The Delayed Effect of Major El Niño Events on Indian Monsoon Rainfall

Hyo-Seok Park Department of Geography, and Berkeley Atmospheric Sciences Center, University of California, Berkeley, Berkeley, California

Search for other papers by Hyo-Seok Park in
Current site
Google Scholar
PubMed
Close
,
John C. H. Chiang Department of Geography, and Berkeley Atmospheric Sciences Center, University of California, Berkeley, Berkeley, California

Search for other papers by John C. H. Chiang in
Current site
Google Scholar
PubMed
Close
,
Benjamin R. Lintner Department of Atmospheric and Oceanic Sciences, and Institute of Geophysics and Planetary Physics, University of California, Los Angeles, Los Angeles, California

Search for other papers by Benjamin R. Lintner in
Current site
Google Scholar
PubMed
Close
, and
Guang J. Zhang Climate, Atmospheric Science, Physical Oceanography Division, Scripps Institution of Oceanography, La Jolla, California

Search for other papers by Guang J. Zhang in
Current site
Google Scholar
PubMed
Close
Full access

Abstract

Previous studies have shown that boreal summer Indian monsoon rainfall is, on average, significantly above normal after major El Niño events. In this study, the underlying causes of this rainfall response are examined using both observational analysis and atmospheric general circulation model (AGCM) simulations. Moist static energy budgets for two strong El Niño events (1982/83 and 1997/98), estimated from monthly 40-yr European Centre for Medium-Range Weather Forecasts Re-Analysis (ERA-40), suggest that stronger low-level moisture transport and reduced moist stability associated with a warmer north Indian Ocean (NIO) can increase monsoon rainfall, despite a weakened monsoon circulation.

The trade-off between a dynamically weaker monsoon and moist processes favoring enhanced monsoonal rainfall is broken during the late monsoon season (August–September) as the warm NIO enhances surface latent heat flux and the monsoon circulation relaxes back to the climatological mean. The monsoon circulation strength and the moist processes work together in the late season, which explains the observed tendency for monsoonal rainfall increases during the late monsoon season after strong winter El Niño conditions.

Idealized AGCM experiments with a fixed-depth ocean mixed layer demonstrate that the remnant but weaker-than-peak warm SSTs in the eastern equatorial Pacific during spring and the early summer following winter El Niños substantially contribute to the NIO warming. The results suggest that local air–sea interactions in the tropical Indian Ocean after winter El Niño are strongly dependent on the details of El Niño’s decaying trend.

Corresponding author address: Hyo-Seok Park, University of California, Berkeley, 531 McCone Hall, Berkeley, CA 94720-4740. Email: hspark@berkeley.edu

Abstract

Previous studies have shown that boreal summer Indian monsoon rainfall is, on average, significantly above normal after major El Niño events. In this study, the underlying causes of this rainfall response are examined using both observational analysis and atmospheric general circulation model (AGCM) simulations. Moist static energy budgets for two strong El Niño events (1982/83 and 1997/98), estimated from monthly 40-yr European Centre for Medium-Range Weather Forecasts Re-Analysis (ERA-40), suggest that stronger low-level moisture transport and reduced moist stability associated with a warmer north Indian Ocean (NIO) can increase monsoon rainfall, despite a weakened monsoon circulation.

The trade-off between a dynamically weaker monsoon and moist processes favoring enhanced monsoonal rainfall is broken during the late monsoon season (August–September) as the warm NIO enhances surface latent heat flux and the monsoon circulation relaxes back to the climatological mean. The monsoon circulation strength and the moist processes work together in the late season, which explains the observed tendency for monsoonal rainfall increases during the late monsoon season after strong winter El Niño conditions.

Idealized AGCM experiments with a fixed-depth ocean mixed layer demonstrate that the remnant but weaker-than-peak warm SSTs in the eastern equatorial Pacific during spring and the early summer following winter El Niños substantially contribute to the NIO warming. The results suggest that local air–sea interactions in the tropical Indian Ocean after winter El Niño are strongly dependent on the details of El Niño’s decaying trend.

Corresponding author address: Hyo-Seok Park, University of California, Berkeley, 531 McCone Hall, Berkeley, CA 94720-4740. Email: hspark@berkeley.edu

1. Introduction

While Indian monsoon rainfall is known to be suppressed during the summer preceding peak winter El Niño conditions, a tendency for above-normal precipitation during the subsequent summer season has been noted (Shukla 1995; Webster et al. 1998; hereafter we refer to the latter as the “delayed effect”). Considered over the entire monsoon season [June–September (JJAS)], the relationship between a common measure of all-Indian monsoon rainfall and ENSO is statistically weak (Fig. 1a). However, looking only at the late monsoon season [August–September (AS)]1 indicates a more robust delayed effect (Fig. 1b): approximately half of the AS periods following strong winter El Niños experience positive rainfall anomalies exceeding one standard deviation.

The delayed effect is often explained in terms of the development of La Niña phase conditions subsequent to strong El Niño events (Shukla 1995). Since the developing La Niña signal is generally stronger in AS relative to June–July (JJ), the contemporaneous effect of developing La Niña conditions may contribute to above-normal rainfall in AS. However, absent in this view is the potential role of residual Indian Ocean warming induced by winter El Niño conditions, that is, the upper ocean’s thermal memory of strong El Niño events.

The seasonal evolution of El Niño–related air–sea interactions in the Indo–western tropical Pacific Ocean, has been extensively investigated. Xie et al. (2002) showed that El Niño–induced westward-propagating oceanic Rossby waves can warm the southwest Indian Ocean. Such warming may be amplified by anomalous anticyclonic surface winds associated with higher surface pressures over the south equatorial Indian Ocean, which lead to deeper thermoclines (Huang and Shukla 2007). Tropospheric warming has also been suggested as another mechanism for the basinwide tropical Indian Ocean warming (Chiang and Lintner 2005), especially in the spring season following major winter El Niño events. The basinwide tropical Indian Ocean warming in boreal spring does not decay immediately, but it may persist well into the summer (active monsoon) season. Although ocean dynamics may contribute to the persistence of Indian Ocean SSTs (Webster et al. 1999), simulations of AGCMs coupled to ocean mixed layer models (e.g., Lau et al. 2005) suggest significant persistence simply from thermodynamic surface flux controls.

