SEARCH

SEARCH BY CITATION

Keywords:

  • equatorial Pacific;
  • El Niño;
  • primary productivity

Abstract

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Overview of the In Situ Data Set
  5. 3. Results
  6. 4. Discussion and Conclusions
  7. Acknowledgments
  8. References
  9. Supporting Information

[1] Using a database of roughly 300 basin-wide conductivity-temperature-depth (CTD) stations per year from 1997 to 2003, we examined interannual and seasonal variability of nutrients and chlorophyll across the equatorial Pacific. During this period, chlorophyll concentrations exceeded the range of previous measurements for this region. Nitrate, silicic acid, and phosphate also varied widely but not necessarily coherently with each other or with chlorophyll. Across the La Niña to El Niño continuum there was nonmonotonic variability in chlorophyll, particulate backscatter (a proxy for phytoplankton carbon), and large size fraction (diatom) chlorophyll. In general, while El Niño was associated with decreased phytoplankton biomass, there was no corresponding increase in phytoplankton or diatoms during La Niña. However, there were increases in macronutrients in response to La Niña. We suggest that the lack of a biological response to these nutrient increases is due to their decoupling from iron supply. Multiple linear regression analysis of the physical factors responsible for vertical nutrient fluxes emphasized the importance of winds in the central and western Pacific, and thermocline depth in the east. We suggest that the role of the winds is not limited to enhancement of upwelling, but perhaps more importantly to increased vertical mixing of nutrients. Seasonal patterns were weak but consistent with previous work which has suggested that enhanced productivity in the second half of the calendar year causes reduced surface pCO2. Enhanced coverage of nutrient sensors on the tropical atmosphere ocean (TAO) array would help to quantify seasonal signals associated with processes such as tropical instability waves and lead to a better understanding of the links between productivity, carbon export, and air-sea CO2 exchange.

1. Introduction

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Overview of the In Situ Data Set
  5. 3. Results
  6. 4. Discussion and Conclusions
  7. Acknowledgments
  8. References
  9. Supporting Information

1.1. Biogeochemical Significance of the Equatorial Pacific

[2] The most prominent feature of the central and eastern equatorial Pacific is the cold tongue of upwelled water which spans approximately one quarter of the Earth's circumference [Wyrtki, 1981]. This upwelled water contains modest concentrations of macronutrients (N, P, Si) and high dissolved inorganic carbon [Chavez and Barber, 1987; Feely et al., 1987; Toggweiler and Carson, 1995]. Primary and new production in the upwelling region are iron-limited and hence remain low relative to the observed levels of macronutrients [Coale et al., 1996a, 1996b]. However, the vastness of the region makes it one of the most productive oceanic ecosystems, accounting for approximately 18% of global new production [Chavez and Toggweiler, 1995]. The partial pressure of carbon dioxide (pCO2) in upwelled waters is ∼50 to 150 μatm higher than atmospheric pCO2 resulting in a flux to the atmosphere of ∼0.5 to 1.1 × 1015 gC a−1 [Takahashi et al., 2002; Feely et al., 2006]. This flux represents the largest oceanic source of CO2 to the atmosphere and is equivalent to ∼10% of the total anthropogenic flux of 6.5 × 1015 gC a−1 [Marland et al., 2000].

[3] Clearly, the equatorial Pacific is a major component of global biogeochemical cycles. The physics, biology and chemistry of the region are perturbed across a broad range of spatial and temporal scales by processes such as Kelvin waves, El Niño events and tropical instability waves (TIWs). Here we document the variability in nutrients and phytoplankton biomass from October 1997 to December 2003 using a growing database of in situ, conductivity-temperature-depth (CTD)–based measurements. We quantify the impact of La Niña (1998–1999) versus strong (1997–1998) and weak El Niño (2002–2003) events and demonstrate that there is no difference between La Niña and “normal” conditions in terms of phytoplankton biomass and community composition. We then construct multiple linear regression models for the temporal variability in nutrients and chlorophyll, to quantify the important physical forcings. We conclude by investigating variability at seasonal timescales.

1.2. Overview of Physical, Chemical, and Biological Variability: 1997 to 2003

[4] The 1997–1998 El Niño was the strongest ever observed. Its physical characteristics and the resulting chemical and biological perturbations have been documented using data from the tropical atmosphere ocean (TAO) array, satellites and in situ observations [Chavez et al., 1999; McPhaden, 1999; Strutton and Chavez, 2000]. The event was initiated by westerly wind bursts over the warm pool, and Kelvin waves that propagated from west to east. These waves deepened the thermocline by 20m or more in the central and eastern Pacific, resulting in decreased surface productivity [Chavez et al., 1998]. At the same time, the thermocline shoaled in the west. The western equatorial warm pool subsequently migrated eastward, and by December 1997, surface waters exceeding 28°C spanned the entire equatorial Pacific, upwelling was essentially shut down and the thermocline in the eastern Pacific was ∼100 m deeper than the climatology. Chlorophyll concentrations at this time were among the lowest ever recorded for the equatorial Pacific (∼0.05 mg m−3; Figure 1d, cf a regional mean of ∼0.2 mg m−3) and for a short time the region became a weak sink of CO2 [Chavez et al., 1999; Strutton and Chavez, 2000]. At the peak of the event, the equatorial undercurrent (EUC) was absent, as has been observed for previous El Niño events [Firing et al., 1983]. The shallow thermocline anomalies in the west migrated into the central Pacific, bringing the source of nutrients close to the surface, but these waters were not upwelled because of a lack of trade wind forcing. When the trades returned in May 1998, the proximity of this cool, nutrient- and CO2-rich water to the surface caused a rapid drop in SST (8°C in a matter of weeks at 0° 140°W), an extremely strong bloom [Ryan et al., 2002] and an intense TIW season [Strutton et al., 2001].

image

Figure 1. Time-longitude plots of (a) zonal wind, (b) SST, and (c) 20°C isotherm depth (Z20°C) calculated from daily tropical atmosphere ocean (TAO) mooring data averaged between 2°N and 2°S. Triangles on the axes indicate the location of the TAO moorings used: data were linearly interpolated at 1° in the longitude direction. Note that changes in Z20°C are mirrored by changes in sea surface height [see McPhaden, 2004, Figure 4b]. (d) Surface chlorophyll concentration from the SeaWiFS ocean color satellite, averaged between 2°N and 2°S. Original data were 9-km spatial resolution and 8-d temporal resolution, averaged to 1° in the longitude direction and interpolated to 1-d in the time direction. SeaWiFS data began in September 1997: hence the gap at the beginning of the time series.

