and featured a semi-circular form (Fig. 7). Fig. 9 shows a reduction in the area from MD2 to MD3. The areas covered with
Fig. 7. Side-scan sonar image from the area around the Eagle well during MD2. On the left side of the navigation line (vertical line) the image shows the well (black dot) and an area with strong reflection associated a pile of drill cuttings.
ARTICLE IN PRESS 10
ˆa et al. / Deep-Sea Research II 56 (2009) 4–11 I.C.S. Corre
cuttings decrease between MD2 and MD3, probably as a result of bottom currents reworking and spreading materials over the 12-month period. Fig. 10, taken from the vicinity of station 05, shows a part of the areas characterized by the strong reflection pattern, due to the presence of cuttings. Medium reflection sonographic pattern: this is the prevailing pattern in the area, with a homogeneous texture (Fig. 11). Because it occurs widely, its reflection intensity served as a standard for comparing the reflection intensity of the acoustic signal for the other patterns observed (Fig. 7). This sonographic pattern possibly marks the intensity of return of the acoustic signal for the prevailing sediment on the continental slope in the mapped area. The results obtained with the side-scan sonar show that the technique was useful for interpretation in this kind of survey. Its use makes it possible to identify specific sampling areas where the images indicate different patterns of reflection and to distinguish between affected and unaffected areas.
Fig. 8. Photograph from side-scan area with weak reflections, near sample station 02, revealing the presence of a light-colored amorphous material.
Fig. 10. Photograph from area with strong side-scan sonar reflection, near station 05, showing a pile of drill cuttings from Eagle well 1 year after drilling.
Fig. 11. View of the sea floor near station 31, where side-scan sonar records show a medium-intensity reflection pattern, suggesting a pattern of homogeneous bioturbation, and no evidence of material associated with the drilling.
7. Conclusions
Fig. 9. Overlap of side-scan sonar images indicates an area where the reflection character is reduced between MD2 and MD3, 1 year after cutting discharge.
The area used for environmental monitoring around the exploratory Eagle well has a bottom morphology cut by numerous valleys and canyons oriented east–west. These valleys were found to be active and are responsible for transporting and distributing sediments across the continental shelf over the continental slope to the abyssal zone. Overall, the surface sediments distributed around the studied area were typically muddy, with some fine to very fine sand. Cuttings associated with the drilling activities were observed only in the area adjacent to the well. The greatest variations in sand grain size distribution were observed at stations located near slopes or the bottom of valleys and canyons. The variation in sand content during the three cruises is associated with bottom currents or gravitational flows that have either washed away the finer material from the deposited sediments, or have brought an influx of sandy material from the continental shelf. The variation in silt and clay contents was not significant; their values fell only in areas where there was a relative increase in sand content. Total organic carbon (TOC) around the well varied slightly between the sampling cruises. The most significant values were observed next to the areas in which there was a variation in sand
ARTICLE IN PRESS ˆa et al. / Deep-Sea Research II 56 (2009) 4–11 I.C.S. Corre
content, because of washing or the influx of coarser material, caused by bottom currents or gravitational flows. Regarding clay minerals, increases in smectite and illite were observed in the study area between the MD1 and MD3 cruises, particularly over an area that agrees with the dispersion model proposed by the results of the cuttings discharge simulation (Pivel et al., 2008). These increases in smectite and illite, which were the opposite of those observed for kaolinite and gibbsite, are related to the contribution of cuttings from the Eagle well. It should be noted that at the six reference stations, there was no significant variation in sand, silt and clay contents, whereas for the clay minerals, there were increases in kaolinite and gibbsite in MD2 and MD3 relative to MD1. The latter increases are a consequence of natural contributions of these clay minerals. The side-scans sonar images show the presence and distribution of drill cuttings in areas where the sonographic reflection pattern was strong to medium-strong. This defines the extent of the area affected by the discharge and shows the main spreading direction of cuttings towards the north. Sonar imaging also show how the area recovered a year after drilling. The areas covered with cuttings decrease and some recovery was observed between MD2 and MD3, probably as a result of bottom currents reworking and spreading materials over the 12-months after drilling.
