Show Summary Details

Page of

date: 12 December 2017

# Interannual Variability of the Indian Monsoon and Its Link to ENSO

## Summary and Keywords

The interannual variability of Indian summer monsoon is probably one of the most intensively studied phenomena in the research area of climate variability. This is because even relatively small variations of about 10% to 20% from the mean rainfall may have dramatic consequences for regional agricultural production. Forecasting such variations months in advance could help agricultural planning substantially. Unfortunately, a perfect forecast of Indian monsoon variations, like any other regional climate variations, is impossible in a long-term prediction (that is, more than 2 weeks or so in advance). The reason is that part of the atmospheric variations influencing the monsoon have an inherent predictability limit of about 2 weeks. Therefore, such predictions will always be probabilistic, and only likelihoods of droughts, excessive rains, or normal conditions may be provided. However, even such probabilistic information may still be useful for agricultural planning. In research regarding interannual Indian monsoon rainfall variations, the main focus is therefore to identify the remaining predictable component and to estimate what fraction of the total variation this component accounts for. It turns out that slowly varying (with respect to atmospheric intrinsic variability) sea-surface temperatures (SSTs) provide the dominant part of the predictable component of Indian monsoon variability. Of the predictable part arising from SSTs, it is the El Niño Southern Oscillation (ENSO) that provides the main part. This is not to say that other forcings may be neglected. Other forcings that have been identified are, for example, SST patterns in the Indian Ocean, Atlantic Ocean, and parts of the Pacific Ocean different from the traditional ENSO region, and springtime snow depth in the Himalayas, as well as aerosols. These other forcings may interact constructively or destructively with the ENSO impact and thus enhance or reduce the ENSO-induced predictable signal. This may result in decade-long changes in the connection between ENSO and the Indian monsoon. The physical mechanism for the connection between ENSO and the Indian monsoon may be understood as large-scale adjustment of atmospheric heatings and circulations to the ENSO-induced SST variations. These adjustments modify the Walker circulation and connect the rising/sinking motion in the central-eastern Pacific during a warm/cold ENSO event with sinking/rising motion in the Indian region, leading to reduced/increased rainfall.

# Introduction

In spring to summer 1997, the strongest El Niño (warm phase of El Niño Southern Oscillation─ENSO) event in recent history developed and peaked in late 1997/early 1998. Since there were precursors of this El Niño event in the western Pacific Ocean subsurface in January 1997, it was predicted 6 to 9 months in advance, which turned out to be one of the biggest success stories in seasonal climate prediction (Ross et al., 1998; Trenberth et al., 2002). By that time, the ENSO-Indian monsoon relationship was well established using observational data and general circulation models (GCMs; Fennessy et al., 1994; Ju & Slingo, 1995; Rasmusson & Carpenter, 1983; Sikka, 1980; Walker, 1924; Webster & Yang, 1992). El Niño is typically associated with drought conditions and La Niña with wet conditions in the Indian monsoon region. On the basis of this there was an expectation for a strong monsoon drought for the 1997 Indian summer monsoon (even though it has to be said that the actual prediction of the Indian Meteorological Department (IMD) at the end of May 1997, based on a multivariate statistical forecasting scheme, was for a normal monsoon). It turned out that the overall Indian monsoon rainfall in summer 1997 was slightly (2%) above normal. This prediction failure has spurred more research into the complex relationship between ENSO and the Indian monsoon (Annamalai et al., 2007; Annamalai & Liu, 2005; Jang & Straus, 2012; Jourdain et al., 2013; Krishna Kumar et al., 2006; Krishnamurthy & Goswami, 2000; Lau & Nath, 2000; Slingo & Annamalai, 2000; Sperber et al., 2014; Wang et al., 2003; Webster et al., 1998). On the other hand, the El Niño event of 2015, which was of similar magnitude as the 1997 event, caused, as predicted, a substantial Indian monsoon drought with a deficit of about 14%, according to the IMD.

Before discussing the ENSO-Indian monsoon relationship, it is useful to review some basic theories related to the Indian monsoon system. Monsoon circulations are traditionally described as continental-scale “sea-breeze” phenomena, induced by different heat capacities between land masses and the ocean (Wallace & Hobbs, 1977). For the Indian summer monsoon, this implies a north–south contrast between the Eurasian land mass and the Indian Ocean as the main driving mechanism (Chou, 2003; Li & Yanai, 1996; Meehl, 1994; Turner & Annamalai, 2012; Webster, 1987). The land–sea contrast can be further enhanced by the orographic insulation provided by the Himalayas and Tibetan plateau (Boos & Kuang, 2010). Chao and Chen (2001) interpreted the monsoon as the intertropical convergence zone (ITCZ) shifted away from the equator, and also highlighted the potential role of the Indo-Pacific heating maximum. The land–sea contrast enhances the northward ITCZ shift. Chen (2003) proposed a more global view of the monsoon systems and suggested in particular that the Indian monsoon is mainly forced by the east–west heating contrast between the tropical western Pacific and the Atlantic region. This contrast is related to the mean global Walker circulation, which has its rising branch in the eastern Indian Ocean/western Pacific region and sinking branches in the tropical eastern Pacific and Atlantic regions. Chen (2003) showed, using observational data, that the Tibetan high may be interpreted as a consequence of Sverdrup balance (Rodwell & Hoskins, 2001) resulting from the upper-level divergence-convergence contrast between the tropical western Pacific and the Atlantic region. Indeed, Kucharski et al. (2011) confirmed with GCM simulations that the southern part of the South Asian monsoon high is mainly induced by the east–west heating contrast. However, they also found that the northern part of the monsoon high is mainly due to the land–sea contrast.

As described in the introduction, there can be deviations of the expected response to an individual ENSO event, and, in a more general sense, multidecadal variations of the correlation between ENSO and the Indian monsoon have been identified. The ENSO–Indian monsoon relationship has weakened in recent decades.

A simple null hypothesis for variations in the relationship between ENSO and the Indian monsoon for single events and also for changes in the long-term relationship is atmospheric internal variability, the unpredictable climate noise on a seasonal time scale (Cash et al., 2016; Gershunov et al., 2001; Kang et al., 2004). But other mechanisms have also been proposed as possible mechanisms for the changes in the ENSO–Indian monsoon relationship. These mechanisms are the interference of ENSO with the Indian Ocean dipole mode (IOD; Saji et al., 1999), interference with SST anomalies in the tropical south Atlantic or Atlantic zonal mode (Kucharski et al., 2007), decadal changes in the atmospheric circulation in the extratropical North Pacific (Kinter et al., 2002), changes in the Atlantic circulation that influence Eurasian snow cover (Chang et al., 2001), changes in the ENSO characteristics (Krishna Kumar et al., 2006), and global warming (Krishna Kumar et al., 1999). Krisnamurthy and Krishnamurthy (2014) proposed that Pacific decadal variability related to the Pacific Decadal Oscillation can also modulate the ENSO–Indian monsoon relationship. The fact that the above-mentioned factors can modify the ENSO–Indian monsoon relation also means that they may individually affect the Indian monsoon interannual variability.

