Evolution of chemical, biological, and physical water properties

in the northern California Current in 2005:

Remote or local wind forcing?

B. Hickey,1 A. MacFadyen,1 W. Cochlan,2 R. Kudela,3 K. Bruland,3 and C. Trick4

Received 2 May 2006; revised 28 July 2006; accepted 1 August 2006; published 6 October 2006.

[1] The spring onset of persistent upwelling-favorablewinds was later than usual in the northern CaliforniaCurrent system in 2005, resulting in delayed provision ofinorganic nutrients to the upper waters of the coastal ocean.This study uses water column measurements to illustrate theevolution of temperature, salinity, nitrate and chlorophyll aprior to and after the onset of persistent local upwelling-favorable winds, including recovery to ‘‘typical’’ conditions.Warm, nutrient- and chlorophyll-depleted surface conditionssimilar to those in an El Nino were observed from VancouverIsland to central Oregon, and extended to depths greater than500 m. Return to typical conditions was more rapid thansuggested by time-integrated local wind stress but consistentin timing with ‘‘remote’’ forcing of water properties in thisregion by upwelling-favorable winds off northern California.Alongshore advection also likely contributed to the observedrecovery, but was much less effective than upwelling.Citation: Hickey, B., A. MacFadyen, W. Cochlan, R. Kudela,

K. Bruland, and C. Trick (2006), Evolution of chemical, biological,

and physical water properties in the northern California Current in

2005: Remote or local wind forcing?, Geophys. Res. Lett., 33,

L22S02, doi:10.1029/2006GL026782.

1. Introduction

[2] Seasonal water properties on the continental shelves ofthe California Current System (CCS) are controlled by thedegree of basin scale advection from the north or southcoupled with the ability of winds to raise this water to theeuphotic zone [Hickey, 1979]. Whatever water is presentbelow the shelf is upwelled onto the shelf in spring andsummer; the degree of upwelling depends on the magnitudeand persistence of upwelling-favorable winds (from thenorth) along the coast. Upwelled water on the continentalshelves of the northern CCS in summer has a remarkabledegree of along-coast uniformity in spite of a factor of threeor more northward decrease in wind stress over the region.Also, at a given location, maximum monthly mean south-ward flow precedes maximum local southward wind stressby 1–2 months [Geier et al., 2006], suggesting that local

wind stress is not solely responsible for observed along-shore currents.[3] Previous extreme conditions have been reported in the

CCS—e.g., El Nino and Subarctic intrusion. During most ElNinos, saltier, warmer and lower nutrient water is present inthe northern CCS together with more southern biota and areduction in planktonic growth [Corwith andWheeler, 2002].During a Subarctic intrusion, fresher, colder, higher nutrientwater is advected from the north and plankton growth isenhanced [Wheeler et al., 2003]. Here, we examine bio/chem/physical conditions during a third type of anomaly—thedelayed onset of persistent local upwelling-favorable winds.

2. Data Set

[4] Water property data were obtained on five cruises inspring and summer 2005. Sampling was conducted alongtransects off central Washington, off Vancouver Is. (only oneof 6 transects is shown) and central Oregon (Figure 1).Timing of data sections relative to wind conditions is shownin Figure 1. All hydrographic profiles were collected with acalibrated SeaBird CTD. Historical CTD profiles calibratedwith bottle salinity data are also used.[5] Inorganic nutrient and chlorophyll a (Chl a) samples

were collected via rosette bottles. Chl awas determined usingstandard in vitro fluorometric analyses [Welschmeyer, 1994]after filtration onto Whatman GF/F filters (0.7 �m nominalpore size). CTD fluorescence measurements were madeusing various in situ instruments from WETLabs, SeaTech,and Chelsea. Voltage was converted to Chl a by regression ofdiscrete bottle samples and coincident fluorometer datafor each cruise (r2 ranged from 0.72 to 0.81). Sampleswere analyzed for nitrate plus nitrite (NO3

� + NO2�; here-

after referred to as nitrate or ‘‘N’’) using a flow-injectionLachat autoanalyzer with the procedure of Smith andBogren [2001].[6] Hourly wind speed and direction were obtained from

the National Data Buoy Center buoys off central Washington(B46041; Figure 1) and northern California (B46014; 39�120N, 124� 00.00W). These data were rotated to the coastlinedirection and alongshore wind stress was computed using thedrag coefficient of Large and Pond [1981]. The data werefiltered using a cosine-Lanczos filter with a half power pointof 46 hr.