11
Acknowledgments This work was supported by FINEP/CTPETRO, IBP, FAURGS and the consortium between Federal Universities of Rio Grande do Sul (UFRGS) and Santa Catarina (UFSC). We thank Robin Clarke and Paul Potter for helpful comments on the manuscript. References Castro, B., 2003. O mar costeiro do Brasil. Revista Scientific American Brasil (12), 33–36. Correˆa, I.C.S., Toldo Jr., E.E., Toledo, F.A., 2004. Geology. In: Ayup Zouain, R.N., Toldo Jr., E.E. (Eds.), MAPEM—Environmental Monitoring of Offshore Drilling for Petroleum Exploration: Deep Waters, MAPEM Technical Report. Porto Alegre UFRGS (CD-ROM). Folk, R.L., Ward, W.C., 1957. Brazos river bar: a study in the significance of grain size parameters. Journal of Sedimentary Petrology (27), 3–27. Pivel, M.A.G., Freitas, A., Carla M.D.S., Comba, J.L.D., 2008. Modeling the discharge of cuttings and drilling fluids in a deepwater environment. Deep-Sea Research II, doi:10.1016/j.dsr2.2008.08.015. Toldo Jr, E.E., Ayup Zouain, R.N., 2004. Modelo de A´guas Profundas do Monitoramento Ambiental em Atividades de Perfurac- a˜o Explorato´ria Marı´tima—MAPEM Technical Report. Porto Alegre UFRGS (CD-ROM). Toledo, F.A., Ayup Zouain, R.N., Toldo Jr., E.E., 2004. Field sampling manual. In: Ayup Zouain, R.N., Toldo Jr, E.E. (Eds.), MAPEM—Environmental Monitoring of Offshore Drilling for Petroleum Exploration: Deep Waters. MAPEM Technical Report. Porto Alegre UFRGS (CD-ROM). Viana, A.R., Faurge`res, J.C., Kowsmannn, R.O., Lima, J.A.M., Caddah, L.F.G., Rizzo, J.G., 1998. Hydrology, morphology and sedimentology of the Campos continental margin, offshore Brazil. Sedimentary Geology (115), 133–157. Wentworth, C.K., 1922. A scale of grade and class terms for clastic sediments. Journal of Geology (30), 377–392.
Contents lists available at ScienceDirect
Deep-Sea Research II journal homepage: www.elsevier.com/locate/dsr2
Seafloor geological impacts associated with drilling disturbance I.C.S. Correˆa a,, E.E. Toldo Jr.a, F.A.L. Toledo b a b
ˆnica, Instituto de Geocieˆncias, Universidade Federal do Rio Grande do Sul, CECO-UFRGS, CP. 15001, Porto Alegre, RS 91509.900, Brazil Centro de Estudos de Geologia Costeira e Ocea ´fico, IO-USP, Pc- a do Oceanogra ´fico, Sa ˜o Paulo, SP 05508.900, Brazil Instituto Oceanogra
a r t i c l e in f o
a b s t r a c t
Article history: Accepted 3 August 2008 Available online 30 October 2008
This study describes the sedimentological composition and morphology of the sea bottom in an area located at 902 m depth in the Campos Basin of Brazil, and compares characteristics before and after drilling activity for oil exploration. The results show no significant sedimentological variation in the area affected by drilling. The most noticeable effects were observed during the second (MD2) of three cruises, in terms of change to grain size distribution, total organic carbon and clay mineral composition around the Eagle well. This impact occurred over the seabed in a direction that corresponds to the nearbottom circulation pattern, predominantly northward. In the third cruise (MD3), 12 months after drilling, some recovery was observed. Side-scan sonar imaging was used to explore the extent of the area affected by drilling, and sedimentological samples from it confirmed the effects of drilling. & 2008 Elsevier Ltd. All rights reserved.
Keywords: Drilling impact Oil exploration Campos Basin Brazil
1. Introduction The deposition of sediments on continental margins is a result of complex interactions between tectonic processes, sea-level change, climatic change, and oceanographic processes. Tectonic and eustatic changes control the sediment deposition. Tectonics, eustatics, climate, and oceanographic processes interact to control the supply and distribution of sediments. During periods of intense activity in the marine environment, such as in storms and when strong currents occur, sediments originating from the continental area can be carried beyond the shore zone and deposited on the continental shelf, often reaching the shelf edge and the continental slope, and sometimes moving down slope by landslide or gravitational flow. Interest in how these slope environments may be affected by drilling disturbance comes from the discovery of large, morphologically complex oil fields, for example, in the Gulf of Mexico, in the North Sea, and in the Campos Basin of Brazil. Assessing the response to different parameters controlling the evolution of such deposits—slope gradient, bottom morphology complexity, sediment flow and size, base level changes—is a task requiring detailed geological knowledge. As a first approach to investigating the effects of small-scale sediment disturbance from deep-sea drilling, the MAPEM Project (Environmental Monitoring of Offshore Drilling for Petroleum Exploration, Toldo and Ayup Zouain, 2004) began in 2001 at 902 m depth in the southwestern Atlantic Ocean in the Campos Corresponding author. Tel./fax: +55 51 3316 9855.
