Oceanography is a scientific discipline concerned with all aspects of the world's oceans and seas, including their physical and chemical properties, their origin and geologic framework, and the life forms that inhabit the marine environment.

Traditionally, oceanography has been divided into four separate but related branches: physical oceanography, chemical oceanography, marine geology, and marine ecology.

Physical oceanography deals with the properties of seawater (temperature, density, pressure, and so on), its movement (waves, currents, and tides), and the interactions between the ocean waters and the atmosphere.

Chemical oceanography has to do with the composition of seawater and the biogeochemical cycles that affect it. Marine geology focuses on the structure, features, and evolution of the ocean basins.

Marine ecology, also called biological oceanography, involves the study of the plants and animals of the sea, including life cycles and food production.

Oceanography is the sum of these several branches. Oceanographic research entails the sampling of seawater and marine life for close study, the remote sensing of oceanic processes with aircraft and Earth-orbiting satellites, and the exploration of the seafloor by means of deep-sea drilling and seismic profiling of the terrestrial crust below the ocean bottom.

Greater knowledge of the world's oceans enables scientists to more accurately predict, for example, long-term weather and climatic changes and also leads to more efficient exploitation of the Earth's resources.

Oceanography also is vital to understanding the effect of pollutants on ocean waters and to the preservation of the quality of the oceans' waters in the face of increasing human demands made on them.

Ocean Basins

The first major undersea survey was undertaken during the 1870s, but it was not until the last half of the 20th century that scientists began to learn what lies beneath the ocean surface in any detail. It has been determined that the ocean basins, which hold the vast quantity of water that covers nearly three-quarters of the Earth's surface, have an average depth of almost four kilometres. A number of major features of the basins depart from this average, as, for example, the mountainous ocean ridges, deep-sea trenches, and jagged, linear fracture zones (see Figure 7). Other significant features of the ocean floor include aseismic ridges, abyssal hills, and seamounts and guyots. The basins also contain a variable amount of sedimentary fill that is thinnest on the ocean ridges and usually thickest near the continental margins.

While the ocean basins lie much lower than sea level, the continents stand high--about one kilometre above sea level. The physical explanation for this condition is that the continental crust is light and thick, whereas the oceanic crust is dense and thin. Both the continental and oceanic crust lie over a more uniform layer called the mantle. As an analogy, one can think of a thick piece of styrofoam and a thin piece of wood floating in a tub of water. The styrofoam rises higher out of the water than the wood.

The ocean basins are transient features over geologic time, changing shape and depth while the process of plate tectonics proceeds. The surface layer of the Earth, the lithosphere, consists of a number of rigid plates that are in continual motion. The boundaries between the lithospheric plates form the principal relief features of the ocean basins: the crests of oceanic ridges are spreading centres where two plates move apart from each other at a rate of several centimetres per year. Molten rock material wells up from the underlying mantle into the gap between the diverging plates and solidifies into oceanic crust, thereby creating new ocean floor. At the deep-sea trenches, two plates converge, with one plate sliding down under the other into the mantle where it is melted. Thus, for each segment of new ocean floor created at the ridges, an equal amount of old oceanic crust is destroyed at the trenches, or so-called subduction zones (see below Deep-sea trenches and also the article plate tectonics). It is for this reason that the oldest segment of ocean floor, found in the far western Pacific, is apparently only about 200 million years old, even though the age of the Earth is estimated to be at least 4.6 billion years.

The dominant factors that govern seafloor relief and topography are the thermal properties of the oceanic plates, tensional forces in the plates, volcanic activity, and sedimentation. In brief, the oceanic ridges rise about two kilometres above the seafloor because the plates near these spreading centres are warm and thermally expanded. In contrast, plates in the subduction zones are generally cooler. Tensional forces resulting in plate divergence at the spreading centres also create block-faulted mountains and abyssal hills, which trend parallel to the oceanic ridges. Seamounts and guyots, as well as abyssal hills and most aseismic ridges, are produced by volcanism. Continuing sedimentation throughout the ocean basin serves to blanket and bury many of the faulted mountains and abyssal hills with time. Erosion plays a relatively minor role in shaping the face of the deep seafloor, in contrast to the continents. This is because deep ocean currents are generally slow (they flow at less than 50 centimetres per second) and lack sufficient power.