3. Results

3.1. How Anomalous were Conditions in 2005?

3.1.1. Comparison with Other Years[7] The spring transition to upwelling conditions, which

is associated with the onset of persistent upwelling-favorable

GEOPHYSICAL RESEARCH LETTERS, VOL. 33, L22S02, doi:10.1029/2006GL026782, 2006ClickHere

for

FullArticle

1School of Oceanography, University of Washington, Seattle, Wa-shington, USA.

2Romberg Tiburon Center for Environmental Studies, San FranciscoState University, Tiburon, California, USA.

3Ocean Sciences Department, University of California, Santa Cruz,California, USA.

4Department of Biology, University of Western Ontario, London,Ontario, Canada.

Copyright 2006 by the American Geophysical Union.0094-8276/06/2006GL026782$05.00

L22S02 1 of 5

winds, was delayed in 2005 from its average date in Marchor April [Strub and James, 1988] until about May 24 [Kosroet al., 2006] (Figure 1). In the following we use the year 2003as a typical year (defined as having no El Nino and noSubarctic intrusion) for comparison to 2005. Physical, chem-ical and biological characteristics for 2003 are similar to longterm summer averages for Oregon and Washington as shownby Landry et al. [1989]. In 2005, stronger than averageupwelling-favorable winds began on about July 14, and fullrecovery to typical physical, biological and chemical con-ditions was observed by early August as shown below.[8] Upper water column nitrate in late May 2005 was

significantly lower than in other years (Figure 2, a compar-ison with 2003; also see Landry et al. [1989, Figure 1.17],long term average surface values >5 �M at mid shelf inMay and June for both Washington and Oregon). In 2003,nitrate was greatest nearshore (>10 �M) and decreasedoffshore such that no detectable nitrate (detection limit

�0.1–0.2 �M) was observed in the upper 10 m of the watercolumn seaward of the 50 m isobath. In contrast, during May2005, the depth of undetectable nitrate extended from thesurface to depths >10–30 m across the entire shelf region,with the deepest depletion at mid shelf.[9] Temperature-salinity (T-S) relationships at mid shelf

(�100 m water depth) show that water properties in 2005were similar to those in the summer of the greatest El Nino ofthe last 30 years, 1983 (Figure 3). Overall, the T-S properties

Figure 1. Locations of sampling sections and wind buoy(B41). Alongshore wind is also shown (positive to right ofaxis), with section timing indicated on the axis.

Figure 2. Concentration of nitrate across the Washingtonshelf and slope in early summer 2005 and 2003. Arrows ontop indicate station locations. On this and subsequent figures,dots indicate bottle sample depths.

Figure 3. Temperature-salinity relationships on theWashington mid shelf (�100 m) in early summer for 2005and selected other years.

Figure 4. Deep hydrographic vertical structure in earlysummer 2005 (GHR) and 2003 (CB).

L22S02 HICKEY ET AL.: 2005 CCS-NORTH PROPERTY EVOLUTION L22S02

2 of 5

indicate a large range of natural variability, with 2005 near thewarm extreme for salinities >32 psu.

3.1.2. Depth Structure[10] Water was warmer from the sea surface to at least

500 m in 2005 compared to 2003, and more saline at depths

near the shelf break (Figure 4). Warmer, saltier water near theshelf break (‘‘spicier’’ water) is consistent with conditionsfrequently observed during El Nino [Huyer et al., 2002]. Fora given temperature, water was saltier, indicating a differencein water mass as opposed to reduced upwelling. Freshersurface water over the shelf in 2005 is due to a northwardplume from the Columbia River (Figure 1), as typicallyoccurs during periods of summer downwelling-favorablewind [Hickey et al., 2005].

3.1.3. Alongshore Extent[11] Comparison of data across 13 sections over a distance

of�450 km demonstrates that warmer, fresher surface watersand nutrient depletion extended from mid Vancouver Islandto central Oregon with nitrate deficits occurring across theentire shelf and slope (most northern and most southerntransects, nitrate only, shown in Figure 5). The Oregonpattern is typical of late May and early June patterns on theother sampling lines (not shown). Corresponding Chl asections (not shown) showed remarkably low concentrations

Figure 5. Nitrate off Vancouver Is. (LD) and centralOregon (NH) in June/July 2005.

