Published data on oceanic and harbor currents, in such form as to be of use for the navigator, are usually confined to the coast pilots of a few maritime nations. The Pacific Coast Pilot of the U. S. Coast and Geodetic Survey, which covers the coasts of California, Oregon, and Washington, states, in the general coastal current data, as follows:
Wind Currents: This subject is very complex. In general, it may be said that along the Pacific Coast of the United States at a distance of five to ten miles offshore, the wind brings about a current having a velocity of 2 per cent that of the wind. The direction of the wind driven currents, however, is not with the wind. With winds from the N.E., S.E., and N.W. quadrants, the current sets about 20° to the right of the wind, while with winds from the S.W. quadrant, the current sets about 20° to the left of the wind. It is evident, however, that these are average values, for strong currents are sometimes experienced when the local winds are light.
In the U. S. Coast and Geodetic Survey Current Tables for the Pacific Coast, 1938, we find:
San Pedro Channel, 7 miles south of Los Angeles Harbor Breakwater. There are two periodic currents here, both of which are rotary, turning clockwise and rather weak. The tidal current has a velocity at strength of about 0.2 knot. The other current, due apparently to land and sea breezes, has a period of 24 hours and an average velocity of ¼ knot. The greatest velocity observed during 5 months of observations is 1.5 knots. Currents greater than 1 knot occur infrequently.
The observations upon which this paragraph is based were made from a lightship moored in about 50 fathoms in Lat. 33°-35'-37" N., Long. 118°-15'-30" N. from June to November, 1935. This position is on the southwestern edge of the silt bank formed by centuries of deposition of debris by the Santa Ana and San Gabriel Rivers, and is as close to the mid-channel as it was possible to anchor without special equipment.
An examination of monthly mean sea level discloses a periodic variation in height for many stations. Figure 1 gives the data for several Pacific coast stations and a conventional representation of north and south Pacific types. From January to June, the monthly mean sea level for Northern Hemisphere stations is below the average value; while during the remainder of the year it is above the average. At Los Angeles Harbor there is a 0.5-foot difference between the lowest month (April) and the highest month (September). A direct comparison with stations in the south Pacific Ocean is not made since actual data are not at hand. However, with the long period tidal components as given by the U. S. Coast and Geodetic Survey for several Northern and Southern Hemisphere stations, it is shown that the same reversal in sea level occurs as those which are well known in the seasons. From this condition it is evident that considerable water must be transferred annually between the hemispheres. This cannot be accomplished without producing currents, however feeble, which are annually reversing. Other currents may mask the currents produced in this manner, and their detection become difficult. Also, it is not essential that they be surface currents.
The charts showing oceanic circulation systems do not give any currents which operate between the hemispheres, so it becomes necessary to seek elsewhere for a plausible method of water transfer to accommodate the changing sea level.
Additional evidence that an exchange of sea water takes place is given in Fig. 2, which shows the monthly surface water salinity at La Jolla, California, and surface water densities in Los Angeles Harbor. The salinities and densities are different manifestations of the same phenomenon—both graphs are given to show that the fresh water flow into Los Angeles Harbor from two flood control channels has only a minor effect on the results, although quite prominent for several days during and following heavy precipitation.
Table I shows the epoch for the solar annual, Sa, component for the two hemispheres in the Pacific Ocean. While no stations were available close to either side of the geographical equator, it is believed that the equator for the sea level will be a function of the land areas of the two hemispheres, modified by corrections for variations in rainfall, snow, and groundwater retention through high mountains and forests, the transfer of water from one ocean to another by river run-off of water acquired in another ocean, and ocean currents produced by winds. This would place it southward of the geographical equator along the Pacific shores of the American continent, and southward also in the Atlantic Ocean.
Note the reversal of the heights of sea level, similar to the change of summer and winter between the latitudes. This is evidence for the viewpoint that the change in sea level is purely a function of the rainfall factors and not of astronomical origin.
The change in sea level is not related to the ordinary tide, i.e., a wave motion, with crests and troughs passing a given point; it is, rather, a definite withdrawal of water and its replacement, similar to the evaporation from a lake or pond, and the replacement by rainfall or stream flow. It affects large areas, although in different ways. For example, in the long, winding, land-locked fjords of southeastern Alaska, where excessive rainfall—and necessarily the resulting heavy stream-flow—pour vast quantities of water into these arms of the sea, it can be seen that their mean level will be a function of the rainfall conditions. At other stations, on tidal estuaries of large river systems, the spring freshets will affect the mean water level. Therefore, in generalizing and stating that the mean sea level of the Northern Hemisphere varies opposite to the Southern Hemisphere, it should be borne in mind that the local levels are affected by local conditions, subject to all the variations in meteorological conditions which take place, and the conditions as found are the result of the several factors entering the problem.
From Fig. 1 it is seen that the rise in mean sea level at Los Angeles Harbor between June and July was 0.28 feet. This water did not come from rainfall or stream flow during these months, because both are extremely light or absent. Since an influx of water was necessary, by a priori reasoning, there must have been ocean currents, not caused directly by winds, differences in atmospheric pressure, earth rotation, or the other usually assigned reasons.
With these data as a guide, we may now examine the Coast Pilot data regarding the direction of the currents in relation to that of the wind.
Although entitled “Wind Currents,” it is apparent that other factors are contributing current components which are independent of the wind. Figure 3 gives a graphical representation of the wind, resulting current, and the necessary addi-ditional current demanded to give the required resultant 20° to the left or right of the wind direction. There is an offshore component current for all quadrants except the SE. Since winds from this quarter are experienced only between October and March, it should be brought out that this period is also the one during which mean sea level is lowering; also, it nearly corresponds to the period when the salinity at La Jolla is decreasing. This condition, then, is in accord with the theory that an offshore current prevails during the time that sea level is ascending.
However, while currents are essential to produce the changes in sea level, they need not necessarily be surface currents, nor need they be continuous in velocity, direction, lateral location, or depth. The currents will flow toward the low place, keeping in mind that winds, differences in atmospheric pressure, and oceanic drifts also produce temporary differences in level. The position with respect to the bottom will be such that the temperature and salinity, which govern the density, will be in equilibrium in relation to the surrounding waters of the rotating spheroid.
While the surface currents are the chief interest of the navigator, it is easily seen that a current which is submerged well beyond the draft of the vessel may, when the conditions of temperature, wind, seasonal sea level, etc., are propitious, become a surface current for a time; or the reverse condition may occur.
H. U. Sverdrup, in an article entitled “On the Process of Upwelling,” Vol. 1, No. 2, April, 1938, issue of the Journal of Marine Research, gives the results of his analysis of data procured on three cruises off Port San Luis in 1937. Figure 4 gives the location of stations occupied, the vertical distribution of density on the three cruises, and the computed circulation which had to take place to bring about the observed variations in density. The location map and the density graphs are self-explanatory. In the circulation diagram, the lines with arrows show the average streamlines for the 41-day interval between the observations, and do not represent trajectories. The dotted lines show the horizontal velocities prevailing during the period, the figures representing cm/sec. The computed maximum offshore velocity is 11 cm/sec. (0.14 kn.).
The cross-hatched area represents a region with a “swift” current parallel to the coast and away from the reader (southeastward). This region or belt also has an offshore movement with a computed velocity of about 3 cm/sec. (0.04 kn.).
The summary states that . . . during the first part of an upwelling the light surface-water is banked up at some distance from the coast. Between the heavier upwelled water and the light offshore water a boundary region forms with a strong current parallel to the coast. This boundary region moves away from the coast, but so slowly that it acts as a barrier against the outward transport of the upwelled water. Within this a cellular circulation develops which, since the offshore boundary moves away, is being fed from below. This appears to be the initial stage, but as the boundary region moves out conditions become more and more unstable and large eddies develop on the coastal side of the boundary.