Download figure to PowerPoint

[5] Several years of La Niña conditions followed, characterized by strong trade winds as far west as ∼160°E (Figure 1a), confinement of the warmest waters to the far western Pacific, greater westward expansion of the equatorial upwelling tongue and cool SST anomalies in the eastern Pacific (Figure 1b and McPhaden [2004]). The zonal gradient in the thermocline was accentuated (visible as deeper than normal Z20°C in the western Pacific; Figure 1c) and the region of extremely low primary productivity (surface chlorophyll concentrations <0.1 mg m−3) was restricted to the far western Pacific (Figure 1d). In May 2002, trade winds across the basin relaxed and a basin-wide warming of ∼1°C developed. During the second half of 2002, westerly wind events in the western Pacific continued to generate downwelling Kelvin waves that progressively led to deeper-than-normal thermocline anomalies penetrating further into the eastern Pacific. At the same time, extremely low chlorophyll concentrations (<0.1 mg m−3) were observed as far east as ∼160°W, while waters with chlorophyll >0.25 mg m−3 were absent from the eastern Pacific (Figure 1d).

[6] At its peak (October to December 2002), the 2002–2003 El Niño event was comparable in magnitude to the moderate El Niños of 1986/1987 and 1991/1992, as measured by SST anomalies in the Niño-3.4 region. Upwelling-favorable winds persisted in the east throughout the peak of the event, but the upwelled waters were warm and nutrient-poor. In early 2003, the anomalies in SST, winds and thermocline depth began to weaken and surface chlorophyll concentrations in the central and eastern Pacific increased (Figure 1d). Similar to 1997–1998, an enhanced La Niña bloom followed the demise of the El Niño event (Figure 1d) [Ryan et al., 2006].

2. Overview of the In Situ Data Set

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Overview of the In Situ Data Set
  5. 3. Results
  6. 4. Discussion and Conclusions
  7. Acknowledgments
  8. References
  9. Supporting Information

[7] Since October 1997, chlorophyll and nutrient profiles (typically eight depths from 0 to 200 m, at ∼300 stations per year) have been obtained on all cruises of the NOAA ship Ka'imimoana: the vessel responsible for servicing the TAO array. Chlorophyll was measured by the standard fluorometric technique [Holm-Hansen et al., 1965; Chavez et al., 1995] and samples for nutrients were frozen at sea for later analysis on an Alpkem Rapid Flow Analyzer [Sakamoto et al., 1990]. Twice per year, more extensive measurements (optical profiles, primary and new production) have been made along the 155°W and 170°W TAO array lines [Strutton and Chavez, 2000]. Figure 2 shows an example of the spatial coverage of CTD stations from 1997 to 2003. In each year after 1997, each of the meridional transects was visited twice, resulting in two complete basin crossings per year approximately 6 months apart, so the station density in Figure 2 is roughly twice what it seems. This coverage enables us to address, at coarse resolution, questions of seasonal variability in chlorophyll and nutrients.

image

Figure 2. Distribution of conductivity-temperature-depth (CTD) stations for 2000 as an illustration of the typical annual coverage. Station locations are plotted over a map of mean SeaWiFS chlorophyll for 2000. Transect (TAO mooring) longitudes from west to east are 165°E, 180° (plotted as a vertical line), 170°W, 155°W, 140°W, 125°W, 110°W, and 95°W. Each line is occupied twice per year, so the density of stations is roughly twice what it appears. The boxes correspond to the regions discussed in the text.

Download figure to PowerPoint

[8] Data describing zonal winds, SST, 20°C isotherm depth, and the speed of the EUC were obtained from NOAA/PMEL (http://www.pmel.noaa.gov/tao/jsdisplay). For SST, Z20°C and winds, daily data from all moorings between 2°N and 2°S within the Wyrtki [1981] box were used. The EUC velocity data came from equatorial ADCP moorings at 110°W, 140°W and 170°W. Section 3.3 describes the results of the multiple linear regression (MLR) analyses that correlated these physical data with chlorophyll and nutrients. Extensive cross-correlation analysis was performed to determine the best lag and window to use. This analysis showed that a window of 30 d and a lag of zero worked best. That is, when creating MLR models to predict chemical and biological data from a TAO line, the SST, Z20°C, wind and EUC data from the 30 d immediately prior to the transect were used.

3. Results

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Overview of the In Situ Data Set
  5. 3. Results
  6. 4. Discussion and Conclusions
  7. Acknowledgments
  8. References
  9. Supporting Information

3.1. Variability of Nutrients and Chlorophyll Through Two El Niño Events

[9] The in situ data set just described allows us to construct a time series of chlorophyll and nutrients in an equatorial “volume”, that is, a box defined by a latitude-longitude-depth criterion. We chose the mixed layer between 5°S and 5°N, 180° and 95°W, which is essentially the Wyrtki [1981] box. To look at regional variability, we have also subdivided this domain into three smaller regions (Figure 2), designated west (180° and 170°W TAO lines), central (155°W, 140°W and 125°W) and east (110°W and 95°W). Note that the “west” box is not the heart of the warm pool near Papua New Guinea, rather the region near the date line that experiences the zonal migration of the front between the upwelling tongue and the western oligotrophic Pacific. The entire Wyrtki box is subsequently referred to as “basin”. The 165°E transect was omitted from the calculations because it was sampled less frequently than the others. With regard to the vertical extent over which we have made our calculations (the mixed layer), the observed patterns are similar if the data are averaged over the euphotic zone, the surface to the 20°C isotherm or the upper 200 m, but the variability is less pronounced because the deeper integration depths begin to include the permanently high nutrient (low chlorophyll) concentrations below about 100m. The mixed layer was defined by a density criterion, that is, the depth at which σt exceeded surface σt by 0.01 kg m−3. Figures 3 and 4show the interannual variability of physics, nutrients and chlorophyll in the central region for the period 1 January 1998 to 31 December 2003. For the physical data (SST, zonal winds, wind speed, Z20°C and EUC velocity) the mean seasonal pattern has been removed.

image

Figure 3. Physical variability for the equatorial Pacific, 1 January 1998 to 31 December 2003. (a) The multivariate ENSO index (MEI [Wolter and Timlin, 1998]) (dimensionless). (b) Sea-surface temperature (°C), (c) zonal (E–W) winds (m s−1), (d) scalar (directionless) winds (m s−1), (e) depth of the 20°C isotherm (meters), and (f) zonal velocity of the equatorial undercurrent (cm s−1). The data presented are for the central Pacific (2°N to 2°S, 125°W, 140°W, and 155°W TAO lines) with the mean annual cycle removed and smoothed with a 6-month running mean.

Download figure to PowerPoint

image

Figure 4. Chemical (nutrients) and biological (chlorophyll) variability for the equatorial Pacific, 1 January 1998 to 31 December 2003. The data presented are for the central Pacific: 5°N to 5°S, 125°W, 140°W, and 155°W TAO lines. (a) The multivariate ENSO index (MEI [Wolter and Timlin, 1998]) (dimensionless). (b) Chlorophyll a (mg m −3) from ship samples (closed circles) and SeaWiFS (solid line). (c) Phosphate, PO4 (mmol m−3), and (d) nitrate, NO3 (mmol m−3), and silicic acid, Si(OH)4 (mmol m−3). All data are mixed layer means, and the horizontal line defining the anomalies is the mean of the data set.

Download figure to PowerPoint

[10] We use the time series of the multivariate ENSO index (MEI; Wolter and Timlin [1998]) as an indication of the El Niño/La Niña status of the tropical Pacific (Figure 3a). Positive (negative) values indicate El Niño (La Niña) conditions. The temporal variability of the MEI is almost identical to that of the southern oscillation index (SOI). We prefer the MEI because its sign is more intuitive (positive for El Niño, same sign as SST anomalies), it has a mean of zero, a standard deviation of one and it is calculated from a suite of variables with more relevance to ocean productivity than the pressure difference between Tahiti and Darwin. The 1997–1998 event is obvious at the beginning of the data set, followed by a rapid and intense transition to La Niña. From 1999 to 2002 there is a gradual weakening of La Niña into the weak 2002–2003 El Niño [McPhaden, 2004]. The other physical parameters (SST (Figure 3b), zonal winds (Figure 3c), wind speed (Figure 3d), 20°C isotherm depth (Figure 3e), and the speed of the EUC (Figure 3f)) all show similar temporal variability consistent with the El Niño phases. The removal of the seasonal cycle was not perfect, so some weak seasonal variability is still evident. The 20°C isotherm depth is notable in that its transition to La Niña conditions in 1998 leads the other physical parameters, consistent with the thermocline variability described by McPhaden [1999].

[11] In the interest of space we have only shown the physical (Figure 3), chlorophyll and nutrient (Figure 4) data for the central region. The patterns for the east and west are similar with the following exceptions. In the west, the La Niña zonal wind anomalies are more strongly negative from mid-1998 through the end of 2001, indicating stronger than average upwelling-favorable conditions. This is reflected in the wind speed as a positive anomaly for the same period. During 2002 and early 2003, upwelling-favorable winds were weaker in the west than in the central region, consistent with the signal that might be expected for a moderate El Niño. In the east the positive SST anomalies of the 1997–1998 El Niño persist perhaps 4–5 months longer than in the central region, and this is correlated with the pattern in the thermocline anomalies, which are more strongly positive (deeper) at the beginning of our time series.

[12] The mean chlorophyll value for the central region (Figure 4b), based on both satellite and in situ measurements, is very close to the 0.21 mg m−3 calculated for 1980–1998 [Chavez et al., 1999]. For the in situ data, there is no large peak in mid-1998 corresponding to the intense La Niña bloom [Ryan et al., 2002] because the ship sampling missed the peak of the bloom, but it did accurately quantify the regional chlorophyll concentrations immediately before and after. Where ship and satellite data cooccur, they mostly resemble each other very closely. The bloom is visible in the SeaWiFS satellite chlorophyll time series for the central and eastern regions, but not the west. In the east there was a slightly more intense bloom in 1998, and chlorophyll concentrations are as much as ∼0.5 mg m−3 greater than the central region on average, but otherwise the time series look very similar. In the west, apart from the absence of the 1998 bloom, chlorophyll concentrations are on average about 0.05 mg m−3 lower than the central region and decrease to less than 0.1 mg m−3 at the end of 2002 associated with El Niño conditions. Note that this was also the period of weaker upwelling-favorable winds in the west.

[13] Nitrate, silicic acid and phosphate varied by up to five-fold over our chosen time period (Figures 4c and 4d). Assuming limiting nitrate and silicic acid values of ∼3 mmol m−3 and ∼2.5 mmol m−3 [Leynaert et al., 2001], respectively, these nutrients were limiting in the central and eastern regions only during the 1997–1998 and 2002–2003 El Niño events. In the west (data not shown) they were always limiting, except perhaps for a period from late 1999 to late 2000. Phosphate was generally between about 0.5 and 0.8 mmol m−3 for the central and eastern regions, except during 1997–1998 and 2002–2003 when it dropped to about 0.3 mmol m−3. In the central Pacific, all three nutrients generally increase from 1998, peak in 2000 and decrease again toward the 2002–2003 El Niño without a clear corresponding pattern in chlorophyll. This emphasizes the importance of iron in this ecosystem, for which there is unfortunately almost no data. In the east the same was generally true except that the increase in NO3 and Si(OH)4 was greater and more rapid than that of PO4 and there was no corresponding signal in chlorophyll. In the west, PO4 was always between 0.2 and 0.5 mmol m−3, NO3 and Si(OH)4 were always limiting and the variability in chlorophyll closely resembled that of PO4. The analysis in subsequent sections attempts to explain this variability in terms of the physical processes determining the vertical nutrient flux and the biological factors consuming surface nutrients.

3.2. Impact of La Niña and Moderate Versus Strong El Niño Events on Nutrients and Chlorophyll

[14] The eqPac database and SeaWiFS satellite data were used to quantify the impact of La Niña and moderate-to-strong El Niño conditions on mixed layer chlorophyll, biogenic particles and nutrients. Figure 5 shows chlorophyll as a function of the MEI, where negative MEI values indicate La Niña conditions, positive values indicate El Niño and values greater than ∼1.5 are associated with the very strong El Niño event of 1997–1998. It is interesting that there is very little variability in basin-averaged chlorophyll through the transition from La Niña to moderate El Niño conditions, but there is a decrease in chlorophyll toward strong El Niño conditions. To quantify the relationship between ENSO status and chlorophyll, we fit a line to all satellite chlorophyll data for which MEI < 0 (La Niña) and all data for which MEI > 0 (El Niño). For MEI < 0 only the western equatorial Pacific showed a significant (negative) slope and correlation, indicating an increase in chlorophyll associated with La Niña. For El Niño (MEI > 0), all regions had significant negative correlations between MEI and chlorophyll, indicating decreasing chlorophyll as El Niño intensifies. We also performed this analysis for 0 < MEI < 1.5, excluding the strong El Niño of 1997–1998. All regions except the central Pacific showed significant negative correlations between chl and ENSO indicating a significant decrease in chl associated with weak-to-moderate El Niño events. To summarize, La Niña does not lead to an increase in chlorophyll, except in the western Pacific. Moderate-to-strong El Niño events decrease chlorophyll, except in the central Pacific, where only strong events reduce chlorophyll concentrations.

image

Figure 5. Basin-averaged and regional chlorophyll concentrations across the La Niña to El Niño continuum. Closed circles represent monthly mean SeaWiFS data (approximately the upper 20 m, September 1997 to September 2004) as a function of MEI for that month. Closed triangles (with standard deviation) represent sporadic cruise data for the 1988–1989 La Niña period, included so as to increase the “sample size” of La Niña conditions sampled. Open circles are the same data from Figure 4 plotted as a function of MEI rather than time. Negative MEI corresponds to La Niña, and positive MEI corresponds to El Niño.

Download figure to PowerPoint

[15] To compliment our investigation of chlorophyll variability associated with ENSO, we have plotted satellite-derived particulate backscatter (bbp) as a function of MEI (Figure 6). The backscatter data were derived from SeaWiFS observations using the Garver-Siegel-Maritorena semianalytical model [Maritorena et al., 2002; Siegel et al., 2005]. The data are available at http://web.science.oregonstate.edu/ocean.productivity. We acknowledge the limitations of these data. They have not been as well validated as the standard SeaWiFS chlorophyll products, but they are incorporated here to aid the discussion of the variability in chlorophyll. Statistically, there is no relationship between bbp and La Niña (MEI < 0) or bbp and moderate El Niño events. In these Case I waters, where terrestrial material makes no appreciable contribution to bbp, this indicates that phytoplankton carbon does not increase with La Niña, nor does it decrease with moderate El Niño events. Taking into account the strong 1997–1998 El Niño event, the eastern and western regions do experience significant changes in bbp as a function of MEI, but the central region does not. Curiously, while bbp decreases with El Niño in the eastern Pacific, as one might expect, it increases in the west.

image

Figure 6. Basin-averaged and regional particulate backscatter (bbp (m−1)). Data are monthly means calculated from SeaWiFS data using the Garver-Siegel-Maritorena model. Negative MEI corresponds to La Niña, and positive MEI corresponds to El Niño.

Download figure to PowerPoint

[16] While bbp is commonly used as an indicator of phytoplankton carbon, it is generally accepted that it cannot provide any information on changes in particle size distribution [Behrenfeld et al., 2005; Siegel et al., 2005]. That is, assuming roughly spherical particles (admittedly a poor assumption), a transition from many small particles (picoplankton) to fewer large particles (diatoms) may not necessarily manifest itself in a change in bbp. When plotted against each other (data not shown), bbp and chlorophyll for the equatorial Pacific conform to the two regimes identified by Behrenfeld et al. [2005]. For chlorophyll greater than approximately 0.15 mg m−3, there is a linear relationship between the two, which Behrenfeld et al. [2005] attributed to changes in phytoplankton biomass. Below ∼0.15 mg m−3, chlorophyll changes with no change in bbp, attributed to physiological changes in cellular pigment concentration. This regime, which Behrenfeld et al. [2005] identified with impoverished ocean provinces, corresponded (for our data) with the western Pacific and El Niño conditions in the central and eastern regions. This decoupling between bbp and chlorophyll at low chlorophyll concentrations is related to the opposing responses for chlorophyll and bbp in the west during El Niño.

[17] To further address this question of pigment and particle changes as a function of ENSO, we have used size fractionated chlorophyll measurements to estimate changes in phytoplankton species composition. Here we consider total chlorophyll to be that portion retained by a Whatman glass fiber filter (GFF). The effective pore size of these filters is at most 0.5 μm, and they have been shown to perform as well as 0.2-μm membrane filters [Chavez et al., 1995]. During TAO cruises we routinely filtered 1-μm and 5-μm surface water samples. Since Prochlorococcus and many Synechococcus cells are <1 μm in diameter, we use the proportion of chlorophyll captured by the 1-μm and 5-μm filters to indicate the proportion of chlorophyll contained in larger cells, likely diatoms [Landry et al., 1997]. Figure 7 shows the fraction of chlorophyll contained in the 1-μm size fraction as a function of ENSO phase, for the central Pacific. The general patterns are the same for the west and east, and for the 5-μm size fraction. Contrary to conventional wisdom, there is no appreciable increase in the large-fraction chlorophyll during La Niña conditions. Statistical analysis revealed a significant decrease in the larger chlorophyll size fraction associated with El Niño conditions (MEI>0) for all three regions (that is, a significant correlation and a slope significantly different from zero). For La Niña, while there appears to be a decrease in “large chlorophyll” with increasingly negative MEI, the correlation was only significant and the slope significantly different from zero for the western Pacific.

image

Figure 7. Variability of the >1-μm chlorophyll size fraction as a function of MEI. Data shown are for the central Pacific. Patterns are similar for other regions and for the >5-μm chlorophyll size fraction.

Download figure to PowerPoint

[18] Next we summarize the temporal variability of nutrients as a function of El Niño/La Niña variability. Figure 8 shows NO3, Si(OH)4 and PO4 variability as a function of the MEI for the western, central and eastern regions. The scatter in the data for a given value of the MEI is because the data span 5°N to 5°S, which includes the mesotrophic waters of the upwelling plume and the oligotrophic waters to the north and south. There is a general appearance of decreasing mixed layer nutrients with increasing MEI (El Niño). Again, we separated the data into El Niño (MEI>0) and La Niña (MEI < 0) and calculated the slope and correlation between nutrients and MEI. In almost all cases, La Niña led to significantly increased nutrients and El Niño to significantly decreased nutrients. This is in contrast to the chlorophyll, bbp and size fractionated chlorophyll data which, while they did show some decreases with El Niño, did not increase with La Niña. Phytoplankton biomass and presumably productivity does not respond to the increased macronutrients available during La Niña.

image

Figure 8. Variability of nutrients as a function of El Niño–La Niña conditions. Columns correspond (left to right) to the western, central, and eastern Pacific, while rows (top to bottom) correspond to mixed layer NO3, Si(OH)4, and PO4.

Download figure to PowerPoint

3.3. Regression Analysis of Mixed Layer Nutrients

[19] The goal of this section is to quantify the processes responsible for surface nutrient variability. The data at our disposal include source nutrient concentrations, mixed layer nutrient concentrations and physical data such as thermocline depth and wind speed that potentially influence nutrient delivery to the surface. Simple (SLR) and multiple linear regression (MLR) analyses were performed using Matlab's matrix algebra functions. The dependent variable was the mean mixed layer concentration of nutrients. The independent variables were chosen from the source nutrient concentration (mean concentration of NO3, Si(OH)4 or PO4 between 100m and 150m depth), depth of the 20°C isotherm (Z20°C), zonal wind speed (i.e., east-west wind component, negative toward the west), scalar wind speed and zonal velocity of the EUC. The expected effects of these independent variables are as follows. Elevated source nutrient concentrations potentially result in elevated mixed layer nutrients; a shallow thermocline helps deliver nutrients to the mixed layer; negative zonal winds favor upwelling; increased scalar winds generate vertical mixing of nutrients and a stronger (faster) EUC helps deliver nutrients to the upwelling tongue. The depth of the EUC core was also calculated from the TAO ADCP data, but it was strongly correlated with Z20°C, so it was not used in the analysis. Regression models were constructed using one (SLR) or two (MLR) independent variables from this list. SST was not used as an independent variable because its spatial and temporal variability is governed by many of the same factors as nutrients, so it tells us little about the physical processes modulating chemistry. Section 2, above, described the physical data in more detail and the window/lag criteria that were used. The coefficient of determination, R2, was used to compare simple linear regressions, while the adjusted coefficient of determination, denoted Ra2 [Zar, 1999], was used to compare MLR models. Tables 1 and 2 summarize the R2 and Ra2 values for the simple and multiple linear regression models, respectively.

Table 1. Success of Simple Linear Regression Models Relating Surface Nutrient Concentration to Source Nutrient Concentration, EUC Velocity (EUCu), Thermocline Depth (Z20C), Zonal Wind Speed (Windu), and Scalar Wind (Wind)a
RegionNutrientParameterR2
  • a

    For each nutrient and region, the parameter(s) with the strongest R2 values are listed. Note that even the highest R2 values are quite low for some parameters, explaining much less than 50% of the variability.

EastNO3Z20C0.34
 Wind0.30
Si(OH)4Wind0.38
 Z20C0.24
PO4PO4, source0.58
CentralNO3EUCu0.39
 Windu0.38
Si(OH)4EUCu0.32
 Windu0.26
PO4PO4, source0.36
 Z20C0.18
WestNO3Windu0.52
 Wind0.50
Si(OH)4Z20C0.08
PO4Wind0.56
 Windu0.45
Table 2. Success of Multiple Linear Regression Models Based on Source Nutrient Concentration and One of Four Physical Variables: EUC Velocity (EUCu), Thermocline Depth (Z20C), Zonal Wind Speed (Windu), and Scalar Wind (Wind)a
RegionNutrientParameterR2a
  • a

    The efficacy of the model was quantified by Ra2 as described in the text, and the models were constructed for distinct regions: eastern, central, and western Pacific.

EastNO3Wind0.56
Si(OH)4Wind0.59
PO4EUCu0.68
 Z20C0.64
 Wind0.58
 Windu0.56
CentralNO3Windu0.55
 EUCu0.48
Si(OH)4Windu0.39
 EUCu0.37
PO4Windu0.52
 EUCu0.49
WestNO3Wind0.49
 Windu0.47
Si(OH)4Z20C0.17
PO4Windu0.69
 Wind0.57

[20] Variability in thermocline depth is important in the east, where the mean thermocline is shallower. In the central and western Pacific, scalar winds and zonal winds, presumably as forcing mechanisms of mixing and upwelling, respectively, become more important. Taking mixed layer nitrate as an example, and considering only simple linear correlation first, the results were as follows. In the east, the most significant correlations occurred with Z20°C and scalar winds with R2 values of 0.34 and 0.30, respectively. In the central region, EUC velocity (R2 = 0.39) and zonal winds (R2 = 0.38) best explained mixed layer nitrate, while in the west, zonal winds (R2 = 0.52) and scalar winds (R2 = 0.50) were most important. The same patterns with similar R2 values occurred for mixed layer silicic acid, except that in the west, the highest correlation (only 0.08) occurred for Z20°C. When constructing MLR models of two independent variables to predict mixed layer nutrients, we combined the source nutrient concentration with a physical mechanism to deliver that nutrient to the mixed layer. In the case of nitrate and silicic acid, the patterns were the same as for the simple linear correlations, in that winds were important everywhere and EUC velocity was important in the central Pacific. For phosphate the source concentration is the dominant factor determining mixed layer concentrations in the east. Elsewhere the scalar and zonal winds play an important role. These results, emphasizing the strong correlation between surface nutrients and local wind forcing, are contrary to some previous work that has emphasized the importance of thermocline variability [Barber et al., 1996; Chavez et al., 1999; Ryan et al., 2002] either locally or remotely forced.

3.4. Seasonal Variability of Nutrients and Chlorophyll

[21] The eqPac CTD database and monthly SeaWiFS data were used to investigate seasonal variability in nutrients and chlorophyll, respectively. Monthly SeaWiFS chlorophyll data spanning 1997 to 2004 were used to generate the mean seasonal cycle depicted in Figure 9, including only months for which the MEI was between −1 and 1 (excluding strong El Niño or La Niña). There are three features of note in the time series: (1) the magnitude of seasonal variability decreases from east to west, (2) the mean chlorophyll concentration decreases from east to west, and (3) the central region behaves very similarly to the basin as a whole. The standard deviation of each data point ranges from 3 to 40% of the mean values plotted (average 11%). Figure 10 shows all of the nutrient data for the central region (for which we have best temporal coverage). The mean value for each month is plotted as open circles. For nitrate and to a lesser extent silicic acid, the data show the same weak double peak (March and September) as the SeaWiFS chlorophyll time series. However, the relative magnitude of the peaks is different for nutrients (slightly larger peak in the first half of the year) and chlorophyll (larger peak in September).

image

Figure 9. Seasonal variability of SeaWiFS chlorophyll for the eastern, central, and western Pacific, as well as the entire basin. Region definitions are described in the text and Figure 2.

Download figure to PowerPoint

image

Figure 10. Seasonal variability of mixed layer nutrients for the central equatorial Pacific. All available data are plotted as closed circles. Monthly means (where data exist for that month) are plotted as open circles joined by the solid line.

Download figure to PowerPoint

4. Discussion and Conclusions

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Overview of the In Situ Data Set
  5. 3. Results
  6. 4. Discussion and Conclusions
  7. Acknowledgments
  8. References
  9. Supporting Information

[22] Perhaps the most general and striking result presented here is the nonmonotonic biological response across the La Niña to El Niño continuum. The expected result would be an almost linear decrease in chlorophyll from La Niña, through normal to El Niño conditions. Behrenfeld et al. [2006] did observe a linear relationship between chlorophyll and the MEI (reanalysis of their supplementary Figure 3, R2 = 0.60), but their analysis used vertically integrated chlorophyll for the permanently stratified ocean (>15°C, roughly 40°N to 40°S). Much of the La Niña signal they observed was likely due to the spatial expansion of the equatorial upwelling tongue as the trade winds strengthen. We have focused on local changes due to physical forcing and nutrient fluxes. We observed a decrease in chlorophyll associated with the transition from normal to strong El Niño conditions, but chlorophyll increased in response to La Niña only in the western Pacific. For all regions considered, there was no increase in bbp (phytoplankton carbon) for La Niña. For El Niño, bbp decreases in the east, does not change significantly in the central region, and (surprisingly) increases in the west. When larger size fractions of chlorophyll (as a measure of diatom biomass) were analyzed, they showed significant decreases with El Niño and no response with La Niña, except for the west where the diatom fraction decreased. Here we explain these results and compare them with previous observations and models.

[23] Synthesis of the JGOFS eqPac data [Landry et al., 1997] emphasizes the widely held view that the equatorial Pacific phytoplankton population can be represented by two general groups: a “background” population of picoplankton whose biomass and productivity varies little as a function of El Niño events, and a diatom community which becomes more dominant when iron limitation is relieved via enhanced upwelling. Strutton and Chavez [2000] quantified the changes in productivity and biomass associated with the >1-μm and >5-μm phytoplankton size fraction (assumed to be diatoms) during the transition from the intense El Niño of 1997–1998 to La Niña conditions of mid-to-late 1998. They observed an increase from 20% to 56% for the > 1-μm fraction's contribution to chlorophyll. The response of the >5-μm fraction was similar, as was the response of both size fractions with respect to their contribution to productivity. Closer examination of the Strutton and Chavez [2000] data shows that most of the increase, in fact all of the increase in the >5-μm fraction occurred in the transition from strong El Niño to weak El Niño conditions. This is confirmed by the data presented here: changes in total chlorophyll and phytoplankton types occurred between strong El Niño and “normal” conditions, but as the system transitions toward La Niña, there was no further increase in both biomass and the relative abundance of larger cells. The exception is the western Pacific.

[24] These observations are entirely consistent with modeling results [Dugdale et al., 2002] which show no change in the relative proportion of diatoms and picoplankton, and therefore no significant change in total phytoplankton biomass, above a mixed layer silicic acid concentration of about 3 mmol m−3 (our reanalysis of their results combined with our data on source and surface nutrient concentrations). Recent modeling results combined with size-fractionated nitrate uptake measurements [Dugdale et al., 2007] further support the idea that diatoms (picoplankton) reach their maximum (minimum) nitrate uptake at about 3 mmol m−3 silicic acid. Together, these results explain the lack of a La Niña response in chlorophyll, bbp and large size fraction chlorophyll, and are presented by Dugdale et al. [2007] as the mechanism governing the low-Si, high-nitrate, low-chlorophyll conditions of the equatorial upwelling zone (EUZ). The 1998 bloom is a possible exception. Figure 4 shows that the peak of this bloom was not well sampled in situ, but satellite data and a model suggest it may have contained a significant diatom population [Bopp et al., 2005].

[25] Results for the western Pacific were different, where La Niña was associated with a decrease in the large size fraction chlorophyll. This is inconsistent with model results (2002; Dugdale et al. [2007]) and the JGOFS synthesis [Landry et al., 1997] based on data from 140°W, indicating that the phytoplankton community in the western Pacific responds differently to that of the EUZ. The increase in chlorophyll for this region associated with La Niña could be caused by increases in the picophytoplankton community, as observed by Mackey et al. [2002], or a decrease in the C:chl ratio. The combined chlorophyll and bbp data certainly indicate a decrease in C:chl, but it does not appear to be a photoacclimation response to lower light. SeaWiFS data (not shown) indicate increasing insolation during La Niña due to the further westward migration of the warm pool's convection (cloud cover). As noted above, the region we refer to as the west in our analysis includes the date line and the area where the 29°C front between the upwelling tongue and the warm pool oscillates east-west during the transition between El Niño and La Niña. The increase in chlorophyll observed in our western box during La Niña is perhaps most simply explained by the expansion of the upwelling tongue (Figure 1). During the strongest La Niña conditions, silicic acid in the west reaches about 3 mmol m−3 (Figure 8), which is where we saw chlorophyll concentrations reach their maximum in the central and eastern Pacific. If the western Pacific phytoplankton community were behaving in a similar way to the central and east, and the models, the diatom contribution to biomass would also be increasing until silicic acid reached 3 mmol m−3, but the size fractionated chlorophyll data suggest otherwise. Closer examination of diatom nutrient uptake kinetics as a function of longitude might show important differences in the west of the upwelling tongue compared to the better studied region near 140°W.

[26] In contrast to the biological data, nutrients in general did increase across the continuum from El Niño to La Niña. In the central region, where we have the best data coverage spanning both La Niña and El Niño, the ratio of mixed layer NO3 to Si(OH)4 increases from<1.0 during strong El Niño to 1.4–1.6 across moderate El Niño to La Niña conditions. This is consistent with enhanced Si uptake by diatoms (relative to N) under iron limited conditions [Takeda, 1998], and indicates that the magnitude of iron stress is essentially unchanged as the MEI increases from −1 (La Niña) to 1 (moderate El Niño). Again, this range of unchanged N:Si corresponds to silicic acid concentrations greater than about 3 mmol m−3. Although we have no data on iron, it seems that for MEI less than about 1 (moderate El Niño through La Niña), the supply of Fe is decoupled from that of the macronutrients, likely because of their different sources (the importance of the EUC as a source for Fe).

[27] There is a distinct lack of variability in source nutrients (data not shown) which indicates that variability in mixed layer nutrients between El Niño and La Niña is governed by the delivery mechanisms to the surface (the physics) and consumption in the mixed layer (the biology). Tables 1 and 2 summarize our results regarding physical processes responsible for vertical nutrient fluxes. Previous work [e.g., Barber et al., 1996; Chavez et al., 1999; Ryan et al., 2002] has emphasized thermocline variability, forced either remotely or locally, as the dominant mechanism impacting the vertical flux of nutrients. Since the thermocline corresponds to the nutricline, it is assumed that shoaling of the thermocline should either bring nutrients into the euphotic zone where they contribute to photosynthesis, or facilitate enhanced vertical flux by being closer to the euphotic zone.

[28] Our simple and multiple linear regression analysis focused on timescales less than one month, and showed that winds and the strength of the EUC are as important as thermocline variability for driving mixed layer nutrient concentrations. When the analysis was broken down into regions, winds and EUC intensity were found to be the most significant drivers of nutrient flux in the central Pacific. In the eastern Pacific, where the thermocline is shallower than the rest of the basin, winds and thermocline variability dominated changes in nutrient flux, while in the west, winds almost exclusively contributed to mixed layer nutrient variability. We conclude that in the eastern Pacific the nutricline is shallow enough that ephemeral (<30 d) shoaling can be sufficient to increase nutrient fluxes to the surface and fuel phytoplankton growth. Elsewhere, a shoaling nutricline may be necessary but not sufficient to increase vertical nutrient fluxes. This is consistent with previous observations, most notably at the end of the 1997–1998 El Niño event, where the thermocline near 140°W was extremely shallow, but did not impact SST or surface productivity until upwelling favorable winds returned around April 1998 [Chavez et al., 1999; McPhaden, 1999].

[29] The impact of zonal winds may not be exclusively via increased upwelling, which would presumably induce thermocline variability through isotherm uplift. Our MLR analysis would then have likely identified thermocline variability as the proximate control on mixed layer nutrient variability. We suggest that increased wind speed might increase nutrient fluxes via enhanced vertical mixing, an aspect that as been neglected in previous discussions of equatorial upwelling. To be most effective, this wind-induced mixing would have to introduce both iron and macronutrients to the surface, by penetrating into the EUC. Recent and ongoing deployments of mixing sensors with bio-optics at 0° 140°W will enable us to quantify the relative importance of mixing versus thermocline variability to SST and productivity.

[30] Finally, we have attempted to quantify seasonal variability in chlorophyll and nutrients using satellite chlorophyll and the in situ data set. Throughout we have emphasized that the in situ data are perhaps best suited to quantifying interannual trends, as in Figures 4, 5, 6, and 8, rather than seasonal variability, because of the coarse ship sampling. Nonetheless, some trends were evident when the data were separated into regions and plotted as a function of year day. The satellite time series showed two modest seasonal increases in chlorophyll, peaking around March and August (larger peak in August), for the eastern and central Pacific (Figure 9). Mixed layer NO3 and Si(OH)4 also show peaks at around the same time, with the smaller peak in the second half of the calendar year (Figure 10). This is the time of year when tropical instability waves (TIWs) are dominant. Cosca et al. [2003] and Feely et al. [2006] noted that the relationship between SST and surface fCO2 changes during the period of TIW dominance, and attributed this difference to enhanced productivity, particularly in the east, during the second half of the year. This notion is supported by the satellite chlorophyll data (Figure 9) and the decreased nutrients during this period could, like CO2, be attributed to enhanced biological uptake. Whether or not the magnitude of the biological and nutrient changes is sufficient to drive the change in the fCO2/SST relationship will be the subject of a future manuscript.

[31] More broadly, the R2 values for many of the fCO2-SST relationships developed by Cosca et al. [2003] and Feely et al. [2006] indicate that the majority of the variability in fCO2 can be attributed to processes other than “the physics” (to the extent that SST represents physical processes such as upwelling). The correlation between SST and fCO2 is particularly weak during El Niño conditions, probably owing to weak and variable upwelling and productivity. Even during non El Niño periods, R2 for fCO2 versus SST ranges from 0.25 to 0.54, depending on the years and seasons considered. To fully understand and predict the variability in equatorial Pacific CO2 fluxes will require continued process studies such as those being synthesized by Dugdale et al. [2007] and others, to look in detail at the efficiency of the biological pump, and mooring-based time series of nutrients and bio-optics, to capture variability at subseasonal timescales.

Acknowledgments

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Overview of the In Situ Data Set
  5. 3. Results
  6. 4. Discussion and Conclusions
  7. Acknowledgments
  8. References
  9. Supporting Information

[32] We would like to thank the following people, mostly undergraduate students from Bloomsburg University, for their assistance in the lab at sea: Kory Angstadt, Dawn Bailey, Kimberly Baldwin, Peter Bernhardt, Mike Burczynski, Ryan Carr, Michael Diamond, Justin Derbes, Eric Drake, Raye Foster, Michael Fountain, Cindy Fraze, Marisel Gonzales-Bustos, Amy Harlan, Crystal Harlan, Mark Harlan, Katrina Hoffman, Adam Horovitz, Jennifer Hunt, Jason Kahn, Mindy Kelley, Jennifer Krapf, Amy MacFadyen, Nadia Meyers, Jerry Nettles, Aaron Norakus, Shawn O'Keefe, Jeff Perry, Tara Stoffel, Adam Thompson, Kate Treese, Chris Urie, Michael Weaver, and Adam Yauch. The volunteer program through which most of these data were collected was organized and partly funded by Cindy Venn, Bloomsburg University. We thank the officers and crew of the Ka'imimoana for their assistance with the CTD program and for their role in maintaining the TAO array. In particular we wish to thank Brian Powers, Dennis Sweeney, and Karen Taylor. Victor Kuwahara and Kevin Mahoney played important roles in organizing and executing the measurement program. Peter Walz and Marguerite Blum ran the nutrient samples, and Reiko Michisaki constructed the database. Mike Behrenfeld, Kirby Worthington, and colleagues took some of the chlorophyll and nutrient samples in the course of their research. Greg Johnson and Kristy McTaggart (NOAA/PMEL) processed the CTD data and made it freely available. Likewise, Eric Firing and Jules Hummon (University of Hawaii) processed and disseminated the ADCP data. The TAO project, Mike McPhaden, Director, provided the mooring data, and the SeaWiFS ocean color data were provided by the SeaWiFS Project, NASA/Goddard Space Flight Center, and ORBIMAGE. Funding for this work was provided by the NASA SIMBIOS program, NOAA's Office of Global Programs, and the David and Lucile Packard Foundation.

