In-situ Observations of the Davidson Current
LCDR Ken Wallace
OC3570 Winter Cruise
23 – 30 January, 2008
The thermal energy from the sun drives most of the circulation in the atmosphere
and beneath the ocean surface on this planet. This is particularly well illustrated in the
North Pacific Ocean. Air heated at the equator rises high into the atmosphere and the
rotation of the Earth on its axis moves the heated air mass westward and northward as it
rises, then eastward once it cools and descends at higher latitudes, creating a huge,
clockwise-rotating mass of air. This rotating air mass frictionally interacts with the ocean
surface, driving the major currents that also flow in a clockwise direction. Known as the
North Pacific Gyre, this ocean-scale rotating mass of water clearly demonstrates the
effects of air-sea interaction on a global scale. Further inspection of this circulation on a
regional and local level reveals many spatially and temporally variable fluctuations that
occur throughout the year.
The California Current System (CCS) forms the eastern branch of the North
Pacific Gyre and consists of three primary components: the California Current (CC), the
California Undercurrent (CUC) and the Davidson Current (DC). The CC transports cold,
subartic surface waters southward from the North Pacific 150 – 1300 km off the U.S.
West Coast and from the surface to a depth of approximately 500 m (Collins et al, 2003).
The CUC and DC both flow inshore of the CCS and poleward. The CUC flows beneath
the surface year round from the continental shelf out to 200 km and generally below the
thermocline from 200 m to over 1000 m deep (Pennington et. al, 2007). The DC is a
seasonal current that flows at shallower depths of less than 200 m in late Fall and Winter
and is the primary focus of this study. Some debate exists among researchers as to the
forcing that drives the DC and whether or not it is simply a surface reflection of the CUC
(Huyer and Smith, 1974).
The CCS is largely influenced by the seasonal wind regimes along the West Coast
of the United States. In summer, the subtropical high dominates the weather pattern with
predominantly northerly winds and long periods of fair weather. As the seasons change,
so too does the behavior of the CCS in response to the variation in the atmospheric
forcing. With the onset of winter, the subtropical high retreats equatorward and the
Aleutian Low becomes anchored over the subarctic waters off of Alaska. This relaxes
the northerly wind stress and allows powerful winter storms with strong southerly flow to
frequently influence the flow of coastal California waters. Under these conditions, during
the late Fall and Winter, the poleward flowing DC becomes established near the surface
over the continental shelf from Northern Baja, CA to Vancouver Island (Chelton, 1982).
The purpose of this study is to compare oceanographic and atmospheric data
collected from 23 – 30 January, 2008 during a Naval Postgraduate School oceanographic
cruise off the Central California Coast and use it to illustrate the characteristics and
position of the DC. Additionally, an attempt will be made to make some conclusions
about the forces responsible for its annual evolution based on the collected data.
Oceanographic and meteorological data will be compared for purposes of explaining the
roles of wind stress and wind stress curl in the forcing of the current. This data will also
be compared to satellite derived dynamic height contours and moored buoy data.
Understanding the physical structure of the DC and the forces that drive it is
important in many regards. From a military perspective, a thorough knowledge of the
location and movement of water masses with different temperature and salinity (ie.
density) characteristics is crucial to understanding how sound propagates through the
ocean. This is obviously important when conducting any manner of undersea warfare
(USW). The currents also affect ship navigation as well as diving and rescue operations.
The principles that govern the intensity, location and duration of the DC are possibly
applicable to other currents worldwide where the Navy operates and therefore warrant
research and scientific explanation.
From an environmental perspective, the DC affects the transport of marine
organisms such as plankton and fish larvae which are important indicators of the health of
the marine ecosystem. The current can also be a transporter of contaminated ballast
water, oil spills, or other pollutants up and down the coast depending on where and at
what depth the subject material is introduced into the ocean. For the fishing industry, the
location and timing of the current has obvious implications to where the desired catch
may be located, for how long it will be there as well as the health and sustainability of the
species in question. Lastly, long-term monitoring of how the current changes over an
annual and interannual time frame can provide information about how the ocean
circulation is changing and possibly affecting the global climate.