E-mail addresses: [email protected] (I.C.S. Correˆa), [email protected] (E.E. Toldo Jr.), [email protected] (F.A.L. Toledo). 0967-0645/$ - see front matter & 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.dsr2.2008.08.014
Basin. This project aims to assess the effect of drilling activity upon oceanic benthic ecosystems, when submitted to the discharge of drill cuttings generated with non-aqueous fluids (NAFs) used in drilling. The present paper describes the texture of the sediments covering a continental slope area situated in the northern part of the Campos Basin. It focuses essentially on characterizing the variation in bottom sediments in this area: prior to drilling activity, 30 days after, and 1 year after the conclusion of a deep-water exploratory well (Correˆa et al., 2004).
2. Location The Campos Basin lies to the southwest of the South Atlantic Ocean, in an area off the Brazilian continental margin between latitudes 201300 S (the region of the Vito´ria high in the State of Espı´rito Santo) and 241S (the region of the Cabo Frio high in the State of Rio de Janeiro). The basin covers over 100,000 km2 (Fig. 1). Over 70% is in water deeper than 200 m. The continental shelf has an average width of 100 km, and the depth of the shelf break ranges from 80 m deep in the northern part of the basin to 130 m in the southern part, with a mean depth of 110 m (Viana et al., 1998). The slope base is shallower in the north of the basin than in the south, with average depths ranging from 1550 to 2000 m, respectively. The transition from continental slope to a continental rise is marked by an intermediate zone, called the Sao Paulo Plateau, a deep ocean area with low gradients controlled by the presence of salt diapers. The Eagle well, around which the environment was monitored, was operated by the company UNOCAL.
ARTICLE IN PRESS ˆa et al. / Deep-Sea Research II 56 (2009) 4–11 I.C.S. Corre
5
Miocene, the most recent having occurred between 85,000 and 53,000 years ago. Coralline deposits have been mapped in waters between 570 and 800 m. The area described in this report is located on the continental slope. It shows a bottom morphology cut by numerous valleys and canyons, mostly oriented east–west (Fig. 1). Water depth ranges from 800 to 1050 m, taking only into account the area included in the 500 m radius from the well, which, as stated above, is at a depth of 902 m. The morphology of the continental slope is regular, with homogeneously distributed isovalue curves. The average declivity of the area is on the order of 4%, typical of many continental slopes.
4. Current patterns on the continental shelf edge
Fig. 1. Location of the study area (white dot) on the SE continental margin of Brazil and 3D view of continental slope morphology in the adjacent area with sampling design for the 54 stations.
In this study, water masses were vertically grouped into: surface and deep water. Surface waters are mainly influenced by mid-latitude atmospheric circulation. They consist of the Superficial Tropical Water (STW) and South Atlantic Central Water (SACW) (Fig. 2). The contour current, also called the Brazil Current (BC), flows along the east, southeast and south coasts as a relatively shallow surface current following the line of the shelf break at a maximum depth of 700 m. This current carries the STW to the southeast between the surface and 200 m water depth. The processes of mass and energy exchange with the atmosphere through intense radiation and excessive evaporation make the
3. Sedimentology and geomorphology Siliciclastic and bioclastic sands occupy the inner to middle part of the continental shelf of the Campos Basin. Predominately siliciclastic sands dominate the outer shelf, ranging from moderately rounded to well-rounded. At the southern area of the shelf, sands are finer than in the central and northern portions. They have quartzose, locally glauconitic, composition, with only a few types of heavy minerals. Bioclastic constituents are primarily molluscs, bryozoans, algae fragments, and foraminifers. In the central and northern portions of the shelf, between its middle and outer parts, sands are well sorted, covered with a thin layer of iron oxide form extensive sub-aqueous dunes that drift to the NE. Those dunes have an average height of 0.5–1 m and are dozens of meters long. From the edge of the shelf seaward, fine sands occur in the southern part of the basin, whereas coarser sands are found to the north and mixed with bioclastics. Muddy sands cover extensive areas and are bioturbated. In the mid-slope, at depths ranging from 550 to 1200 m, these deposits change gradually to fine sands, sandy silts, ferricrusts and deep-water corals. Antarctic Intermediate Water (AAIW), with its high dissolved oxygen content, is responsible for the generation of crusts. The continental shelf is relatively flat, characterized by bedforms from ripples to sand dunes, locally with lithified sediments up to 100 m long and 5 m high, as well as long carbonate bars over 5 km in length and heights up to 10 m. Sets of parallel canyons can be seen beyond the edge of the continental shelf, and, according to Viana et al. (1998), seismic data indicate that these systems were active on the shelf edge during former periods of low sea level. The upper slope is flat, gradually changing to a convex outer form in its intermediate part, at depths ranging from 600 to 1200 m. Its convex shape is attributed to an accumulation of deposits from gravitational flows developing irregular bottom topography. Such deposits occupy an area of approximately 1660 km2, with six flow events identified in high-resolution multi-channel seismic profiles, all after the Upper
Fig. 2. Stratification of water masses near the Brazilian southeastern continental margin, latitude 221S. STW ¼ Superficial Tropical Water, SACW ¼ South Atlantic Central Water, AAIW ¼ Antarctic Intermediate Water, NADW ¼ North Atlantic Deep Water, and AABW ¼ Antarctic Bottom Water (modified from Viana et al., 1998).
