Want to create or adapt books like this? Learn more about how Pressbooks supports open publishing practices.
52 7.7 Thermohaline Circulation
The surface currents we have discussed so far are ultimately driven by the wind, and since they only involve surface water they only affect about 10% of the ocean’s volume. However, there are other significant ocean currents that are independent of the wind, and involve water movements in the other 90% of the ocean. These currents are driven by differences in water density.
Recall that less dense water remains at the surface, while denser water sinks. Waters of different densities tend to stratify themselves into layers, with the densest, coldest water on the bottom and warmer, less dense water on top. It is the movement of these density layers that create the deep water circulation. Since seawater density depends mainly on temperature and salinity (section 6.3), this circulation is referred to as thermohaline circulation.
The main processes that increase seawater density are cooling, evaporation, and ice formation. Evaporation and ice formation cause an increase in density by removing fresh water, leaving the remaining seawater with greater salinity (see section 5.3). The main processes that decrease seawater density are heating, and dilution by fresh water through precipitation, ice melting, or fresh water runoff. Note that all of these processes exert their effects at the surface, but don’t necessarily affect deeper water. However, changing the density of the surface water causes it to sink or rise, and these vertical, density-driven movements create the deep ocean currents. These thermohaline currents are slow, on the order of 10-20 km per year compared with surface currents that move at several kilometers per hour.
Water masses
A water mass is a volume of seawater with a distinctive density as a result of its unique profile of temperature and salinity. As stated above, the processes that affect seawater density really only happen at the surface. Once a water mass has reached its particular temperature and salinity profile due to these surface processes, it may sink below the surface, at which point its density properties won’t really change. We can therefore distinguish particular water masses by taking salinity and temperature measurements at different depths, and looking for the unique combination of these variables that give it its characteristic density. This is often carried out using temperature-salinity diagrams (T-S diagrams, see box below).
There are several well-known water masses in the ocean, particularly in the Atlantic, that are distinguished by their temperature and salinity characteristics. The densest ocean water is formed in two primary locations near the poles, where the water is very cold and highly saline as a result of ice formation. The densest deep water mass is formed in the Weddell Sea of Antarctica, and becomes the Antarctic Bottom Water (AABW). Similar processes in the North Atlantic produce the North Atlantic Deep Water (NADW) in the Greenland Sea (Figure 7.7.1).
Figure 7.7.1 The primary sites of deep water formation; Antarctic Bottom Water is formed int he Weddell Sea, and North Atlantic Deep water is formed in the Greenland Sea (PW).
This cold, dense water sinks, and once it is removed from the surface, its temperature and salinity remain unchanged, so it keeps the same characteristics as it moves throughout the ocean as part of the thermohaline circulation. AABW sinks to the bottom in the Weddell Sea and then moves north along the bottom into the Atlantic, and east through the Southern Ocean. At the same time NADW is sinking in the Greenland Sea. This water mass is less dense than AABW and tends to form a layer above the AABW as it flows across the equator to the south (Figure 7.7.2). As the NADW moves towards the Antarctic continent, it is brought to the surface. Recall that near Antarctica there is the Antarctic divergence, where surface waters move horizontally away from each other, and are replaced by deep water upwelling (bringing nutrients to the surface and leading to high productivity; see section 8.3). Since polar water has a weak thermocline, there isn’t much of a density difference preventing the deep water from reaching the surface, so some NADW rises as part of the upwelling process (Figure 7.7.2).
Figure 7.7.2 The major water masses of the Atlantic Ocean (PW).
As the rising NADW reaches the surface, some travels south where it will eventually contribute to the production of new AABW. The NADW that moves north encounters the Antarctic convergence, which produces downwelling. This sinking NADW becomes a new water mass; Antarctic Intermediate Water (AAIW), which sinks and creates a layer between the surface water and the NADW (Figure 7.7.2). The surface water in the equatorial Atlantic, also called the Central Atlantic Surface Water, is very warm and low density, therefore it remains at the surface and does not contribute much to thermohaline circulation.
In the Atlantic, Mediterranean Intermediate Water(MIW) flows through the Straits of Gibraltar into the open ocean. This water is warm and salty from the warm temperatures and high evaporation characteristic of the Mediterranean Sea, so it is denser than the normal surface water and forms a layer about 1-1.5 km deep. Eventually this water will move north to the Greenland Sea, where it will be cooled and will sink, becoming the dense NADW.
T-S Diagrams
A temperature-salinity (T-S) diagram is used to examine how temperature, salinity, and density change with depth, and to identify the vertical structure of the water column, including the water masses it contains. Water temperature is on the y-axis, and salinity appears on the x-axis. Often, instead of the actual temperature of the water, oceanographers plot potential temperature, which is the temperature the water would achieve if it was brought to the surface and did not get any additional heat through compression at depth. A T-S diagram shows lines of equal density, or isopycnals, for various combinations of temperature and salinity (Figure 7.7.3). You can then plot the values of temperature and salinity on the diagram, and use their point of intersection to calculate the density of the water. In the example in Figure 7.7.3 a temperature of about 11o C and a salinity of 34.6 PSU results in a density of 1.0265 g/cm3.
Figure 7.7.3 Using a T-S diagram to determine density. A temperature of about 11o C (green arrow) and a salinity of 34.6 PSU (red arrow) results in a density of 1.0265 g/cm3.
Since the range of densities in the ocean is rather small, often the density value is shortened and is expressed as sigma-t or σt. Sigma-t is calculated as: (density – 1) x 1000. So it essentially just looks at the last three decimal places of the density value. Thus a density of 1.0275 g/cm3 would have a σt of 27.5.
T-S diagrams can be used to identify water masses. Since each major water mass has its own characteristic range of temperatures and salinities, a deep water sample that falls into that range can presumably have come from that water mass. Figure 7.7.4 shows the typical range of temperature and salinity for the major Atlantic water masses.
Figure 7.7.4 Characteristic ranges of temperature and salinity for the major Atlantic water masses; North Atlantic Central Surface Water (NACSW), Mediterranean Intermediate Water (MIW), Antarctic Intermediate Water (AAIW), North Atlantic Deep Water (NADW), and Antarctic Bottom Water (AABW).
To investigate water masses, oceanographers can take a series of temperature and salinity measurements over a range of depths at a particular location. If the water column was highly stratified and there was no mixing between or within the layers, as the probe was lowered your would get a series of constant temperature and salinity readings as you moved through the first water mass, followed by a sudden jump to another set of different but constant readings as you moved through the next water mass. Plotting temperature vs. salinity on a T-S diagram would result in a distinct and independent point for each water mass. However, in reality, the water masses will show some mixing within and between layers. So as the probes are lowered, they will encounter water that shows traits intermediate between the two points. Therefore, with increasing depth, the points on the T-S diagram will gradually move from one point to the other, creating a line connecting the two points, illustrating mixing between those two water masses.
In the example in Figure 7.7.5, NACSW is present at the surface (0 m depth), and between 0 and about 800 m there is a transition from NACSE into AAIW. Between about 800-2100 m there is a transition from AAIW into the NADW layer just beyond 2000 m. AABW is the deepest water mass, at depths of about 4000 m. The transition between NADW and AABW occurs between about 2100-4000 m.
Figure 7.7.5 Hypothetical T-S diagram for the North Atlantic. Points represent readings taken at the corresponding depths (m). Moving from the surface to the bottom results in water of increasing density, passing through distinct water masses.
Notice that as the recordings get deeper in Figure 7.7.5, the density is always increasing (i.e. moving towards the bottom right corner). This is because the densest water should be located at the bottom, with the other layers stratified according to their density, otherwise the water column would be unstable.
The “Ocean Conveyor Belt”
The bottom water from the Weddell Sea and Greenland Sea does not just circulate through the Atlantic. NADW moves south through the western Atlantic before meeting the AABW north of the Weddell Sea. Together these water masses move eastwards into the Indian and Pacific Oceans. By this time the NADW and AABW have started mixing, to create what is called Common Water. The deep Common Water moves northwards into the Pacific and Indian Oceans and gradually mixes with the warmer water, causing it to eventually rise to the surface. As surface water, it makes its way back to the North Atlantic through the surface currents of the Pacific and Indian Oceans. Once back in the North Atlantic, it cools and once again forms NADW, starting the process anew. This cycle of rising and sinking water transporting water between the surface and deep circulation has been referred to as the global oceanic “conveyor belt”, and may take about 1000-2000 years to complete (Figure 7.7.6).
Figure 7.7.6 The global ocean “conveyor belt.” Cold, dense water sinks in the Greenland and Weddell Seas and circulates over the seafloor into the Indian and Pacific Oceans (blue paths). Eventually the water rises to the surface, and returns to the site of bottom water formation via surface currents (red paths), to start the cycle again (By Robert Simmon, NASA. Minor modifications by Robert A. Rohde also released to the public domain (NASA Earth Observatory) [Public domain], via Wikimedia Commons).
This global circulation pattern has a number of important implications for Earth’s environment. For one, it is vital to the transport of heat around the globe, bringing warm water towards the poles, and cold water to the tropics, stabilizing temperature in both environments.
The conveyor belt also helps deliver oxygen to deep water habitats. The deep water began as cold surface water that was saturated with oxygen, and when it sank it brought that oxygen to depth. Thermohaline circulation carries this oxygen-rich deep water throughout the oceans, where the oxygen will be used by deep water organisms. Bottom water in the Atlantic is relatively high in oxygen, as it still retains much of its original oxygen content, but as it travels over the seafloor the oxygen is used up, so that deep water in the Pacific Ocean has much less oxygen than deep Atlantic water, with Indian Ocean water somewhere in between. At the same time, deep water will accumulate nutrients as organic matter sinks and decomposes. Atlantic bottom water is low in nutrients because it has not had much time to accumulate them, and the original surface water was nutrient-poor. By the time this bottom water reaches the Indian Ocean, and after that the Pacific, it has been accumulating the sinking nutrients for centuries, so deep nutrient concentrations are greater in the Pacific than the Atlantic. We can therefore use the ratios of oxygen to nutrients in the deep water to tell the relative age of a water mass, i.e. how long it has been since it sank from the surface. Younger bottom water should be high in oxygen and low in nutrients, while the opposite would be expected for older bottom water.
The ocean conveyor belt may be significantly impacted by climate change disrupting thermohaline circulation. Increased warming, particularly in the Arctic, could lead to continuing melting of the polar ice caps, adding a large amount of fresh water to the polar surface water. This input of fresh water could create a low density, low salinity surface layer of water that no longer sinks, thus disrupting the deep circulation conveyor belt and preventing oxygen and nutrient transport to bottom communities. The sinking of seawater in the Greenland Sea also helps drive the Gulf Stream; as water sinks, more surface water is pulled northwards in the Gulf Stream. If this polar water stops sinking the Gulf Stream could weaken, reducing heat transport to the poles and cooling the northern climate. It seems counter intuitive, but global warming could lead to colder conditions in Europe and the freezing of ports and cities that are usually ice-free due to the warming effects of the Gulf Stream. Recent evidence has already shown that the strength of the Gulf Stream is waning, likely due to the increased melting of Arctic ice.
All of the salts and ions that dissolve in seawater contribute to its overall salinity. Salinity of seawater is usually expressed as the grams of salt per kilogram (1000 g) of seawater. On average, about 35 g of salt is present in each 1 kg of seawater, so we say that the average salinity of the ocean salinity is 35 parts per thousand (ppt). Note that 35 ppt is equivalent to 3.5% (parts per hundred). Some sources now use practical salinity units (PSU) to express salinity values, where 1 PSU = 1 ppt. The units are not included, so we can refer simply to a salinity of 35.