III. Data Collection and Methods
On 23 January, the R/V Point Sur departed Moss Landing, CA for a week long
round trip research cruise from Monterey Bay to San Francisco. The cruise plan
consisted of gradually transiting to the north and back to Monterey Bay along a track that
made several excursions away from the coast into deeper water (over 2000 m) in order to
obtain cross sectional Acoustic Doppler Current Profiler (ADCP) data of both the near
shore water and water off the continental shelf (Figure 1). Frequent stops were made for
conductivity, temperature and depth (CTD) casts as well as the deployment and recovery
of GPS equipped drifting buoys and other oceanographic instruments. As the DC is
known to typically flow within 100 km of the coast, this cruise plan was expected to
provide data that delineates the waters within the current from the surrounding
The primary instrument used for this study was the Teledyne RD Instruments
(RDI) Broad Band 75 kHz Acoustic Doppler Current Profiler (ADCP). This instrument
uses an acoustic signal and the principle of Doppler shift to measure the velocity of
particles (plankton, suspended debris, fish, etc.) within the water column beneath the hull
of the ship. The particles scatter this signal back to the ADCP receiver and the velocities
are calculated at various depths throughout the water column using the following
Fd = 2 Fs (V/C) cos (A) (Gordon, 1996)
Fd = Doppler shift frequency, Fs = frequency of sound, V = relative velocity ( between
source and receiver), C = sound speed (m/s), A = angle between acoustic beam and
scatter velocity (Figure 2).
The ADCP is ideal for oceanographic research because it generates low-noise,
high resolution data with a minimal consumption of power. The velocities can be depth
averaged from the near surface to the lowest depth or separated by depth ranges into
specific “bins” for a vertical cross section analysis.
The Sea-Bird Electronics, Inc. CTD is used to take water samples and to profile
water characteristics from the surface to near bottom depths. In all, 80 CTD casts were
made at predetermined locations throughout the cruise as indicated in Figure 1. The
primary characteristics are temperature, salinity and pressure (from which density can be
derived) as well as dissolved oxygen and a fluorometer which measures the amount of
chlorophyll in the water. Additionally, the sea surface height data from the CTD casts
can be used to create dynamic topography contours as shown in Figure 3. These contours
are derived from horizontal density differences between two locations and are
representative of the horizontal pressure gradient force. Reference depths of 200 m and
1000 m, assumed to be levels of “no motion”, were used to determine the geostrophic
currents which are referenced to those levels. A program called Surfer 7.0 was used to
interpolate the data between stations and to smooth data sparse regions.
Pacific Gyre Microstar buoys were deployed during the cruise and served as
Lagrangian drifters in the surface current. Each buoy is equipped with a GPS tracking
beacon and is tethered to a drogue which essentially attaches the buoy to a parcel of water
as it moves along the coast. GPS plots of the buoy tracks and their respective speeds of
advance can be determined from the data and give a very good indicator of the movement
of the surface waters.
Monterey Bay Aquarium Research Institute (MBARI) moorings M1 (36.8 N,
122.0 W) and M2 (36.7 N, 122.4 W), deployed outside of Monterey Bay, use solar and
battery powered two-way telemetry to provide real time in-situ data about the water
column and atmospheric forcing (Figure 4). Each buoy is equipped with an ATLAS
sensor to collect meteorological information as well as sea surface temperature once
every ten minutes. An onboard RDI ADCP collects water column speed and direction
every fifteen minutes and CTD sensors deployed at 10 and 20 m depths feed information
upon request to the onboard computer system. These buoys are particularly useful when
examining the effects of wind stress on the properties of the near surface water column
Surface current radar (CODAR) uses high frequency radar signals to measure the
flow in the upper 1 meter of the water column. Several of these stations have been
installed along the Central Coast from Point Sur to north of San Francisco and provide
near-real time analysis of the surface currents (Figure 5). Each CODAR site has two
antennas: the first transmits a radio signal out across the ocean surface and the second
listens for the reflected radio signal after it has bounced off the ocean's waves. By
measuring the change in frequency of the radio signal that returns, the CODAR system
determines how fast the water is moving toward or away from the antenna. The system
can also determine the height and frequency of the waves near the shore. CODAR was
used in this study to locate the DC and estimate its intensity as well as for correlation
with observed atmospheric forcing patterns.