This article revisits some key features of the interannual Indian monsoon variability and its relationship with ENSO.

# Indian Monsoon Climatology and Interannual Variability

The June-to-September (JJAS) Indian monsoon climatology for the period 1981 to 2015, derived from GPCP data (Figure 1a; Adler et al., 2003) is characterized by heavy rainfall on the western Ghats, central parts of India, and the foothills of the Himalayas, reaching more than 7 mm/day in many regions. The interannual rainfall standard deviation (Figure 1b) is only a small fraction of the climatology, and amounts to about 10% to 20% of the total Indian monsoon rainfall.

Click to view larger

Figure 1. (a) GPCP JJAS precipitation climatology, (b) GPCP JJAS interannual standard deviation, (c) NCEP/NCAR re-analysis JJAS 200-hPa velocity potential climatology, and (d) NCEP/NCAR re-analysis JJAS 200-hPa stream function. Units are mm/day for (a) and (b) and 106 m2/s for (c) and (d).

In order to give an idea of the large-scale features associated with the Indian monsoon, the climatologies of the JJAS 200-hPa velocity potential (Figure 1c) and eddy stream function (Figure 1d) are shown (both from the NCEP/NCAR re-analysis; Kalnay et al., 1996). A prominent feature is the upper-level anticyclone north of India, the Tibetan high, which is seen in Figure 1d as maximum of the eddy stream function. The Tibetan high is an integral part of Indian monsoon climatology, and is driven by elevated heating in the Tibetan plateau region (Boos & Kuang, 2010). Also, the upper-level velocity potential distribution is very important for Indian monsoon climatology, although the maxima are out of phase with the Indian monsoon (Figure 1c). The minimum in upper-level velocity potential is located in the western Pacific, representing the center of upper-level divergence, which is the adjustment to the western Pacific heating maximum (rising branch of the Walker circulation). Centers of maximum in velocity potential are located in the African-South Atlantic region and in the eastern Pacific, and mark the sinking branches of the Walker circulation. A part of the upper-level anticylone with its center north of India can be attributed to the dynamical adjustment to the upper-level velocity potential east–west contrast through Sverdrup balance (Chen, 2003; Kucharski et al., 2011). This element is important for the interpretation of Indian monsoon variations induced by ENSO.

# Relationship Between Indian Monsoon Rainfall and Tropical SSTs

In order to investigate interannual rainfall variability patterns in the Indian region that are related to tropical SST forcing, a canonical correlation analysis (CCA) was performed, which was based on a previous principal component analysis in order to reduce the degrees of freedom (Bretherton et al., 1992). A similar analysis (maximum covariance analysis; Bretherton et al., 1992) was applied to find the Indian monsoon rainfall coupled modes with SSTs (Mishra et al., 2012; Syed & Kucharski, 2016). The analysis was between the JJAS monsoon rainfall in the domain 60°–100°E, 0°–35°N and SST reconstructions from the United Kingdom using the Met office’s Hadley Centre Sea Ice and Sea Surface Temperature data set, version 1.1 (HadISST 1.1; Rayner et al., 2003) in the global tropical band from 40°S to 40°N. The period for the analysis was 1981 to 2015. The patterns shown are homogeneous regression maps derived from the covariance of the normalized principal component (PC) time series with the corresponding anomaly fields. The first mode (Figure 2) has a squared covariance fraction of 0.7 and clearly shows a positive rainfall anomaly in the central-western Indian region, whereas a slightly negative anomaly is found in the northeastern Indian region, extending into the Bay of Bengal (Figure 2a) in correspondence to an eastern Pacific cooling (Figure 2b).

Click to view larger

Figure 2. (a) First canonical correlation analysis mode regression pattern for the observed (GPCP) Indian JJAS rainfall for the domain 60°–100°E, 0°–35°N and (b) the corresponding first mode in sea-surface temperature (SST) anomalies for the period 1981–2015. The box shows the Indian summer monsoon rainfall index (ISMRI) for the domain 70°–80°E, 8°–20°N. Units are mm/day for rainfall and K for SST.

Such a pattern is principally in agreement with previous findings for the ENSO–Indian monsoon link, as it suggests an overall drying in the Indian region related to warm ENSO conditions. However, in previous studies using longer data sets (Mishra et al., 2012; Syed & Kucharski, 2016) a more widespread drying in the Indian region was found in relation to El Niño conditions. The less pronounced anomaly in the Indian region seen in the first CCA mode for the more recent period considered here is consistent with the weakening of the ENSO–monsoon relationship in the recent decades, and a similar pattern has also been found in Fig. 13a of Kucharski et al. (2011). The correlation of the corresponding PC time series for the first CCA mode is 0.84. The second mode (0.56 squared covariance fraction) clearly shows important impacts from the eastern pole of the Indian Ocean dipole mode and the tropical south Atlantic SSTs, as also found in Syed and Kucharski (2016), and is well separated from the third mode (0.13 squared covariance fraction; not shown). Given that eastern Pacific SSTs, which are closely related to ENSO, provide the strongest coupled feedback (it can be assumed to be a forcing), the ENSO–Indian monsoon relationship is analyzed.

Based on these results, a simple central-western Indian rainfall index is defined as area-averaged precipitation in the domain 70°–80°E, 8°–20°N, and is referred to as the Indian summer monsoon rainfall index (ISMRI). The ISMRI encompasses the homogeneous region identified in the first CCA mode over the Indian region (see black rectangle in Figure 2a). Figure 3a shows the rainfall regression of the index (calculated from the covariance of the normalized ISMRI with rainfall). The resulting map looks very similar to the first CCA mode shown in Figure 2a.

Click to view larger

Figure 3. Regression of ISMRI onto (a) observed rainfall, (b) the 700-hPa geopotential height (Z700) and 700-hPa winds, (c) 200-hPa eddy stream function, and (d) 200-hPa velocity potential. Units are mm/day for (a), m/s for winds and m for Z700 in (b), and 106 m2/s for (c) and (d). The 95% statistically significant values are shown as shaded.

In order to investigate the circulation anomalies that go along with the precipitation anomalies, the 700-hPa height and wind vector regressions onto ISMRI are shown in Figure 3b. As expected, increased rainfall over India is related to a cyclonic low-level circulation anomaly over India and the Arabian Sea region. The teleconnections related to this cyclonic anomaly may be best investigated by considering upper-level fields that are associated with it. Figure 3c shows the 200-hPa eddy stream function and Figure 3d the velocity potential regressions onto the ISMRI. The upper-level velocity potential clearly shows a minimum in the Indian region, which indicates upper-level divergence and forces rising motion. The upper-level eddy stream function regression indicates a westward shift of the upper-level monsoon high. On a larger scale, the upper-level velocity potential regression in particular suggests a link of the ISMRI with the eastern Pacific, where a maximum of the upper-level velocity potential can be found, indicating upper-level convergence there, which is consistent with downward motion induced by a cold ENSO event.

