1 Introduction

The hydrology of the Indian river basins depends highly on the quantity of Indian summer monsoon (ISM) rainfall (e.g., Rao 1975; Parthasarathy et al. 1987; Dhar and Nandargi 1998, 2000, 2004; Kale 2003, 2004; Jain et al. 2007)—in addition to the melting of snow and glaciers for the Himalayan river basins such as the Indus and the Ganges located in the western Himalayas (WEH; centred over the northern Indian states; Fig. 1), and the Brahmaputra in the central-eastern part of the Himalayas (CEH). The summer monsoon rainfall activity over the Indian subcontinent is closely tied to the intensity of the large-scale monsoon overturning circulation and the monsoon trough (MT) located over the Indo-Gangetic Plains (IGP) (e.g., Krishnamurti and Bhalme 1976; Rao 1976; Rajeevan et al. 2010; Choudhury and Krishnan 2011; Krishnan et al. 2013). During the ISM rainy season, the Himalayan rivers often encounter flood problems in the northern Indian states Uttarakhand, Jammu and Kashmir, Himachal Pradesh (see Fig. 1 for the locations), Uttar Pradesh, Bihar, West Bengal and Assam due to torrential rains and localized intense precipitation occurrences along the southern slopes of the Himalayas, with flooding occasionally being accompanied by landslides (Chalise and Khanal 2001; Sen 2010; Mohapatra and Singh 2003; Vellore et al. 2014).

Fig. 1
figure 1

Topography (shaded units in meters; source United States Geological Survey) map of the western Himalayas (WEH) region. For reference, a few selected foothill stations in the north Indian states of Uttarakhand (UT; see the inset), Himachal Pradesh (HP) and Jammu and Kashmir (JK) are shown in the figure. Also, the location Kedarnath in Uttarakhand is marked by crossed-circle, where vast area around this location is submerged in devastating June 2013 floods (Ray et al. 2013; Dobhal et al. 2013)

Recent decades have experienced an increasing trend of high impact rainfall events over the Asian subcontinent during the summer monsoon season (e.g., Goswami et al. 2006; Rajeevan et al. 2008; Pattanaik and Rajeevan 2010; Guhathakurta et al. 2011; Singh et al. 2014). Heavy precipitation vulnerable to flooding in the CEH region has been reported by earlier studies (e.g., Teegavarapu 2012; Vellore et al. 2014). Nonetheless, some of the unprecedented rain episodes and catastrophic flash floods in the WEH region and neighbourhood have been shown to exhibit extreme severity during the last several decades (e.g., Joshi and Kumar 2006; Houze et al. 2011; Nandargi and Dhar 2012; Rasmussen and Houze 2012; Mujumdar et al. 2012; Rana et al. 2013; Dobhal 2013; Gupta et al. 2013; Martius et al. 2013; Joseph et al. 2014; Kumar et al. 2014). Therefore, under the pretext of climate change impact on hydroclimate variability and water resources management in conjunction with increasing number of catastrophic precipitation events over the Himalayas, the scientific and computational challenges principally lie in the clear understanding of what dynamical processes govern these events and how the frequency and intensity of these events will evolve with global warming (e.g., Khan et al. 2015). From a climate change viewpoint, it has been reported that changes in strength of the upper-level midlatitude jet stream can modulate slow-moving large-amplitude Rossby waves that can have consequences to high impact weather events, such as floods, droughts and heat waves in the northern hemisphere (Francis and Vavrus 2012; Peteoukhov et al. 2013; Screen and Simmonds 2014; Coumou et al. 2014). Enhancement of convective activity over the heat low region of northwest India-Pakistan can induce large-scale circumglobal teleconnection wave-like patterns (Saeed et al. 2011), while there are feedbacks from strong upper tropospheric outflows associated with intense precipitation events over northwest India and Pakistan which can effectively transport heat poleward as far as the Canadian Arctic, and accelerate melting of the Arctic sea-ice (Krishnamurti et al. 2015).

The southern slope of the Himalayas is normally under the influence of both monsoon and midlatitude circulation regimes during the ISM and therefore any sudden circulation changes over this region have the potential to produce very heavy rainfall and can last for a few days. Large-scale atmospheric conditions associated with intense precipitation episodes in the Himalayas generally include the northward and westward advancing low-level moisture-laden tropical monsoon circulation and the southward intruding upper level extratropical circulation. The upper-tropospheric extratropical circulation patterns associated with these episodes bear some resemblance with those observed during monsoon-breaks (e.g., Ramaswamy 1958, 1962; Ramamurthy 1969; Rao 1976; Krishnan et al. 2000, 2009; Sikka 2011, Vellore et al. 2014). Two recent infamous flood-producing precipitation events in the WEH region include: (1) Northwest Pakistan floods in late July 2010 (http://en.wikipedia.org/wiki/2010_Pakistan_floods), and (2) Uttarakhand floods in mid-June 2013 over northwest India (http://en.wikipedia.org/wiki/2013_North_India_floods). These events have become the focal point of recent research on extreme rainfall (rainfall rates exceeding 100 mm day−1) over the Himalayan region. Hong et al. (2011) and Lau and Kim (2012) highlighted the role of persistent European blocking high and downstream midlatitude synoptic disturbances in triggering extreme rainfall events over the northwest Pakistan region during July 2010. Mujumdar et al. (2012) and Priya et al. (2015) attributed the devastating flood-producing precipitation in northwest Pakistan to unusual Indo-Pacific sea surface temperature evolution with strong La Niña conditions in the Pacific as well as to the westward shift of the Pacific anticyclone which favours anomalous moisture transport from the Arabian Sea through the setup of an anomalous trough extending over the subtropical belt in the vicinity of India and Pakistan. Martius et al. (2013) indicated that extratropical wave breaking is one of the intrinsic factors for heavy precipitation episodes in the Himalayan river region. Rasmussen et al. (2014) related the heavy rains to tropospheric moisture advection from the Bay of Bengal to the Himalayan foothills by mid-level easterlies facilitated by quasi-stationary blocking events and synoptic conditions over the Tibetan Plateau.

Despite the aforementioned studies, a clear understanding of the sequence of dynamical mechanisms occurring at various scales of motion resulting in extreme precipitation events over the western Himalayas is thus far lacking. Studies also have emphasized the role of multi-scale interactions (ranging from planetary scale to mesoscale) on extreme precipitation events in the Himalayas (Rasmussen and Houze 2012; Galarneau et al. 2012; Rasmussen et al. 2014; Kumar et al. 2014). Diverse topographical features over the Himalayas and Tibetan Plateau play an essential role in maintenance of the large-scale monsoon circulation and mesoscale precipitation enhancements involving smaller-scale circulations (Hahn and Manabe 1975; Smith 1979; Singh and Kumar 1997; Vellore et al. 2014). Joseph et al. (2014) suggested a conjecture on early advancement of monsoon that could arise from the monsoon-extratropical interactions, however they accredited the June 2013 extreme rainfall over the Uttarakhand region (see Fig. 1 for the location) primarily to orographic uplift. In recognizing the importance of Himalayan foothill precipitation from these studies, it must be mentioned that the inferences are predominantly drawn from the tropical monsoon perspective, while the underlying dynamics of monsoon and midlatitude interactions in these extreme precipitation situations is hitherto unclear. For example, Vellore et al. (2014) associated the flood-producing precipitation in the CEH region to strong mid-tropospheric vertical motion during monsoon-breaks, wherein southward intrusion of large-amplitude midlatitude westerly troughs and dry air subsidence dominates over northwest India. On the other hand, extreme precipitation occurrences in the WEH region (e.g., 2010 northwest Pakistan and 2013 Uttarakhand rain events) tend to co-evolve with southward intruding midlatitude westerly troughs in a prevailing moist environment over the Indian subcontinent.

Clearly, a comprehensive understanding of dynamical mechanisms to the cause of flood catastrophes over the complex terrain of the WEH yet remains elusive. A reliable quantitative scenario of regional-scale rainfall in the Hindu Kush Himalayas is of paramount importance to various research and operational communities (Anders et al. 2006; Bookhagen and Burbank 2010). This study endeavours to provide insights into this aspect in cases of extreme precipitation over the WEH region during the ISM rainy season. The datasets and selection of extreme precipitation events are discussed in the following section.

2 Datasets, selection of extreme precipitation events and model setup

Observational analyses used in this study include the gridded (0.25° × 0.25°) daily precipitation dataset over India archived by the India Meteorological Department (IMD) (Pai et al. 2013, 2014) and the reanalysis products from the European Centre for Medium-Range Weather Forecasts (ECMWF) Interim Reanalysis (ERA-Interim) (Dee et al. 2011; Simmons et al. 2007) available at a 0.75° × 0.75° horizontal resolution. In addition, high-resolution numerical simulations using the Weather Research and Forecasting (WRF) (Skamarock et al. 2008) model results are used in the analysis.

2.1 Selection of extreme precipitation events

Localized as well as marginally widespread precipitation extremes occurring in the WEH region (28°–35° N and 75°–85° E) during the ISM months are only considered in this study. The periods of extreme precipitation are extracted from the gridded IMD precipitation dataset (Pai et al. 2013, 2014) as follows. First, grid point locations are identified where the precipitation rates exceed 100 mm day−1, followed by an additional criterion that the regions encompassing an area of 10,000 km2 around these point locations also display areal mean precipitation rates greater than 100 mm day−1—this gave rise to a total of 1407 precipitation events during the 1951–2013 period. Of these, a subset comprising the most extreme precipitation events, that are of principal interest to this study, are drawn with reference to the third quartile of the Gaussian distribution of the events—there are 353 events associated with rainfall rates exceeding 200 mm day−1. Empirical orthogonal function/principal component analysis (EOF/PC) (Navarra and Simoncini 2010; Wilks 2011) is performed on this subset to obtain the dominant spatio-temporal patterns of the rainfall variability—since we wanted to focus on extreme precipitation events that are spatially confined around the Uttarakhand area in the WEH region (see Fig. 1). Based on the significantly large normed metric of the first two PCs, a final set of 34 precipitation events is selected for this study which occurred during the 1979–2013 period (Table 1). The one-day precipitation accumulation over the WEH region for these case events is as low (high) as 217 mm (460 mm). The composites from these events are used to describe the circulation aspects associated with the extreme rainfall activity in the WEH region. Further, among the 34 events considered for the analysis, only two precipitation events are considered for numerical simulations, as those are computationally expensive, to illustrate the smaller-scale processes. These events represented pockets of unprecedented flood-producing extreme rainfall activity in the northern Indian states of Uttarakhand, Haryana, Jammu and Kashmir and Himachal Pradesh (see Fig. 1). The model setup is described in the following subsection.

Table 1 Maximum rainfall amount (mm) occurred within the box region encompassing 28°N–35°N and 75°–85°E and the day of the occurrence of the extreme event

2.2 Model setup

High-resolution numerical simulations conducted in this study are based on a mass core non-hydrostatic WRF model (Version 3.6.1) built over a parent domain, which covers a region 15°E–125°E and 23°S–60°N, with 54 km grid spacing in the west-east and south-north directions respectively. Two modeling domains are nested into the parent domain set to run at horizontal grid resolutions 18 km (encompassing the south-Asian monsoon region: 46°E–105°E and Equator–46°N) and 6 km (encompassing the WEH region and the adjoining subtropical neighborhood: 69°E–91°E and 24°N–42°N). The WRF modeling domains and the cross-sections along which the vertical structure is investigated are shown in Fig. 2.

Fig. 2
figure 2

WRF nested modeling domains and vertical cross-sections AB, AC, DE, and FG referenced in the study. Also overlain is the model representation of topography (in meters)

The model configuration has 72 levels in the vertical. The model physics includes: (1) momentum and heat fluxes at the surface computed using an Eta surface layer scheme (Janjić 1996, 2001) following Monin–Obukhov similarity theory, (2) turbulence processes following the Mellor–Yamada–Janjić level 2.5 model (Mellor and Yamada 1974, 1982; Janjić 2001), (3) convective processes following a simplified Arakawa-Schubert (SAS) scheme which utilizes mass flux formulation with quasi-equilibrium closure (Arakawa and Schubert 1974; Pan and Wu 1995)—applied only on 54 and 18 km grids, (4) cloud microphysical processes following explicit bulk representation of microphysics (Thompson et al. 2004, 2006), (5) radiative processes using the rapid radiative transfer model for long wave radiation (Mlawer et al. 1997) and Dudhia’s short wave scheme (Dudhia 1989), and (6) the land-surface processes following the Noah land surface model which provides the surface sensible and latent heat fluxes, and upward longwave and shortwave fluxes to the atmospheric model (Chen and Dudhia 2001; Ek et al. 2003). Initialization and boundary value specifications for the model simulations come from the 6-hourly ERA-Interim reanalysis. Model simulations were initialized 4 days prior to the day of extreme precipitation in the WEH region. The model simulations are conducted only to substantiate the evidence that emerges from the observations and reanalysed circulation products. The observed composites of circulation fields (precipitation) diagnosed from the ERA-Interim reanalyses (IMD daily precipitation dataset) are discussed in the following section.