Exploration of the ocean basins

Mapping the characteristics of the ocean basin has been difficult for several reasons. First, the oceans are not easy to travel over; second, until recent times navigation has been extremely crude, so that individual observations have been only loosely correlated with one another; and, finally, the oceans are opaque to light--i.e., the deep seafloor cannot be seen from the ocean surface. Modern technology has given rise to customized research vessels, satellite and electronic navigation, and sophisticated acoustic instruments that have mitigated some of these problems.

The Challenger Expedition, mounted by the British in 1872-76, provided the first systematic view of a few of the major features of the seafloor. Scientists aboard the HMS Challenger determined ocean depths by means of wire-line soundings and discovered the Mid-Atlantic Ridge. Dredges brought up samples of rocks and sediments off the seafloor. The main advance in mapping, however, did not occur until sonar was developed in the early 20th century. This system for detecting the presence of objects underwater by acoustic echo provided marine researchers with a highly useful tool, since sound can be detected over several thousands of kilometres in the ocean (visible light, by comparison, can only penetrate 100 metres or so of water).

Modern sonar systems include the Seabeam multibeam echo sounder and the GLORIA scanning sonar (see undersea exploration: Methodology and instrumentation: Exploration of the seafloor and the Earth's crust). They operate on the principle that the depth (or distance) of the seafloor can be determined by multiplying one-half the elapsed time between a downgoing acoustic pulse and its echo by the speed of sound in seawater (about 1,500 metres per second). Such multifrequency sonar systems permit the use of different pulse frequencies to meet different scientific objectives.

Acoustic pulses of 12 kilohertz (kHz), for example, are normally employed to measure ocean depth, while lower frequencies--3.5 kHz to less than 100 hertz (Hz)--are used to map the thickness of sediments in the ocean basins. Very high frequencies of 100 kHz or more are employed in side-scanning sonar to measure the texture of the seafloor. The acoustic pulses are normally generated by piezoelectric transducers. For determining subbottom structure, low-frequency acoustic pulses are produced by explosives, compressed air, or water-jet implosion.

Near-bottom sonar systems, such as the Deep Tow of the Scripps Institution of Oceanography (in La Jolla, Calif., U.S.), produce even more detailed images of the seafloor and subbottom structure. The Deep Tow package contains both echo sounders and side-scanning sonars, along with associated geophysical instruments, and is towed behind a ship at slow speed 10 to 100 metres above the seafloor. It yields very precise measurements of even finer-scale features than are resolvable with Seabeam and other comparable systems.

Another notable instrument system is ANGUS, a deep-towed camera sled that can take thousands of high-resolution photographs of the seafloor during a single day. It has been successfully used in the detection of hydrothermal vents at spreading centres (see below Oceanic ridges). Overlapping photographic images make it possible to construct photomosaic strips about 10-20 metres wide that reveal details on the order of centimetres.

Three major navigation systems are in use in modern marine geology. These include electromagnetic systems such as loran and Earth-orbiting satellites (see undersea exploration: Basic elements of undersea exploration: Navigation). Acoustic transponder arrays of two or more stations placed on the seafloor a few kilometres apart are used to navigate deeply towed instruments, submersibles, and occasionally surface research vessels when detailed mapping is conducted in small areas. These systems measure the distance between the instrument package and the transponder sites and, using simple geometry, compute fixes accurate to a few metres. Although the individual transponders can be used to determine positions relative to the array with great accuracy, the preciseness of the position of the array itself depends on which system is employed to locate it.

Such Earth-orbiting satellites as SEASAT and GEOSAT have uncovered some significant topographic features of the ocean basins. SEASAT, launched in 1978, carried a radar altimeter into orbit. This device was used to measure the distance between the satellite path and the surfaces of the ocean and continents to 0.1 metre. The measurements revealed that the shape of the ocean surface is warped by seafloor features: massive seamounts cause the surface to bulge over them owing to gravitational attraction. Similarly, the ocean surface downwarps occur over trenches. Using these satellite measurements of the ocean surface, William F. Haxby computed the gravity field there.

The resulting gravity map (Figure 8) provides comprehensive coverage of the ocean surface on a 5' by 5' grid (five nautical miles on each side at the equator). Coverage as complete as this is not available from echo soundings made from ships. Because the gravity field at the ocean surface is a highly sensitive indicator of marine topography, this map reveals various previously uncharted features, including seamounts, ridges, and fracture zones, while improving the detail on other known features. In addition, the gravity map shows a linear pattern of gravity anomalies that cut obliquely across the grain of the topography. These anomalies are most pronounced in the Pacific basin; they are apparently about 100 kilometres across and some 1,000 kilometres long. They have an amplitude of approximately 10 milligals (0.001 percent of the Earth's gravity attraction) and are aligned west-northwest--very close to the direction in which the Pacific Plate moves over the mantle below.