A few recent studies have noted the influence of locally warm SSTs on Indian monsoon rainfall. For example, Terray et al. (2003) suggested that the anomalously warm south Indian Ocean (SIO) SSTs in boreal spring slowly transition to warm north Indian Ocean (NIO) SSTs and strengthen the local Hadley circulation near India and intensify Indian monsoon rainfall during the late monsoon season (AS). On the other hand, Annamalai et al. (2005) found that persistent SIO warming delays Indian summer monsoon onset by a week. In this paper, we provide a comprehensive mechanism for the increasing Indian monsoon rainfall by examining two major El Niño events—1982/83 and 1997/98—which are the strongest twentieth-century El Niño events for which extensive satellite data coverage exists. In our analysis, we focus on the spatial and temporal evolution of SST and low-level wind anomalies in the Indian Ocean prior to and concurrent with the Indian summer monsoon and the impact on the regional moist static energy (MSE) budgets using observational and reanalysis data (sections 3 and 4). From this analysis, and from results of idealized AGCM simulations (section 5), we propose a hypothesis for the delayed effect of El Niño on the Indian monsoon rainfall.

2. Datasets and model

a. Datasets

We investigate the seasonal evolution of the SST and MSE budget using the 40-yr European Centre for Medium-Range Weather Forecasts (ECMWF) Reanalysis (ERA-40; Uppala et al. 2005). The ERA-40 SSTs are based on monthly SST data [Met Office Hadley Centre Sea Ice and Sea Surface Temperature (HadISST); Rayner et al. 2003] up to October 1981 and on the weekly Reynolds optimum interpolation version-2 data (Reynolds et al. 2002) thereafter. Because the long-term SSTs are based on observations, the low-level specific humidity field, which is strongly dependent on the SSTs, will be able to provide reliable and consistent MSE budgets. We use the version of ERA-40 produced by the National Center for Atmospheric Research (NCAR)’s Data Support Section (available online at http://www.cgd.ucar.edu/cas/catalog/ecmwf/era40) consisting of monthly means for 256 × 128 regular Gaussian grids at T85 spectral truncation and 23 pressure levels spanning from September 1957 to August 2002.

We used a 23-yr (1979–2001) monthly long-term mean to calculate the climatological mean of individual variables, such as surface latent heat flux, SST, individual vertical level winds, and specific humidity. The SST, surface latent heat flux (LHF), and specific humidity of the individual vertical levels were linearly detrended because of their increasing trend from the 1960s to the 1990s. However, the individual MSE budgets during major El Niño events that we will present are qualitatively consistent (or, almost identical) regardless of the long-term linear detrending, except that the magnitudes of anomalous SST and the anomalous MSE budget get weaker by 20% for the 1982/83 event in the case where the linear detrending is not applied.

For monsoon precipitation, the Climate Prediction Center (CPC) Merged Analysis of Precipitation (CMAP; Xie and Arkin 1997) dataset was used. The CMAP data consist of monthly means at a horizontal resolution of 2.5° × 2.5° spanning from January 1979 to December 2005. For the CMAP data, climatological means were estimated for the entire period (1979–2005) of coverage, and anomalies were formed as for the ERA-40 fields. The precipitation anomalies of the two major El Niño events are quite robust regardless of the specific averaging periods for the climatological mean calculation.

b. Model

To understand the role of remote forcing from the eastern equatorial Pacific and its temporal evolution, we conducted a set of simplified AGCM experiments. We used the Community Climate Model, version 3.10 (CCM3; Kiehl et al. 1998), at T31 × 15 horizontal resolution (i.e., triangular truncation of 31 basis functions in the meridional and 15 in the zonal) and the standard 18 levels in the vertical. A revised convective parameterization closure (Zhang and Mu 2005, hereafter ZM05) based on the Zhang–McFarlane scheme was implemented. Under the ZM05 scheme, convective closure consists of large-scale forcing of temperature and humidity in the free troposphere rather than convective available potential energy as in the original scheme. We find that the revised convection scheme is superior to the standard CCM3 convection scheme (Zhang and McFarlane 1995) in simulating the Indo-Pacific tropical climate. In particular, the ZM05 scheme increases western Pacific monsoon precipitation, thereby alleviating the negative precipitation bias seen in the original convection scheme by Zhang and McFarlane (1995); it also removes the spurious heavy precipitation over the Arabian Peninsula simulated by the original convection scheme.

The CCM3 was coupled to a 30-m constant mixed layer slab ocean model at all ocean grid points outside of the tropical Pacific (20°S–20°N, 160°–270°E) where SST was prescribed with values corresponding to total SSTs from February of the growing year (1982 or 1997) to November of the decaying year (1983 or 1998). The SST forcing was derived from the gridded 2° × 2° monthly mean NCEP–Reynolds historical global SST field covering January 1950–December 2000 (Reynolds et al. 2002). For the slab ocean grid points, a climatological monthly Q-flux correction was applied to ensure that the simulated SST seasonal climatology closely matches the observed SSTs (Reynolds et al. 2002). For each El Niño event, 12 ensemble members were simulated with each ensemble member initialized from a distinct set of self-consistent atmospheric conditions on 1 February of the growing year. The initial conditions were extracted from a long-term simulation integrated with the climatological seasonal cycle SSTs imposed in the central and eastern tropical Pacific (20°S–20°N, 160°–270°E) and a Q-flux-adjusted uniform 30-m slab ocean elsewhere.

3. North Indian Ocean warming and the monsoonal low-level winds

a. North Indian Ocean warming

A Hovmöller plot of anomalous SSTs (Figs. 2a and 2b) demonstrates that a warming of the NIO occurs in the summer monsoon season after major El Niño events. Here, the longitudinal averaging interval (40°–90°E) covers the entire western Indian Ocean, including the Arabian Sea and part of the Bay of Bengal; this interval was selected because the SSTs and winds over the Arabian Sea and the western part of the Bay of Bengal effectively modulate the Indian monsoon convective strength through the low-level westerly monsoon flow. Also, the monsoonal winds and SST vary more over the Arabian Sea region than the eastern part of the Bay of Bengal or the South China Sea during the major El Niño events.

The 1983 event exhibited strong NIO warming—that is, the SSTs increased rapidly from climatological conditions during late boreal spring (May–June) and persisted until July–August. The 1983 event suggests a northward progression of positive SST anomalies from the south equatorial Indian Ocean during the boreal spring into the NIO region by the summer monsoon season (Fig. 2a). The 1998 event had a warm NIO in the summer monsoon season as well (Fig. 2b), although the signal is weaker compared to the 1983 event. Also, the anomalously warm NIO existed in the spring, leading to a double warming signal (in the spring and the summer) over the NIO (Fig. 2b). This double warming signal is reminiscent of structure in the regression of the winter (November–January) Niño-3.4 index on the subsequent spring-to-summer Indian Ocean SSTs in Du et al. (2009).