# ENSO–Indian Monsoon Relationship

In order to analyze more directly the forcing from ENSO to the Indian monsoon, in the following a simple linear regression analysis is performed (given the relatively short record of satellite-era data, this robust method seems adequate). Figure 4a shows the observed rainfall regression onto the Niño3.4 index (defined here as the area-average SST in the region 190°–240°E, 5°S–5°N) for the period 1981 to 2015 (calculated from the covariance of the normalized Niño3.4 index with rainfall).

Click to view larger

Figure 4. Regression of the Niño3.4 index onto (a) observed rainfall, (b) the 700-hPa geopotential height (Z700) and 700-hPa winds, (c) 200-hPa eddy stream function, and (d) 200-hPa velocity potential. Units are mm/day for (a), m/s for winds and m for Z700 in (b), and 106 m2/s for (c) and (d). The 95% statistically significant values are shown as shaded.

The resulting map looks very similar to the first CCA mode shown in Figure 2a in the Indian region, confirming that the first CCA SST mode is indeed closely related to ENSO. Figure 4b shows the 700-hPa height and wind regression map, and indicates an anticyclone with center over the southern Arabian Peninsula, but extending toward India (which lies on the edge of the low-level high pressure induced by ENSO). Figure 4c and Figure 4d show the 200-hPa eddy stream function and velocity potential regressions, respectively. It is clear from Figure 4d that India lies in the broad maximum of upper-level velocity potential, which is dynamically related to convergent motion and therefore upper-level convergence. On the other hand, in terms of the eddy stream function regression, India lies in the region with a gradient in this field, indicating a substantial shift of the upper-level monsoon high toward the east in warm ENSO years. Such a shift is also seen in the precipitation response in Figure 4a (e.g., the dipole between India and the Bay of Bengal). Kucharski et al. (2011) interpreted this shift in precipitation as a consequence of the shifts in upper-level velocity potential and stream function. As already discussed, whereas India clearly lies in the sinking branch of the upper-level velocity potential adjustment, India also lies in a region with a sharp gradient in the upper-level stream function adjustment. The upper-level stream function adjustment leads to an increased monsoon high as a response to the warm phase of ENSO to the east of India (in the Bay of Bengal, Myanmar, Thailand, South China Sea region). Such an increased upper-level monsoon high is related to increased monsoonal precipitation because the response is baroclinic, and therefore reversed at low levels and will induce cyclonic circulation anomalies. Such low-level cyclonic circulation anomalies induce a precipitation increase due to Ekman pumping. This suggests that small shifts in these responses may lead the cyclonic low-level circulation anomalies (that on average lie to the east of India) to also affect India and cause a contribution to positive rainfall anomalies there. This is indeed seen in Figure 4a and Figure 2a for the period 1981 to 2015 in which the ENSO–Indian monsoon relationship weakened. The complex responses may be part of the reason why the ENSO–Indian monsoon relationship is so fragile and appears to break down in individual ENSO events, or even for longer periods.

Wang et al. (2003) proposed that the strongly enhanced drying response over India could be related to the drying over the Maritime Continent during an El Niño event, which is due to the shifting Walker circulation. The drying in the Maritime Continent means reduced heating in that region, with an equatorial Rossby-wave response to the west, which then induces a tilted band of reduced rainfall extending into India (see Figure 4a). The upper-level stream function adjustment with an increased monsoon high as a response to the warm phase of ENSO in the Bay of Bengal, Myanmar, Thailand, South China Sea region could counteract the overall reduced heating in the eastern Indian Ocean region and thus interfere destructively with the Maritime Continent-induced response. This interference could be another reason for the fragile relationship between ENSO and the Indian monsoon.

For the period 1981 to 2015, it is therefore proposed that the weakened overall ENSO–Indian monsoon relationship compared to the pre-1980 period is likely linked to the rainfall dipole in Figure 1a between the southwestern and northeastern Indian regions. For the warm phase of ENSO, this means that the tendency for a reduction in rainfall induced by the upper-level convergence and sinking in the Indian region related to warm eastern Pacific SSTs (as indicated by the velocity potential adjustments seen in Figure 3d and Figure 4d) is partially offset by the upper-level anticyclonic stream function adjustment with maximum to the east of India (Figure 4c), which extends into the northeastern parts of the Indian region, and could thus strengthen the upper-level monsoon high there. Such a precipitation dipole induced by ENSO is absent if longer periods (that is including pre-1980 data) are considered (Mishra et al., 2012; Syed & Kucharski, 2016). It is beyond the purpose of this article to investigate in detail the reasons for the enhanced competition between the velocity potential and stream function adjustments in the Indian region, but it is possible that several of the proposed mechanisms discussed in the introduction are contemporaneously at work.

# Impacts of Other Forcings on Indian Interannual Rainfall Variability

## Indian Ocean

The Indian Ocean dipole (IOD) has been suggested by many studies as potentially negatively interfering with the ENSO impact (Ashok et al., 2001, 2004; Ashok & Saji, 2007; Cherchi et al., 2007; Clark et al., 2000; Krishnamurthy & Shukla, 2000; Murtugudde et al., 2000; Saji et al., 1999; Webster et al., 1998, 1999). The IOD mode peaks in the boreal autumn, but in its growing phase, in the summer season, it shows a substantial correlation with ENSO, counteracting the ENSO impact (Bracco et al., 2005, 2007; Cherchi & Navarra, 2013). A regression of the IOD index onto rainfall shows a more complex spatial structure than that of ENSO (see, for example, Fig. 1 of Syed & Kucharski, 2016), but overall rainfall is increased. Since the IOD is positively correlated with ENSO, the two effects tend to cancel each other, but in general ENSO provides the stronger impact.

There is another quite interesting, but less studied, impact from the Indian Ocean basin mode (Xie et al., 2009; Yang et al., 2007). This basin mode typically develops during ENSO events in boreal winter and can persist into the following summer to influence the Asian monsoon, including India. El Niño leads to a warm, and La Niña to a cold, basin-wide Indian Ocean. In this respect, one may interpret its impact as a delayed ENSO impact, but with opposite sign to the direct ENSO impact (see, e.g., Fig. 4 of Yang et al., 2007). This delayed impact may lead to some predictability about 6 months after an ENSO event, and could also interfere with the direct ENSO impact.

## Atlantic Ocean

The Atlantic Ocean hosts SST modes that may influence the Indian monsoon. In particular, the tropical South Atlantic (TSA) and the Atlantic zonal mode have been identified as potential drivers for Indian monsoon variability (Barimalala et al., 2011; Kucharski et al., 2007, 2008, 2009; Kucharski & Joshi, 2017; Losada et al., 2010; Pottapinjara et al., 2014, 2015; Wang et al., 2009; Yadav, 2009, 2016). A warm TSA or Atlantic zonal mode (the two are highly positively correlated) induces a Gill-type response to tropical Atlantic heating, with a pair of low-level anticyclonic circulations off the equator to the east. The northern part of the response is over the Arabian Sea and India and induces low-level divergence through Ekman pumping as well as a weakening of the Somali Jet. Both lead to a reduction of rainfall over India (Syed & Kucharski, 2016).