Oceanic crust

Structure and composition

The oceanic crust differs from the continental crust in several ways: it is thinner, denser, younger, of different chemical composition, and created in a different plate-tectonic setting. The oceanic crust is formed at spreading centres on the oceanic ridges, whereas continental crust is formed above the subduction zones. The oceanic crust is about six kilometres thick. It is composed of several layers, not including the overlying sediment. The topmost layer, about 500 metres thick, includes lavas of basaltic composition (i.e., rock material consisting largely of plagioclase [feldspar] and pyroxene).

The lavas are generally of two types: pillow lavas and sheet flows. Pillow lavas appear to be shaped exactly as the name implies--like large overstuffed pillows about one metre in cross section and one to several metres long. They commonly form small hills tens of metres high at the spreading centres. Sheet flows have the appearance of wrinkled bed sheets. They commonly are thin (only about 10 centimetres thick) and cover a broader area than pillow lavas. There is evidence that sheet flows are erupted at higher temperatures than those of the pillow variety. On the East Pacific Rise at 8 S latitude, a series of sheet flow eruptions (possibly since the mid-1960s) have covered more than 220 square kilometres of seafloor to an average depth of 70 metres.

Below the lava is a layer composed of feeder, or sheeted, dikes that measures more than one kilometre thick. Dikes are fractures that serve as the plumbing system for transporting magmas (molten rock material) to the seafloor to produce lavas. They are about one metre wide, subvertical, and elongate along the trend of the spreading centre where they formed, and they abut one another's sides--hence the term sheeted. These dikes are also of basaltic composition. There are two layers below the dikes totaling about 4.5 kilometres in thickness. Both of these include gabbros, which are essentially basalts with coarser mineral grains.

These gabbro layers are thought to represent the magma chambers, or pockets of lava, that ultimately erupt on the seafloor. The upper gabbro layer is isotropic (uniform) in structure.

In some places, this layer includes pods of plagiogranite, a differentiated rock richer in silica than gabbro. The lower gabbro layer has a stratified structure and evidently represents the floor or sides of the magma chamber. This layered structure is called cumulate, meaning that the layers (which measure up to several metres thick) result from the sedimentation of minerals out of the liquid magma. The layers in the cumulate gabbro have less silica but are richer in iron and magnesium than the upper portions of the crust. Olivine, an iron-magnesium silicate, is a common mineral in the lower gabbro layer.

The oceanic crust lies atop the Earth's mantle, as does the continental crust. Mantle rock is composed mostly of peridotite, which consists primarily of the mineral olivine with small amounts of pyroxene and amphibole.

Investigations of the oceanic crust

Knowledge of the structure and composition of the oceanic crust comes from several sources. Bottom sampling during early exploration brought up all varieties of the above-mentioned rocks, but the structure of the crust and the abundance of the constituent rocks were unclear. Simultaneously, seismic refraction experiments enabled researchers to determine the layered nature of the oceanic crust.

These experiments involve measuring the travel times of seismic waves generated by explosions (e.g., dynamite blasts) set off over distances of several tens of kilometres. The results of early refraction experiments revealed the existence of two layers beneath the sediment cover. More sophisticated experiments and analyses led to dividing these layers into two parts, each with a different seismic wave velocity, which increases with depth. The seismic velocity is a kind of fingerprint that can be attributed to a limited number of rock types. Sampled rock data and seismic results were combined to yield a model for the structure and composition of the crust.

Study of ophiolites

Great strides in understanding the oceanic crust were made by the study of ophiolites. These are slices of the ocean floor that have been thrust above sea level by the action of plate tectonics. In various places in the world, the entire sequence of oceanic crust and upper mantle is exposed. These areas include, among others, Newfoundland and the Pacific Coast Ranges of California, the island of Cyprus in the Mediterranean Sea, and the mountains in Oman on the southeastern tip of the Arabian Peninsula.