The geostationary satellite GOES-11 and Quikscat satellites were used to provide
information about the synoptic scale weather conditions during the cruise. The GOES-11
satellite is geostationary over the Pacific and provides 1.5 km resolution visible and IR
images at half-hourly intervals throughout the day from which the clouds and flow
patterns can be determined. The Quikscat scatterometer is a polar orbiting satellite that
uses microwave pulses to detect sea surface roughness at a 12.5 km resolution. These
roughness measurements are correlated using an algorithm and converted to surface wind
stress values displayed by color coded wind barbs over water. Together, the satellite data
provides information about the large scale forcing from the dominant wind regimes that
impacted sea height conditions during the cruise.
IV. DATA Analysis
An ADCP with an acoustic frequency of 75 Hz was used to profile the speed and
direction of the near surface water particles beneath the research vessel. A review of the
27 – 59 meter depth-averaged lateral transects running perpendicular to the coast of
California clearly reveals the position and intensity of a poleward flowing current evident
from Monterey Bay to the northernmost transect outside of San Francisco Bay (Figure 6).
Off of Monterey Bay, the core of the current appears to be approximately located 30 NM
off the coast near 122.25 W and is flowing to the Northwest. It then appears to turn and
flow inland slightly as it approaches the mouth of the bay. Once north of the Bay (north
of 37 N), the flow then appears to move farther offshore again. A study of the
bathymetry in the area reveals that the continental shelf extends further offshore to the
north of Monterey Bay (Figure 7), suggesting that the current is influenced somewhat by
the extension of the shelf and roughly follows the 200 meter depth contour. The longest
transect was directly off of Monterey Bay (line 67) and shows the width of the poleward
flow to be approximately 35 NM with a maximum velocity about 10 – 12 NM outside the
Bay. Beyond this distance, the magnitude of the current decreases significantly and even
reverses direction to the south around 122.6 W. Transects to the North also reflect this
decrease in current velocity to the west but are shorter in length and so it is impossible to
estimate the true width of the flow of point of reversal in these regions. On the last day
of the cruise, approximately 48 hours after the storm had passed and northerly flow
associated with a high pressure ridge to the West had been established, a transect was
conducted along track 67 (Figure 8). The 27 – 59 m data along this track shows very
strong poleward flow with a maximum near 122.2 W and decreasing poleward flow to
the west as far as 122.8 W.
Additional information about the position and strength of the current can be
obtained by viewing the ADCP data in vertically segmented slices. By separating the
ADCP data collected on 23 – 25 January outside of Monterey Bay into bins of specified
depth ranges from the near surface to deep water, it is obvious that the poleward flow is
strongest between the near surface depths to approximately 125 meters (Figures 6 and 7).
Evidence of northward flow can be seen down to 325 – 375 meters but at a reduced
velocity and over a much narrower lateral distance. The vertical cross section to the
north off of San Francisco Bay (SF track) again shows the current to be most intense
from near the surface to approximately 125 meters (Figure 9).
B. Water Properties
Easily measured water properties such as temperature and salinity are often used to
distinguish masses of water or currents from the surrounding environment. By displaying
the interpolated CTD data obtained along the same transects used to collect ADCP
profiles, a concentrated area of poleward flow is easily recognizable. The current is
characterized in these profiles as a warmer and less saline anomaly in the upper 200
meters near the coast (Figure 10) along both the line 67 track and the SF track. A notable
increase in the dissolved oxygen and fluorometer profiles is also evident in these profiles,
characterizing the poleward flowing current as a layer of oxygen and chlorophyll rich
water as compared to the rest of the water column. The line 67 profile compares
favorably with the ADCP data in positioning the western boundary of the current
approximately 30 NM seaward from station #80 in Monterey Bay. The location of the
current is more difficult to distinguish along the SF track due to the broad shelf that
extends further west of the coast line. This vast expanse of shallow water is more easily
mixed and any distinction between water masses could be obscured by strong wind stress
and associated mechanical mixing at the surface. The only notable difference in
properties of the water in this region is in the salinity profile which is slightly less saline
in an area near the coast. As revealed by the ADCP data, the Davidson current appears to
be further from the coast near the edge of the continental shelf. This slight difference in
salinity could very well be attributed to freshwater run off from the storm that passed
over the region from 24 – 26 January.
C. Lagrangian Drifters
Perhaps the most obvious indicator of the northerly flowing near-surface waters
during the cruise is the tracks of the Microstar GPS equipped drifting buoys. From their
deployment site near the mouth of the bay, the buoys moved initially to the northeast
toward the bay before moving in a steady northwesterly direction along the coast. The
first buoy was recovered approximately 15 NM to the Northwest of where it was
deployed after almost 48 hours of drifting. The three buoys moved together along a very
similar course and at virtually the same speed and were recovered outside of San
Francisco Bay after almost 94 hours of drifting on the 27th of January (Figure 11). Their
average speed of advance was approximately 40 cm/s. It is important to note that during
the time of their northward transit, a powerful storm was impacting the coast with strong
southerly winds and associated southerly seas and swell. This alone would obviously
move objects near the surface to the north. However, their paths also coincide with the
ADCP data which indicated a concentrated area of poleward flow along the coast at
depths of over 300 meters.
D. CODAR Data
Nine coastal radar sites were used for this study: spread out from Point Sur and
into Monterey Bay, then scattered up the coast to the north side of the San Francisco Bay
outlet. For the most part, the resulting plots show a fair correlation with the observed
winds from scatterometry, the MBARI moorings, and the COAMPS wind field plots.
The strongest poleward flow is observed on the 27th of January, during the height of the
southerly winds produced by the transiting storm. However, even during periods
dominated by high pressure and northerly winds at the beginning and end of the cruise,
the poleward flow is much weaker but still evident throughout much of the Central Coast
and is especially obvious off of Pt Sur (Figure 12), suggesting that the current is not
entirely driven by surface winds.
E. Meteorological Data
It is important to note that the atmospheric forcing prior to and during the period
of the cruise encompassed both extremes of the typical seasonal weather. For several
days prior to the 23rd of January, a high pressure system dominated the coastal wind flow
with light to moderate northwesterly winds. On the second day of the cruise, however, a
strong low pressure system approached from the West, generating gale force winds from
the southwest that persisted for several days and generated 8 – 11 ft southerly seas
(Figure 13). The storm passed to the north by the evening of the 27th as high pressure
ridged in from the west and influenced the weather pattern for the remainder of the cruise
with predominantly northerly flow.
The aforementioned storm that transited the Central California Coast from the 24th
through the 28th was very well illustrated in the GOES-11 visible imagery, the Quikscat
scatterometry, and the COAMPS fields. The most obvious impact to the area of study
was the very strong southerly surface winds that persisted for the duration of the storm.
As indicated in Figure 14, these winds were not fetch limited and extended to the south
and west for approximately 1000 NM. The impacts of this kind of surface stress are three
fold: movement of the layer of water in contact with the wind, mechanical mixing of
water near the surface to some depth, and the effects of wind stress curl (Figure 15).