Kucharski et al. (2007) showed that in the 1980s and 1990s ENSO and the TSA SSTs were anticorrelated (this anticorrelation can also be identified in the first SST CCA mode shown in Figure 2a). Since both modes in their positive phases induce overall negative rainfall anomalies over India, their anticorrelation means that there is a negative interference of the two.

Decadal variability in the North Atlantic related to the Atlantic Multidecadal Oscillation (AMO) has also been proposed to influence the Indian monsoon (Chattopadhyay et al., 2015; Goswami et al., 2006; Joshi & Rai, 2015; Zhang & Delworth, 2006). The physical mechanism for such an influence is through the modification of the tropospheric temperature gradient: In its warm phase, AMO leads to a warming of the Eurasian continent and thus to an increase of the tropospheric temperature gradient, enhancing the Indian monsoon circulation (Goswami et al., 2006).

## Pacific Ocean

SST anomalies related to Pacific decadal variability in the tropical and extratropical Indo-Pacific region expressed by either the Pacific Decadal Oscillation (PDO) or the physically closely related Interdecadal Pacific Oscillation (IPO) have also been proposed to impact the Indian monsoon (Chattopadhyay et al., 2015; Joshi & Kucharski, 2016; Joshi & Rai, 2015; Krishnamurthy & Goswami, 2000; Krishnamurthy & Krishamurthy, 2014; Kucharski et al., 2006). As for the ENSO impact on the Indian monsoon, the proposed mechanism for the PDO/IPO impact on the Indian monsoon involves a modification of the Walker circulation and also a local Indian Ocean Hadley cell adjustment (Krishnamurthy & Krishamurthy, 2014). SST anomalies in the Indian Ocean, which are part of the PDO/IPO pattern, also play a role in this teleconnection (Kucharski et al., 2006). A warm phase of the PDO/IPO leads to a drying in the Indian monsoon region, similar to the ENSO impact.

The PDO/IPO influence can interfere with the ENSO influence and can thus lead to particularly strong Indian monsoon droughts/floods if they are in phase, or to weak impacts if they are out of phase (Krishnamurthy & Krishamurthy, 2014), thus potentially modulating the ENSO–Indian monsoon relation.

## Other Influences

Many other influences not discussed in detail here may prove in future to be major sources of Indian monsoon predictability. Examples of such influences are land-surface conditions (Halder et al., 2016), Eurasian snow cover (Halder & Dirmeyer, 2016; Turner & Slingo, 2011), and aerosols (D’Errico et al., 2015; Wang et al., 2009).

# Predictability

The basis for long-range atmospheric predictability has been investigated by Shukla (1981) and Charney and Shukla (1981), who proposed that atmospheric predictability at seasonal time scale should be related to slowly varying boundary conditions. An important question is how predictable the Indian monsoon rainfall year-to-year variations are, given that there are influences from several ocean basins. There are also atmospheric internal variability patterns that influence Indian rainfall, and the atmospheric internal variability patterns may be induced by extratropical atmospheric variability (Syed et al., 2012) as well as intraseasonal monsoon variations and daily rainfall variations in active and break phases, all of which can influence the seasonal mean rainfall (see Goswami & Chakravorty, 2017). Such influences are not predictable or are much less predictable than those related to the more slowly varying tropical SSTs, although some evidence has been provided that, for example, active and break phases of the Indian monsoon could be modulated by ENSO, and thus be at least in part predictable (Dwivedi et al., 2015). In any case, only a part of the total rainfall anomaly that occurs in a particular year’s monsoon season may be predictable in a seasonal (long-range) prediction framework. The predictable part is usually referred to as the signal and the unpredictable part as the noise (Kang et al., 2004; Mohan & Goswami, 2003). It is assumed that an ensemble of multiple GCM realizations is used to predict Indian monsoon rainfall anomalies (at each grid point, or the ISMRI). In this case, the common signal in each simulation can be derived from the ensemble mean, and its time variance is referred to as the signal, S. Each individual member of such simulations will deviate from the ensemble mean and the variance (in time and for all ensemble members) of these deviations is called the noise, N. One way to measure predictability is the signal-to-noise ratio (Kang & Shukla, 2006). One may argue that a signal-to-noise ratio of 1 is a useful threshold for predictability, because the signal is just as large as the noise. Such a value is, however, rarely reached on a seasonal time scale outside the tropical Pacific region (Saha et al., 2016). Another useful measure for predictability is the theoretical limit of predictability (Abid et al., 2016; Kang & Shukla, 2006), which is based on a simple transformation of the signal-to-noise ratio:

$Display mathematics$
(1)

Rlimit varies between 0 and 1, where the result 0 means no predictability and the result 1 means perfect predictability. A signal-to-noise ratio of 1 results in an Rlimit value of about 0.7. Rlimit can be interpreted as the maximum correlation between the model ensemble mean and the observations that may be reached if the model was perfect. Please note that the Rlimit measure, as well as signal and noise, are model-dependent measures, and may also provide an underestimation of the real predictability (Eade et al., 2014). The two measures introduced here, the signal-to-noise ratio and the theoretical limit of predictability, are based on the analysis of variance derived from model ensembles. There are other measures of predictability based on model ensembles, one of which is the perfect model correlation. In this case, each ensemble member is treated as if it were an observation, and the ensemble mean, generated by all other ensemble members, is correlated with it. This will generate an ensemble of correlation coefficients (at each grid point or for an index) that is just as large as the ensemble of model simulations. The properly averaged correlation coefficients then give the perfect model correlation (Ehsan et al., 2013; Saha et al., 2016). In practice, the resulting values are typically very close to the Rlimit values, and indeed their physical interpretation is similar. There are also measures of predictability related to information theory (Delsole & Tippett, 2007), but it is beyond the scope of this review to discuss them in detail. In general, the GCM used for seasonal predictions is a coupled ocean–atmosphere model, because future SSTs also have to be predicted and interactive ocean–atmosphere coupling has been shown to be crucial for the modelling of the ENSO–monsoon relationship (Bracco et al., 2007). In order to assess the potential predictability of Indian monsoon rainfall, the hindcast data set from ECMWF-Sys4 for the period 1981–2012 was analyzed as an example. ECMWF-Sys4 is a fully coupled atmosphere–ocean general-circulation model used for seasonal mean prediction (for details, see Abid et al., 2016; Kim et al., 2012; Molteni et al., 2011). The ECMWF-Sys4 hindcast data set initialized from May 1 each year is used because this initial date is close to the monsoon but the monsoon has not yet started. A total of 15 ensemble members were simulated for 7 months each, with different atmospheric initial conditions. Figure 5 shows the Rlimit values for simulated rainfall for the Indian region.

Click to view larger

Figure 5. Rlimit values for JJAS interannual rainfall variability over the Indian domain derived from the ECMWF sys4.