Figure 6. Seasonal evolution of water properties in 2005 off the Washington coast.

L22S02 HICKEY ET AL.: 2005 CCS-NORTH PROPERTY EVOLUTION L22S02

3 of 5

in these regions, consistent with satellite-derived distribu-tions given by Thomas and Brickley [2006].

3.2. Seasonal Evolution of Ocean Properties

[12] In a section off central Washington on April 19 (notshown) 8 and 9 �C isotherms tended downward toward thecoast over the slope, consistent with average winter con-ditions rather than average April conditions [seeHuyer et al.,2002]. Water colder than 7 �C was deeper than 225 m,indicating that upwelling had not yet begun. Upwelling ofthis deep water from the slope to the shelf occurred betweenApril 19 and May 31 (Figure 6), likely associated with theMay 24 spring transition. Water in the ranges 7.5–8 �Cand 33.5–33.7 psu progressed shoreward over the shelfbottom as the season advanced. This water reached theinner shelf (bottom depth of 30 m) by July 17 (Figure 6),following the onset of strong local upwelling-favorablewinds on July 14. The upward slope of temperature andsalinity isopleths in the upper 50 m increased and theregion of upward slope expanded offshore as the seasonadvanced. The thickness of the layer of colder water alsoincreased.[13] The salty, cold tongue along the bottom brought with

it higher nitrate (between �25 and 30 �M). However nitratein the upper layers began to increase even before the coldtongue reached the inner shelf (e.g., July 9). This suggeststhat vertical mixing or three dimensional effects also play arole in nitrate supply. Note that differences in layer thicknessin the contoured nitrate sections are due in part to the differentsampling density between surveys (black dots indicate bottlelocations).[14] Chl a was generally low at the beginning of the

season (<2 �g l�1) and concentration increased as the seasonprogressed, exceeding 15 �g l�1 by August 5, a highvalue for this period [Landry et al., 1989; see also R. M.Kudela et al., 2006]. Elevated Chl a concentrations on July 9are related to the plume from the Columbia River, which

supports elevated biomass under some conditions [Hickey,2006].

4. Discussion

[15] Data presented here show that both Chl a and nitratewere suppressed from central Oregon to Vancouver Is. inearly summer relative to ‘‘typical’’ conditions. Recovery toaverage surface concentrations on the inner shelf did notoccur until early August, following the onset of strong localupwelling-favorable wind stress. However, physical andchemical conditions below the surface layer were progressingsteadily toward normal conditions much earlier in the season(Figure 6). This progression is evaluated with two metrics:first, the distance of the ‘‘nose’’ of 8 �C water from the coast,and second, the thickness of the water <8 �C at mid shelf(100 m depth). Both are compared to time-integrated along-shore wind stress assuming linearized equations of motion,and a two dimensional, frictional (surface and bottom stressequal) mass balance with constant density. If we also depthaverage velocity over a bottom boundary layer with as-sumed constant thickness H, then distance D across the shelf

bottom can be expressed as D =R �

�Hfdt, where � is

alongshore wind stress, � is density, t is time and f is theCoriolis parameter. Thus the location of the nose of theupwelled water might be expected to be at least roughlyrelated to time-integrated local wind stress. For localwinds we use both total integrated wind stress (WA)and upwelling-favorable integrated stress only [WA(uf)].Use of the latter may be more reasonable becausealongshore currents on the mid to outer shelf after thespring transition do not reverse under downwelling windconditions [MacFadyen et al., 2005]. Thus transport isalways onshore in the bottom boundary layer in thoseregions until an undercurrent develops. The integration wasstarted on April 15.[16] Cold water advanced up the shelf in May and June

more persistently than suggested by the rate of changeof local time-integrated upwelling-favorable wind stress

Figure 7. (left) Distance from the coast of the nose of 8�Cwater and (right) thickness of the nose at midWA shelf relative tolocal (WA), local upwelling-favorable only [WA(uf)] and remote (CA) alongshore wind stress. The black symbol on the rightpanel is the value interpolated from stations on either side of the�100 m station, whose data appeared to be biased by a smallscale anomaly.