The detailed longitude–latitude structure of the SST anomalies for the two events contains some substantial differences. The 1983 event has positive SST anomalies over the southeast Indian Ocean in spring (Fig. 3c) and a strong warming signal appears over the Arabian Sea in the summer (Fig. 3a). On the other hand, the 1998 event has very strong warm SST anomalies in the equatorial and south equatorial Indian Ocean in spring (Fig. 3d), which may have been supported by stronger equatorial Ocean wave activities (Webster et al. 1999) or large-scale tropical troposphere warming (Chiang and Lintner 2005). The south equatorial warm SST anomalies during the spring of 1998 dissipate by the summer monsoon season, whereas the warm NIO signal slightly strengthens (Fig. 3b). The maximum warm SST anomalies exist over the western Pacific in the summer of 1998, whereas the 1983 event has its maximum warm anomalies over the Arabian Sea.

Although there are clearly differences between the 1983 and 1998 events, we emphasize here two common features, namely, (i) the distinctly warm NIO throughout the summer monsoon season and (ii) the dissipation of the south equatorial Indian Ocean warming signal by the monsoon onset.

b. Monsoonal low-level winds and surface latent heat flux

Hovmöller plots of anomalous surface wind speeds (50°–83°E mean; mostly over the Arabian Sea region, where strong monsoonal low-level winds blow) indicate weakening of NIO-region surface wind speeds during the early phase of the monsoon followed by a later rebound toward climatological values for both El Niño events (contour lines in Figs. 4a and 4b). Since the area-averaged (50°–83°E mean) monthly standard deviations of surface wind speed during the summer monsoon season (JJAS mean from the monthly mean ERA-40) is only ∼0.3 m s−1, the observed early season weakening (0.5 m s−1) is rather substantial. Although both events consistently show the weakening of the low-level winds and the surface latent heat suppression from the late spring to the midmonsoon season, the timing of wind speed recovery toward climatology slightly differs in each case. For the 1983 case, the weakening starts in May and peaks in June, maintaining the weakening signal by July. The wind speed begins to rebound in August and grows substantially stronger than the seasonal mean in September. On the other hand, for the 1998 event, the weakening persists longer, peaking in August, with some recovery evident in September.

Figures 4a and 4b also show that a suppression of the LHF is coincident with the weakening of winds both for the 1983 and 1998 events, although the latter slightly lags the LHF. Here, we hypothesize that the surface wind weakening might be the principal driver of the NIO warming. To verify this, we decomposed the surface evaporation perturbations into dynamic and thermodynamic contributions, following Chikamoto and Tanimoto (2006). The surface latent heat flux anomaly is linearized as
i1520-0442-23-4-932-e1
where q* is the saturation specific humidity at the sea surface, qa is the specific humidity at a reference level near the sea surface, W is the surface wind speed, and κ is a bulk coefficient. The first term on the rhs (the dynamic contribution) is associated with wind speed changes, whereas the second term (the thermodynamic contribution) is associated with the humidity difference changes. Assuming constant κ, the dynamic contribution dominates the NIO LHF suppression during the monsoon onset period (not shown), though since the assumption of constant κ can be unrealistic and the ocean dynamics are neglected, we are not confident in providing the exact magnitude of wind effect. Rather, we suggest that the contribution from the weakening of monsoonal winds is a nontrivial factor for the NIO warming.

The persistent NIO warming leads to an increased LHF during the decaying period of the monsoon. The LHF signal leads the surface wind signal because the warmer NIO increases the LHF to the atmosphere, compensating for the effect from the weaker surface wind speed. Additionally, the rebound of surface wind strength during the late monsoon season may also contribute to increasing the LHF. One caveat worth noting is that the LHF rebound is limited to the Arabian Sea, whereas the Bay of Bengal consistently experiences LHF suppression because of persistent weakening of surface winds throughout the summer monsoon season (not shown).

4. Moisture transport and moist stability over the NIO

While the occurrence of positive rainfall anomalies during the late monsoon season (Fig. 1b) is consistent with anomalous LHF, how do we explain the above-normal rainfall in the early season? In a dynamical sense, the substantially weaker surface winds and the reduced LHF during the early seasons of both events suggest a weakened monsoon, although the rainfall is slightly above normal or close to normal (Figs. 5a and 5b).2 The paradox of the increased monsoon rainfall associated with the weaker monsoon circulation has been previously addressed by a few GCM studies under global warming scenarios (Kitoh et al. 1997; Stowasser et al. 2009).

As we presented, the paradox seems to appear at the interannual time scale as well. To explain the mechanism, we examine the anomalous moist processes associated with the warmer NIO, specifically changes to horizontal moisture transport and moist stability over the NIO.

a. MSE budget methodology

Following Chou and Neelin (2004) and Chou et al. (2006), the anomalous vertically integrated MSE 〈h〉′ can be written as
i1520-0442-23-4-932-e2a
where the prime denotes a perturbation, the angle brackets represents vertical integration from the surface layer (1000 hPa) to near the tropopause height (150 hPa), and v and ω are the horizontal wind and pressure velocity individually, respectively; Fnet denotes net energy input into the atmosphere column, including both net radiative flux and surface heat flux. Because of rapid atmospheric adjustment, (∂/∂t)〈h〉′ can be neglected on monthly time scales. Given our interest in convective energy variations associated with anomalous vertical motions, the second term on the lhs of (2a)—that is, vertical advection of MSE—can be linearized by separating perturbations and climatological means. Equation (1) is then rearranged to
i1520-0442-23-4-932-e2b
where the perturbation moisture and dry static energy s = CpT + gZ terms have been explicitly separated. In this framework, the lhs of (2b), which is proportional to the anomalous vertical motion, provides a measure of the anomalous large-scale convective activity; the terms on the rhs represent a diagnostic budget for these motions. The first term on the rhs of (2b), 〈−ω(∂q′/∂p)〉, is analogous to the anomalous moist stability Mq′ (Chou et al. 2006). Over the equatorial ocean region, the moisture transport 〈−v · q〉′ is usually much larger than the dry static energy advection 〈−v · s〉′, although the magnitude of the dry static energy advection, especially temperature advection, becomes nontrivial where tropospheric temperature gradients become large (e.g., near land–ocean interfaces). The anomalous dry static stability 〈−ω(∂s′/∂p)〉 tends to compensate the Mq′; however, the magnitude of compensation is limited, around one-third of Mq′ over the NIO (not shown).

b. Moist stability and moisture transport over the NIO

The JJA mean Mq′ is mostly positive over the vast warm pool region, covering the NIO and the western Pacific for both events (Figs. 6a and 6b). Here, Mq′ > 0 occurs with strengthened moist convection over the ocean regions adjacent to the Indian subcontinent, although its relationship to continental convection looks less obvious. Even though the Arabian Sea has higher SST anomalies than the Bay of Bengal region (see the 1983 event in Fig. 3a), Mq′ is more strongly positive over the Bay of Bengal than the Arabian Sea because upward motions are stronger and the moisture stratification effect is more efficient over the former.