Many different substances are dissolved in the ocean, but six ions comprise about 99.4% of all the dissolved ions in seawater. These six major ions are (Table 5.3.1):
Table 5.3.1 The six major ions in seawater
g/kg in seawater
% of ions by weight
Chloride Cl-
19.35
55.07%
Sodium Na+
10.76
30.6%
Sulfate SO42-
2.71
7.72%
Magnesium Mg2+
1.29
3.68%
Calcium Ca2+
0.41
1.17%
Potassium K+
0.39
1.1%
99.36%
Figure 5.3.1 The relative proportions of ions in seawater. (By derivative work: Tcncv (talk) Sea_salt-e_hg.svg: Hannes Grobe, Alfred Wegener Institute for Polar and Marine Research, Bremerhaven, Germany; SVG version by Stefan Majewsky (Sea_salt-e_hg.svg) [CC BY-SA 2.5], via Wikimedia Commons)
Chloride and sodium, the components of table salt (sodium chloride NaCl), make up over 85% of the ions in the ocean, which is why seawater tastes salty (Figure 5.3.1). In addition to the major constituents, there are numerous minor constituents; radionucleotides, organic compounds, metals etc. These minor constituents are found in concentrations of ppm (parts per million) or ppb (parts per billion), unlike the major ions that are far more abundant (ppt) (Table 5.3.2). To put this into perspective, 1 ppm = 1 mg/kg, or the equivalent of 1 teaspoon of sugar dissolved in 14,000 cans of soda. 1 ppb = 1 μg/kg, or the equivalent of 1 teaspoon of a substance dissolved in five Olympic-sized swimming pools! These minor constituents represent numerous substances, but together they make up less than 1% of the ions in the seawater. Some of these may be important as minerals and trace elements vital to living organisms, but they don’t have much impact on the overall salinity. But given the vast size of the oceans, even materials found in trace abundance can represent fairly large reservoirs. For example gold is a trace element in seawater, found in concentrations of parts per trillion, yet if you could extract all of the gold in just one km3 of seawater, it would be worth about $20 million!
Table 5.3.2 Concentrations of some minor elements in seawater
g/kg in seawater
g/kg in seawater
Carbon
0.028
Iron
2 x 10-6
Nitrogen
0.0115
Manganese
2 x 10-7
Oxygen
0.006
Copper
1 x 10-7
Silicon
0.002
Mercury
3 x 10-8
Phosphorous
6 x 10-5
Gold
4 x 10-9
Uranium
3.2 x 10-6
Lead
5 x 10-10
Aluminum
2 x 10-6
Radon
6 x 10-19
Because the six major ions in seawater comprise over 99% of the total salinity, changes in abundance of the minor constituents have little effect on overall salinity. Furthermore, the rule of constant proportions states that even though the absolute salinity of ocean water might differ in different places, the relative proportions of the six major ions within that water are always constant. For example, no matter the total salinity of a seawater sample, 55% of the total salinity will be due to chloride, 30% due to sodium, and so on. Since the proportion of these major ions does not change, we call these conservative ions.
Given these constant proportions, in order to calculate total salinity you can simply measure the concentration of just one of the major ions and use that value to calculate the rest. Traditionally chloride has been the ion measured because it is the most abundant, and thus the simplest to measure accurately. Multiplying the concentration of chloride by 1.8 gives the total salinity. For example, looking at Figure 5.3.1, 19.25 g/kg (ppt) chloride x 1.8 = 35 ppt. Today, for rapid measurements of salinity, electrical conductivity is often used rather than determining chloride concentrations (see box below).
Measuring salinity
There are a number of methods available for measuring the salinity of water. The most precise measurements utilize direct chemical analysis of the seawater in a lab setting, but there are a number of ways to get immediate salinity measurements in the field. For a quick estimate of salinity, a hand-held refractometer can be used (right).
Handheld refractometer (Photo by CEphoto, Uwe Aranas / CC-BY-SA-3.0 via Wikimedia Commons)
This instrument measures the degree of bending, or refraction, of light rays as they pass through a fluid. The greater the amount of dissolved salts in the sample, the greater the degree of light refraction. The observer traps a drop of water on the blue screen, and looks through the eyepiece. The dividing line between the blue and white sections of the scale (inset) can be used to read the salinity.
For more accurate measurements, most oceanographers use an instrument that measures electrical conductivity. An electrical current is passed between two electrodes immersed in water, and the higher the salinity, the more readily the current will be conducted (the ions in seawater conduct electrical currents). Conductivity probes are often bundled into an instrument called a CTD, which stands for Conductivity, Temperature, and Depth, which are the most commonly-measured parameters. Modern CTDs can be outfitted with an array of probes measuring parameters like light, turbidity (water clarity), dissolved gases etc. CTDs can be large instruments (below), but small hand-held salinity probes are also widely available.
For large-scale salinity measurements, oceanographers can use satellites, such as the Aquarius satellite, which was able to measure surface salinity differences as small as 0.2 PSU as it mapped the ocean surface every seven days (below).
(NASA, Public Domain)
It is important to be aware that while the rule of constant proportions applies to most of the ocean, there may be certain coastal areas where lots of river discharge may alter these proportions slightly. Furthermore, it is important to remember that the rule of constant proportions only applies to the major ions. The proportions of the minor ions may fluctuate, but remember that they make a very minor contribution to overall salinity. Because the concentrations of the minor ions are not constant, these are referred to as non-conservative ions.
Why are the major ions found in constant proportions? There is constant input of ions from river runoff and other processes, usually in very different proportions from what is found in the ocean. So why don’t the proportions in the oceans change? Most of the ions discharged by rivers have fairly low residence times (see section 5.2) compared to ions in seawater, usually because they are used in biological processes. These low residence times do not allow the ions to accumulate and alter salinity. Also, the mixing time of the world ocean is around 1000 years, which is very short compared to the residence times of the major ions, which may be tens of millions of years long. So during the residence time of a single ion the ocean has mixed numerous times, and the major ions have become evenly distributed throughout the ocean.
Variations in Salinity
Total salinity in the open ocean averages 33-37 ppt, but it can vary significantly in different locations. But since the major ion proportions are constant, the regional salinity differences must be due more to water input and removal rather than the addition or removal of ions. Fresh water input comes through processes like precipitation, runoff from land, and melting ice. Fresh water removal primarily comes from evaporation and freezing (when seawater freezes, the resulting ice is mostly fresh water and the salts are excluded, making the remaining water even saltier). So differences in rates of precipitation, evaporation, river discharge, and ice formation play a significant role in regional salinity variations. For example, the Baltic Sea has a very low surface salinity of around 10 ppt, because it is a mostly enclosed body of water with lots of river input. Conversely, the Red Sea is very salty (around 40 ppt), due to the lack of precipitation and the hot environment which leads to high levels of evaporation.
One of the saltiest large bodies of water on Earth is the Dead Sea, between Israel and Jordan. Salinity in the Dead Sea is around 330 ppt, which is almost ten times saltier than the ocean. This extremely high salinity is a result of the hot, arid conditions in the Middle East that lead to high rates of evaporation. In addition, in the 1950s the flow from the Jordan River was diverted away from the Dead Sea, so there is no longer significant fresh water input. With no input and high evaporation, the water level in the Dead Sea is receding at a rate of about 1 m per year. The high salinity makes the water very dense, which creates buoyant forces that allow people to easily float at the surface. But the high salinity also means that the water is too salty for most living organisms, so only microbes are able to call it home; hence the name the Dead Sea. But as salty as the Dead Sea may be, it is not the saltiest body of water on Earth. That distinction currently belongs to Gaet’ale Pond in Ethiopia, with a salinity of 433 ppt!
Latitudinal Variations
While local conditions are important for determining salinity patterns in any single location, there are some global patterns that bear further investigation. Temperature is highest at the equator, and lowest near the poles, so we would expect higher rates of evaporation, and therefore higher salinity, in equatorial regions (Figure 5.3.2). This is generally the case, but in the figure below salinity right along the equator seems to be a little lower than at slightly higher latitudes. This is because equatorial regions also get a high volume of rain on a regular basis, which dilutes the surface water along the equator. So the higher salinities are found at subtropical, warm latitudes with high evaporation and less precipitation. At the poles there is little evaporation, which, coupled with ice and snow melting, produces a relatively low surface salinity. The image below shows high salinity in the Mediterranean Sea; this is located in a warm region with high evaporation, and the sea is largely isolated from mixing with the rest of the North Atlantic water, leading to high salinity. Lower salinities, such as those around southeast Asia, are the result of precipitation and high volumes of river input.
Figure 5.3.2 Annual mean global sea surface salinity (Data from World Ocean Atlas 2009. By Plumbago - Own work, [CC BY-SA 3.0], via Wikimedia Commons).
Figure 5.3.3 shows the mean global differences between evaporation and precipitation (evaporation - precipitation). Green colors represent areas where precipitation exceeds evaporation, while brown regions are where evaporation is greater than precipitations. Note the correlation between precipitation, evaporation, and surface salinity as seen in Figure 5.3.2.
Figure 5.3.3 Annual mean global differences in evaporation and precipitation (J.Fasullo. National Center for Atmospheric Research Staff (Eds). "The Climate Data Guide: ERA-Interim: derived components." Retrieved from https://climatedataguide.ucar.edu/climate-data/era-interim-derived-components).
Vertical Variation
In addition to geographical variation in salinity, there are also changes in salinity with depth. As we have seen, most differences in salinity are due to variations in evaporation, precipitation, runoff, and ice cover. All of these process occur at the ocean surface, not at depth, so the most pronounced differences in salinity should be found in surface waters. Salinity in deeper water remains relatively uniform, as it is unaffected by these surface processes. Some representative salinity profiles are shown in Figure 5.3.4. At the surface, the top 200 m or so show relatively uniform salinity in what is called the mixed layer. Winds, waves, and surface currents stir up the surface water, causing a great deal of mixing in this layer and fairly uniform salinity conditions. Below the mixed layer is an area of rapid salinity change over a small change in depth. This zone of rapid change is called the halocline, and it represents a transition between the mixed layer and the deep ocean. Below the halocline, salinity may show little variation down to the seafloor, as this region is far removed from the surface processes that impact salinity. In the figure below, note the low surface salinity at high latitudes, and higher surface salinity at low latitudes as discussed above. Yet despite the surface differences, salinity at depth in both locations may be very similar.
Figure 5.3.4 Salinity profiles from two hypothetical sites in the open ocean, one from high latitude and one from low latitude (PW).
All of the salts and ions that dissolve in seawater contribute to its overall salinity. Salinity of seawater is usually expressed as the grams of salt per kilogram (1000 g) of seawater. On average, about 35 g of salt is present in each 1 kg of seawater, so we say that the average salinity of the ocean salinity is 35 parts per thousand (ppt). Note that 35 ppt is equivalent to 3.5% (parts per hundred). Some sources now use practical salinity units (PSU) to express salinity values, where 1 PSU = 1 ppt. The units are not included, so we can refer simply to a salinity of 35.
Many different substances are dissolved in the ocean, but six ions comprise about 99.4% of all the dissolved ions in seawater. These six major ions are (Table 5.3.1):
Table 5.3.1 The six major ions in seawater
g/kg in seawater
% of ions by weight
Chloride Cl-
19.35
55.07%
Sodium Na+
10.76
30.6%
Sulfate SO42-
2.71
7.72%
Magnesium Mg2+
1.29
3.68%
Calcium Ca2+
0.41
1.17%
Potassium K+
0.39
1.1%
99.36%
Figure 5.3.1 The relative proportions of ions in seawater. (By derivative work: Tcncv (talk) Sea_salt-e_hg.svg: Hannes Grobe, Alfred Wegener Institute for Polar and Marine Research, Bremerhaven, Germany; SVG version by Stefan Majewsky (Sea_salt-e_hg.svg) [CC BY-SA 2.5], via Wikimedia Commons)
Chloride and sodium, the components of table salt (sodium chloride NaCl), make up over 85% of the ions in the ocean, which is why seawater tastes salty (Figure 5.3.1). In addition to the major constituents, there are numerous minor constituents; radionucleotides, organic compounds, metals etc. These minor constituents are found in concentrations of ppm (parts per million) or ppb (parts per billion), unlike the major ions that are far more abundant (ppt) (Table 5.3.2). To put this into perspective, 1 ppm = 1 mg/kg, or the equivalent of 1 teaspoon of sugar dissolved in 14,000 cans of soda. 1 ppb = 1 μg/kg, or the equivalent of 1 teaspoon of a substance dissolved in five Olympic-sized swimming pools! These minor constituents represent numerous substances, but together they make up less than 1% of the ions in the seawater. Some of these may be important as minerals and trace elements vital to living organisms, but they don’t have much impact on the overall salinity. But given the vast size of the oceans, even materials found in trace abundance can represent fairly large reservoirs. For example gold is a trace element in seawater, found in concentrations of parts per trillion, yet if you could extract all of the gold in just one km3 of seawater, it would be worth about $20 million!