L22S02 HICKEY ET AL.: 2005 CCS-NORTH PROPERTY EVOLUTION L22S02

4 of 5

(Figure 7). Both the cold water advance and the increase inthickness of the cold layer have a temporal structure (i.e.,changes in slope are similar) more related to time-integratedwind stress off California than offWashington (Figure 7; r2 =0.98 vs. 0.91, respectively). The correlation with local windstress is not improved (0.84) if the early season data areremoved from the Washington dataset, suggesting thatchanges are more strongly related to remote than local windstress throughout the season.[17] Bottom velocity estimated from the advance of the

cold water nose is an order of magnitude less (�0.5 cm s�1)than that calculated from the expression above. However, wenote that observed cross-shelf bottom velocities in thisregion are frequently below 1 cm s�1, in part because themajority of the upwelling return flow occurs above thebottom boundary layer [Hickey, 1989]. The failure of simplelocal stress balance and the better statistical relationship withtemporal changes in time-integrated alongshore wind stressfrom northern California suggests that remote wind forcingand subsequent internal Kelvin wave propagation play asignificant role in seasonal setup of the summertime densityfield, as suggested by model studies [McCreary et al., 1987].[18] Both cold water advancement along the shelf bottom

and the thickness of the cold water layer also likely reflectsome contribution from alongshore advection of Subarctic

water (@T

@t= �v

@T

@y, where T is temperature, v is alongshore

velocity and y is alongshore distance). To estimate the orderof magnitude of this effect we use an average alongshoretemperature gradient of 2 �C per 1000 km [Huyer, 2003]and velocities near 50, 74 and 100 m over both the mid andouter shelf in June and July (maximum velocity ��10, �5and �2.5 cm s�1, respectively [from Hickey, 1989; Geier etal., 2006]). The advective temperature contributions are�1.2, 0.6 and 0.3 �C at mid shelf, much less than theobserved differences of �4.0, 2.5 and 1.0 �C at mid shelfand about half the values observed on the outer shelf. Thisresult, combined with a salinity increase rather than thedecrease expected from alongshore advection (��0.2 psufor a velocity of 10 cm s�1) suggest that upwelling dominatesthe observed seasonal changes.[19] Thus we conclude that the delicate balance between

upwelling-favorable winds, nutrient supply and planktongrowth in the northern CCS likely depends not only on localwinds but also on remote forcing by winds along the northernCalifornia coast. These latter winds and the internal Kelvinwaves they generate appear critical for uplifting deeper,nutrient-rich water to the shelf and transporting it from theshelf break to the inner shelf, where it can be upwelledto the sea surface by the local winds. The flow thataccompanies the resulting density field transports upwelledwater southward along the coast and likely also contributes tothe seasonal water property evolution.

[20] Acknowledgments. Water property data collection was sup-ported by the National Science Foundation (NSF) as part of the CoOPRISE Program (OCE0239089 to B. Hickey and OCE0238347 to K. Brulandand R. Kudela) as well as by the Coastal Ocean Program of the NationalAtmospheric and Oceanic Administration (NOAA) (NA17OP2789 toB. Hickey) and NSF (OCE0234587 to B. Hickey and W. Cochlan) as part ofECOHAB PNW, and to B. Hickey as part of ORHAB (NA07OA0310).Analysis was supported by these grants as well as by the GLOBECNortheast

Pacific CCS program (OCE0001034 to B. Hickey). This is contribution 9 ofthe ECOHAB PNW program, 188 of the ECOHAB program and 315 of theU.S. GLOBEC program. The findings and conclusions are those of theauthors and do not necessarily reflect those of NOAA or the Department ofCommerce. Our many thanks to the seagoing RISE, ECOHAB PNW andORHAB teams for their critical contribution to the data collection.

ReferencesCorwith, H. L., and P. A. Wheeler (2002), El Nino related variations innutrient and chlorophyll distributions off Oregon, Prog. Oceanogr., 54,361–380.

Geier, S. L., B. M. Hickey, S. R. Ramp, P. M. Kosro, N. B. Kachel, andF. Bahr (2006), Interannual variability in water properties and velocityin the U.S. Pacific Northwest coastal zone, Eos Trans. AGU, 87(36),Ocean Sci. Meet. Suppl., Abstract 36D-23.

