SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Methodology
  5. 3. Results
  6. 4. Discussion
  7. 5. Conclusion
  8. Acknowledgments
  9. References

[1] Processes regulating OMZs persistence in the oxygenated ocean remain poorly understood. Four cruises (21°–30°S) and fixed-point monitoring (36°S) between 2000 and 2002 using techniques adapted to O2 conditions as low as 1 μM allow a preliminary analysis of the entire Chilean OMZ structure. A shallow OMZ is observed in the three studied areas, although its structure differs. Off northern and central Chile, the OMZ is a permanent feature, more pronounced at the coast than further offshore. On the shelf, it forms in spring and erodes in fall. A conceptual model of two intermittent active or passive phases (intense or low biogeochemical O2 consumption) is proposed as a key mechanism for the local OMZ maintaining. The highest O2 consumptions are paradoxically favoured at the oxycline when the OMZ is less intense as offshore and on the shelf in spring and fall, suggesting a control by O2 availability of the OMZ remineralization.

1. Introduction

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Methodology
  5. 3. Results
  6. 4. Discussion
  7. 5. Conclusion
  8. Acknowledgments
  9. References

[2] Oxygen minimum zones (OMZs) are subsurface suboxic layers, several hundred meters deep, which in the generally oxygenated present-day ocean can potentially be considered as persistent traces of the primitive anoxic ocean. OMZs in the open ocean are found in the Eastern Tropical North Pacific (ETNP), in the Arabian Sea (AS) and off Peru [Anderson et al., 1982]. They have been studied mainly for their fundamental role in the global nitrogen cycle through the processes of denitrification and nitrification [Anderson et al., 1982], and the Anammox reactions [Thamdrup et al., 2006]. However, the oxygen distribution in OMZs is poorly characterized, and the root causes of their existence and persistence remain an open question.

[3] The Eastern South Pacific (ESP) OMZ, one of the most extended OMZs (cf map of hypoxia [Kamykowski and Zentara, 1990; Helly and Levin, 2004]), is today a permanent feature covering the areas offshore Equator, Peru and Chile (Figure 1). Little is known about the structure and variability of the southernmost component, the Chilean OMZ. Morales et al. [1999] studied the OMZ off northern Chile focusing on the variability of the 40 μM O2 isoline; Escribano et al. [2004] concentrated their study on waters presenting O2 contents higher than 40 μM. Both studies dealt only with the upper ocean (0–100 m) only. The entire Chilean OMZ thickness has not yet been documented. O2 concentrations can go lower than 40 μM, reaching O2<10 μM as in Peru, and can extend down to 400 m depth [Anderson et al., 1982]. The Chilean OMZ exhibits variability on the interannual scale: a deepening (50–100 m) of the 40 μM O2 isoline has been observed during El Niño periods, associated with coastally-trapped Kelvin waves and a reduction in upwelling activity [Morales et al., 1999; Ulloa et al., 2001]. On the seasonal scale, only one study reports a marked seasonal variability in O2 distribution, on the continental shelf (36°S, Concepción Bay), with low O2 of 10 μM in summer and one order of magnitude higher in fall-winter [Graco et al., 2001; Graco, 2002; Graco et al., 2006]. The possible dynamical and biogeochemical mechanisms contributing to the maintaining of the OMZ remain poorly understood. The existence of the ESP OMZ regionally depends on advection of low-O2 (<40–80 μM) waters from the equator by the Peru-Chile Undercurrent (PCU) [Strub et al., 1998]. The role of the intense productivity (9 gC/m2/d [Daneri et al., 2000]), 20 times higher than the oceanic average, leading to a strong subsurface organic matter degradation, may drive an intense O2 consumption [Reid, 1965] and be an important contributor to the OMZ maintaining. Our objective is to analyze the biogeochemical role of O2 consumption on the maintaining of the local Chilean OMZ. To do this, the OMZ changes in intensity, depth and thickness will be determined in coastal and open areas a priori subject to different productivities and dynamical regimes.

image

Figure 1. (a) O2 distribution at 250 m depth (1992–95 WOCE data, dotted lines). Sampling sites and stations depths: (b) 21°S at Iquique (80–1880 m); (c) 30°S at Coquimbo (700–4000 m); (d) 36°S at Concepción (down to 60 m).