As can be seen, largest values of Rlimit are seen over the ocean regions and decrease over land regions, which is a typical result (Wang et al., 2009). The values over the Indian land mass are around 0.4 to 0.5. A similar distribution of the Rlimit values has also been reported in other studies (Saha et al., 2016). Such values indicate that the explained variance of Indian monsoon rainfall variability is only about 15% to 25%. These values are somewhat smaller than the ones derived assuming perfect SST boundary conditions in AGCM multimodel intercomparison studies (typically around 0.6; Kucharski et al., 2009), or with AGCM pacemaker experiments, in which SSTs in the tropical Pacific region are prescribed, but interactive air–sea coupling in the Indian Ocean is permitted (Bracco et al., 2007; Kucharski et al., 2007). However, such a large predictability is typically not reached with real state-of-the-art forecast systems (DelSole & Shukla, 2012; Johnson et al., 2016; Li et al., 2016; Ramu et al., 2016; Wang et al., 2009) in which also SSTs are predicted, although some models do show a potential prediction skill of about 0.6 or more for an all Indian rainfall index (Saha et al., 2016).

# Outstanding Issues

Given that the actual seasonal prediction skill for the Indian summer monsoon (as for other monsoon regions) is much lower than the potential prediction skill estimated by AGCM simulations, or by AGCM pacemaker experiments, there is a lot of potential to improve simulation of the Indian monsoon interannual variability. Such improvements could be achieved by improvements of tropical SST predictions as well as their teleconnections with the Indian monsoon. The parameterization of tropical convection remains one of the main challenges in this regard; to resolve convection requires simulations with a horizontal resolution in the 1 km range. The predictability of ENSO-related SSTs is in general large compared to other SST modes, although the so-called spring barrier remains a major obstacle for the prediction of summer ENSO SST anomalies (Chen et al., 2004; Duan & Wei, 2012). Other SST modes, such as the Indian Ocean dipole or the Atlantic zonal mode, may influence the Indian monsoon, especially in non-ENSO years. However, the ability to forecast Indian monsoon rainfall in seasonal prediction systems is almost entirely due to the teleconnection with ENSO (DelSole & Shukla, 2012; Johnson et al., 2016). This indicates that there could also be a potential for improving Indian monsoon forecasts by improving the representation and prediction of SST modes other than ENSO, such as the Indian Ocean dipole or the Atlantic zonal mode, as well as their teleconnection to the Indian monsoon.

Another area for future research is the weakening of the ENSO–Indian monsoon relationship from the 1980s onward. One aspect of this weakening already highlighted is the competing influences on rainfall of the large-scale velocity potential and stream function adjustments due to ENSO in the Indian region. Further research on why this competition has increased from the 1980s onward is required, and the reason for the recently observed weakening may involve several of the previously proposed mechanisms for the weakening of the ENSO–monsoon relationship.

# Acknowledgments

The ECMWF-Sys4 data set has been provided as a contribution to the ICTP Targeted Training Program. We also thank Franco Molteni for his introduction to the ECMWF-Sys4 data set during this activity.

Wang, B. (2006). The Asian monsoon. Springer-Praxis.Find this resource:

## References

Abid, M. A., Kucharski, F., Almazroui, M., & Kang I.-S. (2016). Interannual rainfall variability and ECMWF-Sys4-based predictability over the Arabian peninsula winter monsoon region. Quarterly Journal of the Royal Meteorological Society, 142(694), 233–242.Find this resource:

Adler, R. F., Huffman, G. J., Chang, A., Ferraro, R., Xie, P., Janowiak, J., … Arkin, P. (2003). The Version 2 Global Precipitation Climatology Project (GPCP) monthly precipitation analysis (1979–present). Journal of Hydrometeorology, 4, 1147–1167.Find this resource:

Alexander, M. A., Blade, I., Newman, M., Lanzante, J. R., Lau, N. C., & Scott, J. D. (2002). The atmospheric bridge: The influence of ENSO teleconnections on air-sea interaction over the global oceans. Journal of Climate, 15, 2205–2231.Find this resource:

Annamalai, H., Hamilton, K., & Sperber, K. (2007). The South Asian summer monsoon and its relationship with ENSO in the IPCC AR4 simulations. Journal of Climate, 20, 1071–1092.Find this resource:

Annamalai, H., & Liu, P. (2005). Response of the Asian summer monsoon to changes in El Nino properties. Quarterly Journal of the Royal Meteorological Society, 131, 805–831.Find this resource:

Ashok, K., Guan, Z., Saji, N. H., & Yamagata, T. (2004). Individual and combined influences of ENSO and the Indian Ocean dipole on the Indian summer monsoon. Journal of Climate, 17, 3141–3155.Find this resource:

Ashok, K., Guan, Z., & Yamagata, T. (2001). Impact of the Indian Ocean dipole on the relationship between the Indian monsoon rainfall and ENSO. Geophysical Research Letters, 28(23), 4499–4502.Find this resource:

Ashok, K., & Saji, N. H. (2007). On the impacts of ENSO and Indian Ocean dipole events on sub-regional Indian summer monsoon rainfall. Natural Hazards, 42(2), 273–285.Find this resource:

Barimalala, R., Bracco, A., & Kucharski, F. (2011). The representation of the south tropical Atlantic teleconnection to the Indian Ocean in the AR4 coupled models. Climate Dynamics, 38(5–6), 1147–1166.Find this resource:

Boos, W., & Kuang, Z. (2010). Dominant control of the South Asian monsoon by orographic insulation versus plateau heating. Nature, 463, 218–222.Find this resource:

Bracco, A., Kucharski, F., & Molteni, F. (2005). Internal and forced modes of variability in the Indian Ocean. Geophysical Research Letters, 32, L12707.Find this resource:

Bracco, A., Kucharski, F., Molteni, F., Hazeleger, W., & Severijns, C. (2007). A recipe for simulating the interannual variability of the Asian summer monsoon and its relation with ENSO. Climate Dynamics, 28.Find this resource:

Bretherton, C. S., Smith, C., & Wallace, J. M. (1992). An intercomparison of methods for finding coupled patterns in climate data. Journal of Climate, 5(6), 541–560.Find this resource:

Cash, B. A., Barimalala, R., Kinter, J. L. III, Altshuler, E. L., Fennessy, M. J., Manganello, J. V., … Vitart, V. (2016). Sampling variability and the changing ENSO–monsoon relationship. Climate Dynamics, 48, 4071–4079.Find this resource:

Chang, C.-P., Harr, P., & Ju, J. (2001). Possible roles of Atlantic circulations on the weakening Indian monsoon rainfall–ENSO relationship. Journal of Climate, 14, 2376–2380.Find this resource:

Chao, W. C., & Chen, B. (2001). The origin of monsoons. Journal of the Atmospheric Sciences, 58, 3497–3507.Find this resource:

Charney, J. G., & Shukla, J. (1981). Predictability of monsoons. Monsoon dynamics (pp. 99–109). Cambridge, U.K.: Cambridge University Press.Find this resource:

Chattopadhyay, R., Phani, R., Sabeerali, R. P., Dhakate, C. T., Salunke, A. R., Mahapatra, K. D., … Goswami, B. N. (2015). Influence of extratropical sea-surface temperature on the Indian summer monsoon: An unexplored source of seasonal predictability. Quarterly Journal of the Royal Meteorological Society, 141, 2760–2775.Find this resource:

Chen, D., Cane, M. A., Kaplan, A., Zebiak, Z. E., & Huang, D. (2004). Predictability of El Nino over the past 148 years. Nature, 428, 733–736.Find this resource:

Chen, T.-C. (2003). Maintenance of summer Monsoon circulations: A planetary perspective. Journal of Climate, 16, 2022–2037.Find this resource:

Cherchi, A., Gualdi, S., Behera, S., Luo, J. J., Masson, S., Yamagata, T., & Navarra, A. (2007). The influence of tropical Indian Ocean SST on the Indian summer monsoon. Journal of Climate, 20(13), 3083–3105.Find this resource:

Cherchi, A., & Navarra, A. (2013). Influence of ENSO and of the Indian Ocean dipole on the Indian summer monsoon variability. Climate Dynamics, 41(1), 81–103.Find this resource:

Chou, C. (2003). Land–sea heating contrast in an idealized Asian summer monsoon. Climate Dynamics, 21, 11–25.Find this resource:

Clark, C., Cole, J., & Webster, P. (2000). Indian Ocean SST and Indian summer rainfall: Predictive relationships and their decadal variability. Journal of Climate, 13, 2503–2519.Find this resource:

DelSole, T., & Shukla, J. (2012). Climate models produce skillful predictions of Indian summer monsoon rainfall. Geophysical Research Letters, 39, L09703.Find this resource:

Delsole, T., & Tippett, M. K. (2007). Predictability: Recent insight from information theory. Reviews of Geophysics, 45, RG4002.Find this resource:

D’Errico, M., Cagnazzo, C., Fogli, P. G., Lau, W. K. M., von Hardenberg, J., Fierli, F., & Cherchi, A. (2015). Indian monsoon and the elevated‐heat‐pump mechanism in a coupled aerosol‐climate model. Journal of Geophysical Research, 120, 8712–8723.Find this resource:

Duan, W., & Wei, C. (2012). The “spring predictability barrier” for ENSO predictions and its possible mechanism: Results from a fully coupled model. International Journal of Climatology, 33, 1280–1292.Find this resource:

Dwivedi, S., Goswami, B. N., & Kucharski, F. (2015). Unraveling the missing link of ENSO control over the Indian monsoon rainfall. Geophysical Research Letters, 42, 8201–8207.Find this resource:

Eade, R., Smith, D., Scaife, A., Wallace, A., Dunstone, N., Hermanson, L., & Robinson, N. (2014). Do seasonal-to-decadal climate predictions underestimate the predictability of the real world? Geophysical Research Letters, 41, 5620–5628.Find this resource:

Ehsan, M. A., Kang, I.-S., Almazroui, M., Abid, M. A., & Kucharski, F. (2013). A quantitative assessment of changes in seasonal potential predictability for the twentieth century. Climate Dynamics, 41, 2697–2709.Find this resource:

Fennessy, M. J., Kinter, J. L. III, Kirtman, B., Marx, L., Nigam, S., Schneider, E., … Zhou, J. (1994). The simulated Indian monsoon: A GCM sensitivity study. Journal of Climate, 7, 33–43.Find this resource:

Gill, A. (1980). Some simple solutions for heat-induced tropical circulations. Quarterly Journal of the Royal Meteorological Society, 106, 447–462.Find this resource:

Gershunov, A., Schneider, N., & Barnett, T. (2001). Low-frequency modulation of the ENSO–Indian monsoon rainfall relationship: Signal or noise? Journal of Climate, 14, 2486–2492.Find this resource:

Goswami, B. N., & Chakravorty, S. (2017). Dynamics of the Indian summer monsoon climate. Oxford research encyclopedias (Climate science; Dynamics of the climate system).Find this resource:

Goswami, B. N., Madhusoodanan, M. S., Neema, C. P., & Sengupta, D. (2006). A physical mechanism for North Atlantic SST influence on the Indian summer monsoon. Geophysical Research Letters, 33, L02706.Find this resource:

Goswami, B. N. (1998). Interannual variations of Indian summer monsoon in a GCM: External conditions versus internal feedbacks. Journal of Climate, 11, 501–521.Find this resource:

Goswami, B. N., & Xavier, P. K. (2005). ENSO control on the south Asian monsoon through the length of the rainy season. Geophysical Research Letters, 32, L18717.Find this resource:

Halder, S., & Dirmeyer, P. A. (2016). Relation of Eurasian snow cover and Indian summer monsoon rainfall: Importance of the delayed hydrological effect. Journal of Climate, 30, 1273–1289.Find this resource:

Halder, S., Saha, S. K., Dirmeyer, P. A., Chase, T. N., & Goswami, B. N. (2016). Investigating the impact of land-use land-cover change on Indian summer monsoon daily rainfall and temperature during 1951–2005 using a regional climate model. Hydrology and Earth System Sciences, 20, 1765–1784.Find this resource:

Jang, Y., & Straus, D. M. (2012). The Indian Monsoon Circulation Response to El Niño Diabatic Heating. Journal of Climate, 25, 7487–7508.Find this resource:

Johnson, S. J., Turner, A., Woolnough, S., Martin, G., & MacLachlan, C. (2016). An assessment of Indian monsoon seasonal forecasts and mechanisms underlying monsoon interannual variability in the Met Office GloSea5-GC2 system. Climate Dynamics, 48, 1447–1465.Find this resource:

Joshi, M. K., & Kucharski, F. (2016). Impact of Interdecadal Pacific Oscillation on Indian summer monsoon rainfall: An assessment from CMIP5 climate models. Climate Dynamics, 48, 2375–2391.Find this resource:

Joshi, M. K., & Rai, A. (2015). Combined interplay of the Atlantic multidecadal oscillation and the Interdecadal Pacific Oscillation on rainfall and its extremes over Indian subcontinent. Climate Dynamics, 44(11–12), 3339–3359.Find this resource:

Jourdain, N. C., Gupta, A., Taschetto, A. S., Ummenhofer, C. C., Moise, A. F., & Ashok, K. (2013). The Indo-Australian monsoon and its relationship to ENSO and IOD in re-analysis and the CMIP3/CMIP5 simulations. Climate Dynamics, 41, 3073–3102.Find this resource:

Ju, J., & Slingo, J. M. (1995). The Asian summer monsoon and ENSO. Quarterly Journal of the Royal Meteorological Society, 121, 1133–1168.Find this resource:

Kalnay, E., Kanamitsu, M., Kistler, R., Collins, W., Deaven, D., Gandin, L., … Joseph, D. (1996). The NCEP/NCAR 40-Year Reanalysis Project. Bulletin of the American Meteorological Society, 77, 437–471.Find this resource:

Kang, I.-S., Lee, J., & Park, C. (2004). Potential predictability of summer mean precipitation in a dynamical seasonal prediction system with systematic error correction. Journal of Climate, 17, 834–844.Find this resource:

Kang, I.-S., & Shukla, J. (2006). Dynamic seasonal prediction and predictability of the monsoon. The Asian monsoon. Springer-Praxis.Find this resource:

Kim, H. M., Webster, P. J., & Curry, J. A. (2012). Seasonal prediction skill of ECMWF System 4 and NCEP CFSv2 retrospective forecast for the Northern Hemisphere winter. Climate Dynamics, 39, 2957–2973.Find this resource:

Kinter, J. L., III, Miyakoda, K., & Yang, S. (2002). Recent change in the connection from the Asian monsoon to ENSO. Journal of Climate, 15, 1203–1215.Find this resource:

Krishna Kumar, K., Rajagopalan, B., & Cane, M. A. (1999). On the weakening relationship between the Indian monsoon and ENSO. Science, 284, 2156–2159.Find this resource:

Krishna Kumar, K., Rajagopalan, B., Hoerling, M., Bates, G., & Cane, M. A. (2006). Unraveling the mystery of Indian monsoon failure during El Nino. Nature, 314, 115–119.Find this resource:

Krishnamurthy, V., & Goswami, B. N. (2000). Indian monsoon-ENSO relationship on interdecadal timescale. Journal of Climate, 13, 579–595.Find this resource:

Krishnamurthy, L., & Krishnamurthy, V. (2014). Influence of PDO on South Asian summer monsoon and monsoon–ENSO relation. Climate Dynamics, 42, 2397–2410.Find this resource:

Krishnamurthy, V., & Shukla, J. (2000). Intraseasonal and interannual variability of rainfall over India. Journal of Climate, 13(24), 4366–4377.Find this resource:

Kucharski, F., Bracco, A., Barimalala, R., & Yoo, J. H. (2011). Contribution of the east–west thermal heating contrast to the South Asian monsoon and consequences for its variability. Climate Dynamics, 37, 721–735.Find this resource:

Kucharski, F., Bracco, A., Yoo, J. H., & Molteni, F. (2007). Low-frequency variability of the Indian monsoon–ENSO relationship and the tropical Atlantic: The “weakening” of the 1980s and 1990s. Journal of Climate, 20(16), 4255–4266.Find this resource:

Kucharski, F., Bracco, A., Yoo, J. H., & Molteni, F. (2008). Atlantic forced component of the Indian monsoon interannual variability. Geophysical Research Letters, 35(4), L04706.Find this resource:

Kucharski, F., Bracco, A., Yoo, J. H., Tompkins, A. M., Feudale, L., & Ruti, P. (2009). A Gill–Matsuno-type mechanism explains the tropical Atlantic influence on African and Indian monsoon rainfall. Quarterly Journal of the Royal Meteorological Society, 579, 569–579.Find this resource:

Kucharski, F., & Joshi, M. K. (2017). Influence of tropical South Atlantic sea surface temperatures on the Indian summer monsoon in CMIP5 models. Quarterly Journal of the Royal Meteorological Society, 143, 1351–1363.Find this resource:

Kucharski, F., Molteni, F., & Yoo, J. H. (2006). SST forcing of decadal Indian monsoon rainfall variability. Geophysical Research Letters, 33, L03709.Find this resource:

Kucharski, F., et al. (2009). The CLIVAR C20C Project: Skill of simulating Indian monsoon rainfall on interannual to decadal timescale. Does GHG forcing play a role? Climate Dynamics, 33, 615–627.Find this resource:

Lau, N. C., & Nath, M. J. (2000). Impact of ENSO on the variability of the Asian-Australian monsoon as simulated in GCM experiments. Journal of Climate, 13, 4287–4309.Find this resource:

Li, C.-F., & Yanai, M. (1996). The onset and interannual variability of the Asian summer monsoon in relation to land–sea thermal contrast. Journal of Climate, 9, 358–375.Find this resource:

Li, J., Wang, B., & Yang, Y. M. (2016). Retrospective seasonal prediction of summer monsoon rainfall over west central and peninsular India in the past 142 years. Climate Dynamics, 48, 2581–2596Find this resource:

Losada, T., Rodriguez-Fonseca, B., Polo, I., Janicot, S., Gervois, S., Chauvin, F., & Ruti, P. (2010). Tropical response to the Atlantic equatorial mode: AGCM multimodel approach. Climate Dynamics, 35, 45–52.Find this resource:

Meehl, G. A. (1994). Influence of the land surface in the Asian summer monsoon: External conditions versus internal feedbacks. Journal of Climate, 7, 1033–1049.Find this resource:

Mishra, V., Smoliak, B. V., Lettenmaier, D. P., & Wallace, J. M. (2012). A prominent pattern of year-to-year variability in Indian summer monsoon rainfall. Proceedings of National Academy of Sciences of United States of America, 109(19), 7213–7217.Find this resource:

Mohan, R. S. A., & Goswami, B. N. (2003). Potential predictability of the Asian summer monsoon on monthly and seasonal time scales. Meteorology and Atmospheric Physics, 84, 83–100.Find this resource:

Molteni, F., Stockdale, T., Balmaseda, M., Balsamo, G., Buizza, R., Ferranti, L., … Vitart, F. (2011). The new ECMWF Seasonal Forecast System (System 4), Technical Memorandum 656. Reading, U.K.: ECMWF.Find this resource:

Murtugudde, R., McCreary, J. P., & Busalacchi, A. J. (2000). Oceanic processes associated with anomalous events in the Indian Ocean with relevance to 1997–1998. Journal of Geophysical Research, 105(C2), 3295.Find this resource:

Pottapinjara, V., Girishkumar, M. S., Ravichandran, M., & Murtugudde, R. (2014). Influence of the Atlantic zonal mode on monsoon depressions in the Bay of Bengal during boreal summer. Journal of Geophysical Research: Atmospheres, 119(11), 6456–6469.Find this resource:

Pottapinjara, V., Girishkumar, M. S., Sivareddy, S., Ravichandran, M., & Murtugudde, R. (2015). Relation between the upper ocean heat content in the equatorial Atlantic during boreal spring and the Indian monsoon rainfall during June–September. International Journal of Climatology, 36, 2469–2480.Find this resource:

Ramu, D. A., Sabeerali, C. T., Chattopadhyay, R., Rao, D. N., George, G., Dhakate, A. R., … Rao, S. A. (2016). Indian summer monsoon rainfall simulation and prediction skill in the CFSv2 coupled model: Impact of atmospheric horizontal resolution. Journal of Geophysical Research: Atmospheres, 121, 2205–2221.Find this resource:

Rasmusson, E. M., & Carpenter, T. H. (1983). The relationship between the eastern Pacific sea surface temperature and rainfall over India and Sri Lanka. Monthly Weather Review, 111, 517–528.Find this resource:

Rayner, N. A., Parker, D. E., Horton, E. B., Folland, C. K., Alexander, L. V., Rowell, D. P., … Kaplan, A. (2003). Global analyses of sea surface temperature, sea ice, and night marine air temperature since the late nineteenth century. Journal of Geophysical Research, 108, 4407.Find this resource:

Rodwell, M. J., & Hoskins, B. J. (2001). Subtropical anticyclones and summer monsoons. Journal of Climate, 14, 3192–3211.Find this resource:

Ross, T., Lott, N., McCown, S., & Quinn, D. (1998) The El Nino Winter of ’97–’98. National Climatic Data Center Technical Report No. 98–02.Find this resource:

Saha, S. K., Halder, S., Kumar, K. K., & Goswami, B. N. (2011). Pre-onset land surface processes and “internal” interannual variabilities of the Indian summer monsoon. Climate Dynamics, 36, 2077–2089.Find this resource:

Saha, S. K., Pokhrel, S., Salunke, K., Dhakate, A., Chaudhari, H. S., Rahaman, H., … Sikka, D. R. (2016). Potential predictability of Indian summer monsoon rainfall in NCEP CFSv2. Journal of Advances in. Modeling Earth Systems, 8, 96–120.Find this resource:

Saji, N. H., Goswami, B. N., Vinayachandran, P. N., & Yamagata, T. (1999). A dipole mode in the tropical Indian Ocean. Nature, 401(6751), 360–363.Find this resource:

Shukla, J. (1981). Dynamical predictability of monthly means. Journal of the Atmospheric Sciences, 38, 2547–2572.Find this resource:

Sikka, D. R. (1980). Some aspects of the large-scale fluctuations of summer monsoon rainfall over India in relations to fluctuations in the planetary and regional scale circulation parameter. Journal of Earth System Sciences, 89, 179–195.Find this resource:

Slingo, J. M., & Annamalai, H. (2000). The El Niño of the Century and the Response of the Indian Summer Monsoon. Monthly Weather Review, 128, 1778–1796.Find this resource:

Syed, F. S., Kornlich, H., & Kucharski, F. (2012). Extratropical influences on the inter-annual variability of South-Asian monsoon. Climate Dynamics, 38, 1661–1674.Find this resource:

Syed, F. S., & Kucharski, F. (2016). Statistically related coupled modes of South Asian summer monsoon interannual variability in the tropics. Atmospheric Science Letters, 17, 183–189.Find this resource:

Sperber, K. R., Annamalai, H., Kang, I.-S., Kitoh, A., Moise, A., Turner, A., Wang, B., & Zhou, T. (2014). The Asian summer monsoon: an intercomparison of CMIP5 vs. CMIP3 simulations of the late 20th century. Climate Dynamics, 41, 2711–2744.Find this resource:

Trenberth, K. E., Branstator, G. W., Karoly, D., Kumar, A., Lau, N. C., & Ropelewski, C. (1998). Progress during TOGA in understanding and modeling global teleconnections associated with tropical sea surface temperatures. Journal of Geophysical Research, 14, 291–324.Find this resource:

Trenberth, K. E., Caron, J. M., Stepaniak, D. P., & Worley, S. (2002). Evolution of El Niño–Southern Oscillation and global atmospheric surface temperatures. Journal of Geophysical Research, 107(D8), AAC 5-1–AAC 5-17.Find this resource:

Turner, A. G., & Annamalai, H. (2012). Climate change and the South Asian summer monsoon. Nature Climate Change, 2, 587–595.Find this resource:

Turner, A. G., & Slingo, J. M. (2011). Using idealized snow forcing to test teleconnections with the Indian summer monsoon in the Hadley Centre GCM. Climate Dynamics, 36, 1717.Find this resource:

Walker, G. T. (1924). Correlations in seasonal variations of weather (Mem. 24, pp. 275–332). New Delhi: Indian Meteorological Department.Find this resource:

Wallace, J. M., & Hobbs, P. V. (1977). Atmospheric Science. An Introductory Survey (p. 467). Academic Press.Find this resource:

Wang, C., Kim, D., Ekman, A. M. L., Barth, M. C., & Rasch, P. J. (2009). Impact of anthropogenic aerosols on Indian summer monsoon. Geophysical Research Letters, 36, L21704.Find this resource:

Wang, C., Kucharski, F., Barimalala, R., & Bracco, A. (2009). Teleconnections of the tropical Atlantic to the tropical Indian and Pacific Oceans: A review of recent findings. Meteorologische Zeitschrift, 18, 445–454.Find this resource:

Wang, B., Wu, R., & Li, T. (2003). Atmosphere–warm ocean interaction and its impacts on Asian–Australian monsoon variation. Journal of Climate, 16, 1195–1211.Find this resource:

Wang, B., et al. (2009). Advance and prospectus of seasonal prediction: Assessment of the APCC/CliPAS 14-model ensemble retrospective seasonal prediction (1980–2004). Climate Dynamics, 33(1), 93–117.Find this resource:

Webster, P., Magana, V., Palmer, T., Shukla, J., Tomas, R., Yanai, Y., & Yasunari, T. (1998). Monsoons: processes, predictability, and the prospects for prediction. Journal of Geophysical Research, 103(C7), 14451–14510.Find this resource:

Webster, P. J. (1987). The elementary monsoon. In J. S. Fein & P. L. Stephens (Eds.), Monsoons (pp. 3–32). London: Wiley.Find this resource:

Webster, P. J., Moore, A. M., Loschnigg, J. P., & Leben, R. R. (1999). Coupled ocean-atmosphere dynamics in the Indian Ocean during 1997–98. Nature, 401(6751), 356–360.Find this resource:

Webster, P. J., & Yang, S. (1992). Monsoon and ENSO: Selectively interactive systems. Quarterly Journal of the Royal Meteorological Society, 118, 877–926.Find this resource:

Xie, S., Hu, K., Hafner, J., Tokinaga, H., Du, Y., Huang, G., & Sampe, T. (2009). Indian Ocean capacitor effect on Indo-western Pacific climate during the summer following El-Nino. Journal of Climate, 22, 730–747.Find this resource:

Yadav, R. K. (2009). Role of equatorial central Pacific and northwest of North Atlantic 2-metre surface temperatures in modulating Indian summer monsoon variability. Climate Dynamics, 32(4), 549–563.Find this resource:

Yadav, R. K. (2016). On the relationship between east equatorial Atlantic SST and ISM through Eurasian wave. Climate Dynamics, 48, 281–295.Find this resource:

Yang, J., Liu, Q., Xie, S.-P., Liu, Z., & Wu, L. (2007). Impact of the Indian Ocean SST basin mode on the Asian summer monsoon. Geophysical Research Letters, 34, L02708.Find this resource:

Zhang, R., & Delworth, T. L. (2006). Impact of Atlantic multidecadal oscillations on India/Sahel rainfall and Atlantic hurricanes. Geophysical Research Letters, 33, L17712.Find this resource: