Marine Geosciences

Combining Heat Flow Measurements and Paleo Climatic studies using R/V Marion Dufresne's unique coring capabilities

¹Ifremer, BP 70, 29280 Plouzané, France, ²Université de Bordeaux-I, Avenue des Facultés, 33405 Talence, France, ³Institut français de Recherche Polaire, BP 75, 29280 Plouzané, ⁴Lamont Doherty Geological Observatory, Palisades, 10964, New York, USA, ⁵Université de Bretagne Occidentale, Institut Universitaire Européen de la Mer, Place Nicolas, Copernic, 29280 Plouzané, France, ⁶Laboratoire des Sciences de l'Environnement, CNRS-CEA, 91198, Gif s/ Yvette, France, ⁷Max Planck Institut für Chimie, Postfach 3060, 55020 Mainz, Germany.


For the first time, heat flow measurements and paleo oceanographic studies were combined during the MD120-ANTAUS expedition of R/V Marion Dufresne, the vessel of the French Polar Institute (Fig.1). This program, which was conducted between October 12th and November 7th, from Fremantle (Australia) to La Réunion Island, results from a collaboration between Ifremer, IFRTP, the University of Bordeaux and the French Laboratory for Climate and Environment Sciences (LSCE : Laboratoire des Sciences de l'Environnement et du Climat). It also includes research scientists from Brest and Mainz Universities, and from Lamont Doherty Geological Observatory.

Figure 1a : R/V Marion Dufresne (photograph : courtesy of IFRTP - See also Fig. 6) Named after Marc-Joseph Marion Dufresne (1724-1772), R/V Marion Dufresne is a multi-purpose, 130 m long, research and supply vessel having two main missions : logistics for the French Austral Islands (Crozet and Kerguelen archipelagoes, Amsterdam and Saint-Paul Islands) and oceanographic research. Specifically designed for very severe weather conditions, she benefits from an exceptional sea behaviour allowing full performance in rough seas. The vessel is equipped with the full suite of geophysical facilities (including multi beam bathymetry and imagery) and has unique capabilities for raising long sediment cores, up to 60 meters (her record core being 58 meters), with a capacity for 110 passengers, allowing large scientific parties to embark for multidisciplinary programmes. Facing an increasing scientific demand, the French Ministry of Research decided in 1999 to reduce the ship time devoted to logistical operations to 120 days/year and allow the French Polar Institute (IFRTP : Institut Français pour la Recherche et la Technologie Polaires) to conduct research in all oceans, world-wide, 245 days/year. Therefore, the ship is no longer confined to the Indian Ocean (for example, several months in 1999 were spent coring for paleo-climatic purposes in the North Atlantic). This paves the road for new approaches and the development of integrated, multidisciplinary programs, as recently evidenced by the MD120-ANTAUS expedition.

The primary objective of this cruise was to study terrestrial heat flow variations along an isochron paralleling the South-East Indian Ridge (SEIR) between the Saint-Paul/Amsterdam hotspot and the Australian-Antarctic Discordance (AAD), an anomalously deep section of the Mid-Ocean Ridge between Australia and Antarctica, often attributed to a mantle " cold spot " (Fig. 3). Heat flow measurements were obtained in the region of the AAD as part of reconnaissance surveys in the 1960's and 1970's [Von Herzen and Langseth, 1966, Langseth and Taylor, 1967, Anderson et al, 1977]. In some cases, conductivity measurements were obtained from cores collected at the same location and at a few sites, the maximum depths to which temperature was measured was 3 to 10 meters. In addition, near-surface temperature gradients are modified by bottom water temperature variations, calling for temperature measurements deeper in the sedimentary column.
In order to interpret heat flow variations in this region of thin and patchy sediment cover, it is not only necessary to obtain more data, but to collect data that can also be used to trace water circulation and discriminate between the conductive and the convective components of the measured heat flow (Anderson et al, 1979). To accomplish this, we collected long sediment cores, along with heat flow data to greater depths. The cores will be used to study the physical properties of the sediments and also for 3He/4He studies, in order to investigate the role of water cirulation using helium isotopic ratios as tracers of hydrothermal activity. Our need for long cores led naturally to an interdisciplinary program in which the heat flow objectives are complemented by a study of paleoclimate variations. A total of 25 terrestrial heat flow measurements were collected, using 9 autonomous digital temperature probes fitted on a 18 m-long, 13 cm-diameter, gravity corer (Fig. 2 and Table II).

Table II : Site locations and conductivity values. Note that the average thermal conductivity increases progressively from east to west, due to the change from siliceous to carbonate content, as previously observed by Anderson et al (1977).

Core number	Latitude	Longitude	Depth	Core length	Conductivity W/m/K
MD002362	S45°27,13	E123°27,31	4 740 m	06,66 m	0,687
MD002363	S45°26,58	E123°32,16	4 520 m	06,52 m	0,729
MD002364	S45°31,18	E123°32,94	4 380 m	15,87 m	0,717
MD002365	S44°54,95	E125°34,29	4 700 m	08,32 m	0,701
MD002366	S44°56,61	S125°25,54	4 400/ 4 450 m ?	05,73 m	0,789
MD002367	S45°53,62	E129°57,77	4 380 m	06,80 m	0,8
MD002368	S45°58,07	E130°01,08	4 500 m	07,80 m	0,764
MD002369	S45°16,75	E120°01,59	4 430 m	10,52 m	0,717
MD002370	S45°23,00	E119°51,96	4 240 m	06,26 m	0,788
MD002371	S45°23,85	E113°29,50	4 120 m	06,42 m	0,782
MD002372	S44°08,23	E105°49,48	3 840 m	06,58 m	0,883
MD002373	S43°29,98	E99°58,99	3 750 m	06,76 m	0,956
MD002374	S46°02,508	E96°29,493	3 320 m	42,46 m
MD002375	S45°42,76	E86°45,01	3 500 m	17,24 m	0,797

Full penetration of 18 m was regularly achieved. In some places, we used the " PO-GO " multiple-entry technique, allowing series of 3 to 4 closely spaced measurements to be made for assessing the local variability of heat flow. Fourteen gravity cores (with lengths ranging between 6 m and 17.80 m) were collected. In addition, one 42-m long core was obtained (the longest ever in this part of the world ocean) using the R/V Marion Dufresne giant piston corer, named Calypso, providing information on the global climate changes that occured during the last 400 000 years. These sediment cores form a good complement to the database of the IMAGES (International Marine Global Change Study) program, a part of the International Global Change Program (IGCP).

Sediment redistribution in the ocean basin south of the SEIR is controlled by relatively strong currents associated with the Antarctic Circumpolar Current and the Antarctic Bottom Water. These currents have a strong erosive capacity, resulting in a large crestal area with a non uniform, thin (if any) sediment cover. The heat flow measurement sites were located along a 14 Ma isochron, parallel to the SEIR (Fig. 3). Seismic data from USNS Eltanin and R/V Vema collected in the 60's and early 70's indicate that these sites are all located in " grade C " or " grade D " environment, according to the classification of Sclater et al (1976) : areas of rough topography (ocean floor with greater than 400 m relief and wavelength of 5-10 km) with a thin, variable sediment cover (< 200 m) and basement outcropping within 15 km of the station. In this type of environment, most of the scatter in closely spaced heat flow measurements results from hydrothermal circulation. To characterize and tentatively model this circulation, sediment samples clamped in 25 cm-long copper tubes were systematically taken at different depths in each sediment core for 3He/4He studies. Samples were taken at 15 cm from the extremity of each 1.5 m-long core section, 2 to 4 hours after the core was onboard, in order to avoid the diffusion of Helium out ot the sediments. If water has circulated within the crust, then the 3He/4He isotopic signature should theoretically be that of the crust and upper mantle ; if water circulation has been confined to the sediment layer, then the 3He/4He isotopic signature will be that of the ocean and atmosphere.

Figure 3 : Site locations (see exact coordinates in table II).

 

The quality of the heat flow data is illustrated by the example in Figure 4, showing depth-temperature data for the longest core recovered (17.25 m). The temperature gradient is almost perfectly linear and equal to G = 69.8 ± 0.83 mK/m (Fig. 4a). Thermal conductivity, which was measured onboard every 25 cm along the core length, exhibits important variations, up to 30 % (note however that when the " equivalent conductivity (k) " is considered - 1/k=1/nxS(1/ki), where n is the number of 25 cm-long segments -, the uncertainty on k is only 8 %). In order to account for these variations, we introduce, following Bullard (1939), a thermal resistance x, such as : dx=dz/k. Heat flow, obtained as the slope of T(x), integrates the variations in thermal conductivity. Its value is : qBullard=dT/dx= 55.4 ± 4.5 mWm-1. The product of G by the " equivalent conductivity " value yields : q = 55.6 mWm-2 ± 6.7 mWm-2, which is very close to qBullard. Because they are of short wavelength (< 1 m), thermal conductivity variations do not significantly affect the estimation of heat flow, which is based on temperature probes spaced every 2 meters. In the few cases where heat flow measurements have been made in DSDP and ODP drill holes and compared to nearby surface heat flow measurements, good agreement has been found, suggesting that standard heat flow measurements made with typical penetrations of 3 to 4 m are reliable enough to model deep seated thermal processes [e.g. Erickson et al, 1975 ; Hyndman et al, 1984 ; Horai and Von Herzen, 1985 ; Fischer et al, 1997]. Data collected on the South-East Indian flanks with R/V Marion Dufresne using autonomous temperature probes welded onto a 18 m long core show that this is not always the case : in the study area, the agreement between surface heat and deep seated heat flow is rarely met. Most often, indeed, the temperature gradients that we obtained are not linear.   Figure 4 : 4a. Temperature vs depth curve at core site MD-002375 ; Thermal conductivity vs depth for core MD-002375 corrected from temperature and pressure ; 4c : Temperature vs integrated thermal resistance (x) for core MD-002375.

Figure 5 shows three examples of temperature data collected at three different sites. Each temperature measurement being systematically duplicated by two sensors spaced by 64 mm, experimental effects cannot explain the observed non-linearity (in addition, tests performed onboard after recovery clearly preclude systematic errors resulting from errors of calibration or from drift of a thermistor). The observed non linearity, which is too important to be explained by variations in thermal conductivity, results from either one or from the combination of two natural effects : variations in bottom water temperature and vertical advection of water into the sediments. At two sites (MD002369 and MD002371), the existence of vertical temperature gradients between two successive temperature sensors indicates the probable presence of an aquifer within the sediment column, the effect of which needs to be taken into account to estimate the conductive component of heat flow. Further analysis, including hydrology and He results, is needed to interpret the data and determine the contribution of each different effect. The above points out some advantages of combining coring and heat flow measurements at great depths within the sediments . R/V Marion Dufresne has the capacity to take ultra long cores (up to 60 meters) using the Calypso giant corer. Theoretically, the longer the core, the more information on the heat flow at depth. In practice, however, there are some limitations in core length. Tests performed on the Meriadzeck Plateau in 1999 using 25-m-long piston corers have clearly shown that adding temperature probes on the tube multiplies by a factor of 2 the required traction for pulling the corer out of the sediments. Heat flow measurements at depths of 30 to 40 m is thus an objective that seems feasible, although difficult to reach in routine. But for all R/V Marion Dufresne's coring programmes that require gravity cores for preserving surface sediments, heat flow measurements could be routinely and systematically performed in the future without any additional shiptime using 15 m to 25 m-long pipes. This would dramatically increase the number of heat flow measurements in the world ocean.       Figure 5 : Temperature vs depth curve at core sites : 5a) MD-002369. Note that between 10.2 m and 12.8 m, temperature is constant ; 5b) MD-002370 ; 5c) MD002371. Note that between 2.15 m and 4.35 m and between 8 .75 m and 10.94 m, temperature is constant.

Acknowledgments : We acknowledge the effective collaboration of Captain R. Gauthier which gave much freedom for initiatives by the scientific team enabling collection of record gravity and CALYPSO cores ; the professional efficiency of the team handling the cores on the fantail ; and the excellent station-holding capability of the R/V Marion Dufresne, even in very bad weather.

• Anderson, R. N., Langseth, M. G., and Sclater, J. G., The mechanism of heat transfer through the floor of the Indian Ocean, J. Geophys. Res., 82, 3391, 1977
  
• Anderson, R. N., Hobarth, M. A., and Langseth, M. G., Geothermal Convection Through Oceanic Crust, Science, 204, 823, 1979
  
• Bullard, E. C., Heat flow in South Africa, Proc. R. Soc. London Ser., A, 173, 474, 1939
  
• Erickson, A.J., Von Herzen, R.P., Sclater, J. G., Girdler, R.W., Marshall, B.V., Hyndman, R., Geothermal Measurements in Deep-Sea Drill Holes, J. Geophys. Res., 17, 2515-2528, 1975
  
• Langseth, M. A., and Taylor, P. I., Recent measurements in the Indian Ocean, J. Geophys. Res., 72, 6249, 1967
  
• Horai, K., Von Herzen, R.P., Measurement of Heat Flow on the Leg 86 of the Deep Sea Drilling Project, Initial Report of Deep Sea Drilling Project, 1985
  
• Hyndman, D.R., Langseth, M.G., Von Herzen, R.P., Review of deep sea drilling project geothermal measurements through leg 71, Initial Report of Deep Sea Drilling Project, 78B, 813-823, 1984
  
• Duyker, E., An Officer of the Blue, Marc-Joseph Marion Dufresne, South Sea Explorer, 1724-1772, Melbourne University Press, 229 p., 1994
  
• Sclater, J. G., Crowe, J., and Anderson, R. N., On the reliability of oceanic heat flow averages, J. Geophys. Res., 81, 2997, 1976
  
• Von Herzen, R.P., Langseth, M.G., Present Status of Oceanic Heat-Flow Measurements, Phys. Chim. Earth, 365, 1966
  
• Fisher, A.T., Becker, K., and Davis, E.E., The permeability of young oceanic crust east of Juan de Fuca Ridge determined using borehole thermal measurements. Geophys. Res. Lett., 24, 1311-1314, 1997

Figure 6 : Death of Nicolas-Joseph Marion Dufresne (crayon, pencil and chalk sketch of Charles Meryon, 1821-1868). Born in Saint-Malo, northern France, to a family of merchants, Marion Dufresne embarked at the age of 11 with the French East India Company. In 1771, he sailed from Mauritius to Tahiti in command of two ships of the French Navy that were returning Tahitian chief Ahu-toru from France to Tahiti. Soon after leaving Mauritius, Ahu-toru died of smallpox. Consequently, instead of continuing on to Tahiti, Dufresne set out to explore the extreme southern part of the Indian Ocean in search of Terra Australis, and discovered Marion, Crozet and Prince Edward Islands. While in New Zealand, he undertook coastal explorations and established friendly contacts with the native Maori. However, on January 12, 1772, Dufresne and some of his men were attacked and killed by the Maori. Dufresne's surviving officers made important contributions to the study of Maori life and culture, noting the similarities between the Maori language and that of the Polynesian inhabitants of Tahiti. Moreover, scientists with the expedition made some of the earliest studies of ocean temperatures and salinity levels at varying depths (See book of Duyker, 1994).