Table 5.3.2 Concentrations of some minor elements in seawater
g/kg in seawater
g/kg in seawater
Carbon
0.028
Iron
2 x 10-6
Nitrogen
0.0115
Manganese
2 x 10-7
Oxygen
0.006
Copper
1 x 10-7
Silicon
0.002
Mercury
3 x 10-8
Phosphorous
6 x 10-5
Gold
4 x 10-9
Uranium
3.2 x 10-6
Lead
5 x 10-10
Aluminum
2 x 10-6
Radon
6 x 10-19
Because the six major ions in seawater comprise over 99% of the total salinity, changes in abundance of the minor constituents have little effect on overall salinity. Furthermore, the rule of constant proportions states that even though the absolute salinity of ocean water might differ in different places, the relative proportions of the six major ions within that water are always constant. For example, no matter the total salinity of a seawater sample, 55% of the total salinity will be due to chloride, 30% due to sodium, and so on. Since the proportion of these major ions does not change, we call these conservative ions.
Given these constant proportions, in order to calculate total salinity you can simply measure the concentration of just one of the major ions and use that value to calculate the rest. Traditionally chloride has been the ion measured because it is the most abundant, and thus the simplest to measure accurately. Multiplying the concentration of chloride by 1.8 gives the total salinity. For example, looking at Figure 5.3.1, 19.25 g/kg (ppt) chloride x 1.8 = 35 ppt. Today, for rapid measurements of salinity, electrical conductivity is often used rather than determining chloride concentrations (see box below).
Measuring salinity
There are a number of methods available for measuring the salinity of water. The most precise measurements utilize direct chemical analysis of the seawater in a lab setting, but there are a number of ways to get immediate salinity measurements in the field. For a quick estimate of salinity, a hand-held refractometer can be used (right).
Handheld refractometer (Photo by CEphoto, Uwe Aranas / CC-BY-SA-3.0 via Wikimedia Commons)
This instrument measures the degree of bending, or refraction, of light rays as they pass through a fluid. The greater the amount of dissolved salts in the sample, the greater the degree of light refraction. The observer traps a drop of water on the blue screen, and looks through the eyepiece. The dividing line between the blue and white sections of the scale (inset) can be used to read the salinity.
For more accurate measurements, most oceanographers use an instrument that measures electrical conductivity. An electrical current is passed between two electrodes immersed in water, and the higher the salinity, the more readily the current will be conducted (the ions in seawater conduct electrical currents). Conductivity probes are often bundled into an instrument called a CTD, which stands for Conductivity, Temperature, and Depth, which are the most commonly-measured parameters. Modern CTDs can be outfitted with an array of probes measuring parameters like light, turbidity (water clarity), dissolved gases etc. CTDs can be large instruments (below), but small hand-held salinity probes are also widely available.
For large-scale salinity measurements, oceanographers can use satellites, such as the Aquarius satellite, which was able to measure surface salinity differences as small as 0.2 PSU as it mapped the ocean surface every seven days (below).
(NASA, Public Domain)
It is important to be aware that while the rule of constant proportions applies to most of the ocean, there may be certain coastal areas where lots of river discharge may alter these proportions slightly. Furthermore, it is important to remember that the rule of constant proportions only applies to the major ions. The proportions of the minor ions may fluctuate, but remember that they make a very minor contribution to overall salinity. Because the concentrations of the minor ions are not constant, these are referred to as non-conservative ions.
Why are the major ions found in constant proportions? There is constant input of ions from river runoff and other processes, usually in very different proportions from what is found in the ocean. So why don’t the proportions in the oceans change? Most of the ions discharged by rivers have fairly low residence times (see section 5.2) compared to ions in seawater, usually because they are used in biological processes. These low residence times do not allow the ions to accumulate and alter salinity. Also, the mixing time of the world ocean is around 1000 years, which is very short compared to the residence times of the major ions, which may be tens of millions of years long. So during the residence time of a single ion the ocean has mixed numerous times, and the major ions have become evenly distributed throughout the ocean.
Variations in Salinity
Total salinity in the open ocean averages 33-37 ppt, but it can vary significantly in different locations. But since the major ion proportions are constant, the regional salinity differences must be due more to water input and removal rather than the addition or removal of ions. Fresh water input comes through processes like precipitation, runoff from land, and melting ice. Fresh water removal primarily comes from evaporation and freezing (when seawater freezes, the resulting ice is mostly fresh water and the salts are excluded, making the remaining water even saltier). So differences in rates of precipitation, evaporation, river discharge, and ice formation play a significant role in regional salinity variations. For example, the Baltic Sea has a very low surface salinity of around 10 ppt, because it is a mostly enclosed body of water with lots of river input. Conversely, the Red Sea is very salty (around 40 ppt), due to the lack of precipitation and the hot environment which leads to high levels of evaporation.