We suggest that the warmer NIO-induced Mq′ has some similarities to the rich-gets-richer mechanism under global warming (Chou et al. 2006, 2009). As shown in Chou et al. (2009), the application of the MSE budget to 10 coupled global climate models under a global warming scenario indicates that Mq′ > 0 occurs in all models when the mean vertical motion field is upward. That is, over oceanic convecting regions, where the mean large-scale vertical motion is upward, the large-scale lower-troposphere moistening under global warming tends to enhance tropical precipitation. Similarly, the spatial patterns of Mq′ between the 1983 and 1998 events are consistent with each other, showing strong consistency with the seasonal mean vertical motion field (contour lines in Figs. 6a and 6b), regardless of the detailed SST patterns.

The Indian Ocean warming and the subsequent moisture effect to the continental region become more obvious when the moisture transport is considered. Figures 7a and 7b indicate that the combined effect of Mq′ and the moisture transport 〈−v · q〉′ contributes to increasing the basinwide monsoon rainfall. The JAS and AS mean total moist processes also indicate positive monsoon convective strength with similar magnitude with the JJA mean (not shown). The box-averaged (10°–25°N, 60°–100°E) JJAS mean combined moist energy is about 30 (34) W m−2 for the 1983 (1998) event. These values appear significant given that the JJAS mean standard deviation of the combined moist energy during the period of 1979–2001 is about 18 W m−2. However, these values should be interpreted as rough approximations, especially given the lack of information about finer spatial and temporal scales. For example, using daily ERA-40 and National Centers for Environmental Prediction (NCEP) reanalysis datasets, Back and Bretherton (2006) demonstrated substantial magnitude of nonconserving small source terms (or residuals) in the MSE budget. Therefore, the MSE budget estimated from monthly mean data likely has a larger bias. Here, we only suggest that the warmer NIO-induced reduced moist stability and stronger moisture transport may be major mechanisms for compensating the weaker monsoon circulation effect during the early–midmonsoon season as well as for increasing the rainfall during the late monsoon season.

5. The effects of El Niño’s residual signal on the north Indian Ocean warming

In the absence of NIO warming, the warmer equatorial Indian Ocean alone would reduce the monsoon rainfall over the Indian subcontinent. For example, Chung and Ramanathan (2006) prescribed a warmer equatorial Indian Ocean in their AGCM and found that an anomalous northward-decreasing SST profile weakens Indian monsoon rainfall, whereas uniform Indian Ocean warming strengthens it. Given that the warm off-equatorial SIO signal largely dissipates by the early monsoon season, how are the warm NIO SSTs maintained and even strengthened (up to 1.5 K) during the summer monsoon season?

As discussed in section 3, the ERA-40 SSTs (Figs. 2a and 2b) indicate a peak NIO warming signal in July–August, or two seasons after the El Niño peak. This lag is intriguing because the Niño-3 (or Niño-3.4) SST anomalies are significantly attenuated by this time. Is the contemporaneous influence of (weak) eastern equatorial Pacific conditions persisting through spring or early summer monsoon season following El Niño sufficient to drive the peak NIO warming? Or, does some other mechanism—for example, the effect of local Indian ocean–atmosphere feedback processes—provide a bridge to the peak winter forcing? To address this, we introduce an idealized experimental setup.

a. Experimental design

Although the 1982/83 and 1997/98 forcing experiments have their peak forcing during boreal winter, some warming of the eastern equatorial Pacific persists through the end of the monsoon season (Fig. 8, solid line). In the idealized experiment discussed here, we simply eliminated the anomalous SST forcing immediately preceding (and during) the monsoon season by relaxing the SST anomalies to zero over February 1983 (or 1998) to May, as in Fig. 8 (dotted line). This experiment (hereafter referred to as the “no-spring El Niño” case) effectively removes the influence of El Niño’s “tail” on the monsoon. By comparing the “no-spring El Niño” case to the case considered previously (referred to hereafter as the “spring El Niño” case; solid line in Fig. 8), it is possible to isolate the effect of contemporaneous El Niño forcing from the delayed effect of winter El Niño conditions. Since the simulations for both events produced similar results, we focus here on the 1997/98 results only.

b. Role of El Niño’s tail on the NIO warming

The CCM3 perturbation experiment simulates the NIO warming in July–August, with values up to 1.5°C warmer than the climatological July–August mean (Fig. 9a). The difference between the July–August SST and the late-spring (May–June) SST anomalies illustrates the meridional evolution of the anomalous SST field, with the NIO (SIO) warming (cooling) in July–August relative to May–June (Fig. 9b). The suppression of NIO LHF during the onset period of the monsoon (Fig. 10a) and its rebound during the late season (Fig. 10b) are also captured.

Although our idealized El Niño experiment does simulate the Arabian Sea warming in the early monsoon season, which is similar to the ERA-40 1983 SST anomalies, the detailed spatial patterns of the NIO warming are quite different from the ERA-40 SSTs. For example, the observed western Pacific warming during the 1998 is not captured. Also, the simulated latent heat flux rebound occurs in both the Arabian Sea and the Bay of Bengal (Fig. 10b), whereas ERA-40 indicates consistent suppression of the LHF over the Bay of Bengal (not shown). Despite these detailed differences between the model and the reanalysis, we argue that our idealized El Niño model setup captures the salient mechanistic features, such as the weakening of monsoonal winds, premonsoon suppression of LHF, and the subsequent NIO warming during monsoon onset.

Examination of the no-spring El Niño case does show NIO warming at the onset of the monsoon, suggesting a role for the peak El Niño forcing, although the amplitude is less than half when spring El Niño conditions are included (Figs. 9c and 9d). Figures 10c and 10d illustrate the differences of the low-level winds and the LHF over the NIO between the spring El Niño and no-spring El Niño experiments. The magnitudes of low-level wind speed anomalies and the LHF anomalies between Fig. 10a and Fig. 10c are quite close, suggesting that early monsoon season weakening of low-level winds and LHF suppression are mostly from spring El Niño conditions. Because the spring El Niño strongly suppresses the LHF in the early period and thereby warms the NIO, there is a larger rebound of the LHF during the late period (Figs. 10b and 10d).