Download figure to PowerPoint

2. Methodology

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Methodology
  5. 3. Results
  6. 4. Discussion
  7. 5. Conclusion
  8. Acknowledgments
  9. References

[4] To capture the strong O2 gradients (variations of 200 μM over 20–30 m) and to cover the entire OMZ, a high vertical resolution sampling, every 5–10 m between 0 and 110 m and every 20–100 m below 110 m, has been carried out. The OMZ off Chile was sampled during different non El Niño periods between 2000 and 2002, and at 3 sites (in the North: Iquique, 21°S; in the Center: Coquimbo, 30°S; on the shelf, Concepción, 36°S). Four cruises off the continental shelf at 21°S and 30°S (Figures 1b and 1c) allowed a description of different biogeochemical and dynamical regimes: e.g. a higher primary production (between 0.5–20 mgC/m2/d [Daneri et al., 2000]) and a deeper PCU [Strub et al., 1998] in the Center than in the North. A monthly monitoring at a fixed station at 36°S outside of an enclosed bay (Figure 1d) for nearly one year from September 2000 to May 2001 allowed a documentation of the seasonal variability.

[5] The samples were analyzed according to the Winckler method, improved and adapted to low concentration detection (O2<40 μM) by the PROFC following Broenckow and Cline [1969]. Avoiding any interference with NO2 by NaN3 addition, the precision over the whole set (154) of samples ranges between 0.5 and 1 μM [Paulmier, 2005]. The obtained reproducibility over 77 duplicates is 1.8 μM, in agreement with the WOCE Operation Manual [World Ocean Circulation Experiment, 1994] recommendations.

[6] To obtain estimates of O2 consumption contributions within the OMZ, we performed a water mass analysis based on an inversion method [Maamaatuaiahutapu et al., 1992] to determine mixing proportions between the source water types present in the area between 0 and 1000 m (STW-Surface Tropical, SAW-Surface Antarctic, ESSW-Equatorial Subsurface, AIW-Antarctic Intermediate Waters). The measured O2 concentrations can be expressed as the sum of a contribution due to mixing processes called O2mixing, and a remaining contribution or residual called ΔO2 due only to biogeochemical processes. This hydrological model is usually applied regionally and the ΔO2 corresponds to the integrated biogeochemical effect during history of mixing between the theoretical source water types up to the considered hydrographic station. To compare the biogeochemical effect on O2 of the OMZ communities at each sampled profile, we chose to apply locally our method to infer a local biogeochemical residual within the OMZ and in particular at the oxycline level. This implies a minor effect of the local horizontal mixing and advection which does not significantly affect the vertical O2 mixing [Paulmier, 2005]. The biogeochemical residuals can be interpreted as the in situ O2 consumption or production of the OMZ communities for each location, depth and period.

3. Results

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Methodology
  5. 3. Results
  6. 4. Discussion
  7. 5. Conclusion
  8. Acknowledgments
  9. References

[7] A new criterion to characterize and analyze the entire OMZ changes between 21°S and 30°S, from the shelf to offshore (21°S) and at the seasonal scale (36°S) is defined (Figure 2) from all O2 profiles. This criterion can be defined due to the high vertical resolution and the improved precision (40 times higher) over previous studies on the Chilean OMZ [Escribano et al., 2004]. The O2 data set collected off Chile shows an OMZ structure off the shelf: O2<20 μM starting at 140 m depth, with a thickness of ∼220 m at 21°S and 30°S (Figures 2a, 2b, and 2c).This structure has an upper O2 gradient (oxycline), a core and a lower O2 gradient. The oxycline, with an upper limit defined by the level where O2< 4% of O2 surface is intense (>1 μM/m: up to 10 μM/m at 21°S, Figure 2a), ∼10 times more than a classical O2 minimum outside of the OMZ. The oxycline exhibits a ∼200 μM O2 decrease over tens of meters, i.e., an O2 continuum from oxic (>200 μM) to suboxic (<20 μM) conditions associated with a strong stratification (0.5 kg/m3 increase over ∼40 m between 26 and 26.5 kg/m3 isopycnals, Figure 2c). The oxycline is close to the sea surface (5–200 m), and can thus intercept the euphotic layer between 60 and 120 m (Figures 2a and 2b). The OMZ core is defined by a central part with O2<20 μM over a 100–320 m thickness. The O2 concentrations in the core are among the weakest found in the global ocean [Helly and Levin, 2004], reaching the detection limit (<1 μM at 21°S, Figure 2a). The OMZ core is always located below the thermocline (by more than 25 m: Figures 2a and 2b) and below the 26.5 kg/m3 isopycnal (by more than 5 m: Figures 2c and 2d). These physical characteristics tend to indicate that the upper boundary of the core is delimited by the intense isopycnal mixing with the surface oxygenated waters. The lower O2 gradient, an order of magnitude less intense (∼ +0.3 μM/m) than the oxycline, is delimited above by the core and below by the depth where the O2 slope changes significantly (slope break >0.1 μM/m). Consequently, the lower O2 gradient may extend from 300 to 800 m depth, and the total OMZ layer from 5 to 800 m.

image

Figure 2. (a, b) O2 profiles at 21°S and 30°S for four representative stations far and near the coast (S2-S7 and S5-S8). Th = thermocline depth. O2 vertical sections, (c) along an East-West gradient and (d) versus time. Points indicate sampling locations. October (O) and December (D) correspond to year 2001.

Download figure to PowerPoint

[8] This OMZ structure, observed at the two sampled sites at 21°S and 30°S (Figures 2a and 2b), corresponds to a permanent structure off northern and central Chile. The core at 30°S (Figure 2b) is located below 100 m (S8) and 300 m (S5), i.e., 75 and 250 m deeper than at 21°S in S7 and S2 (Figure 2a). This observation confirms the southward deepening of the 40 μM O2 isoline between 18° and 24°S [Morales et al., 1999], ∼500 km more in the North. The OMZ is between 50 and 200 m thinner at 30°S than at 21°S. The core minima (10–20 μM) and the oxycline (∼1–1.25 μM/m) at 30°S are 2–3 and 3–4 times less intense than at 21°S, respectively. At 21°S, the core (white, Figure 2c) is truncated by the continental shelf without change in its intensity. The continental slope induces a 30–40 m core shoaling following the 20 μM O2 isoline. This shoaling is not constant, sharp (40 m) between S1 and S3 and slow (10 m) between S3 and S7. The reduction towards the coast of the vertical distance between the 200 μM and 20 μM O2 isolines, the upper and lower oxycline limits, yields an oxycline intensification of 40% corresponding to a higher stratification.

[9] At 36°S on the shelf, the OMZ becomes established in austral spring-summer (October to March) with O2 concentrations lower than 20 μM from 20 to 50 m of depth (Figure 2d). The OMZ starts to settle in spring during the high upwelling activity period [Ahumada et al., 1983]. In fall (April-May), the 180 μM O2 isoline extends from the surface to the bottom, indicative of a homogenized and reoxygenated water column with a complete OMZ destruction, confirming previous observations in the same area [Graco, 2002].

[10] Off northern and central Chile, the OMZ is shallower and more intense at 21°S and the coast than at 30°S and offshore. On the shelf at 36°S, the setting is distinct and a seasonal suboxia develops: the OMZ forms in spring, gets established in summer, and is destroyed at the end of fall. Despite these modulations, the OMZ structure has been observed in all the three studied areas off the coast of Chile.

4. Discussion

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Methodology
  5. 3. Results
  6. 4. Discussion
  7. 5. Conclusion
  8. Acknowledgments
  9. References

[11] After having characterized the vertical extension of the Chilean OMZ, we examine the possible causes of its maintaining. In a “classical” O2 minimum, typical of all intermediate waters of the world ocean [Wyrtki, 1962], both biogeochemical and dynamical processes contribute. Considering the relatively small area documented by our data (<2% of the ESP surface, Figure 1), we will focus here on the local O2 contributions. Results of the hydrological method applied locally to each profile provide the ΔO2 = O2–O2mixing (Figure 3) residuals associated with the biogeochemical contributions. In all profiles, the computed ΔO2 are negative, indicating a O2 consumption. The comparison between O2 and O2mixing profiles suggests that the core and oxycline would be 10–20 μM lower and 3 times less intense without this O2 consumption, respectively.

image

Figure 3. O2, O2mixing and ΔO2 profiles evaluated from the hydrological method for an illustrative station (S2). Points represent the ΔO2 values for samples of all cruises. Errors on O2mixing and ΔO2 are < 2 and 4 μM, respectively [Paulmier, 2005].

Download figure to PowerPoint

[12] This consumption does not take place uniformly over the OMZ depth, but has peaks (∣ΔO2max) in the oxycline on average three times more intense than in the OMZ core, with a biogeochemical contribution which can reach 100 μM (Figure 3). Because of this strong O2 consumption and the potential ventilation effect from the oxygenated surface layers, the oxycline can be regarded as the local engine maintaining the OMZ. On the cross-shore transect and time-series, the peaks ∣ΔO2max offshore, in October and April (between 22 and 101 μM) are 5 and 9 times higher than at the coast and in summer, respectively (Figure 4). To analyze the role of these ∣ΔO2max variations on the OMZ changes, we propose a conceptual model with two phases, passive and active. A passive (active) phase corresponds to an OMZ whose ∣ΔO2max is lower (larger) than 20 μM at the oxycline. This threshold of 20 μM corresponds to the O2 consumption associated with the most intense remineralization estimated in the oxygenated ocean [Rivkin and Legendre, 2001], calculated for a mean annual primary production off Chile of 2.6 gC/m2/d [Daneri et al., 2000]. The OMZ is active offshore (Figure 4a) where the oxycline associated with a lower stratification is less intense than at the coast (Figure 2c). On the shelf, the OMZ is active in spring and fall (Figure 4b) when the OMZ is formed during the upwelling period and destroyed at the beginning of the strong winter overturning (Figure 2d). To understand the regional and temporal OMZ differences, the alternation of active and passive phases is not sufficient. The spatial and seasonal changes in mixing, rather than in the O2 consumption at the oxycline, contribute to the intensification, establishment and destruction of the OMZ. The OMZ in regions in the North and at the coast (S4-S7 at 21°S; S8 at 30°S) is under the stronger influence of the less-oxygenated STW, while the Center and offshore (S1-S3 at 21°S; S5 at 30°S) feel the stronger influence of the more-oxygenated SAW [Paulmier, 2005]. In addition, the PCU and upwelling advection of less-oxygenated waters (STW and ESSW, respectively) are higher in the North and at the coast than in the Center and offshore [Strub et al., 1998]. Consequently, the northern and coastal parts of the OMZ should be more intense and dynamically-maintained than the Centre and offshore parts of the OMZ. At 36°S on the shelf, the OMZ should also be more pre-formed and dynamically-maintained in spring and summer during upwelling favourable periods.

image

Figure 4. (a) Cross-shore and (b) seasonal variations, for maximal O2 consumptions (∣ΔO2max) and O2 at ∣ΔO2max depth in the oxycline, mean surface chlorophyll from SeaWiFS climatology and stratification intensity. O2 ∣ΔO2max in May (225 μM) is not reported because of the complete OMZ destruction. Biogeochemical OMZ activity is defined by ∣ΔO2max > 20 μM (grey background).

Download figure to PowerPoint

[13] Why is this local O2 consumption at the oxycline both intermittent and up to 5 times more intense than in the oxygenated ocean? On one hand, O2 consumption is classically related to biomass. However here, the OMZ activity (Figure 4) is associated with O2 availability (oxycline less intense offshore; OMZ not yet formed in spring or eroded in fall: Figures 2c and 2d). During the active phases, O2 is provided by mixing of oxygenated surface layers (O2>30–40 μM). The sea surface biomass is then 3 times lower than at the coast and in summer (passive conditions: Figure 4). The high O2 consumption would agree with a non-local “productivity-driven” denitrification in the OMZ on a paleoceanographic scale [De Pol-Holz et al., 2006], and rather suggests an “oxygen-driven” activity of both remineralization and denitrification in the OMZ. Only a small occurrence of an active phase (35% of the sampling: Figure 4) - which by high O2 consumption would induce a shift to a passive phase (a less-oxygenated oxycline) - should be enough to maintain the OMZ. Then the changes in productivity level and distribution do not directly affect the changes in O2 consumption, and thus in the local OMZ maintaining. On the other hand, the oxycline has a well-lit O2 continuum going from oxic (>200 μM near the surface) to suboxic (<20 μM near the core) concentrations with sufficient high stratification (Figure 2) and shifts between the passive (suboxic) and active (oxic) phases. These unique conditions allow a potential co-existence of anaerobic/aerobic (e.g. denitrification/nitrification) and photic/aphotic (e.g. phototrophy/heterotrophy) processes. The high O2 consumption found at the oxycline is in agreement with previous experimental work, showing that remineralization with alternation of oxic and anoxic conditions could be ∼10 times more efficient than in oxic conditions only [Sun et al., 2002].

[14] Despite the fact that oxycline O2 consumption is vital in maintaining such an intense O2 minimum, another essential factor in the ESP OMZ formation is the O2 deficit pre-formation by the PCU. Because the PCU is confined to within 40 km of the coast [Strub et al., 1998], while the OMZ extends 2000 km further offshore (Figure 1), we can here infer a rough estimate of the PCU influence on the OMZ maintaining off Chile only for regions close to the coast. Consider a typical O2 profile from our data set, with concentrations lower than 20 μM in the core. This Chilean OMZ characteristic profile is compared with a historical O2 profile, from the South Eastern Pacific zone outside the OMZ, with O2 = 100 ± 5 μM at the OMZ core depth of between 200 and 400 m (B, Figure 1). We make the assumption that O2 concentrations of water masses advected by the PCU correspond to those upstream close to the PCU formation zone, i.e. 40 ± 20 μM (A, Figure 1). The mean PCU contribution is then 100 ± 5 − 40 ± 20 = 60 ± 25 μM. Compared to the total OMZ O2 deficit (100 − 20 = 80 μM), the PCU then accounts for ∼75% of the OMZ. The remaining contribution of 25% must be due to the local O2 consumption through the intense remineralization discussed above. Thus the local biogeochemical effect is significant in understanding the existence of the Chilean OMZ, complementary to the important regional role of the PCU, which may also control the OMZ vertical position and thickness.

5. Conclusion

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Methodology
  5. 3. Results
  6. 4. Discussion
  7. 5. Conclusion
  8. Acknowledgments
  9. References

[15] The Chilean OMZ structure, between 5 and 800 m depth, is maintained everywhere off the shelf with an intense oxycline (>1 μM/m) and minimum concentrations reaching O2<1 μM in the OMZ core. The maintaining of the OMZ requires a local high remineralization at the oxycline (“activity”), 3 times more intense than in the OMZ core. This activity, associated with the remineralization process, leads us to consider the oxycline as the OMZ ”engine“. Without this remineralization of up to 5 times higher than in the oxygenated areas, a ”classical“ O2 minimum would form rather than an OMZ. The OMZ activity requires a minimal oxygenation (O2>30–40 μM) which can occur only in the oxycline. Consequently, the remineralization is thought to be more sensitive to O2 availability than to surface biomass, suggesting an ”oxygen-driven“ remineralization. This strong OMZ remineralization should be further explored, considering the oxycline oxic/suboxic/well-lit specificity which allows a possible co-existence of aerobic/anaerobic and photic/aphotic processes, usually occurring at different depths. Despite its persistence off northern and central Chile, the OMZ is shallower and stronger at the coast and 21°S than offshore and at 30°S. Off southern Chile, on the shelf, a seasonal suboxia develops: it is formed in austral spring by upwelling and destroyed at the end of fall by winter overturning. The highest O2 consumption occurs preferentially offshore, in spring and fall, during periods of weak stratification (weaker oxycline), and then shifts to passivity (lower O2 consumption). The OMZ maintaining cannot be due only to the activity. Another prerequisite, not detailed here, is regional dynamical transport by the PCU of O2−depleted waters; this dynamically pre-formed OMZ gets intensified by the local biogeochemical O2 consumption.

[16] Both climatic and environmental anthropogenic perturbations in the coming decades will likely drive OMZ changes. It is therefore urgent to characterize the structure of all OMZs, evaluate their extension and analyze the details of their chemical and ecosystem nature and dynamics. The adapted sampling and ultra low-O2 technique used to capture the Chilean OMZ structure could support future OMZs study strategies.

Acknowledgments

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Methodology
  5. 3. Results
  6. 4. Discussion
  7. 5. Conclusion
  8. Acknowledgments
  9. References

[17] This study was supported by a CNRS French Ph.D. fellowship to A. P., by the Chilean FONDAP project, and ECOS from the Ministry of Foreign Affairs. We thank C. Riviera for the O2 analysis, the R/V Vidal Gormaz, Carlos Porter and Kay-Kay crews, C. Provost from LOCEAN for encouragement, O. Ulloa from PROFC for critical reading of an early version of this manuscript and A. Fischer for correcting English.

References

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Methodology
  5. 3. Results
  6. 4. Discussion
  7. 5. Conclusion
  8. Acknowledgments
  9. References
  • Ahumada, R., A. Rudolph, and V. Martinez (1983), Circulation and fertility of waters in Concepción Bay, Estuarine Coastal Shelf Sci., 16, 95105.
  • Anderson, J. J., A. Okubo, A. Robbins, and A. Richards (1982), A model for nitrite and nitrate distributions in oceanic oxygen minimum zones, Deep Sea Res., 29, 11131140.
  • Broenckow, W. W., and J. D. Cline (1969), Colorimetric determination of dissolved oxygen at low concentrations, Limnol. Oceanogr., 14, 450454.
  • Daneri, G., V. Dellarosa, R. Quiñones, B. Jacob, P. Montero, and O. Ulloa (2000), Primary production and community respiration in the Humboldt current system off Chile and associated areas, Mar. Ecol. Prog. Ser., 197, 4149.
  • De Pol-Holz, R., O. Ulloa, L. Dezileau, J. Kaiser, F. Lamy, and D. Hebbeln (2006), Melting of the Patagonian Ice Sheet and deglacial perturbations of the nitrogen cycle in the eastern South Pacific, Geophys. Res. Lett., 33, L04704, doi:10.1029/2005GL024477.
  • Escribano, R., et al. (2004), Biological and chemical consequences of the 1997–1998 El Niño in the Chilean coastal upwelling system: A synthesis, Deep Sea Res., Part II, 51, 23892411.
  • Graco, M. (2002), Variación temporal de la producción de amonio en un area de surgencias de Chile Central (36°S): Factores y procesos involucrados en su reciclaje, Ph.D. thesis, 223 pp., Univ. Pierre et Marie Curie, Paris.
  • Graco, M., L. Farías, V. Molina, D. Gutiérrez, and L. P. Nielsen (2001), Massive development of microbial mats following phytoplankton blooms in an eutrophicated bay: Implications on nitrogen cycling, Limnol. Oceanogr., 46, 821832.
  • Graco, M., L. Farías, and D. Gutiérrez (2006), Inter-annual variability of the Pelagic-Benthic coupling in the upwelling system off central Chile, Adv. Geosci., 6, 127132.
  • Helly, J. J., and L. A. Levin (2004), Global distribution of naturally occurring marine hypoxia on continental margins, Deep Sea Res., Part I, 51, 11591168.
  • Kamykowski, D., and S. Zentara (1990), Hypoxia in the world ocean as recorded in the historical data set, Deep Sea Res., 37, 18611874.
  • Maamaatuaiahutapu, K., V. Garçon, C. Provost, M. Boulahdid, and A. P. Osiroff (1992), Brazil-Malvinas confluence: Water mass composition, J. Geophys. Res., 97, 94939505.
  • Morales, C., S. Hormazabal, and J. L. Blanco (1999), Interannual variability in the mesoscale distribution of the depth of the upper boundary of the oxygen minimum layer off northern Chile (18°–24°S): Implications for the pelagic system and biogeochemical cycling, J. Mar. Res., 57, 909932.
  • Paulmier, A. (2005), Zones de Minimum d'Oxygène (OMZs) de l'océan moderne: Une étude focalisée dans le Pacifique Sud Est, Ph.D thesis, 248 pp., Univ. Pierre et Marie Curie, Paris.
  • Reid, J. (1965), Intermediate Waters of the Pacific Ocean, 85 pp., Johns Hopkins Press, Baltimore, Md.
  • Rivkin, R., and L. Legendre (2001), Biogenic carbon cycling in the upper ocean: Effects of microbial respiration, Science, 291, 23982400.
  • Strub, P.. J. Mesías, V. Montecino, J. Rutllant, and S. Salinas (1998), Coastal ocean circulation off Western South America, in The Sea, vol. 11, Regional Studies and Synthesis, edited by A. R. Robinson, and K. H. Brink, pp. 273313, John Wiley, Hoboken, N. J.
  • Sun, M., R. Aller, C. Lee, and S. Wakeham (2002), Effects of oxygen and redox oscillation on degradation of cell-associated lipids in superficial marine sediments, Geochem. Cosmochim. Acta, 66, 20032012.
  • Thamdrup, B., T. Dalsgaard, M. M. Jensen, O. Ulloa, L. Farías, and R. Escribano (2006), Anaerobic ammonium oxidation in the oxygen-deficient waters off northern Chile, Limnol. Oceanogr., 55, 21452156.
  • Ulloa, O., R. Escribano, S. Hormazabal, R. Quiñones, R. Gonzalez, and M. Ramos (2001), Evolution and biological effects of the 1997–98 El Niño in the upwelling ecosystem off northern Chile, Geophys. Res. Lett., 28, 15911594.
  • World Ocean Circulation Experiment (1994), WOCE Operation Manual, vol. 3, The Observational Programme: Part 3.1.3—WHP Operations and Methods, Rep. 68/91, Geneva.
  • Wyrtki, K. (1962), The oxygen minima in relation to ocean circulation, Deep Sea Res., 9, 1123.