ARTICLE IN PRESS 6
ˆa et al. / Deep-Sea Research II 56 (2009) 4–11 I.C.S. Corre
STW a warm water mass, with temperatures above 20 1C, salinities above 36, and low concentrations of dissolved nutrients (Castro, 2003). Besides the STW water mass, the South Atlantic Central Water (SACW) is found along the whole Brazilian continental margin, formed by sinking of surface water in the region of subtropical convergence (Fig. 2). Starting at the region of subtropical convergence, the SACW flows to the east and northeast. Along the southeast and southern coasts of Brazil, it is carried immediately below at depths ranging from 200 to 500 m, with a maximum of 700 m. The SACW is a cold-water mass with temperatures below 20 1C, salinities between 34.6 and 36, and high nutrient concentrations. Deep-water masses found below the thermocline are Antarctic Intermediate Water (AAIW), North Atlantic Deep Water (NADW), and Antarctic Bottom Water (AABW). The AAIW, which occupies the mid-slope between 550 and 1200 m, has temperatures ranging from 6 to 2 1C, high dissolved oxygen contents, and a minimum salinity of 34.2. This current flows to the north from Cabo Frio. The NADW corresponds to a 2000-m thick layer that flows to the south, with high salinity, low nutrient levels, but rich in oxygen. The dense and cold AABW develops below the NADW, in waters deeper than 4000 m (Fig. 2). Data obtained from the study area show a complex hydrodynamic context between the continental shelf edge and the upper slope, influenced by different strata of water masses, tidal currents, storm waves, vortices, and meanders generated by BC. These processes cause erosion, transportation, and deposition of sediments along the area, as well as towards the deep sea (Viana et al., 1998). Bottom currents measured at the edge of the Campos Basin shelf indicate the presence of bidirectional currents crossing the continental shelf with mean values of 25 cm s1 and peaks of 50 cm s1. A spectral analysis of the data reveals the predominance of cycles with periods longer than 31 h, which can be influenced by the predominant wind regime from NE and prevailing winds from SW, associated with the passage of coldfront storms. Cycles of astronomical tides, with minimum and maximum amplitudes of 0.5 and 1.5 m, respectively, also can generate a pattern of bidirectional currents. Statistical analyses show that the pattern of currents running parallel to the beach, or essentially northeastward, is mainly related to the passage of cold fronts. Satellite images showing ocean surface temperature indicate the presence of vortices with diameters greater than 50 km, associated with the propagation of BC meanders. Those vortices are displaced from their original place of formation in the central area of the shelf to its border, at an average speed of 4–35 cm s1. The frequency of such occurrences, though not well established, varies from 15 to 45 days, with residence time ranging from a few days to less than a month. Preliminary analyses of current records obtained close to the bottom at a water depth of 400 m indicate that the currents associated to those vortices may reach speeds above 35 cm s1. During the entire discharge period the maximum velocity measured near the bottom, 28 m above the seabed, varied between 15 and 20 cm s1 and did not present a defined pattern, although the currents had a predominantly northward direction. For a detailed description of the currents for different depths along the water column for the period of discharged cuttings, see Pivel et al. (2008).
obtained from the bottom samples collected in three sampling operations in the studied area of the continental slope. Three oceanographic cruises were made, denoted by MD1, April 19–24, 2001, prior to drilling; MD2, July 23–27, 2001, 30 days after drilling; and MD3, June 22–26, 2002, 12 months after drilling. The field program included bottom-sampling, onboard laboratory analyses, and storage of collected samples, ending with the subsequent transfer of samples and data for the other analyses in biological, geological and chemical laboratories. Field operations related to the management of the vessel Satro 25, as well as recording and interpreting the sonar data, were carried out by the Petroleum and Environment Geo-Service (PEG). The vessel was equipped with a global positioning system and the software Hydropro, which were used for navigation and positioning for the duration of the fieldwork, the precision of the system being 71 m. As described by Toledo et al. (2004), the positioning of the sample stations was recorded as UTM coordinates, whose reference datum was WGS 84. The positioning of the box-cores and the side-scan sonar during the MD2 and MD3 cruises was carried out by a hydroacoustic positioning system (USBL) manufactured by SonarDynes. This system consists of a transceiver attached to the hull of the vessel and a beacon attached to a cable close to the sensor. This hydro-acoustic system provides the positioning of the towed equipment in relation to the previous position of the vessel on the surface. The sampling plan around the Eagle well involved 54 stations (Fig. 1), with 54 box-core samples planned for each sampling cruise. A concentric radial sampling grid was adopted, with six sediment samples collected in circles situated at 50 and 100 m from the well, and with 12 other samples collected at circles situated at 150, 300 and 500 m from it (Fig. 3). A distance scale was thus established between the sampled spots and the density of the sampling grid so as to adequately cover the processes affecting the ocean bottom. Six additional reference samples were collected at 2500 m from the drilling area. Each sample collected during each cruise was analyzed for grain size, composition of the 2-mm fraction and total organic
5. Methodology The methodology used in the MAPEM project has been described by Toledo et al. (2004). The present report gives results
Fig. 3. Plan view of the morphology on the continental slope area and location of 48 sample stations.
ARTICLE IN PRESS ˆa et al. / Deep-Sea Research II 56 (2009) 4–11 I.C.S. Corre
carbon (TOC) determination. Samples were collected using a 50 50 cm box-corer, with a maximum sampling height of 70 cm, resulting in a maximum total sample volume of 1.75 m3. Samples collected for geological purposes were sieved and pipetted, and their grain size assessed and classified according to the Wentworth scale (1922). A part of each sample collected was used to identify the clay mineral components of the o2-mm fraction using X-ray diffractometry. With grain size data obtained from the sediment analysis, the statistical parameters of the samples were calculated using the Folk and Ward formulas (1957). The same spots, or others just a few meters distant, were sampled to obtain an integrated statistical model based on the concept of repeated measurements. The sonar images were obtained through the EdgeTech DF 1000 digital sonar, based on three survey routes, on both the MD2 and the MD3 cruises. The equipment was mounted on an EdgeTech 560 processing unit. The whole system was connected and towed by means of a 3500 m-long electromechanical cable specifically suited for towing such a system. The sonar acquisition software was ISIS, 5.0 version, by Triton Elics. During the survey, the towfish sonar was towed above the seabed at a height of approximately 10–20% of the slant range used to collect the best possible information. Near the Eagle well, lines with a 100 m ground range were obtained. The positioning of the towfish on the sea bottom was carried out by means of USBL described above.
7
Table 1 shows minimum silt percentages of 26% on MD1 and 29% on MD3, with maximum values of 48% on MD1 and 51% on MD3. MD2 shows greater variation between the minimum and maximum silt values, the minimum being of the order of 17%, and the maximum 51%. The mean value among all three sampling operations was consistent, ranging from 38% on MD1 to 42% on MD2. Clay presents significant variations when comparing the MD2 cruise to the others (Table 1). On MD1, the minimum content was 32%, while the maximum was 61%; on MD2, those values became more extreme, the minimum content being 10% and the maximum 56%. The clay content observed on MD3 was very similar to that seen on MD1, the minimum being 23% and the maximum 60%. The mean clay content was 51% on MD1, 45% on MD2 and 50% on MD3. On MD1, median grain size (Md) ranged from medium silt (0.0180 mm) to coarse clay (0.0023 mm); on MD2, the median ranged from fine sand (0.1450 mm) to coarse clay (0.0028 mm). On MD3, it ranged from coarse silt (0.0580 mm) to coarse clay (0.0022 mm). Average median grain size was very fine silt (0.008 and 0.004 mm), reflecting different sand contents in MD1 and MD2. Thus changes in sediment texture were minimal through the time between MD1 and MD2, for sand, silt and clay contents, except for the site within the submarine canyon. The variations in grain size distribution, at the six reference stations (49, 50, 51, 52, 53 and 54), were not significant enough to
6. Results 6.1. Grain size distribution Samples described here are from stations located within a 500 m radius from the well. The amount of gravel in the surface sediments from MD1, MD2 and MD3 showed very little variation, as seen in Table 1. The small amounts of gravel were predominantly bioclastic and not cuttings (Fig. 4). In fact, cuttings generally break down during sample preparation, particularly those composed of fine material. The sand content in the bottom-surface samples from the three sampling cruises shows a significant variation in texture. A considerable variation can be seen between the first and second samplings (Table 1), with values increasing from 29% to 73%, then decreasing to 48% on MD3. Sand content is higher in the southern areas, along an east–west axis cut by the canyon. This increase in sand, more specifically at stations 29, 31, 33 and 41 (Figs. 3 and 5), is probably due to bottom currents (gravitational flows) in the canyon area and winnowing of finer sediments washed through the valley system.