These results confirm that the residual El Niño region SSTs in the eastern equatorial Pacific, although attenuated relative to peak winter conditions, play a crucial role for the NIO warming in the CCM3 simulations. However, the configuration of these experiments has some limitation in elucidating the effect of the internal feedback process between the Indian Ocean and the northwestern subtropical Pacific. For example, Watanabe and Jin (2003) found in their simplified moist baroclinic model that anomalous surface highs over the Philippine Sea are maintained by the local tropical Indian Ocean warming as well as El Niño’s direct subsidence effects over the Philippine Sea. Based on various GCM experiments, Annamalai et al. (2005) infers that the El Niño–induced warm tropical Indian Ocean SSTs explains more than 50% of the anomalous surface anticyclones over the Philippine and South China Seas. A recent GCM study by Xie et al. (2009) suggests advanced dynamics on the interactions between the warmer NIO and the Philippine anticyclones—that is, a warmer NIO generates eastward-propagating atmospheric Kelvin waves, which would maintain surface highs in the northwestern subtropical Pacific.

In our case, it is not clear if the remnant warm El Niño condition in the eastern equatorial Pacific directly reduces the monsoonal low-level winds, or if the warm tropical Indian Ocean induced by the persistent spring El Niño weakens the low-level winds and generates the anomalous Philippine anticyclone. Regardless of the causality, we suggest that the spring El Niño signal, persisting into the onset period, is a potential source for weakening the monsoonal winds and maintaining the anomalously warm NIO throughout the summer monsoon season.

6. Summary and discussion

We have demonstrated that the influence of a major winter El Niño event persists into the Indian summer monsoon season via an anomalously warm NIO, with the latter in turn driving an increase of rainfall over the NIO and neighboring Indian subcontinent. The weakening of low-level monsoonal winds and the suppression of surface latent heat flux during the late spring and the onset period of the monsoon consistently occur for both the 1983 and 1998 El Niño events. The weakening of winds persists into the midmonsoon season, but the monsoon rainfall remains close to its climatological mean. Anomalous moist processes tied to the warmer NIO, such as the reduced moist stability and the increased horizontal moisture advection, seem able to compensate for the weakened monsoon circulation effect. Idealized CCM3 experiments demonstrate that the strong NIO warming present at the onset of the monsoon season arises in response to residual warm SSTs in the eastern equatorial Pacific, which persist into the late spring and early summer seasons. This remote warming appears to be linked to low-level monsoonal wind weakening and substantial suppression of the surface LHF (∼30% lower than the seasonal mean), thereby warming the NIO during the onset period of monsoon. However, the detailed pathway connecting monsoonal wind weakening to residual eastern Pacific SSTs is unclear. Specifically, we are unable to quantify the relative importance of the remote atmospheric subsidence effect associated with the remnant warm El Niño condition (an anomalous forced Walker circulation) and internal circulation feedback processes initiated by the warm NIO. A possible example of the latter, the anomalous Philippine anticyclone, contribute to weakening the low-level winds, which in turn warms the NIO. The NIO-induced internal feedback may also be related to the precise timing of the monsoonal low-level wind rebound during the late monsoon season. Specifically, we still do not know why weaker monsoonal winds persisted into August during the 1998 event, whereas they rebounded a month earlier during the 1983 event. Further investigation on the feedback processes between the NIO and the Philippine surface anticyclones may elucidate the mechanisms as well as the timing of the monsoonal low-level winds rebound during the late monsoon season.

The most significant insight we can provide at this stage is that El Niño’s decay, if it occurs sufficiently slowly to maintain eastern Pacific SSTs into the spring season, is a useful index for predicting the Indian monsoon rainfall and the monsoon circulation strength. Another interesting result is that the monsoonal low-level wind strength does not necessarily represent the monsoonal convective strength. Rather, moist processes, such as moist stability and moisture transport associated with the warmer NIO, can compensate for the wind effect. In particular, the weakening of winds can warm the NIO, with the latter increasing the monsoon region rainfall by reducing the moist stability and increasing the horizontal moisture advection. This compensation points to the complexity of the Indian monsoon climate system, for which the interannual variability may not be readily categorized by a simplified circulation index or SSTs. In terms of prediction, we argue for the importance of considering the slowly varying time evolution of the SSTs adjacent the Indian subcontinent (Shukla 2007), specifically if it is in the warming phase or in the cooling (i.e., warm SSTs relax back to normal condition) phase. As we indicated, the delayed effect is most obvious during the cooling phase when both the moist process and low-level winds act in concert to strengthen monsoon convection.

Acknowledgments

HSP would like to thank Dr. C-Y Chang and Mr. A. Friedman for their useful discussions. A conversation with Dr. J-Y Yu (UCI) was particularly helpful. HSP and JCHC acknowledge the financial support of NSF ATM-0438201 and the Gary Comer Science and Education Foundation. BRL acknowledges the financial support of NSF ATM-0645200 and GJZ acknowledges the financial support of NSF ATM-0601781.

REFERENCES

  • Annamalai, H., P. Liu, and S-P. Xie, 2005: Southwest Indian Ocean SST variability: Its local effect and remote influence on Asian monsoons. J. Climate, 18 , 41504167.

    • Search Google Scholar
    • Export Citation
  • Back, L. E., and C. S. Bretherton, 2006: Geographic variability in the export of moist static energy and vertical motion profiles in the tropical Pacific. Geophys. Res. Lett., 33 , L17810. doi:10.1029/2006GL026672.

    • Search Google Scholar
    • Export Citation
  • Chiang, J. C. H., and B. R. Lintner, 2005: Mechanism of remote tropical surface warming during El Niño. J. Climate, 18 , 41304149.

  • Chikamoto, Y., and Y. Tanimoto, 2006: Air-sea humidity effects on the generation of tropical Atlantic SST anomalies during the ENSO events. Geophys. Res. Lett., 33 , L19702. doi:10.1029/2006GL027238.

    • Search Google Scholar
    • Export Citation
  • Chou, C., and J. D. Neelin, 2004: Mechanisms of global warming impacts on regional tropical precipitation. J. Climate, 17 , 26882701.

  • Chou, C., J. D. Neelin, J-Y. Tu, and C-T. Chen, 2006: Regional tropical precipitation change mechanisms in ECHAM4/OPYC3 under global warming. J. Climate, 19 , 42074223.

    • Search Google Scholar
    • Export Citation
  • Chou, C., J. D. Neelin, C-A. Chen, and J-Y. Tu, 2009: Evaluating the “rich-get-richer” mechanism in tropical precipitation change under global warming. J. Climate, 22 , 19822005.

    • Search Google Scholar
    • Export Citation
  • Chung, E. C., and V. Ramanathan, 2006: Weakening of north Indian SST gradients and the monsoon rainfall in India and the Sahel. J. Climate, 19 , 20362045.

    • Search Google Scholar
    • Export Citation
  • Du, Y., S-P. Xie, G. Huang, and K. Hu, 2009: Role of air–sea interaction in the long persistence of El Niño–induced north Indian Ocean warming. J. Climate, 22 , 20232038.