The MBARI M1 and M2 buoy wind, temperature and salinity time series plots
show a very clear and direct correlation between the increased southerly winds and the
deepening of the mixed layer. As would be expected, there is a slight lag between the
atmospheric forcing and the associated affects within the water column but the
relationship is consistent for both the northerly and southerly wind regimes. Winds from
the north leading up to the cruise had clearly upwelled cooler, subsurface waters and
stratified the water column and the opposite affect occurred with the strong southerly
winds during the storm (Figure 16).
An additional but less obvious affect associated with the wind speed and direction
is forcing due to wind stress curl. The wind stress initially moves surface water in the
direction of the winds but Coriolis due to the rotation of the earth accelerates the water
proportionally and to the right in the Northern Hemisphere. The resulting Ekman
transport is 90 degrees to the right of the wind direction. The vertical motion within the
Ekman layer (WE) can be represented as the curl of the wind stress:
Where Ty and Tx are the North-South and East-West wind stresses, f is the Coriolis
parameter and rho is the density. From this equation, we can see that positive wind stress
curl (northerly winds along the coast of California) cause divergence in the Ekman layer
and negative (southerly winds) wind stress curl causes convergence. Convergence at the
surface, as with southerly winds, would essentially pile water up along the coast and
divergence would tend to cause upwelling. This correlates nicely with the mooring time
series plots which indicated the vertical movement of the mixed layer.
F. Dynamic Height
The dynamic topography for the central coast was determined from the CTD casts
from Point Sur north to approximately 37.5 N. Figure 3 depicts the dynamic topography
in the upper layer which gives a first guess of what the surface currents should look like.
Clearly, the highest heights are located along the coast and in Monterey bay, decreasing
offshore to the west and southwest. Overall, a maximum difference of approximately 0.5
J/kg is observed. This difference implies a westward pressure gradient force, which,
when combined with the Coriolis effect as the water parcels are accelerated would result
in a southerly flow of water at the surface (as observed).
The 0/1000 depiction, which assumes the 1000 m depth is a level of “no motion”
and compares the overall density difference of the entire water level down to this level.
The result is very similar to but with even less variation that that of the 0/200 depiction
and again illustrates higher heights along the coast and in the bay, decreasing steadily
offshore. The maximum difference is approximately 0.4 J/kg, suggesting that a density
difference exists deeper than just the surface layer. The pressure gradient force is the
same as at the 0/200 level and again results in a northward geostrophic current, as
No one would argue after a review of all of the data above, that surface and sub-
surface poleward flow is indeed evident off of the central coast during the timeframe of
the cruise. The ADCP indicates that this flow extends to over 300m in the water column
and is continuous from south to north throughout the region surveyed. The CODAR data,
as a non-survey ship source, consistently confirms the movement of the surface layer
water to the north although with some variability. The dynamic height contours derived
from the CTD casts also match what would be expected of geostrophic forcing for a
Although little distinction can be made between the CUC and the DC, it is
difficult to imagine that the wind stress alone could be responsible for the poleward flow
below depths of 50m. For waters nearer to the surface, however, it makes sense that the
widespread southerly winds associated with the powerful storm system that hit the coast
could have driven much of the motion of the layer. Therefore, whether or not the surface
flow is actually an extension of the CUC or an entirely separate and distinct current with
different forcing functions is a complicated argument to make. From a dynamic
topography perspective, both opinions can be argued. One would expect that for a
northerly current to exist in the absence of atmospheric forcing, a pressure gradient force
would still be necessary and, hence, higher dynamic topography would have to be
observed inshore. On the other hand, a storm with considerable southerly winds would
naturally be expected to pile up water to the right of the direction of the wind flow,
which, in this case, would also be along the coast. With only a week of data to rely on,
no conclusive evidence of either case can be demonstrated.
