• Carton, J. A., 1991: Effect of seasonal surface freshwater flux on sea surface temperature in the tropical Atlantic Ocean. J. Geophys. Res., 96 , 1259312598.

    • Search Google Scholar
    • Export Citation
  • Gill, A. E., 1982: Atmosphere–Ocean Dynamics. Academic Press, 662 pp.

  • Huang, B., , and V. M. Mehta, 2004: The response of the Indo-Pacific warm pool to interannual variations in net atmospheric freshwater. J. Geophys. Res., 109 , C06022. doi:10.1029/2003JC002114.

    • Search Google Scholar
    • Export Citation
  • Huang, B., , and V. M. Mehta, 2005: Response of the Pacific and Atlantic Oceans to interannual variations in net atmospheric freshwater. J. Geophys. Res., 110 , C08008. doi:10.1029/2004JC002830.

    • Search Google Scholar
    • Export Citation
  • Huang, B., , V. M. Mehta, , and N. S. Schneider, 2005: Oceanic response to idealized net atmospheric freshwater in the Pacific at decadal timescales. J. Phys. Oceanogr., 35 , 24672486.

    • Search Google Scholar
    • Export Citation
  • Jacob, R. L., 1997: Low frequency variability in a simulated atmosphere ocean system. Ph.D. thesis, University of Wisconsin—Madison, 155 pp.

  • Liu, Z., , J. Kutzbach, , and L. Wu, 2000: Modeling climate shift of El Niño variability in the Holocene. Geophys. Res. Lett., 27 , 22652268.

    • Search Google Scholar
    • Export Citation
  • Lukas, R., , and E. Lindstrom, 1991: The mixed layer of the western equatorial Pacific Ocean. J. Geophys. Res., 96 , 33433357.

  • Murtugudde, R., , and A. J. Busalacchi, 1998: Salinity effects in a tropical ocean model. J. Geophys. Res., 103 , 32833300.

  • Reason, C., 1992: On the effect of ENSO precipitation anomalies in a global ocean GCM. Climate Dyn., 8 , 3947.

  • Wu, L., , and Z. Liu, 2006: Decadal variability in the North Pacific: The eastern North Pacific mode. J. Climate, 16 , 31113131.

  • Wu, L., , Z. Liu, , R. Gallimore, , R. Jacob, , Y. Zhong, , and D. Lee, 2003: Pacific decadal variability: The tropical Pacific mode and the North Pacific mode. J. Climate, 16 , 11011120.

    • Search Google Scholar
    • Export Citation
  • Yang, S., , K-M. Lau, , and P. S. Schopf, 1999: Sensitivity of the tropical Pacific ocean to precipitation-induced freshwater flux. Climate Dyn., 15 , 737750.

    • Search Google Scholar
    • Export Citation
  • Yu, L., , and R. A. Weller, 2007: Objectively Analyzed Air–Sea Heat Fluxes (OAFlux) for the global ice-free oceans (1981–2005). Bull. Amer. Meteor. Soc., 88 , 527539.

    • Search Google Scholar
    • Export Citation
  • Zhang, R-H., , and A. J. Busalacchi, 2009: Freshwater flux (FWF)-induced oceanic feedback in a hybrid coupled model of the tropical Pacific. J. Climate, 22 , 853879.

    • Search Google Scholar
    • Export Citation
  • View in gallery

    Salinity and temperature anomalies in response to an idealized negative freshwater forcing in the western tropical Pacific. (a) SSS (contours in dark-shaded area at intervals of 0.2 psu), (b) SST (contours in dark-shaded area at intervals of 0.1°C), and surface wind (vectors, m s−1). Longitude–depth distribution of (c) salinity (contours at intervals of 0.1 psu) and (d) temperature averaged between 5°S and 5°N (contours in dark-shaded area at intervals of 0.1°C). The dark line in (d) denotes the 20°C isotherm in the CTRL. Area shaded exceeds the 95% statistical significance using a Student’s t test.

  • View in gallery

    Annual mean (a) net surface heat flux (positive downward, contours in dark-shaded area at intervals of 2 W m−2), (b) evaporation (contours in dark-shaded area at intervals of 0.05 mm day−1), (c) precipitation, and (d) PmE anomaly (contours in dark-shaded area at intervals of 0.1 mm day−1). Area shaded with dark gray exceeds 90% statistical significance using a Student’s t test. Area shaded with light gray represents positive anomaly area.

  • View in gallery

    Annual mean buoyancy flux anomaly (contours in dark-shaded area at intervals of 1 × 10−9 m2 s−3). Area shaded with dark gray exceeds 90% statistical significance using a Student’s t test, and area shaded with light gray represents the positive anomaly area.

  • View in gallery

    Seasonality of SST (contours in shaded area at intervals of 0.1°C) and surface wind (vectors, m s−1) responses in the Pacific. (a) Longitude–time distribution averaged over the equatorial band (5°S, 5°N). Maps for (b) June–August and (c) December–February. Area shaded exceeds 90% statistical significance using a Student’s t test.

  • View in gallery

    Heat budget analyses for the (left) surface and (right) subsurface in the (top) western and (middle) eastern equatorial Pacific. See the regional boxes in Fig. 1d. The unit for each budget term is 1 × 107 W m−2. (bottom) Latitude–depth distribution of (e) meridional velocity anomalies in the western Pacific averaged from 120° to 140°E, and (f) zonal and vertical velocity averaged between 5°S and 5°N. Unit for velocity: m s−1. The magnitude of horizontal and vertical velocity are multiplied 1 × 102 and 2 × 106, respectively.

  • View in gallery

    (a) Annual mean sea surface height anomaly (contours in shaded area at intervals of 5 mm), (b) seasonal evolution of sea surface height anomaly along the equator averaged over the equatorial band (5°S, 5°N; contours at intervals of 3 mm) induced by the imposed freshwater flux. Area shaded exceeds 90% statistical significance using a Student’s t test.

  • View in gallery

    Oceanic changes in the FWIND experiment where the wind stress in the tropics is fixed to the model climatology. Longitude–time distribution of (a) SST (contours at intervals of 0.05°C), (b) temperature averaged over the equatorial band (5°S, 5°N; contours in shaded area at intervals of 0.1°C), (c) meridional velocity anomalies in the western Pacific averaged from 120° to 160°E (contours at intervals of 0.2 × 10−2 m s−1), and (d) zonal and vertical velocity averaged between 5°S and 5°N. Unit for velocity: m s−1. The magnitude of horizontal and vertical velocity multiplied 1 × 102 and 3 × 106, respectively. Area shaded exceeds 90% statistical significance using a Student’s t test.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 10 10 2
PDF Downloads 1 1 0

Coupled Ocean–Atmosphere Response to Idealized Freshwater Forcing over the Western Tropical Pacific

View More View Less
  • 1 Physical Oceanography Laboratory, Ocean University of China, Qingdao, China
  • | 2 Department of Natural History Science, Graduate School of Science, Hokkaido University, Hokkaido, Japan
  • | 3 Physical Oceanography Laboratory, Ocean University of China, Qingdao, China
  • | 4 Department of Natural History Science, Graduate School of Science, Hokkaido University, Hokkaido, Japan
© Get Permissions
Full access

Abstract

The coupled ocean–atmosphere responses to idealized freshwater forcing in the western tropical Pacific are studied using a fully coupled climate model. The model explicitly demonstrates that freshwater forcing in the western tropical Pacific can lead to a basinwide response with the pattern resembling the Pacific decadal oscillation. In the tropics, a negative (positive) freshwater forcing over the western tropical Pacific decreases (increases) sea surface height locally, and sets up a positive (negative) zonal pressure gradient anomaly, which accelerates (decelerates) the meridional overturning circulation and equatorial surface westward flow. This leads to an intensification (reduction) of meridional heat divergence and vertical cold advection, and thus a development of La Niña (El Niño)–like responses in the tropics. The tropical responses are further substantiated by the positive Bjerknes feedback, and subsequently force significant changes in the extratropical North Pacific through atmospheric teleconnection. The local freshwater response also reinforces the imposed forcing, forming a positive feedback loop. Applications to Pacific climate changes are discussed.

Corresponding author address: Dr. Lixin Wu, 5 Yushan Road, Physical Oceanography Laboratory, Ocean University of China, Qingdao 266003, China. Email: lxwu@ouc.edu.cn

Abstract

The coupled ocean–atmosphere responses to idealized freshwater forcing in the western tropical Pacific are studied using a fully coupled climate model. The model explicitly demonstrates that freshwater forcing in the western tropical Pacific can lead to a basinwide response with the pattern resembling the Pacific decadal oscillation. In the tropics, a negative (positive) freshwater forcing over the western tropical Pacific decreases (increases) sea surface height locally, and sets up a positive (negative) zonal pressure gradient anomaly, which accelerates (decelerates) the meridional overturning circulation and equatorial surface westward flow. This leads to an intensification (reduction) of meridional heat divergence and vertical cold advection, and thus a development of La Niña (El Niño)–like responses in the tropics. The tropical responses are further substantiated by the positive Bjerknes feedback, and subsequently force significant changes in the extratropical North Pacific through atmospheric teleconnection. The local freshwater response also reinforces the imposed forcing, forming a positive feedback loop. Applications to Pacific climate changes are discussed.

Corresponding author address: Dr. Lixin Wu, 5 Yushan Road, Physical Oceanography Laboratory, Ocean University of China, Qingdao 266003, China. Email: lxwu@ouc.edu.cn

1. Introduction

The western Pacific warm pool plays a critical role in both tropical and global climate by contributing a large amount of diabatic heating to the tropical atmosphere. The large precipitation associated with deep convection imposes a significant amount of buoyancy flux to the warm pool, which can be as important as net surface heat flux in affecting ocean dynamics and thermodynamics (e.g., Lukas and Lindstrom 1991; Huang and Mehta 2005). Recent observations demonstrate a significant trend of freshwater flux change, defined as precipitation minus evaporation (PmE), over the Indo-Pacific warm pool region. For example, Huang and Mehta (2004) found that from 1989 to 2000 the PmE in the warm pool region is decreased by about one-third in magnitude from that in 1988.

Oceanic modeling studies have shown that the freshwater flux can cause changes in ocean salinity, current, and temperature fields (e.g., Carton 1991; Reason 1992; Murtugudde and Busalacchi 1998; Yang et al. 1999; Huang et al. 2005). For example, Yang et al. (1999) showed that the SST in the warm pool region can be increased by about a half-degree when the climatological freshwater flux is ignored. Such a warming effect has been suggested to be associated with changes of vertical mixing and horizontal advection in response to freshwater flux (e.g., Yang et al. 1999; Huang and Mehta 2004). At interannual time scales, modeling studies also showed that the ocean response to El Niño– and La Niña–type precipitation is different (e.g., Reason 1992), and the freshwater flux anomalies associated with El Niño may sustain the warming in the eastern equatorial Pacific (e.g., Yang et al. 1999). With an OGCM coupled to an empirical statistical atmospheric model, Zhang and Busalacchi (2009) demonstrate a positive feedback between SST and freshwater flux in the tropical Pacific, supporting early studies of Yang et al. (1999).

The modeling studies above are helpful in understanding the mechanisms of freshwater flux–induced oceanic adjustment, but they are unable to explicitly demonstrate the coupled response, including freshwater responses. Furthermore, OGCM studies may underestimate the effects of the freshwater flux changes in the fully coupled ocean–atmosphere system. In this paper, we attempt to use a fully coupled GCM to understand the mechanisms of the coupled ocean–atmosphere responses to freshwater flux change in the western tropical Pacific. Our study demonstrates that a negative (positive) freshwater forcing over the western tropical Pacific decreases (increases) sea surface height locally and sets up a positive (negative) zonal pressure gradient anomaly, which accelerates (decelerates) the meridional overturning circulation and equatorial surface westward flow. This leads to an intensification (reduction) of meridional heat divergence and vertical cold advection, and thus a development of La Niña (El Niño)–like response in the tropics. The tropical responses are further substantiated by the positive Bjerknes feedback, and subsequently force significant changes in the extratropical North Pacific through atmospheric teleconnection.

The paper is constructed as follows: A brief description of model and experimental design is given in section 2. Sections 3 and 4 are devoted to discussions of the coupled ocean–atmosphere response and possible mechanisms. The paper is concluded with a summary and some further discussions.

2. Model and experiment design

We use the Fast Ocean Atmosphere Model (FOAM), version 1.5, a fully coupled global model developed at the University of Wisconsin. This is the improved version of the original FOAM (version 1.0), which is described in detail in Jacob (1997). The atmospheric model is a parallel version of the National Center for Atmospheric Research (NCAR) Community Climate Model, version 2 (CCM2) but the atmospheric physics is replaced with those of CCM, version 3 (CCM3). The ocean model was developed following the Geophysical Fluid Dynamics Laboratory (GFDL) Modular Ocean Model, version 3 (MOM3). The version used here has an atmospheric resolution of R15 with 18 vertical levels, and an oceanic resolution of 1.4° latitude × 2.8° longitude with 32 vertical levels. Without flux adjustment, the fully coupled control simulation has been integrated for over 2000 yr, without any apparent climate shift. The model captures major features of the observed climatology, as in most state-of-the-art climate models, and also produces reasonable climate variability, such as ENSO (Liu et al. 2000) and Pacific decadal climate variability (Wu et al. 2003). Also as in most state-of-the-art climate models, the model has a double-ITCZ problem, characterized by an unrealistic rainfall band south of the equator.

To elucidate the mechanisms of the coupled ocean–atmosphere response to the western tropical Pacific freshwater flux forcing, we impose an idealized PmE forcing over the region 10°S–10°N, 60°E–180° in the coupled model. The forcing has a sinusoidal distribution at both zonal and meridional directions and a maximum magnitude of about −3.456 mm day−1 in the center. This amounts to a total freshwater flux deficit of 0.49 Sv (1 Sv ≡ 106 m3 s−1), equivalent to about one-third of the observed climatological PmE. The amplitude of the imposed freshwater flux is roughly twice that during the 1997 El Niño year, and also about three times that of the increment of freshwater loss from 1981 to 2005 over the Indo-Pacific warm pool based on the Climate Prediction Center (CPC) Merged Analysis of Precipitation (CMAP) data [provided by the National Oceanic and Atmospheric Administration/Office of Oceanic and Atmospheric Research/Earth Systems Research Laboratory/Physical Sciences Division (NOAA/OAR/ESRL/PSD] and Woods Hole Oceanographic Institution (WHOI) evaporation data (Yu and Weller 2007). It should be noted that the forcing region in this study is somewhat larger than the ENSO-type freshwater perturbation in the western tropical Pacific. The sensitivity of the responses to different forcing magnitudes as well as forcing spatial scales is also tested, and the responses appear to be linear. The reason to select a stronger and simplified forcing in this modeling study is to increase the signal-to-noise ratio and better isolate the controlling physical mechanisms.

The fully coupled model has been integrated for 2000 yr to reach an equilibrium state. The PmE anomaly is then added at each time step and held constant for 30 yr. This experiment is named FORCE. A parallel experiment is conducted from the same equilibrium state but with no PmE anomaly added; this experiment is taken as the control run (CTRL). The difference of the last 20-yr average between the FORCE and CTRL runs is taken as the response. It should be noted that although the forcing also covers the eastern Indian Ocean warm pool, our analyses here will focus on the tropical Pacific responses.

3. Coupled ocean–atmosphere responses

A reduction of PmE in the warm pool leads to an increase of sea surface salinity (SSS) nearly over the entire western Pacific, with the maximum located off the equator (Fig. 1a). The meridional spreading of SSS anomaly is due to the advection of the oceanic gyre circulation. Along the equator, the salinification extends to a depth of about 300 m in the far western Pacific because of the effects of vertical mixing (Fig. 1c), although the penetration is somewhat deeper than that in some OGCM studies (e.g., Yang et al. 1999). While the SSS shows signals mainly in the western Pacific, the SST exhibits a basinwide response (Fig. 1b). In the tropics, the cooling dominates the western and central equatorial Pacific with a magnitude of −0.3°C. The cooling induces surface easterly anomalies (Fig. 1b), reflecting a Gill-type response in the lower troposphere. The tropical cooling is damped by the surface heat flux, suggesting the dominant role of oceanic dynamics in producing the cooling (Fig. 2a). More significant temperature changes can be seen in the equatorial thermocline (Fig. 1d). A large subsurface warming develops in the west, which has an opposite polarity to, and magnitudes three times stronger than, the SST change. A strong cooling is also seen in the east. This implies a deepening (shoaling) of equatorial thermocline in the west (east). In the North Pacific, a horseshoe-like SST response develops, with warming in the west and central North Pacific surrounded by cooling along the North America coast (Fig. 1b). The midlatitude jet is weakened, which reduces the southward Ekman cold advection as well as surface turbulent heat flux to sustain the warming (Fig. 2a; Wu et al. 2003). In the east, the anomalous anticyclonic wind intensifies both the coastal upwelling and southward cold advection as well as oceanic turbulent heat loss to favor the cooling in the Gulf of Alaska (e.g., Wu et al. 2003). The eastern subtropical North Pacific cooling is predominantly associated with an intensification of the ocean-to-atmosphere turbulent heat loss resulting from an acceleration of the northeasterly trades (Fig. 2a). A similar cooling is also seen in the eastern subtropical South Pacific, coupled with an intensification of the southeasterly trades. Overall, the basin-scale pattern of SST responses shares some similarities with the Pacific decadal oscillation (PDO).

The negative PmE forcing also induces a local positive feedback to reinforce the forcing (Figs. 2b–d). While the evaporation is reduced slightly (Fig. 2b), the reduction of the precipitation over the western Pacific is notable (Fig. 2c), which leads to a further loss of freshwater over the forcing region, with a maximum of about −0.6 mm day−1 located in the far western Pacific and a total accounting for about 20% of the imposed forcing. This positive feedback is due to the development of cooling in the western Pacific in response to the imposed negative PmE forcing.

Based on the freshwater and heat flux responses, we further calculate the response of the buoyancy flux Bo, given as
i1520-0442-23-7-1945-eq1
where Cp is the specific heat of water, ρo is a reference density, α = −ρ−1(∂ρ/∂T)s,p is the surface value of the expansion coefficient of water at fixed salinity s and pressure p, Q is the net heat flux into the ocean, and β = ρ−1(∂ρ/∂S)T,P (Gill 1982). Positive Bo corresponds to a buoyancy loss by ocean. It can be seen that a negative PmE forcing over the warm pool region leads to a net gain of buoyancy in the tropics, but with a weak magnitude (Fig. 3). This comes from the fact that the buoyancy loss resulting from the imposed freshwater forcing and response is nearly compensated by the buoyancy gain resulting from the net heat flux anomaly into the tropical ocean (Fig. 2a).

The coupled wind and SST response also exhibits a distinct seasonality, showing an apparent westward propagation from summer to winter (Fig. 4a). The cooling appears to initiate in the eastern equatorial Pacific in summer (Fig. 4b), and subsequently grows and propagates toward the west, forming a La Niña–like pattern in the following fall with a maximum cooling of −0.6°C around 150°W (not shown). After that, the cooling propagates farther westward, but damps quickly in the east (Fig. 4c). The cooling in the warm pool region peaks in the winter, with the same magnitude as that in the eastern equatorial Pacific. The entire tropical responses virtually disappear in the spring. Over the North Pacific, the most significant responses are seen in winter and spring (not shown), lagging the tropics by about one season. Therefore, the North Pacific response can be presumably attributed to the atmospheric teleconnection from the tropics.

In short, a negative PmE anomaly in the western tropical Pacific can induce not only a La Niña–like response in the tropics, but also a significant response in the extratropics. It is noted that, in spite of different freshwater forcings, the tropical oceanic changes are broadly similar to some early OGCM studies (Yang et al. 1999; Huang and Mehta 2004; Huang et al. 2005). Here, the coupled model study can further demonstrate the atmospheric response. The following section will discuss the mechanisms of the coupled responses.

4. Possible mechanisms

To understand the mechanisms controlling the tropical coupled responses, a heat budget analysis is carried out for the surface and subsurface ocean in the western and eastern equatorial Pacific, respectively. In the western equatorial Pacific, the surface cooling is largely associated with the anomalous meridional advection and damped by surface heat flux, while in the subsurface the warming is mainly associated with the anomalous meridional advection and mean horizontal advection, and damped by the vertical advection (Figs. 5a,b). This is consistent with an acceleration of the upper-ocean meridional cell in both hemispheres (Fig. 5e), which intensifies the heat transport divergence in the surface and convergence in the subsurface in the western tropical Pacific. It will be demonstrated that the acceleration of the meridional overturning cells is largely driven by the buoyancy flux anomalies in response to the change of salinity, while the local easterly wind anomalies play a secondary role. In the eastern equatorial Pacific, the annual mean surface cooling is generally weak in spite of a strong seasonal response (Figs. 1d and 4a). The heat budget analysis demonstrates a strong damping role of the surface heat flux (Fig. 5c). In the subsurface, the cooling is associated with anomalous meridional and vertical advection, but is damped by mean vertical advection (Fig. 5d). This is consistent with an intensification of the equatorial upwelling (Fig. 5f) and the meridional overturning circulation (not shown).

The heat budget analyses lead to the following hypothesis for the mechanisms of the tropical response: a salinification in the western tropical Pacific reduces the sea surface height locally and sets up a positive zonal pressure gradient anomaly, inducing anomalous poleward surface flow and in turn subsurface equatorward-compensating flow. This leads to an intensification of meridional heat divergence and thus the tropical surface cooling. Along the equator, the positive zonal pressure gradient anomaly also intensifies the surface westward current, and thus the equatorial upwelling. The cooling is subsequently amplified by the positive wind stress–upwelling feedback.

The annual mean sea surface height anomaly (SSHA) and the seasonal evolution of SSHA along the equator are demonstrated in Fig. 6. In the tropics, a negative SSHA with a maximum magnitude of 15 mm is clearly identified in the western Pacific (Fig. 6a). The positive zonal pressure gradient anomaly persists year-round (Fig. 6b), except in the fall season when the substantial cooling develops in the central and eastern equatorial Pacific (Fig. 4a). In the fall, the vigorous development of easterly wind anomalies coupled with a cold SST anomaly leads to surface anomalous Ekman divergence and thus negative SSHA in the eastern and central Pacific. It is noted that a strong negative SSHA also develops in the western subtropical North Pacific (Fig. 6a). This is due to the effect of local salinification (Fig. 1a) resulting from the oceanic advection of the tropical salinification.

To further demonstrate the coupled mechanisms of SST and wind stress responses, we conducted a sensitivity experiment, which is the same as the FORCE experiment, but with the tropical wind stress (15°S–15°N) fixed to the model climatology. This experiment is named as FWIND, in which the effects of the Bjerknes feedback and the wind-induced oceanic circulation changes are eliminated. In the absence of dynamic coupling, cooling still develops in the tropics with thermocline deepening (shoaling) in the west (east; Figs. 7a,b). The oceanic circulation changes are broadly similar to those in the fully coupled experiment: an acceleration of the meridional overturning flow in the western Pacific (Fig. 7c), an equatorial surface westward flow, and upwelling (Fig. 7d). The heat budget analyses also reveal the same oceanic processes involving in the tropical oceanic temperature changes (not shown). All of these suggest fundamental roles of buoyancy-induced oceanic circulation changes in driving both equatorial SST and thermocline changes. However, it also demonstrates that the dynamic coupling substantiates the responses dramatically. The dynamic coupling not only doubles the amplitude, but also intensifies the seasonality of the surface cooling (Fig. 4a versus Fig. 7a). The westward-coupled propagation of SST and surface winds virtually disappears in the absence of dynamic coupling. Because the surface heat flux acts to damp the SST (Figs. 5a,c), the coupled development of SST and surface wind anomalies largely indicates a positive feedback of wind stress–upwelling as follows: a cold (warm) equatorial SST anomaly induces a Gill-type response to enhance (reduce) equatorial easterly and in turn upwelling to the west, leading to a westward intensification and propagation of SST anomaly.

5. Summary and discussion

The coupled ocean–atmosphere response to an idealized freshwater flux in the western tropical Pacific is studied using a fully coupled climate model. The model explicitly demonstrates that a negative freshwater flux in the western equatorial Pacific can lead to a La Niña–like SST response in the tropics. This response is attributed to buoyancy-induced intensification of the meridional heat divergence and vertical cold advection, and the positive wind stress–upwelling feedback. The tropical cooling also forces a significant response in the extratropical North Pacific through atmospheric teleconnection. A similar experiment with the same forcing amplitude but opposite polarity is also completed. The responses turn out to be virtually the same, but with an opposite sign, implying that the responses are largely linear.

The similarity of the vertical structure in the western tropical Pacific between the FORCE and the FWIND experiment may be associated with the buoyancy-forced high-order baroclinic mode. It is known that the buoyancy forcing can excite higher vertical modes, in contrast to the wind forcing. The strongest amplitude for the forced vertical mode roughly has a vertical wavelength equivalent to twice that of the forcing depth, which may be approximated by the mixed layer depth. Although the temperature anomalies are the largest in the thermocline, the vertical displacement [T ′/(dT/dz)] is largest around the nodal point of the meridional velocity anomalies.

There is an issue whether the responding salinity and temperature fields may compensate with each other relative to their effects on density. The significant increase of density appears in the western Pacific and the eastern subtropical North Pacific. In the western Pacific, the density increase is dominated by the salinification resulting from the imposed forcing and oceanic gyre advection. In the eastern subtropical Pacific, the density increase is associated with the intensification of both freshwater (Fig. 2d) and oceanic heat loss (Fig. 2a).

The coupled modeling study here, although idealized, demonstrates potential roles of the freshwater flux in the tropical coupled ocean–atmosphere interaction. During an El Niño event, the convection center is shifted to the central Pacific, leading to a negative freshwater flux anomaly in the western tropical Pacific. This anomalous freshwater flux can produce a negative feedback to El Niño through changes of tropical ocean circulation and coupled feedbacks. On the other hand, it is noted that the positive freshwater flux anomaly in the central and eastern equatorial Pacific during El Niño may enhance warming in the eastern equatorial Pacific (Yang et al. 1999). The net impacts of the freshwater flux changes on ENSO depend on the competitions of these two opposite effects.

Our study here indicates that a long-lasting freshwater flux in the western tropical Pacific not only forms a positive feedback to reinforce the freshwater forcing, but also induces a basin-wide SST response that shares some similarities with the PDO phase. The study may have some implications for the PDO and global warming. Many studies indicate a warming trend in the Pacific resembling the positive-phase PDO over the past several decades. In the tropics, the dry trend in the western tropical Pacific (e.g., Huang and Mehta 2004) can force a negative PDO-like warming anomaly, which may potentially offset the positive PDO-like warming trend, although such an opposite effect may not be dominant. A multimodel assessment with the observed freshwater forcing will be helpful in understanding the oceanic and climatic responses to the hydrological changes in the western tropical Pacific.

Acknowledgments

This work is supported by Chinese National Key Basic Research Program (2007CB411800) and Chinese National Science Foundation Outstanding Young Investigator Program (NSFC 40788002), and Major Research Project (NSFC 40890155). We are indebted to Drs. Boyin Huang, Ping Chang, Lisan Yu, and Ronghua Zhang for helpful discussions. Comments from both anonymous reviewers improve the paper substantially in many aspects.

REFERENCES

  • Carton, J. A., 1991: Effect of seasonal surface freshwater flux on sea surface temperature in the tropical Atlantic Ocean. J. Geophys. Res., 96 , 1259312598.

    • Search Google Scholar
    • Export Citation
  • Gill, A. E., 1982: Atmosphere–Ocean Dynamics. Academic Press, 662 pp.

  • Huang, B., , and V. M. Mehta, 2004: The response of the Indo-Pacific warm pool to interannual variations in net atmospheric freshwater. J. Geophys. Res., 109 , C06022. doi:10.1029/2003JC002114.

    • Search Google Scholar
    • Export Citation
  • Huang, B., , and V. M. Mehta, 2005: Response of the Pacific and Atlantic Oceans to interannual variations in net atmospheric freshwater. J. Geophys. Res., 110 , C08008. doi:10.1029/2004JC002830.

    • Search Google Scholar
    • Export Citation
  • Huang, B., , V. M. Mehta, , and N. S. Schneider, 2005: Oceanic response to idealized net atmospheric freshwater in the Pacific at decadal timescales. J. Phys. Oceanogr., 35 , 24672486.

    • Search Google Scholar
    • Export Citation
  • Jacob, R. L., 1997: Low frequency variability in a simulated atmosphere ocean system. Ph.D. thesis, University of Wisconsin—Madison, 155 pp.

  • Liu, Z., , J. Kutzbach, , and L. Wu, 2000: Modeling climate shift of El Niño variability in the Holocene. Geophys. Res. Lett., 27 , 22652268.

    • Search Google Scholar
    • Export Citation
  • Lukas, R., , and E. Lindstrom, 1991: The mixed layer of the western equatorial Pacific Ocean. J. Geophys. Res., 96 , 33433357.

  • Murtugudde, R., , and A. J. Busalacchi, 1998: Salinity effects in a tropical ocean model. J. Geophys. Res., 103 , 32833300.

  • Reason, C., 1992: On the effect of ENSO precipitation anomalies in a global ocean GCM. Climate Dyn., 8 , 3947.

  • Wu, L., , and Z. Liu, 2006: Decadal variability in the North Pacific: The eastern North Pacific mode. J. Climate, 16 , 31113131.

  • Wu, L., , Z. Liu, , R. Gallimore, , R. Jacob, , Y. Zhong, , and D. Lee, 2003: Pacific decadal variability: The tropical Pacific mode and the North Pacific mode. J. Climate, 16 , 11011120.

    • Search Google Scholar
    • Export Citation
  • Yang, S., , K-M. Lau, , and P. S. Schopf, 1999: Sensitivity of the tropical Pacific ocean to precipitation-induced freshwater flux. Climate Dyn., 15 , 737750.

    • Search Google Scholar
    • Export Citation
  • Yu, L., , and R. A. Weller, 2007: Objectively Analyzed Air–Sea Heat Fluxes (OAFlux) for the global ice-free oceans (1981–2005). Bull. Amer. Meteor. Soc., 88 , 527539.

    • Search Google Scholar
    • Export Citation
  • Zhang, R-H., , and A. J. Busalacchi, 2009: Freshwater flux (FWF)-induced oceanic feedback in a hybrid coupled model of the tropical Pacific. J. Climate, 22 , 853879.

    • Search Google Scholar
    • Export Citation

Fig. 1.
Fig. 1.

Salinity and temperature anomalies in response to an idealized negative freshwater forcing in the western tropical Pacific. (a) SSS (contours in dark-shaded area at intervals of 0.2 psu), (b) SST (contours in dark-shaded area at intervals of 0.1°C), and surface wind (vectors, m s−1). Longitude–depth distribution of (c) salinity (contours at intervals of 0.1 psu) and (d) temperature averaged between 5°S and 5°N (contours in dark-shaded area at intervals of 0.1°C). The dark line in (d) denotes the 20°C isotherm in the CTRL. Area shaded exceeds the 95% statistical significance using a Student’s t test.

Citation: Journal of Climate 23, 7; 10.1175/2009JCLI3009.1

Fig. 2.
Fig. 2.

Annual mean (a) net surface heat flux (positive downward, contours in dark-shaded area at intervals of 2 W m−2), (b) evaporation (contours in dark-shaded area at intervals of 0.05 mm day−1), (c) precipitation, and (d) PmE anomaly (contours in dark-shaded area at intervals of 0.1 mm day−1). Area shaded with dark gray exceeds 90% statistical significance using a Student’s t test. Area shaded with light gray represents positive anomaly area.

Citation: Journal of Climate 23, 7; 10.1175/2009JCLI3009.1

Fig. 3.
Fig. 3.

Annual mean buoyancy flux anomaly (contours in dark-shaded area at intervals of 1 × 10−9 m2 s−3). Area shaded with dark gray exceeds 90% statistical significance using a Student’s t test, and area shaded with light gray represents the positive anomaly area.

Citation: Journal of Climate 23, 7; 10.1175/2009JCLI3009.1

Fig. 4.
Fig. 4.

Seasonality of SST (contours in shaded area at intervals of 0.1°C) and surface wind (vectors, m s−1) responses in the Pacific. (a) Longitude–time distribution averaged over the equatorial band (5°S, 5°N). Maps for (b) June–August and (c) December–February. Area shaded exceeds 90% statistical significance using a Student’s t test.

Citation: Journal of Climate 23, 7; 10.1175/2009JCLI3009.1

Fig. 5.
Fig. 5.

Heat budget analyses for the (left) surface and (right) subsurface in the (top) western and (middle) eastern equatorial Pacific. See the regional boxes in Fig. 1d. The unit for each budget term is 1 × 107 W m−2. (bottom) Latitude–depth distribution of (e) meridional velocity anomalies in the western Pacific averaged from 120° to 140°E, and (f) zonal and vertical velocity averaged between 5°S and 5°N. Unit for velocity: m s−1. The magnitude of horizontal and vertical velocity are multiplied 1 × 102 and 2 × 106, respectively.

Citation: Journal of Climate 23, 7; 10.1175/2009JCLI3009.1

Fig. 6.
Fig. 6.

(a) Annual mean sea surface height anomaly (contours in shaded area at intervals of 5 mm), (b) seasonal evolution of sea surface height anomaly along the equator averaged over the equatorial band (5°S, 5°N; contours at intervals of 3 mm) induced by the imposed freshwater flux. Area shaded exceeds 90% statistical significance using a Student’s t test.

Citation: Journal of Climate 23, 7; 10.1175/2009JCLI3009.1

Fig. 7.
Fig. 7.

Oceanic changes in the FWIND experiment where the wind stress in the tropics is fixed to the model climatology. Longitude–time distribution of (a) SST (contours at intervals of 0.05°C), (b) temperature averaged over the equatorial band (5°S, 5°N; contours in shaded area at intervals of 0.1°C), (c) meridional velocity anomalies in the western Pacific averaged from 120° to 160°E (contours at intervals of 0.2 × 10−2 m s−1), and (d) zonal and vertical velocity averaged between 5°S and 5°N. Unit for velocity: m s−1. The magnitude of horizontal and vertical velocity multiplied 1 × 102 and 3 × 106, respectively. Area shaded exceeds 90% statistical significance using a Student’s t test.

Citation: Journal of Climate 23, 7; 10.1175/2009JCLI3009.1

Save