Ophiolites reveal the structure and composition of the oceanic crust in astonishing detail. Also, the process of crustal formation and hydrothermal circulation, as well as the origin of marine magnetic anomalies (see below), can be studied with comparative clarity. Although it is clear that ophiolites are of marine origin, there is some controversy as to whether they represent typical oceanic crust or crust formed in settings other than an oceanic spreading centre--behind island arcs, for example.

The age of the oceanic crust does not go back farther than about 200 million years. Such crust is being formed today at oceanic spreading centres. Many ophiolites are much older than the oldest oceanic crust, demonstrating continuity of the formation processes over hundreds of millions of years. Methods that may be used to determine the age of the crustal material include direct dating of rock samples by radiometric dating (measuring the relative abundances of a particular radioactive isotope and its daughter isotopes in the samples) or by the analyses of fossil evidence, marine magnetic anomalies, or ocean depth. Of these, magnetic anomalies deserve special attention.

A marine magnetic anomaly is a variation in strength of the Earth's magnetic field caused by magnetism in rocks of the ocean floor. Marine magnetic anomalies typically represent 1 percent of the total geomagnetic field strength. They can be stronger ("positive") or weaker ("negative") than the average total field. Also, the magnetic anomalies occur in long bands that run parallel to spreading centres for hundreds of kilometres and may reach up to a few tens of kilometres in width.

Marine magnetic anomalies

Marine magnetic anomalies were first discovered off the coast of the western United States in the late 1950s and completely baffled scientists. The anomalies were charted from southern California to northern Washington and out several hundred kilometres. Victor Vacquier, a geophysicist, noticed that these linear anomalies ended at the fracture zones mapped in this area. In addition, he noticed that they had unique shapes, occurred in a predictable sequence across their trends, and could be correlated across the fracture zones.

Soon thereafter, linear magnetic anomalies were mapped over the Reykjanes Ridge south of Iceland. They were found to occur on both sides of the ridge crest and parallel to it. Simultaneously, Alan Cox and several other American geophysicists documented evidence that the Earth's magnetic field had reversed in the past: the north magnetic pole had been the south magnetic pole about 700,000 years ago, and there were reasons to believe older reversals existed. Also at this time, Robert S. Dietz and Harry H. Hess were formulating the theory of seafloor spreading--the hypothesis that oceanic crust is created at the crests of the oceanic ridges and consumed in the deep-sea trenches.

It remained for Frederick J. Vine and Drummond H. Matthews of Great Britain and Lawrence W. Morley of Canada to put these observations together in a theory that explained marine magnetic anomalies. The theory rests on three assumptions: (1) that the Earth's magnetic field periodically reverses polarity; (2) that seafloor spreading occurs; and (3) that the oceanic crust is permanently magnetized as it forms and cools at spreading centres.

The theory expresses the assumptions--namely, that the oceanic crust records reversals of the Earth's field as it is formed during seafloor spreading. Positive anomalies result when the crust is magnetized in a "normal" polarity parallel to the ambient field of the Earth, and negative anomalies result when the crust is "reversely" magnetized in an opposite sense. As the magnetized crust moves down the flanks of a ridge away from the spreading centre, it remains permanently magnetized and "carries" the magnetic anomalies along with it. (For further details about paleomagnetism and seafloor spreading, see plate tectonics: Historical overview: Renewed interest in continental drift.)

A brilliant leap in understanding was now possible. If the age of the field reversals were known, the age of the ocean crust could be predicted by mapping the corresponding anomaly. By the mid-1960s, Cox and his colleagues had put together a schedule of reversals for the last four or five million years by studying the ages and magnetic polarities of lava flows found on land. Vine and the Canadian geologist J. Tuzo Wilson applied the time scale to marine magnetic anomalies mapped over the Juan de Fuca Ridge, a spreading centre off the northwest United States.

They thus dated the crust there and also computed the first seafloor spreading rate of about 30 millimetres per year. The rate is computed by dividing the distance of an anomaly from the ridge crest by the age of the anomaly twice. Thus the oceanic crust at the Juan de Fuca Ridge is moving at about 15 millimetres per year away from the ridge crest and at about 60 millimetres per year away from the crustal segment on the opposite side of the crest.

During the 1960s and '70s marine magnetic anomalies were mapped over wide areas of the ocean basins. By using estimates of the ages of oceanic crust obtained from core samples by deep-sea drilling, a magnetic anomaly time scale was constructed, and at the same time the spreading history for the ocean basins covering the last 200 million years or so was proposed.

It is thought that the most important contributor to marine magnetic anomalies is the layer of lavas in the upper oceanic crust. A secondary contribution originates in the upper layer of gabbros. The dike layer is essentially demagnetized by the action of hydrothermal waters at the spreading centres.

The dominant mechanism of permanent magnetization is the thermoremanent magnetization (or TRM) of iron-titanium oxide minerals. These minerals lock in a TRM as they cool below 200 to 300 C in the presence of the Earth's magnetic field. Although several processes are capable of altering the TRM, including reheating and oxidation at the seafloor, it is remarkably robust, as is evidenced by magnetic anomalies as old as 165 million years in the far western equatorial Pacific.

Oceanic ridges

The largest features of the ocean basin are the oceanic ridges. In the past these features were referred to as mid-ocean ridges, but, as will be seen, the largest oceanic ridge, the East Pacific Rise, is far from a mid-ocean location, and the nomenclature is thus inaccurate. Oceanic ridges are not to be confused with aseismic ridges, which have an entirely different origin (see below).

Principal characteristics

Oceanic ridges are linear mountain chains comprising the largest features on Earth. They are found in every ocean basin and appear to girdle the Earth. The ridges rise from depths near 5 kilometres to an essentially uniform depth of about 2.6 kilometres and are roughly symmetrical in cross section. They can be thousands of kilometres wide. In places, the crests of the ridges are offset across transform faults, or fracture zones, which can be followed down the flanks of the ridges. (Transform faults are those along which lateral movement occurs.) The flanks are marked by sets of mountains and hills that are elongate and parallel to the ridge trend.

New oceanic crust (and part of the upper mantle, which, together with the crust, makes up the lithosphere) is formed at seafloor spreading centres at the crests of the oceanic ridges. Because of this, certain unique geologic features are found there. Fresh basaltic lavas are exposed on the seafloor at the ridge crests. These lavas are progressively buried by sediments as the seafloor spreads away from the site. The flow of heat out of the crust is many times greater at the crests than elsewhere in the world. Earthquakes are common along the crests and in the transform faults that join the offset ridge segments. Analysis of earthquakes occurring at the ridge crests indicates that the oceanic crust is under tension there. A high-amplitude magnetic anomaly is centred over the crests because fresh lavas at the crests are being magnetized in the direction of the present geomagnetic field.

The depths over the oceanic ridges are rather precisely correlated with the age of the ocean crust; specifically, it has been demonstrated that the ocean depth is proportional to the square root of crustal age. The theory explaining this relationship holds that the increase in depth with age is due to the thermal contraction of the oceanic crust and upper mantle as they are carried away from the seafloor spreading centre in an oceanic plate. Because such a plate is ultimately about 100 kilometres thick, contraction of only a few percent predicts the entire relief of an oceanic ridge. It then follows that the width of a ridge can be defined as twice the distance from the crest to the point where the plate has cooled to a steady thermal state.

Most of the cooling takes place within 70 or 80 million years, by which time the ocean depth is about 5 to 5.5 kilometres. Because this cooling is a function of age, slow-spreading ridges, such as the Mid-Atlantic Ridge, are narrower than faster-spreading ridges, like the East Pacific Rise (see below). Further, a correlation has been found between global spreading rates and the transgression and regression of ocean waters onto the continents. During the Early Cretaceous period about 100 million years ago, when global spreading rates were uniformly high, oceanic ridges occupied comparatively more of the ocean basins, causing the ocean waters to transgress (spill over) onto the continents, leaving marine sediments in areas now well away from coastlines.

Besides ridge width, other features appear to be a function of spreading rate. Global spreading rates range from 10 millimetres per year (mm/yr total rate) or less up to 160 mm/yr. Oceanic ridges can be classified as slow (up to 50 mm/yr), intermediate (up to 90 mm/yr), and fast (up to 160 mm/yr). Slow-spreading ridges are characterized by a rift valley at the crest. Such a valley is fault-controlled. It is typically 1.4 kilometres deep and 20 to 40 kilometres wide. Faster-spreading ridges lack rift valleys. At intermediate rates, the crest regions are broad highs with occasional fault-bounded valleys no deeper than 200 metres. At fast rates, an axial high is present at the crest. The slow-spreading rifted ridges have rough faulted topography on their flanks, while the faster-spreading ridges have much smoother flanks.