VI. Future Work
In light of the above discussion, it is obvious that long-term data collection from
in-situ sensors at various depths and for prolonged periods of time would be beneficial
for better modeling the complex interactions of the CCS. Use of glider and RAFOS
floats at appropriate depths in the water column could significantly improve the amount
and resolution of data collected, further enhancing the understanding of the dynamics
involved with the DC. A multiyear time series comparison of dynamic topography
contours and measurements of both the subsurface and surface currents, while filtering
out the seasonal effects of changing wind stress regimes, could also help resolve the
question of what other factors are involved with driving the currents. Seasonal effects of
the changing weather patterns could also be factored in to match the actual observed
Chelton, Dudley B., 1982: Large Scale Response of the California Current to Forcing by
the Wind Stress Curl. CalCOFI Rep., Vol. XXII.
Collins et al., 2003: The California Current system off Monterey, California: physical and
biological coupling. Deep-Sea Research, 50, 2389 - 2404.
Huyer, A. and R. L. Smith, 1974. A subsurface ribbon of cool water over the continental
shelf off Orgeon. J. Phys. Oceanogr., 4, 381-391.
Pennington, J. T. et al., 2007: Ocean Observing in the Monterey Bay National Martine
Sanctuary: CalCOFI and the MBARI time series. A report to the Sanctuary Integrated
Monitoring Network (SIMoN) Monterey Bay Sanctuary Foundation.
OC3570, Pacific Ocean Activities
23-26 Jan. 2008 (Leg I)
27-30 Jan. 2008 (Leg II)
S17 = KiteSonde
37.75 S8 = Mooring
X79 63 Drifters
X83 * = 6, S3, S6, X70
37.00 65 X84,S19
70,X85,S20 79 1
200, 1000, 2000, 2,X66 S1 80
36.75 and 3000m
77,X91 7 78,15,X77
isobaths shown 72,X87
X90 75 14 8,X71
36.50 74,X89 13,S5,X76
12,X75 S4 10,X73
Figure 1. OC3570 Winter Cruise Track
Figure 2. Principles of ADCP operation illustrated (Gordon, 1996).
Figure 3. Dynamic Height Contours derived from CTD casts.
Figure 4. MBARI Moorings (M1 and M2).
Figure 5. CODAR site at Point Sur, CA.
Mean 27-59 m, OS75 VMADCP, OC3570 Winter 2008
-123.4 -123.2 -123 -122.8 -122.6 -122.4 -122.2 -122 -121.8
Figure 6. ADCP Transects 23 – 30 January, 2008.
Figure 7. California shelf extension off the coast north of Monterey Bay.
Mean 27-59 m, OS75 VMADCP, OC3570 Jan 30 2008 0630-1842 Line 67
-123.4 -123.2 -123 -122.8 -122.6 -122.4 -122.2 -122 -121.8
Figure 8. Line 67 ADCP transect, 30 January, 2008.
Line SF 29 Jan 2008
10 10 20
20 20 20 240
0-4 0 0 210
-5 0 - - 20
4 -3 00
-10 10 0 0-50 20
10 20 60
0 10 0 20
0 -50 -20
0 0-50 --4 0
-400 10 0
-4 0 0 -10 -5 0 --3
-5 0 0 -2 4-0
0 -110 0
-123.4 -123.35 -123.3 -123.25 -123.2 -123.15 -123.1 -123.05
Figure 9. Vertical cross section of SF ADCP track.
Figure 10. Smoothed CTD profile along track 67.
Figure 11. Lagrangian drifter tracks.
Figure 12. January 23rd CODAR image showing northerly flow off of Pt. Sur
despite northerly wind flow.
Figure 13. Quikscat imagery from 25 January, illustrating the large storm that
impacted the Winter cruise.
Figure 14. COAMPS field indicating the wind strength of the storm.
Figure 15. Wind Stress Curl defined.
Figure 16. MBARI M2 buoy profile showing downwelling associated with the
southerly storm winds January 25 – 28th.