    • Search Google Scholar
    • Export Citation
  • Huang, B., and J. Shukla, 2007: On the mechanisms for the interannual variability in the tropical Indian Ocean. Part I: The role of remote forcing from tropical Pacific. J. Climate, 20 , 29172936.

    • Search Google Scholar
    • Export Citation
  • Kiehl, J. T., J. J. Hack, G. B. Bonan, B. A. Boville, D. L. Williamson, and P. J. Rasch, 1998: The National Center for Atmospheric Research Community Climate Model: CCM3. J. Climate, 11 , 11311149.

    • Search Google Scholar
    • Export Citation
  • Kitoh, A., S. Yukimoto, A. Noda, and T. Motoni, 1997: Simulated changes in the Asian summer monsoon at times of increased atmospheric CO2. J. Meteor. Soc. Japan, 75 , 10191031.

    • Search Google Scholar
    • Export Citation
  • Lau, N-C., A. Leetmaa, M. J. Nath, and H-L. Wang, 2005: Influences of ENSO-induced Indo–western Pacific SST anomalies on extratropical atmospheric variability during the boreal summer. J. Climate, 18 , 29222942.

    • Search Google Scholar
    • Export Citation
  • Parthasarathy, B., A. A. Munot, and D. R. Kothawale, 1995: Monthly and seasonal rainfall series for All-India homogeneous regions and meteorological subdivisions: 1871-1994. Indian Institute of Tropical Meteorology Research Rep. RR-065, 113 pp.

    • Search Google Scholar
    • Export Citation
  • Rayner, N. A., D. E. Parker, E. B. Horton, C. K. Folland, L. V. Alexander, D. P. Rowell, E. C. Kent, and A. Kaplan, 2003: Global analyses of sea surface temperature, sea ice, and night marine air temperature since the late nineteenth century. J. Geophys. Res., 108 , 4407. doi:10.1029/2002JD002670.

    • Search Google Scholar
    • Export Citation
  • Reynolds, R. W., N. A. Rayner, T. M. Smith, D. C. Stokes, and W. Wang, 2002: An improved in situ and satellite SST analysis for climate. J. Climate, 15 , 16091625.

    • Search Google Scholar
    • Export Citation
  • Shukla, J., 1995: Predictability of the tropical atmosphere, the tropical oceans and TOGA. Proceedings of the International Scientific Conference on Tropical Ocean Global Atmosphere (TOGA) Programme, Vol. 2, WCRP-91, WMO/TD 717, 725–730.

    • Search Google Scholar
    • Export Citation
  • Shukla, J., 2007: Monsoon mysteries. Science, 318 , 204205.

  • Stowasser, M., H. Annamalai, and J. Hafner, 2009: Response of the South Asian summer monsoon to global warming: Mean and synoptic systems. J. Climate, 22 , 10141036.

    • Search Google Scholar
    • Export Citation
  • Terray, P., P. Delecluse, S. Labattu, and L. Terray, 2003: Sea surface temperature associations with the late Indian summer monsoon. Climate Dyn., 21 , 593618.

    • Search Google Scholar
    • Export Citation
  • Uppala, S. M., and Coauthors, 2005: The ERA-40 Re-Analysis. Quart. J. Roy. Meteor. Soc., 131 , 29613012.

  • Watanabe, M., and F. F. Jin, 2003: A moist linear baroclinic model: Coupled dynamical–convective response to El Niño. J. Climate, 16 , 11211139.

    • Search Google Scholar
    • Export Citation
  • Webster, P. J., V. O. Magaña, T. N. Palmer, J. Shukla, R. A. Tomas, M. Yanai, and T. Yasunari, 1998: Monsoons: Processes, predictability, and the prospects for prediction. J. Geophys. Res., 103 , 1445114510.

    • Search Google Scholar
    • Export Citation
  • Webster, P. J., A. M. Moore, J. P. Loschnigg, and R. R. Leben, 1999: Coupled oceanic-atmospheric dynamics in the Indian Ocean during 1997-8. Nature, 401 , 356360.

    • Search Google Scholar
    • Export Citation
  • Xie, P., and P. A. Arkin, 1997: Global precipitation: A 17-year monthly analysis based on gauge observations, satellite estimates, and numerical model outputs. Bull. Amer. Meteor. Soc., 78 , 25392558.

    • Search Google Scholar
    • Export Citation
  • Xie, S-P., H. Annamalai, F. A. Schott, and J. P. McCreary, 2002: Structure and mechanisms of south Indian Ocean climate variability. J. Climate, 15 , 864878.

    • Search Google Scholar
    • Export Citation
  • Xie, S-P., K. Hu, J. Hafner, H. Tokinaga, Y. Du, G. Huang, and T. Sampe, 2009: Indian Ocean capacitor effect on Indo–western Pacific climate during the summer following El Niño. J. Climate, 22 , 730747.

    • Search Google Scholar
    • Export Citation
  • Zhang, G. J., and N. A. McFarlane, 1995: Role of convective-scale momentum transport in climate simulation. J. Geophys. Res., 100 , 14171426.

    • Search Google Scholar
    • Export Citation
  • Zhang, G. J., and M. Mu, 2005: Effects of modifications to the Zhang-McFarlane convection parameterization on the simulation of the tropical precipitation in the National Center for Atmospheric Research Community Climate Model, version 3. J. Geophys. Res., 110 , D09109. doi:10.1029/2004JD005617.

    • Search Google Scholar
    • Export Citation

Fig. 1.
Fig. 1.

(a) Standardized JJAS mean all-Indian monsoon rainfall index (Parthasarathy et al. 1995; solid line) and the years with strong winter El Niños before the summer monsoon season (squares). Strong winter El Niño years are defined as those November–January means with a standardized Niño-3 value >1.4. (b) Same as in (a), but for late season (August–September) Indian monsoon rainfall.

Citation: Journal of Climate 23, 4; 10.1175/2009JCLI2916.1

Fig. 2.
Fig. 2.

Hovmöller diagram (40°–90°E longitudinal mean) of anomalous ERA-40 SST (K) following the (a) 1982/83 and (b) 1997/98 major winter El Niño events.

Citation: Journal of Climate 23, 4; 10.1175/2009JCLI2916.1

Fig. 3.
Fig. 3.

Seasonal transitions of anomalous ERA-40 SSTs (K) during the summer monsoon season (June–August) for the (a) 1983 and (b) 1998 events. (c),(d) Same as in (a),(b), but for the spring season (March–May).

Citation: Journal of Climate 23, 4; 10.1175/2009JCLI2916.1

Fig. 4.
Fig. 4.

(a) Hovmöller diagram (50°–83°E longitudinal mean) of anomalous ERA-40 surface latent heat flux (shadings: warm colors imply more evaporation from the surface than the seasonal mean; W m−2) and anomalous surface wind speed (contours: m s−1) for the (a) 1982/83 and (b) 1997/98 events.