Fig. 4. Cuttings from sample MD2-05, taken from a box-core.
Table 1 Descriptive analysis of textural properties.
MD1 Minimum Maximum Mean MD2 Minimum Maximum Mean MD3 Minimum Maximum Mean
Gravel (%)
Sand (%)
Silt (%)
Clay (%)
Md (mm)
Mz (mm)
s
SkI
TOC (%)
0 1 0
3 29 11
26 48 38
32 61 51
0.0180 0.0023 0.0039
0.0096 0.0028 00043
2.50 3.45 2.85
0.25 0.35 0.09
0.95 1.60 1.27
0 1 0
5 73 14
17 51 42
10 56 45
0.1450 0.0028 0.0058
0.0666 0.0028 0.0055
2.50 3.53 2.87
0.22 0.61 0.01
0.39 2.00 1.19
0 1 0
3 48 10
29 51 40
23 60 50
0.0580 0.0022 0.0043
0.0272 0.0022 0.0045
2.20 3.47 2.78
0.27 0.55 0.05
0.68 2.10 1.25
Md ¼ median (mm), Mz ¼ mean size (mm), standard deviation (s), SkI ¼ asymmetry, TOC ¼ total organic carbon (%).
ARTICLE IN PRESS 8
ˆa et al. / Deep-Sea Research II 56 (2009) 4–11 I.C.S. Corre
6.3. Description of clay minerals
Fig. 5. Sand concentration variability (%), between MD1 (upper) and MD2 (lower).
characterize an increase or decrease of the sediments texture in this area, except at station 49, where sand increased from 11% on MD1 to 23% on MD2. This may be related to winnowing of finer sediments by bottom currents as this station was located it edge of a submarine canyon. 6.2. Total organic carbon (TOC) The TOC observed in Table 1 shows variations over the three sampling cruises. On MD1, TOC varied from 0.95% to 1.60%; on MD2, it varied from 0.39% to 2.00%, and on MD3, TOC ranged from 0.68% to 2.10%. There was a drop in total organic carbon on MD2, while those of MD1 and MD3 were similar. The areas with a drop in organic carbon relative to MD1 were the same areas as those in which greater variations in contents of sand, silt and clay were seen.
The clay minerals reported below are related to a semiquantitative analysis of surface samples in the fraction smaller than 2 mm. Assessment of proportions for different phases of the identified clay minerals (and gibbsite) was by integration of the areas of peaks 001 normalized to 100. It was then possible to make relative internal parameterizations, thus obtaining a semiquantitative approximation for the identified phases. Features of the program EVA-DIFFPLUSs were used to obtain the area values. Table 2 shows the variation in kaolinite content. There was a reduction in its minimum content between MD1and MD3, while the maximum values were the same. On MD1, the areas close to the well presented slightly higher kaolinite contents, reaching values close to 70%, and decreasing to around 60% near the edges of the area. After the well was drilled, an inversion in values seems to have occurred. In the central part of the area, close to the well, the contents are lower, around 60%, while at the edges, they increased to around 75%. Comparison of results from the cruises shows that some material produced by the drilling of the well may have entered the central part of the area, leading to a reduction in the kaolinite content there. The increase observed at the edges of the area may be related to an influx of material through natural processes, such as gravitational flows or bottom currents. Table 2 shows a decrease in minimum values of smectite from MD1 to MD3 and an increase of a similar order in maximum values. On MD1, the areas closer to the well present lower smectite contents, reaching values below 25%, and increasing to values above 35% near the edges of the area. The distribution is the opposite of the distribution seen for kaolinite. After the well was drilled, values became inverted: in the central part of the area, close to the well, the contents increased a little to around 30–35%, while at the edges to the west they decreased to values lower than 15%. To the east, they varied and reached up to 30%. Twelve months after drilling the smectite values increased considerably in the central part of the area, reaching values higher than 50%. Near the western edge, these values decrease to around 15%, remaining around 30% near the eastern edge. Comparing all sampling results shows a general decrease in smectite contents from the central area to the edges, the opposite of the kaolinite distribution. The increase seen in the central part of the area close to the well may be related to the deposit by decantation of clay material coming from the well or even from the drilling fluid, rich in bentonite (Fig. 6). Regarding the illite content (Table 2) the minimum values remained constant between MD1 and MD3, whilst there was an increase in maximum values. This clay mineral, like smectite, comes from material originating from the well as cuttings and
Table 2 Descriptive analysis of clay minerals.
MD1 Minimum Maximum Mean MD2 Minimum Maximum Mean MD3 Minimum Maximum Mean
Kaolinite (%)
Smectite (%)
Ilite (%)
Gibbsite (%)
50 80 65
10 50 30
1 8 3
0 3 1
40 80 70
5 60 20
2 20 5
1 5 2
20 80 60
2 70 30
2 40 6
0 8 3
ARTICLE IN PRESS ˆa et al. / Deep-Sea Research II 56 (2009) 4–11 I.C.S. Corre
drilling fluid. Table 2 shows that gibbsite content varied in the same way as illite, the mean generally remaining between 1% and 3%, and increasing from MD1 to MD3. It should be noted that the behavior of gibbsite mirrored that of kaolinite. Where there was an increase in gibbsite content, an increase in kaolinite was also seen. The origin of gibbsite is the same as that for kaolinite. The material entering the area under study was probably related to gravitational flows or bottom currents. Overall, there was an increase in gibbsite and kaolinite between MD2 and MD3, from the well to the edges of the area. This relative increase in kaolinite and gibbsite was related to the natural contribution made by clay minerals, probably brought in by the currents. The increases observed in smectite and illite, in inverse proportion to those of kaolinite and gibbsite, were linked to cuttings and material from the well.
Fig. 6. Variability of smectite concentrations in the clay fraction (%), between MD1 (upper) and MD2 (lower).
9
At the six reference stations, the prevailing clay mineral is kaolinite, followed by smectite, with smaller but similar values for illite and gibbsite. There was a general increase in kaolinite and gibbsite on MD2 and MD3 relative to MD1. They were probably carried by currents running through the study area, and are likely to be of continental origin. 6.4. Side-scan sonar images The side-scan sonar images were obtained during the second and third cruise, and additional underwater video footage was produced targeting part of the studied area. This material was processed and interpreted to generate supplementary information for biological, geological and chemical analysis in order to establish relationships between the features recorded by the side-scan sonar images and those depicted in video and photographs. The use of images generated by side-scan sonar allows a visualization of the bottom in the area studied. Using the different types of underwater reflection patterns, it was possible to identify the type of sea bottom and the deposition of the material produced by drilling activity. The reflection patterns traced on the side-scan sonar image were subsequently compared with the sediments collected through the box-corer, as well as with photographs obtained by digital video. Based on the data around the Eagle well (Fig. 7), sonographic patterns were seen that indicate the presence of different materials, on MD2 and MD3. These patterns were grouped into four types: Weak reflection sonographic pattern: this pattern occurred around the Eagle well, and was extended in the ENE direction. Its clear aspect, indicates a low return of the acoustic signal (Fig. 8). The extent of the area of this sonographic pattern on MD3 was noticeably smaller than that from MD2 (Fig. 9). Medium-to-strong reflection sonographic pattern: this pattern occurred in the quadrants situated W and NW from the well, characterized by a reflection of medium to strong intensity (Fig. 7). The main spreading direction of cuttings in this direction is in agreement to the near-bottom circulation pattern, predominantly to northward. In the MD3 cruise, the area with this sonographic pattern was reduced (Fig. 9). Strong reflection sonographic pattern: this pattern occurred within the limits of the medium-to-strong reflection sonographic pattern and featured a semi-circular form (Fig. 7). Fig. 9 shows a reduction in the area from MD2 to MD3. The areas covered with
Fig. 7. Side-scan sonar image from the area around the Eagle well during MD2. On the left side of the navigation line (vertical line) the image shows the well (black dot) and an area with strong reflection associated a pile of drill cuttings.
ARTICLE IN PRESS 10
ˆa et al. / Deep-Sea Research II 56 (2009) 4–11 I.C.S. Corre
cuttings decrease between MD2 and MD3, probably as a result of bottom currents reworking and spreading materials over the 12-month period. Fig. 10, taken from the vicinity of station 05, shows a part of the areas characterized by the strong reflection pattern, due to the presence of cuttings. Medium reflection sonographic pattern: this is the prevailing pattern in the area, with a homogeneous texture (Fig. 11). Because it occurs widely, its reflection intensity served as a standard for comparing the reflection intensity of the acoustic signal for the other patterns observed (Fig. 7). This sonographic pattern possibly marks the intensity of return of the acoustic signal for the prevailing sediment on the continental slope in the mapped area. The results obtained with the side-scan sonar show that the technique was useful for interpretation in this kind of survey. Its use makes it possible to identify specific sampling areas where the images indicate different patterns of reflection and to distinguish between affected and unaffected areas.
Fig. 8. Photograph from side-scan area with weak reflections, near sample station 02, revealing the presence of a light-colored amorphous material.
Fig. 10. Photograph from area with strong side-scan sonar reflection, near station 05, showing a pile of drill cuttings from Eagle well 1 year after drilling.
Fig. 11. View of the sea floor near station 31, where side-scan sonar records show a medium-intensity reflection pattern, suggesting a pattern of homogeneous bioturbation, and no evidence of material associated with the drilling.
7. Conclusions
Fig. 9. Overlap of side-scan sonar images indicates an area where the reflection character is reduced between MD2 and MD3, 1 year after cutting discharge.
The area used for environmental monitoring around the exploratory Eagle well has a bottom morphology cut by numerous valleys and canyons oriented east–west. These valleys were found to be active and are responsible for transporting and distributing sediments across the continental shelf over the continental slope to the abyssal zone. Overall, the surface sediments distributed around the studied area were typically muddy, with some fine to very fine sand. Cuttings associated with the drilling activities were observed only in the area adjacent to the well. The greatest variations in sand grain size distribution were observed at stations located near slopes or the bottom of valleys and canyons. The variation in sand content during the three cruises is associated with bottom currents or gravitational flows that have either washed away the finer material from the deposited sediments, or have brought an influx of sandy material from the continental shelf. The variation in silt and clay contents was not significant; their values fell only in areas where there was a relative increase in sand content. Total organic carbon (TOC) around the well varied slightly between the sampling cruises. The most significant values were observed next to the areas in which there was a variation in sand
ARTICLE IN PRESS ˆa et al. / Deep-Sea Research II 56 (2009) 4–11 I.C.S. Corre
content, because of washing or the influx of coarser material, caused by bottom currents or gravitational flows. Regarding clay minerals, increases in smectite and illite were observed in the study area between the MD1 and MD3 cruises, particularly over an area that agrees with the dispersion model proposed by the results of the cuttings discharge simulation (Pivel et al., 2008). These increases in smectite and illite, which were the opposite of those observed for kaolinite and gibbsite, are related to the contribution of cuttings from the Eagle well. It should be noted that at the six reference stations, there was no significant variation in sand, silt and clay contents, whereas for the clay minerals, there were increases in kaolinite and gibbsite in MD2 and MD3 relative to MD1. The latter increases are a consequence of natural contributions of these clay minerals. The side-scans sonar images show the presence and distribution of drill cuttings in areas where the sonographic reflection pattern was strong to medium-strong. This defines the extent of the area affected by the discharge and shows the main spreading direction of cuttings towards the north. Sonar imaging also show how the area recovered a year after drilling. The areas covered with cuttings decrease and some recovery was observed between MD2 and MD3, probably as a result of bottom currents reworking and spreading materials over the 12-months after drilling.
11
Acknowledgments This work was supported by FINEP/CTPETRO, IBP, FAURGS and the consortium between Federal Universities of Rio Grande do Sul (UFRGS) and Santa Catarina (UFSC). We thank Robin Clarke and Paul Potter for helpful comments on the manuscript. References Castro, B., 2003. O mar costeiro do Brasil. Revista Scientific American Brasil (12), 33–36. Correˆa, I.C.S., Toldo Jr., E.E., Toledo, F.A., 2004. Geology. In: Ayup Zouain, R.N., Toldo Jr., E.E. (Eds.), MAPEM—Environmental Monitoring of Offshore Drilling for Petroleum Exploration: Deep Waters, MAPEM Technical Report. Porto Alegre UFRGS (CD-ROM). Folk, R.L., Ward, W.C., 1957. Brazos river bar: a study in the significance of grain size parameters. Journal of Sedimentary Petrology (27), 3–27. Pivel, M.A.G., Freitas, A., Carla M.D.S., Comba, J.L.D., 2008. Modeling the discharge of cuttings and drilling fluids in a deepwater environment. Deep-Sea Research II, doi:10.1016/j.dsr2.2008.08.015. Toldo Jr, E.E., Ayup Zouain, R.N., 2004. Modelo de A´guas Profundas do Monitoramento Ambiental em Atividades de Perfurac- a˜o Explorato´ria Marı´tima—MAPEM Technical Report. Porto Alegre UFRGS (CD-ROM). Toledo, F.A., Ayup Zouain, R.N., Toldo Jr., E.E., 2004. Field sampling manual. In: Ayup Zouain, R.N., Toldo Jr, E.E. (Eds.), MAPEM—Environmental Monitoring of Offshore Drilling for Petroleum Exploration: Deep Waters. MAPEM Technical Report. Porto Alegre UFRGS (CD-ROM). Viana, A.R., Faurge`res, J.C., Kowsmannn, R.O., Lima, J.A.M., Caddah, L.F.G., Rizzo, J.G., 1998. Hydrology, morphology and sedimentology of the Campos continental margin, offshore Brazil. Sedimentary Geology (115), 133–157. Wentworth, C.K., 1922. A scale of grade and class terms for clastic sediments. Journal of Geology (30), 377–392.