Dr. Sverdrup assigns the wind as the cause of the upwelling, but points out that other possibilities exist. The writer, of course, is assigning the variation of “mean sea level” as the primary cause of the phenomenon. The studies of upwelling have not yet assigned definite seasons during which it occurs in Southern California, but, drawing conclusions from the scanty evidence of the three cruises, it may be observed that the upwelling was greatest in early May and had almost ceased on June 26-27, and therefore coincides with the period when water must be “imported” from elsewhere to make the July to September maximum sea levels. Needless to say, the winds continue to blow with nearly the same velocities during the remainder of the year, and can therefore be only a secondary factor in the cause of upwelling—perhaps only to affect the directions of the resulting currents—for were only the winds responsible for the upwelling, the phenomenon would, of necessity, be almost continuous.
Furthermore, Dr. Sverdrup states, by letter, that:
It appears that upwelling occurs regularly only north of Point Conception during the time of year when northwesterly winds prevail. Local upwelling may, however, occur as far south as Point Fermin, but probably not farther south. A comparison between variation in sea-level and sea-surface temperature shows a close relationship from 14 years of observations. Generally, the surface temperature will be low during periods in which the sea level (tides eliminated) is low, and vice versa. The probable explanation is that when sea level is low the current flows south along the coast and the coldest water is found on the shoreward side of the current.
That upwelling is simultaneous with the increase in sea level height was shown conclusively by Dr. Tage Skogsberg of the Hopkins Marine Station, Pacific Grove, California, in his extremely detailed and thorough discussion of thermal conditions in Monterey Bay in the December, 1936, Transactions of the American Philosophical Society.
Figure 6, compiled from these data, is self-explanatory. The year divides itself into periods as shown, viz., from November to February when the Davidson Current prevails, the upwelling period from March to July, and the oceanic period from August to October. The oceanic period is one in which the conditions resemble those of the offshore portions of the ocean. During this period the water, according to Dr. Skogsberg, is of a deeper blue and of a greater transparency than at other seasons. This agrees with Captain Ray Wall’s empiric color differentiation given on page 694. Furthermore, this variation is reflected in the drop in salinity during August, September, and October, shown in Fig. 2, which indicates the intrusion of low salinity water.
It is possible that the same dip in the salinity as shown for September-October for the Northern Hemisphere occurs during the similar season in the Southern Hemisphere, reducing the pressure to which the upwelling off the California coast is herein assigned. From this consideration, the “8°C water” smooth curve as drawn by Dr. Skogsberg should be modified to follow more closely the observed values.
It is evident that, in order to produce the long continued periods of upwelling, regardless of their variations from year to year, a similar long continued force must be present somewhere to account for the process. While winds have been assigned as the primary cause, various investigators have found that this reason is insufficient. McEwen (Scripps Institute of Oceanography) implied evaporation, turbulence, solar radiation, as well as barometric pressure differences as additional causes. None of these is of sufficient duration to do more than slightly vary the total effect caused by the hydrostatic pressure due to the difference in level of the entire ocean as shown herein.
The variations noted in the annual upwelling can easily be explained by the variations in the differences in the mean sea level existing between the North and South Pacific—not the absolute height of only one portion or the other—but their difference.
The height of mean sea level is a function of the rainfall, chiefly by withdrawal of water to form the precipitation, much of which is retained as snow, ice, and ground water for some time after the withdrawal, thereby causing the observed variations in the monthly and annual sea level heights. Since the seasons in the two hemispheres are reversed, the difference in the mean sea-level heights will be a function of the rainfall conditions of both hemispheres. Thus, heavy withdrawal accompanied by slow return through streamflow (long, cold winters) in both hemispheres would not cause a large difference in mean sea levels and up- welling would therefore be limited. However, if a long, cold winter with heavy precipitation occurs in only one hemisphere while the other is relatively dry, a large difference in sea level will take place, with much upwelling. The result will then be increases in the strength of the coastal currents, the Davidson and California currents along the California coast, and the Peruvian (Humboldt) current along the South American coast.
The high salinity of the upwelled water is again reduced by the inflow of the northern waters of the California current. This indicates tropical origin for the up-welled water and arctic origin for the low salinity California current water.
The cause of the upwelling, according to Skogsberg,
. . . should be sought not only in the offshore deflection of the superficial, wind driven, southerly current along the coast, but also in a pressure from below caused by an unknown, deep water movement directed at approximately right angles to the coast and not connected with the water movement caused by the offshore deflection of the surface current.
In view of the fact that all moving bodies in the Northern Hemisphere are deflected to the right by the rotation of the earth, and to the left in the Southern Hemisphere, the aforementioned north and south current must also be deflected, with a maximum value of 45°, or NE. for our latitudes and SE. for the South American coast. This causes the water of the entire Pacific basin, moving in response to the difference in head shown to exist between the hemispheres, to flow nearly at right angles against the California coast during the upwelling period, and presumably against the South American coast approximately six months later, although no data are at hand for the latter conjecture.
Since the balancing of the waters of the Pacific Ocean need not, of necessity, take place as a surface phenomenon, and upwelling has been shown by Sverdrup, Skogsberg, and others to take place from about 100 to 200 meters deep, the requirements of Skogsberg are fully met.
Furthermore, considering the upwelling as the predominating factor operating during the period from March to July, when this pressure is released by the balancing of oceanic heads the normal forces can then function. This may be the Japan Current, the attendant California Current, or local wind driven surface currents. From October to February, the flow of water should be southward to give the South American upwelling.
The Davidson Current is a wind current produced by the southerly winds, and is shown on the December, January, and February, U. S. Hydrographic Office Pilot charts of the North Pacific Ocean as a northerly current. At other times of the year it is absent, the current inshore being southerly.
Since winds are encountered throughout the entire year, in order to account for the seasonal upwelling purely as a result of winds, a period of several months is postulated to produce the current, and a similar long period to subside. No such adjustments to conform to the conditions as found are necessary if we accept the annual variation in sea level as the cause of upwelling.
From the consideration that upwelling corresponds in time to the change in sea level, the current observations from the light vessel—made during the period when there is no upwelling with its attendant currents—cannot reflect conditions for the entire year but only for the months during which observations were made. There is no fixed seasonal limit for the upwelling; it may commence a month or more sooner, or end later, therefore the navigator should not dismiss allowance for current during the fall and winter months on account of their feebleness.
Upwelling so affects local currents that the U. S. Coast and Geodetic Survey is not warranted in making harmonic analyses of the observed currents and correlating them to tides, position of tide producing astronomic bodies, etc., since the results would not include the most important and purely meteorological cause of currents—the upwelling. Upwelling has been observed off the coast of West Africa, both northward and southward of the Gulf of Guinea; along the Atlantic coast of the United States and off the Argentine coast; and in the Pacific Ocean off the Peruvian, Central American, and California coasts.
The present theory regarding upwelling has the less saline but colder, therefore denser, water from the arctic and antarctic regions, produced by the melting of glacier and sea ice, traveling slowly along the ocean floor to the sites of the upwelling phenomenon. At these points, usually in or near the tropics, the trade winds forming the equatorial currents carry off the surface waters, permitting the upward flow of the less saline but colder water from the arctic regions. From this theory, the upwelled water is less saline than that normally found in the vicinity. Compare this statement with the densities found by Dr. Sverdrup off the California coasts. It is the denser water that upwells.
During the spring months there is a marked increase in the salinity (density) of the water. Obviously this cannot be attributed to the fresher arctic waters; the influx of water must come from the more saline tropical regions. That such a transfer of water takes place is substantiated by the so-called “mean sea level” heights of the tide for the two hemispheres. Figure 1 shows conventional graphs for the two oceans.
It should be remembered, however, that the solar annual component (Sa) cannot be considered a true tide. It is the variation in water height of the ocean in response to the withdrawal of water from the oceans to form the rainfall; the adjustment in height which must ensue in adjacent portions of the ocean due to the withdrawal of water elsewhere, the addition by river run-off and rainfall produced by withdrawal of water from other oceans, and the various combinations of these factors which take place in nature. In view of the known vagaries of the weather, it is easily seen that the yearly variation in “mean sea level” is a very erratic quantity, and the values shown reflect only that period covered by the tide analysis.
Current data for Los Angeles and Long Beach Harbor channels were obtained from a tidal and current survey made by the U. S. Coast and Geodetic Survey in co-operation with the City of Los Angeles Harbor Department between October, 1935, and May, 1936. The currents reflect the seiche phenomena described in the U. S. Naval Institute Proceedings for June, 1937, page 788. The currents rarely exceed 1½ knots, set fair with the channels, and are met with everywhere in the inner harbor channels.
Figure 5 gives currents for an entire day, measured with a current pole where they run strongest—in the narrow entrance to Long Beach Harbor.
The tide, in its rise and fall, produces only feeble currents, as shown by the interpolated dashed line. At the crests and troughs of the seiches, the surges (currents induced by the seiches) are minimum. At the mid-points between crests and troughs (where tide curve crosses the interpolated smooth curve on the marigram record), the surges are greatest. The reversal of the current is rapid, as shown on the current graph. Very small amplitude seiches riding the hourly seiche likewise produce rather severe surges at times.
These currents offer little hazard for the navigator except, perhaps, when approaching bridges or mooring at a wharf and failing to properly allow for the current. The flood enters the harbor simultaneously from both entrances. This causes the surges to flow southwestward through Cerritos Channel while it is setting northward in the Los Angeles Main Channel. The currents meet in the East Basin.
The rapid change in direction of the currents, and the fact that they are independent of the tides make any prediction difficult. Fortunately, their small velocity reduces probability of mishaps, but caution should be exercised against sudden reversal when mooring.
The result of inquiries regarding currents among the many boatmen frequenting the local waters may be best described by quoting Captain Ray Wall of the Los Angeles Harbor Department, who for many years has been master of coastwise steamships, steam schooners, towboats, fishing boats, and Catalina Island ferries:
During the months of April to June, a southeasterly set is experienced on the crossing from Catalina Island to Los Angeles Harbor, although the set is not continuous. When the water is green and cold, the southeasterly set will be in evidence; if the water is the usual blue and warmer than the green, the ordinary westerly set will be met. The blue water is clear, and only in this warmer blue water are albicore ever found. On the shallow shelf near the shore, the green water lingers after the offshore water is again blue, although streaks are never encountered. The color, and sets, change quickly; green one day and perhaps blue the next, or the reverse. The strength of the currents was such that the S.S. Cabrillo, in the run from Avalon, Catalina Island, to Los Angeles Harbor, held up 1° on its course—to the left for the southeast set (green water) or to the right for the westerly set (blue water). The speed of the vessel was 12 knots. These phenomena are not confined to the local waters, but hold true for the coast from Lower California to Cape Flattery.
However, other competent observers concluded, after studying careful records of five years of crossing the channel under variable weather conditions, that no method for gauging and allowing for currents was possible. That there were strong currents at times, no current at others, and no constant direction for the currents under similar conditions, was observed. The only conclusion these observers made was, that from April to June, there was a strong inshore set, at times, between San Diego and Los Angeles Harbor. From these observations, it is evident that current predictions are an intricate problem requiring much additional study.
In view of this discontinuity of the southeasterly set, and, since Dr. Sverdrup based his computations on a continuous upwelling between the cruise dates, it is reasonable to presume that the offshore velocities as determined by him would be much greater in nature, as well as the southeastward current which his analysis requires, had his observations been limited to the actual period during which upwelling took place. In other words, the time element would have been greatly reduced.
All evidence regarding currents as herein presented is in agreement, but the subject is still far from exhausted, and the introductory sentence in “Wind Currents” as quoted, still holds true—“This subject is very complex.”