Citation: Journal of Climate 23, 4; 10.1175/2009JCLI2916.1

Fig. 5.
Fig. 5.

June–August mean CMAP precipitation anomalies (shadings: mm day−1) and ERA-40 surface wind anomalies (vectors: m s−1) for the (a) 1983 and (b) 1998 events.

Citation: Journal of Climate 23, 4; 10.1175/2009JCLI2916.1

Fig. 6.
Fig. 6.

June–August mean anomalous moist stability (warm colors contribute to strengthening the moist convection; W m−2) for the (a) 1983 and (b) 1998 events. Green contours indicate the pressure velocity (positive values imply upward motion). The units are 0.01 Pa s−1.

Citation: Journal of Climate 23, 4; 10.1175/2009JCLI2916.1

Fig. 7.
Fig. 7.

Total moist process: anomalous moisture transport plus anomalous moist stability (W m−2) for the (a) 1983 and (b) 1998 events. Light (dark) shadings contribute to strengthening (weakening) of the moist convection. Absolute values greater than 20 are shaded, and the contour interval is 20 W m−2.

Citation: Journal of Climate 23, 4; 10.1175/2009JCLI2916.1

Fig. 8.
Fig. 8.

SST anomalies (20°S–20°N, 210°–260°E mean) imposed in idealized El Niño experiments using CCM3. The solid curve is the original spring El Niño experiment and the dotted curve is the no-spring El Niño sensitivity experiment (see section 5a).

Citation: Journal of Climate 23, 4; 10.1175/2009JCLI2916.1

Fig. 9.
Fig. 9.

Anomalous SSTs simulated by the spring El Niño experiment. (a) July–August mean (K) and (b) the difference between the July–August mean and the May–June mean (July–Aug minus May–June). (c),(d) Same as in (a),(b), but for the no-spring El Niño experiment.

Citation: Journal of Climate 23, 4; 10.1175/2009JCLI2916.1

Fig. 10.
Fig. 10.

Anomalous surface latent heat flux (warm colors imply more evaporation from the surface than seasonal mean; W m−2) simulated by the spring El Niño experiment during (a) May–June and (b) August–September. Vectors indicate surface wind anomalies (m s−1). Also shown are differences in the surface latent heat flux and surface wind anomalies between the spring El Niño experiment and no-spring El Niño experiments during (c) May–June and (d) August–September.

Citation: Journal of Climate 23, 4; 10.1175/2009JCLI2916.1

1

The definition of “early” versus “late” monsoon seasons is somewhat arbitrary. However, since the anomalous surface latent heat flux begins rebounding in August (see section 3b) and has maximum effect in September, we contrast the AS period with the JJ period.

2

We present June–August mean precipitation to smooth out monsoon intraseasonal variability. Use of the July–September or August–September mean precipitation indicates stronger monsoon rainfall over the Indian subcontinent for both the 1983 and 1998 events (not shown).

Save
  • Annamalai, H., P. Liu, and S-P. Xie, 2005: Southwest Indian Ocean SST variability: Its local effect and remote influence on Asian monsoons. J. Climate, 18 , 41504167.

    • Search Google Scholar
    • Export Citation
  • Back, L. E., and C. S. Bretherton, 2006: Geographic variability in the export of moist static energy and vertical motion profiles in the tropical Pacific. Geophys. Res. Lett., 33 , L17810. doi:10.1029/2006GL026672.

    • Search Google Scholar
    • Export Citation
  • Chiang, J. C. H., and B. R. Lintner, 2005: Mechanism of remote tropical surface warming during El Niño. J. Climate, 18 , 41304149.

  • Chikamoto, Y., and Y. Tanimoto, 2006: Air-sea humidity effects on the generation of tropical Atlantic SST anomalies during the ENSO events. Geophys. Res. Lett., 33 , L19702. doi:10.1029/2006GL027238.

    • Search Google Scholar
    • Export Citation
  • Chou, C., and J. D. Neelin, 2004: Mechanisms of global warming impacts on regional tropical precipitation. J. Climate, 17 , 26882701.

  • Chou, C., J. D. Neelin, J-Y. Tu, and C-T. Chen, 2006: Regional tropical precipitation change mechanisms in ECHAM4/OPYC3 under global warming. J. Climate, 19 , 42074223.

    • Search Google Scholar
    • Export Citation
  • Chou, C., J. D. Neelin, C-A. Chen, and J-Y. Tu, 2009: Evaluating the “rich-get-richer” mechanism in tropical precipitation change under global warming. J. Climate, 22 , 19822005.

    • Search Google Scholar
    • Export Citation
  • Chung, E. C., and V. Ramanathan, 2006: Weakening of north Indian SST gradients and the monsoon rainfall in India and the Sahel. J. Climate, 19 , 20362045.

    • Search Google Scholar
    • Export Citation
  • Du, Y., S-P. Xie, G. Huang, and K. Hu, 2009: Role of air–sea interaction in the long persistence of El Niño–induced north Indian Ocean warming. J. Climate, 22 , 20232038.

    • Search Google Scholar
    • Export Citation
  • Huang, B., and J. Shukla, 2007: On the mechanisms for the interannual variability in the tropical Indian Ocean. Part I: The role of remote forcing from tropical Pacific. J. Climate, 20 , 29172936.

    • Search Google Scholar
    • Export Citation
  • Kiehl, J. T., J. J. Hack, G. B. Bonan, B. A. Boville, D. L. Williamson, and P. J. Rasch, 1998: The National Center for Atmospheric Research Community Climate Model: CCM3. J. Climate, 11 , 11311149.

    • Search Google Scholar
    • Export Citation
  • Kitoh, A., S. Yukimoto, A. Noda, and T. Motoni, 1997: Simulated changes in the Asian summer monsoon at times of increased atmospheric CO2. J. Meteor. Soc. Japan, 75 , 10191031.

    • Search Google Scholar
    • Export Citation
  • Lau, N-C., A. Leetmaa, M. J. Nath, and H-L. Wang, 2005: Influences of ENSO-induced Indo–western Pacific SST anomalies on extratropical atmospheric variability during the boreal summer. J. Climate, 18 , 29222942.

    • Search Google Scholar
    • Export Citation
  • Parthasarathy, B., A. A. Munot, and D. R. Kothawale, 1995: Monthly and seasonal rainfall series for All-India homogeneous regions and meteorological subdivisions: 1871-1994. Indian Institute of Tropical Meteorology Research Rep. RR-065, 113 pp.