References

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Overview of the In Situ Data Set
  5. 3. Results
  6. 4. Discussion and Conclusions
  7. Acknowledgments
  8. References
  9. Supporting Information
  • Barber, R. T., M. P. Sanderson, S. T. Lindley, F. Chai, J. Newton, C. C. Trees, D. G. Foley, and F. P. Chavez (1996), Primary productivity and its regulation in the equatorial Pacific during and following the 1991–1992 El Niño, Deep Sea Res., Part II, 43, 933969.
  • Behrenfeld, M. J., E. Boss, D. A. Siegel, and D. M. Shea (2005), Carbon-based ocean productivity and phytoplankton physiology from space, Global Biogeochem. Cycles, 19, GB1006, doi:10.1029/2004GB002299.
  • Behrenfeld, M. J., R. T. O'Malley, D. A. Siegel, C. R. McClain, J. L. Sarmiento, G. C. Feldman, A. J. Milligan, P. G. Falkowski, R. M. Letelier, and E. S. Boss (2006), Climate-driven trends in contemporary ocean productivity, Nature, 444, 752755.
  • Bopp, L., O. Aumont, P. Cadule, S. Alvain, and M. Gehlen (2005), Response of diatoms distribution to global warming and potential implications: A global model study, Geophys. Res. Lett., 32, L19606, doi:10.1029/2005GL023653.
  • Chavez, F. P., and R. T. Barber (1987), An estimate of new production in the equatorial Pacific, Deep Sea Res., Part A, 34, 12291243.
  • Chavez, F. P., and J. R. Toggweiler (1995), Physical estimates of global new production: The upwelling contribution, in Upwelling in the Ocean: Modern Processes and Ancient Records, edited by C. P. Summerhayes et al., J. Wiley, Chichester, UK.
  • Chavez, F. P., K. R. Buck, R. R. Bidigare, D. M. Karl, D. Hebel, M. Latasa, M. E. Ondrusek, L. Campbell, and J. Newton (1995), On the chlorophyll a retention properties of glass-fiber GF/F filters, Limnol. Oceanogr., 40, 428433.
  • Chavez, F. P., P. G. Strutton, and M. J. McPhaden (1998), Biological-physical coupling in the central equatorial Pacific during the onset of the 1997–98 El Niño, Geophys. Res. Lett., 25, 35433546.
  • Chavez, F. P., P. G. Strutton, G. E. Friederich, R. A. Feely, G. C. Feldman, D. G. Foley, and M. J. McPhaden (1999), Biological and chemical response of the equatorial Pacific Ocean to the 1997–98 El Niño, Science, 286, 21262131.
  • Coale, K. H., S. E. Fitzwater, R. M. Gordon, K. S. Johnson, and R. T. Barber (1996a), Control of community growth and export production by upwelled iron in the equatorial Pacific Ocean, Nature, 379, 621624.
  • Coale, K. H., et al. (1996b), A massive phytoplankton bloom induced by an ecosystem-scale iron fertilization experiment in the equatorial Pacific Ocean, Nature, 383, 495501.
  • Cosca, C. E., R. A. Feely, J. Boutin, J. Etcheto, M. J. McPhaden, F. P. Chavez, and P. G. Strutton (2003), Seasonal and interannual CO2 fluxes for the central and eastern equatorial Pacific Ocean as determined from fCO2-SST relationships, J. Geophys. Res., 108(C8), 3278, doi:10.1029/2000JC000677.
  • Dugdale, R. C., R. T. Barber, F. Chai, T. H. Peng, and F. P. Wilkerson (2002), One dimensional ecosystem model of the equatorial Pacific upwelling system. part II: Sensitivity analysis and comparison with JGOFS EqPac data, Deep Sea Res., Part II, 49, 27472768.
  • Dugdale, R. C., F. P. Wilkerson, F. Chai, and R. A. Feely (2007), Size-fractionated nitrogen uptake measurements in the equatorial Pacific and confirmation of the low Si–high-nitrate low-chlorophyll condition, Global Biogeochem. Cycles, 21, GB2005, doi:10.1029/2006GB002722.
  • Feely, R. A., R. H. Gammon, B. A. Taft, P. E. Pullen, L. S. Waterman, T. J. Conway, and J. F. Gendron (1987), Distribution of chemical tracers in the eastern equatorial Pacific during and after the 1982–83 El Niño-Southern oscillation event, J. Geophys. Res., 92, 65456558.
  • Feely, R. A., T. Takahashi, R. Wanninkhof, M. J. McPhaden, C. E. Cosca, S. C. Sutherland, and M.-E. Carr (2006), Decadal variability of the air-sea CO2 fluxes in the equatorial Pacific Ocean, J. Geophys. Res., 111, C08S90, doi:10.1029/2005JC003129.
  • Firing, E., R. Lukas, J. Sadler, and K. Wyrtki (1983), Equatorial undercurrent disappears during 1982–83 El Niño, Science, 222, 11211123.
  • Holm-Hansen, O., C. J. Lorenzen, R. W. Holmes, and J. D. Strickland (1965), Fluorometric determination of chlorophyll, J. Cons. Cons. Int. Explor. Mer., 30, 315.
  • Landry, M. R., et al. (1997), Iron and grazing constraints on primary production in the central equatorial Pacific: An EqPac synthesis, Limnol. Oceanogr., 42, 405418.
  • Leynaert, A., P. Tréguer, C. Lancelot, and M. Rodier (2001), Silicon limitation of biogenic silica production in the equatorial Pacific, Deep Sea Res., Part I, 48, 639660.
  • Mackey, D. J., J. Blanchot, H. W. Higgins, and J. Neveaux (2002), Phytoplankton abundances and community structure in the equatorial Pacific, Deep Sea Res., Part II, 49, 25612582.
  • Maritorena, S., D. A. Siegel, and A. Peterson (2002), Optimization of a semi-analytical ocean color model for global scale applications, Appl. Opt., 41, 27052714.
  • Marland, G., T. A. Boden, and R. J. Andres (2000), Global, regional, and national fossil fuel CO2 emissions, in Trends: A Compendium of Data on Global Change, Carbon Dioxide Inf. Anal. Cent., Oak Ridge Natl. Lab., U.S. Dep. of Energy, Oak Ridge, Tenn.
  • McPhaden, M. J. (1999), Genesis and evolution of the 1997–98 El Niño, Science, 283, 950954.
  • McPhaden, M. J. (2004), Evolution of the 2002/03 El Niño, Bull. Am. Meteorol. Soc., 85, 677695.
  • Ryan, J. P., P. S. Polito, P. G. Strutton, and F. P. Chavez (2002), Unusual large-scale phytoplankton blooms in the equatorial Pacific, Progr. Oceanogr., 55, 263285.
  • Ryan, J. P., I. Ueki, Y. Chao, H. Zhang, P. S. Polito, and F. P. Chavez (2006), Western Pacific modulation of large phytoplankton blooms in the central and eastern equatorial Pacific, J. Geophys. Res., 111, G02013, doi:10.1029/2005JG000084.
  • Sakamoto, C., G. E. Friederich, and L. A. Codispoti (1990), MBARI procedures for automated nutrient analyses using a modified Alpkem series 300 rapid flow analyzer, MBARI Tech. Rep. 90-2, Monterey Bay Aquarium Res. Inst., Moss Landing, Calif.
  • Siegel, D. A., S. Maritorena, N. B. Nelson, and M. J. Behrenfeld (2005), Independence and interdependencies among global ocean color properties: Reassessing the bio-optical assumption, J. Geophys. Res., 110, C07011, doi:10.1029/2004JC002527.
  • Strutton, P. G., and F. P. Chavez (2000), Primary productivity in the equatorial Pacific during the 1997–98 El Niño, J. Geophys. Res., 105, 26,08926,101.
  • Strutton, P. G., J. P. Ryan, and F. P. Chavez (2001), Enhanced chlorophyll associated with tropical instability waves in the equatorial Pacific, Geophys. Res. Lett., 28, 20052008.
  • Takahashi, T., et al. (2002), Global sea-air CO2 flux based on climatological surface ocean pCO2, and seasonal biological and temperature effects, Deep Sea Res., Part II, 49, 16011622.
  • Takeda, S. (1998), Influence of iron availability on nutrient consumption ratio of diatoms in oceanic waters, Nature, 393, 774777.
  • Toggweiler, J. R., and S. Carson (1995), What are upwelling systems contributing to global nutrient and carbon budgets? in Upwelling the Ocean: Modern Processes and Ancient Records, edited by C. Summerhayes et al., John Wiley, Chichester, UK.
  • Wolter, K., and M. S. Timlin (1998), Measuring the strength of ENSO: How does 1997/98 rank? Weather, 53, 315324.
  • Wyrtki, K. (1981), An estimate of equatorial upwelling in the Pacific, J. Phys. Oceanogr., 11, 12051214.
  • Zar, J. H. (1999), Biostatistical Analysis, 4th ed., 664 pp., Prentice Hall, Upper Saddle River, N. J.

Supporting Information

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Overview of the In Situ Data Set
  5. 3. Results
  6. 4. Discussion and Conclusions
  7. Acknowledgments
  8. References
  9. Supporting Information
FilenameFormatSizeDescription
gbc1460-sup-0001-t01.txtplain text document1KTab-delimited Table 1.
gbc1460-sup-0002-t02.txtplain text document1KTab-delimited Table 2.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.