Hickey, B. M. (1979), The California current system—Hypotheses andfacts, Prog. Oceanogr., 8, 191–279.

Hickey, B. M. (1989), Patterns and processes of circulation over the Wa-shington shelf and slope, in Coastal Oceanography of Washington andOregon, edited by M. R. Landry, and B. M. Hickey pp. 41–115, Elsevier,New York.

Hickey, B. M. (2006), Complexity of a large freshwater plume, Eos Trans.AGU, 87(36), Ocean Sci. Meet. Suppl., Abstract 323-01.

Hickey, B., S. Geier, N. Kachel, and A. MacFadyen (2005), A bi-directionalriver plume: The Columbia in summer, Cont. Shelf Res., 25, 1631–1656.

Huyer, A. (2003), Preface to special section on enhanced Subarctic influ-ence in the California Current, 2002, Geophys. Res. Lett., 30(15), 8019,doi:10.1029/2003GL017724.

Huyer, A., R. L. Smith, and J. Fleischbein (2002), The coastal ocean offOregon and northern California during the 1997–8 El Nino, Prog. Ocea-nogr., 54, 311–341.

Kosro, P. M., W. T. Peterson, B. M. Hickey, R. K. Shearman, and S. D.Pierce (2006), Physical versus biological spring transition: 2005, Geo-phys. Res. Lett., 33, L22S03, doi:10.1029/2006GL027072.

Kudela, R. M., et al. (2006), Impacts of phytoplankton biomass and pro-ductivity in the Pacific Northwest during the warm conditions of 2005,Geophys. Res. Lett., doi:10.1029/2006GL026772, in press.

Landry, M. R., J. R. Postel, W. K. Peterson, and J. Newman (1989), Broad-scale patterns in the distribution of hydrographic variables, in CoastalOceanography of Washington and Oregon, edited by M. R. Landry, andB. M. Hickey. pp. 1–41, Elsevier, New York.

Large, W. G., and S. Pond (1981), Open ocean momentum flux measure-ments in moderate to strong winds, J. Phys. Oceanogr., 11, 324–336.

MacFadyen, A., B. M. Hickey, and M. G. G. Foreman (2005), Transport ofsurface waters from the Juan de Fuca eddy region to the Washingtoncoast, Cont. Shelf Res., 25, 2008–2021.

McCreary, J. P., P. K. Kundu, and S. Chao (1987), Dynamics of the Cali-fornia current system, J. Mar. Res., 45, 1–32.

Smith, P., and K. Bogren (2001), Determination of nitrate and/or nitrite inbrackish or seawater by flow injection analysis colorimeter: QuickChemMethod 31-107-04-1-E, in Saline Methods of Analysis, report, 12 pp.,Lachat Instrum., Milwaukee, Wis.

Strub, P. T., and C. James (1988), Atmospheric conditions during thespring and fall transitions in the coastal ocean off western UnitedStates, J. Geophys. Res., 93(C12), 15,561–15,584.

Thomas, A. C., and P. Brickley (2006), Satellite measurements of chloro-phyll distribution during spring 2005 in the California Current, Geophys.Res. Lett., 33, L22S05, doi:10.1029/2006GL026588.

Welschmeyer, N. A. (1994), Fluorometric analysis of chlorophyll a in thepresence of chlorophyll b and pheopigments, Limnol. Oceanogr., 39,1985–1992.

Wheeler, P. A., A. Huyer, and J. Fleischbein (2003), Cold halocline, in-creased nutrients and higher chlorophyll off Oregon in 2002, Geophys.Res. Lett., 30(15), 8021, doi:10.1029/2003GL017395.

�����������������������K. Bruland and R. Kudela, Ocean Sciences Department, University of

California, Santa Cruz, Santa Cruz, CA 95064, USA.W. Cochlan, Romberg Tiburon Center for Environmental Studies, San

Francisco State University, 3152 Paradise Drive, Tiburon, CA 95064, USA.B. Hickey and A. MacFadyen, School of Oceanography, University of

Washington, Box 355351, Seattle, WA 98195, USA. (bhickey@u.washington.edu)C. Trick, Department of Biology, University of Western Ontario, London,

Ontario, N6A5B7, Canada.

L22S02 HICKEY ET AL.: 2005 CCS-NORTH PROPERTY EVOLUTION L22S02

5 of 5