    • Search Google Scholar
    • Export Citation
  • Rayner, N. A., D. E. Parker, E. B. Horton, C. K. Folland, L. V. Alexander, D. P. Rowell, E. C. Kent, and A. Kaplan, 2003: Global analyses of sea surface temperature, sea ice, and night marine air temperature since the late nineteenth century. J. Geophys. Res., 108 , 4407. doi:10.1029/2002JD002670.

    • Search Google Scholar
    • Export Citation
  • Reynolds, R. W., N. A. Rayner, T. M. Smith, D. C. Stokes, and W. Wang, 2002: An improved in situ and satellite SST analysis for climate. J. Climate, 15 , 16091625.

    • Search Google Scholar
    • Export Citation
  • Shukla, J., 1995: Predictability of the tropical atmosphere, the tropical oceans and TOGA. Proceedings of the International Scientific Conference on Tropical Ocean Global Atmosphere (TOGA) Programme, Vol. 2, WCRP-91, WMO/TD 717, 725–730.

    • Search Google Scholar
    • Export Citation
  • Shukla, J., 2007: Monsoon mysteries. Science, 318 , 204205.

  • Stowasser, M., H. Annamalai, and J. Hafner, 2009: Response of the South Asian summer monsoon to global warming: Mean and synoptic systems. J. Climate, 22 , 10141036.

    • Search Google Scholar
    • Export Citation
  • Terray, P., P. Delecluse, S. Labattu, and L. Terray, 2003: Sea surface temperature associations with the late Indian summer monsoon. Climate Dyn., 21 , 593618.

    • Search Google Scholar
    • Export Citation
  • Uppala, S. M., and Coauthors, 2005: The ERA-40 Re-Analysis. Quart. J. Roy. Meteor. Soc., 131 , 29613012.

  • Watanabe, M., and F. F. Jin, 2003: A moist linear baroclinic model: Coupled dynamical–convective response to El Niño. J. Climate, 16 , 11211139.

    • Search Google Scholar
    • Export Citation
  • Webster, P. J., V. O. Magaña, T. N. Palmer, J. Shukla, R. A. Tomas, M. Yanai, and T. Yasunari, 1998: Monsoons: Processes, predictability, and the prospects for prediction. J. Geophys. Res., 103 , 1445114510.

    • Search Google Scholar
    • Export Citation
  • Webster, P. J., A. M. Moore, J. P. Loschnigg, and R. R. Leben, 1999: Coupled oceanic-atmospheric dynamics in the Indian Ocean during 1997-8. Nature, 401 , 356360.

    • Search Google Scholar
    • Export Citation
  • Xie, P., and P. A. Arkin, 1997: Global precipitation: A 17-year monthly analysis based on gauge observations, satellite estimates, and numerical model outputs. Bull. Amer. Meteor. Soc., 78 , 25392558.

    • Search Google Scholar
    • Export Citation
  • Xie, S-P., H. Annamalai, F. A. Schott, and J. P. McCreary, 2002: Structure and mechanisms of south Indian Ocean climate variability. J. Climate, 15 , 864878.

    • Search Google Scholar
    • Export Citation
  • Xie, S-P., K. Hu, J. Hafner, H. Tokinaga, Y. Du, G. Huang, and T. Sampe, 2009: Indian Ocean capacitor effect on Indo–western Pacific climate during the summer following El Niño. J. Climate, 22 , 730747.

    • Search Google Scholar
    • Export Citation
  • Zhang, G. J., and N. A. McFarlane, 1995: Role of convective-scale momentum transport in climate simulation. J. Geophys. Res., 100 , 14171426.

    • Search Google Scholar
    • Export Citation
  • Zhang, G. J., and M. Mu, 2005: Effects of modifications to the Zhang-McFarlane convection parameterization on the simulation of the tropical precipitation in the National Center for Atmospheric Research Community Climate Model, version 3. J. Geophys. Res., 110 , D09109. doi:10.1029/2004JD005617.

    • Search Google Scholar
    • Export Citation
  • Fig. 1.

    (a) Standardized JJAS mean all-Indian monsoon rainfall index (Parthasarathy et al. 1995; solid line) and the years with strong winter El Niños before the summer monsoon season (squares). Strong winter El Niño years are defined as those November–January means with a standardized Niño-3 value >1.4. (b) Same as in (a), but for late season (August–September) Indian monsoon rainfall.

  • Fig. 2.

    Hovmöller diagram (40°–90°E longitudinal mean) of anomalous ERA-40 SST (K) following the (a) 1982/83 and (b) 1997/98 major winter El Niño events.

  • Fig. 3.

    Seasonal transitions of anomalous ERA-40 SSTs (K) during the summer monsoon season (June–August) for the (a) 1983 and (b) 1998 events. (c),(d) Same as in (a),(b), but for the spring season (March–May).

  • Fig. 4.

    (a) Hovmöller diagram (50°–83°E longitudinal mean) of anomalous ERA-40 surface latent heat flux (shadings: warm colors imply more evaporation from the surface than the seasonal mean; W m−2) and anomalous surface wind speed (contours: m s−1) for the (a) 1982/83 and (b) 1997/98 events.

  • Fig. 5.

    June–August mean CMAP precipitation anomalies (shadings: mm day−1) and ERA-40 surface wind anomalies (vectors: m s−1) for the (a) 1983 and (b) 1998 events.

  • Fig. 6.

    June–August mean anomalous moist stability (warm colors contribute to strengthening the moist convection; W m−2) for the (a) 1983 and (b) 1998 events. Green contours indicate the pressure velocity (positive values imply upward motion). The units are 0.01 Pa s−1.

  • Fig. 7.

    Total moist process: anomalous moisture transport plus anomalous moist stability (W m−2) for the (a) 1983 and (b) 1998 events. Light (dark) shadings contribute to strengthening (weakening) of the moist convection. Absolute values greater than 20 are shaded, and the contour interval is 20 W m−2.

  • Fig. 8.

    SST anomalies (20°S–20°N, 210°–260°E mean) imposed in idealized El Niño experiments using CCM3. The solid curve is the original spring El Niño experiment and the dotted curve is the no-spring El Niño sensitivity experiment (see section 5a).

  • Fig. 9.

    Anomalous SSTs simulated by the spring El Niño experiment. (a) July–August mean (K) and (b) the difference between the July–August mean and the May–June mean (July–Aug minus May–June). (c),(d) Same as in (a),(b), but for the no-spring El Niño experiment.

  • Fig. 10.

    Anomalous surface latent heat flux (warm colors imply more evaporation from the surface than seasonal mean; W m−2) simulated by the spring El Niño experiment during (a) May–June and (b) August–September. Vectors indicate surface wind anomalies (m s−1). Also shown are differences in the surface latent heat flux and surface wind anomalies between the spring El Niño experiment and no-spring El Niño experiments during (c) May–June and (d) August–September.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 772 214 32
PDF Downloads 577 107 12