One of the saltiest large bodies of water on Earth is the Dead Sea, between Israel and Jordan. Salinity in the Dead Sea is around 330 ppt, which is almost ten times saltier than the ocean. This extremely high salinity is a result of the hot, arid conditions in the Middle East that lead to high rates of evaporation. In addition, in the 1950s the flow from the Jordan River was diverted away from the Dead Sea, so there is no longer significant fresh water input. With no input and high evaporation, the water level in the Dead Sea is receding at a rate of about 1 m per year. The high salinity makes the water very dense, which creates buoyant forces that allow people to easily float at the surface. But the high salinity also means that the water is too salty for most living organisms, so only microbes are able to call it home; hence the name the Dead Sea. But as salty as the Dead Sea may be, it is not the saltiest body of water on Earth. That distinction currently belongs to Gaet’ale Pond in Ethiopia, with a salinity of 433 ppt!
Latitudinal Variations
While local conditions are important for determining salinity patterns in any single location, there are some global patterns that bear further investigation. Temperature is highest at the equator, and lowest near the poles, so we would expect higher rates of evaporation, and therefore higher salinity, in equatorial regions (Figure 5.3.2). This is generally the case, but in the figure below salinity right along the equator seems to be a little lower than at slightly higher latitudes. This is because equatorial regions also get a high volume of rain on a regular basis, which dilutes the surface water along the equator. So the higher salinities are found at subtropical, warm latitudes with high evaporation and less precipitation. At the poles there is little evaporation, which, coupled with ice and snow melting, produces a relatively low surface salinity. The image below shows high salinity in the Mediterranean Sea; this is located in a warm region with high evaporation, and the sea is largely isolated from mixing with the rest of the North Atlantic water, leading to high salinity. Lower salinities, such as those around southeast Asia, are the result of precipitation and high volumes of river input.
Figure 5.3.2 Annual mean global sea surface salinity (Data from World Ocean Atlas 2009. By Plumbago - Own work, [CC BY-SA 3.0], via Wikimedia Commons).
Figure 5.3.3 shows the mean global differences between evaporation and precipitation (evaporation - precipitation). Green colors represent areas where precipitation exceeds evaporation, while brown regions are where evaporation is greater than precipitations. Note the correlation between precipitation, evaporation, and surface salinity as seen in Figure 5.3.2.
Figure 5.3.3 Annual mean global differences in evaporation and precipitation (J.Fasullo. National Center for Atmospheric Research Staff (Eds). "The Climate Data Guide: ERA-Interim: derived components." Retrieved from https://climatedataguide.ucar.edu/climate-data/era-interim-derived-components).
Vertical Variation
In addition to geographical variation in salinity, there are also changes in salinity with depth. As we have seen, most differences in salinity are due to variations in evaporation, precipitation, runoff, and ice cover. All of these process occur at the ocean surface, not at depth, so the most pronounced differences in salinity should be found in surface waters. Salinity in deeper water remains relatively uniform, as it is unaffected by these surface processes. Some representative salinity profiles are shown in Figure 5.3.4. At the surface, the top 200 m or so show relatively uniform salinity in what is called the mixed layer. Winds, waves, and surface currents stir up the surface water, causing a great deal of mixing in this layer and fairly uniform salinity conditions. Below the mixed layer is an area of rapid salinity change over a small change in depth. This zone of rapid change is called the halocline, and it represents a transition between the mixed layer and the deep ocean. Below the halocline, salinity may show little variation down to the seafloor, as this region is far removed from the surface processes that impact salinity. In the figure below, note the low surface salinity at high latitudes, and higher surface salinity at low latitudes as discussed above. Yet despite the surface differences, salinity at depth in both locations may be very similar.
Figure 5.3.4 Salinity profiles from two hypothetical sites in the open ocean, one from high latitude and one from low latitude (PW).
The primary surface current along the east coast of the United States is the Gulf Stream, which was first mapped by Benjamin Franklin in the 18th century (Figure 7.2.1). As a strong, fast current, it reduced the sailing time for ships traveling from the United States back to Europe, so sailors would use thermometers to locate its warm water and stay within the current.
Figure 7.2.1 Benjamin Franklin's original map of the Gulf Stream (Public domain, via Wikimedia Commons).
The Gulf Stream is formed from the convergence of the North Atlantic Equatorial Current bringing tropical water from the east, and the Florida Current that brings warm water from the Gulf of Mexico. The Gulf Stream takes this warm water and transports it northwards along the U.S. east coast (Figure 7.2.2). As a western boundary current, the Gulf Stream experiences western intensification (section 7.4), making the current narrow (50-100 km wide), deep (to depths of 1.5 km) and fast. With an average speed of 6.4 km/hr, and a maximum speed of about 9 km/hr, it is the fastest current in the world ocean. It also transports huge amounts of water, more than 100 times greater than the combined flow of all of the rivers on Earth.
Figure 7.2.2 Sea surface temperature map illustrating the Gulf Stream. Warmer water is shown in red, colder water in blue and violet. Meanders and eddies are visible where the current moves towards the northeast (NASA, Public Domain via Wikimedia Commons).
As the Gulf Stream approaches Canada, the current becomes wider and slower as the flow dissipates and it encounters the cold Labrador Current moving in from the north. At this point, the current begins to meander, or change from a fast, straight flow to a slower, looping current (Figure 7.2.2). Often these meanders loop so much that they pinch off and form large rotating water masses called rings or eddies, that separate from the Gulf Stream. If an eddy pinches off from the north side of the Gulf Stream, it entraps a mass of warm water and moves it north into the surrounding cold water of the North Atlantic. These warm core rings are shallow, bowl-shaped water masses about 1 km deep, and about 100 km across, that rotate clockwise as they carry warm water in to the North Atlantic (Figure 7.2.3). If the meanders pinch off at the southern boundary of the Gulf Stream, they form cold core rings that rotate counterclockwise and move to the south. Cold core rings are cone-shaped water masses extending down to over 3.5 km deep, and may be over 500 km wide at the surface.
Figure 7.2.3 Formation of warm and cold core rings from meanders in the Gulf Stream. As the Gulf Stream flows to the northeast (1), it starts to meander as it slows, forming warm or cold water extensions on either side of the current (2). If the meanders pinch off the extensions, they trap pockets of warm or cold water (3), that can separate from the Gulf Stream and travel north or south. Warm core rings rotate clockwise, while cold core rings rotate counterclockwise (4) (PW).
After the Gulf Stream meets the cold Labrador Current, it joins the North Atlantic Current, which transports the warm water towards Europe, where it moderates the European climate. It is estimated that Northern Europe is up to 9o C warmer than expected because of the Gulf Stream, and the warm water helps to keep many northern European ports ice-free in the winter.
In the east, the Gulf Stream merges into the Sargasso Sea, which is the area of the ocean within the rotation center of the North Atlantic gyre. The Sargasso Sea gets its name from the large floating mats of the marine algae Sargassum that are abundant on the surface (Figure 7.2.4). These Sargassum mats may play an important role in the early life stages of sea turtles, who may live and feed within the algae for many years before reaching adulthood.
Figure 7.2.4 Map of the Sargasso Sea (left), and closeup photo of Sargassum algae (right) (Map by USFWS; photo by Bogdan Giușcă; Public Domain via Wikimedia Commons).
Differential heating of the Earth's surface results in equatorial regions receiving more heat than the poles (section 6.1). As air is warmed at the equator it becomes less dense and rises, while at the poles the cold air is denser and sinks. If the Earth was non-rotating, the warm air rising at the equator would reach the upper atmosphere and begin moving horizontally towards the poles. As the air reached the poles it would cool and sink, and would move over the surface of Earth back towards the equator. This would result in one large atmospheric convection cell in each hemisphere (Figure 6.2.1), with air rising at the equator and sinking at the poles, and the movement of air over the Earth's surface creating the winds. On this non-rotating Earth, the prevailing winds would thus blow from the poles towards the equator in both hemispheres (Figure 6.2.1).
Figure 6.2.1 Hypothetical atmospheric convection cells on a non-rotating Earth. Air rises at the equator and sinks at the poles, creating a single convection cell in each hemisphere. The prevailing winds moving over the Earth's surface blow from the poles towards the equator in both hemispheres (Modified by PW from globe image by Location_of_Cape_Verde_in_the_globe.svg: Eddo derivative work: Luan fala! [CC BY-SA 3.0], via Wikimedia Commons).
The non-rotating situation in Figure 6.2.1 is of course only hypothetical, and in reality the Earth's rotation makes this atmospheric circulation a bit more complex. The paths of the winds on a rotating Earth are deflected by the Coriolis Effect. The Coriolis Effect is a result of the fact that different latitudes on Earth rotate at different speeds. This is because every point on Earth must make a complete rotation in 24 hours, but some points must travel farther, and therefore faster, to complete the rotation in the same amount of time. In 24 hours a point on the equator must complete a rotation distance equal to the circumference of the Earth, which is about 40,000 km. A point right on the poles covers no distance in that time; it just turns in a circle. So the speed of rotation at the equator is about 1600 km/hr, while at the poles the speed is 0 km/hr. Latitudes in between rotate at intermediate speeds; approximately 1400 km/hr at 30o and 800 km/hr at 60o. As objects move over the surface of the Earth they encounter regions of varying speed, which causes their path to be deflected by the Coriolis Effect.
To explain the Coriolis Effect, imagine a cannon positioned at the equator and facing north. Even though the cannon appears stationary to someone on Earth, it is in fact moving east at about 1600 km/hr due to Earth's rotation. When the cannon fires the projectile travels north towards its target; but it also continues to move to the east at 1600 km/hr, the speed it had while it was still in the cannon. As the shell moves over higher latitudes, its momentum carries it eastward faster than the speed at which the ground beneath it is rotating. For example, by 30o latitude the shell is moving east at 1600 km/hr while the ground is moving east at only 1400 km/hr. Therefore, the shell gets "ahead" of its target, and will land to the east of its intended destination. From the point of view of the cannon, the path of the projectile appears to have been deflected to the right (red arrow, Figure 6.2.2). Similarly, a cannon located at 60o and facing the equator will be moving east at 800 km/hr. When its shell is fired towards the equator, the shell will be moving east at 800 km/hr, but as it approaches the equator it will be moving over land that is traveling east faster than the projectile. So the projectile gets "behind" its target, and will land to the west of its destination. But from the point of view of the cannon facing the equator, the path of the shell still appears to have been deflected to the right (green arrow, Figure 6.2.2). Therefore, in the Northern Hemisphere, the apparent Coriolis deflection will always be to the right.
In the Southern Hemisphere the situation is reversed (Figure 6.2.2). Objects moving towards the equator from the south pole are moving from low speed to high speed, so are left behind and their path is deflected to the left. Movement from the equator towards the south pole also leads to deflection to the left. In the Southern Hemisphere, the Coriolis deflection is always to the left from the point of origin.
The magnitude of the Coriolis deflection is related to the difference in rotation speed between the start and end points. Between the poles and 60o latitude, the difference in rotation speed is 800 km/hr. Between the equator and 30o latitude, the difference is only 200 km/hr (Figure 6.2.2). Therefore the strength of the Coriolis Effect is stronger near the poles, and weaker at the equator.
Figure 6.2.2 The Coriolis Effect. Objects moving from the equator towards the poles (red arrows) move into a region of slower rotational speed and their paths are deflected "ahead" of their point of origin. Movement from high latitudes to low latitudes (green arrows) goes from a region of low speed to a region of higher rotation speed, and there is deflection "behind" their point of origin. In the Northern Hemisphere this deflection is always to the right from the point of origin, and in the Southern Hemisphere the deflection is always to the left (Modified by PW from globe image by Location_of_Cape_Verde_in_the_globe.svg: Eddo derivative work: Luan fala! [CC BY-SA 3.0], via Wikimedia Commons).
Because of the rotation of the Earth and the Coriolis Effect, rather than a single atmospheric convection cell in each hemisphere, there are three major cells per hemisphere. Warm air rising at the equator cools as it moves through the upper atmosphere, and it descends at around 30o latitude. The convection cells created by rising air at the equator and sinking air at 30o are referred to as Hadley Cells, of which there is one in each hemisphere. The cold air that descends at the poles moves over the Earth's surface towards the equator, and by about 60o latitude it begins to rise, creating a Polar Cell between 60o and 90o. Between 30o and 60o lie the Ferrel Cells, composed of sinking air at 30o and rising air at 60o (Figure 6.2.3). With three convection cells in each hemisphere that rotate in alternate directions, the surface winds no longer always blow from the poles towards the equator as in the non-rotating Earth in Figure 6.2.1. Instead, surface winds in both hemispheres blow towards the equator between 90o and 60o latitude, and between 0o and 30o latitude. Between 30o and 60o latitude, the surface winds blow towards the poles (Figure 6.2.3).
Figure 6.2.3 On a rotating Earth, there are three atmospheric convection cells in each hemisphere, leading to alternating bands of surface winds (red arrows) (Modified by PW from globe image by Location_of_Cape_Verde_in_the_globe.svg: Eddo derivative work: Luan fala! [CC BY-SA 3.0], via Wikimedia Commons).
The surface winds created by the atmospheric convection cells are also influenced by the Coriolis Effect as they change latitudes. The Coriolis Effect deflects the path of the winds to the right in the Northern Hemisphere and to the left in the Southern Hemisphere. Adding this deflection leads to the pattern of prevailing winds illustrated in Figure 6.2.4. Between the equator and 30o latitude are the trade winds; the northeast trade winds in the Northern Hemisphere and the southeast trade winds in the Southern Hemisphere (note that winds are named based on the direction from which they originate, not where they are going). The westerlies are the dominant winds between 30o and 60o in both hemispheres, and the polar easterlies are found between 60o and the poles.
Figure 6.2.4 The prevailing wind patterns of Earth (Modified by PW from globe image by Location_of_Cape_Verde_in_the_globe.svg: Eddo derivative work: Luan fala! [CC BY-SA 3.0], via Wikimedia Commons).
In between these wind bands lie regions of high and low pressure. High pressure zones occur where air is descending, while low pressure zones indicate rising air. Along the equator the rising air creates a low pressure region called the doldrums, or the Intertropical Convergence Zone (ITCZ)(convergence zone because this is where the trade winds converge). At 30o latitude there are high pressure zones of descending air known as the horse latitudes, or the subtropical highs. Finally, at 60o lies another low pressure region called the polar front. It should be noted that these high and low pressure zones are not fixed in place; their latitude fluctuates depending on the season, and these fluctuations have important implications for regional climates.
Doldrums? Horse latitudes? Trade winds?
These may seem like some odd names for these atmospheric phenomena, but many of them can be traced back to maritime traditions and lore.
The doldrums refer to regions of low pressure around the equator. In these areas, air is rising rather than moving horizontally, so these regions commonly encounter very light winds. The lack of wind could leave sailing ships becalmed for days or weeks at a time, which was not good for the morale of the ship's crew.
Like the doldrums the horse latitudes are also areas with light winds, this time due to descending air, which could leave ships becalmed. One explanation for the term "horse latitudes" is that when these ships became stranded they ran the risk of running out of food or water. To conserve these resources, sailors would throw their dead or dying horses overboard, hence the "horse latitudes." Another explanation is that many sailors received part of their pay before a voyage, and often spent it before departing. This meant that they would spend the first part of the voyage working without pay and in debt, a period called the "dead horse" time, which might last for a few months. When they started earning their pay once again, they had a "dead horse" ceremony and threw a pretend horse overboard. The timing of this ceremony often coincided with reaching the horse latitudes, leading to the association of the ceremony with the location. A third explanation is that a ship was referred to as "horsed" when winds were weak and the ship instead had to rely on ocean currents to move them. This could be a common occurrence in the high pressure zones around 30o latitude, so they were referred to as the horse latitudes.
The term trade winds may have originally derived from the terms for "track" or "path", but the term may have become more common during European exploration and commercialization of the New World. Mariners sailing from Europe to the New World could sail south until they reached the trade winds, which would then propel their ships across the Atlantic to the Caribbean. To return to Europe, ships could sail to the northeast until they entered the westerlies, which would then steer them back to Europe.
It may seem odd to be discussing coral reefs in a section about geology, but due to the stony calcium carbonate skeletons secreted by many coral species, coral reefs are as interesting as geological features as they are biological ones. Corals grow best in warm, clear, tropical water, that is close enough to the surface for light to support photosynthesis by the algae living in the coral tissues. Because of this need for light, new coral will often grown on top of the stony skeletons of older corals.
In the 1830s Charles Darwin made some observations about different types of coral reefs, and hypothesized that they represent a progression from one form to the next. The types of reefs he examined were fringing reefs, barrier reefs, and atolls, which are associated with oceanic islands (Figure 2.10.1). Fringing reefs are reefs that are close to or are connected to shore. Barrier reefs are offshore reefs that are separated from the land by an expanse of water, such as a lagoon. Atolls are circular or oval reefs surrounding a lagoon, without any central land mass in the lagoon. Darwin speculated that reefs progressed from fringing, to barrier, to atolls as the land mass subsided. However, he had no explanation for how volcanic islands could sink. Today we know that Darwin was correct, and that islands can sink as oceanic crust subsides as it moves away from a spreading center, or as sea level rises as glaciers melt.
Figure 2.10.1 A fringing reef (left), barrier reef (center), and atoll (right) (Public domain, via Wikimedia Commons).
The progression starts with a fringing reef built against the shores of island (Figure 2.10.2). If sea level doesn't change or the land doesn't sink, many reefs will not progress beyond this stage. But if the land does subside, the corals would eventually sink too deep for light penetration and they would die. So as the reef gets deeper, the corals continue to grow upwards, at a rate of about 3-5 m per 1000 years, and eventually a lagoon develops between the reef and the island; the reef is now a barrier reef. If the land continues to subside until it is completely submerged, all that is left is a ring of coral that has been growing upwards around the central lagoon; an atoll.
Figure 2.10.2 Steps in the development of coral reefs (Steven Earle, "Physical Geology").