3 Observed composites of precipitation and circulation fields

3.1 Rainfall

The observed patterns of mean and anomalous rainfall amounts for the chosen periods are shown in Fig. 3. It is noticeable that mean (anomalous) precipitation amounts exceeding 40 (25) mm day−1 are more prevalent and confined to the WEH region while they are less profound over the summer monsoon core region (in central part of India; within the box region 75°E–88°E; 18°N–27°N; Rao 1976) of the Indian subcontinent. The spatial distribution of precipitation is clearly different from the localized flood-producing extreme rainfall activity that is commonly observed over the CEH foothills during monsoon-breaks (e.g., Ramamurthy 1969; Raghavan 1973; De et al. 1998; Krishnan et al. 2000; Rajeevan et al. 2010; cf. Figure 3 of Vellore et al. 2014). Also, the anomalous rainfall pattern shown in Fig. 3 is distinctly different from the anomalous rainfall conditions in the WEH region during the active rainy periods in the ISM (cf. Figure 8 of Rajeevan et al. 2008). The spatially averaged precipitation in the monsoon core region during the active periods is about 16.1 mm day−1—which is nearly twice the amount (7.5 mm day−1) during the case periods chosen for this study. This suggests that extreme precipitation events considered in this study predominantly occurred during moderate monsoon conditions, i.e., neither during very active monsoon phase periods when the axis of the MT is distinctly located over the central part of India nor during the weak monsoon phase periods when the MT axis migrates to the CEH foothills.

Fig. 3
figure 3

A composite map of observed a mean and b anomalous precipitation (mm day−1) from the precipitation events shown in Table 1 (Source India Meteorological Department)

3.2 Circulation

Active rainy periods of the ISM (Rao 1976; Wang 2005; Rajeevan et al. 2010; Krishnamurti et al. 2013) are commonly identified through a synoptic setup which comprises of a low-level jet (Joseph and Raman 1966; Findlater 1969) that brings the moisture-laden air from the Indian Ocean into the Indian subcontinent, while more tropospheric warmth over the Indian subcontinent is typically aided by the presence of the poleward migrating jet stream in the subtropics and the formation of a strong easterly jet over peninsular India at the upper levels. A planetary scale anticyclonic circulation forms over the Tibetan Highlands during the ISM due to diabatic heating associated with Southeast Asian deep convective monsoon rainfall as well as due to sensible and convective heating over the elevated plateau. The northern flank of this anticyclone generally borders with subtropical jet stream (STJ; Ramaswamy 1958; Archer and Caldeira 2008), while the southern flank borders with tropical easterly jet (Koteswaram 1958) where the Himalayas act as a barrier in their path during the ISM. Nevertheless, eastward moving extratropical circulation patterns disrupt the synoptic setup from time to time by drifting the STJ equatorward and its interaction with monsoon circulation often produces anomalous circulation conditions across the Himalayas resulting in heavy precipitation in the foothills (e.g., Vellore et al. 2014).

A cursory examination indicates that large-scale atmospheric flow is comparable for the chosen periods and the events occurred in the presence of northwestward moving weak monsoon low pressure system over the IGP region (see Fig. 1; see also Fig. 6). The large-scale mean circulation fields, shown in Fig. 4, generally displayed an archetypal summer monsoonal flow (Rao 1976) associated with a large-amplitude upper level trough embedded in the midlatitude westerlies located northwest of the Tibetan anticyclone (TA) centered at 28°N, 82°E and the TA intensifies and elongates westward in time. The 500 hPa circulation shows a poleward protrusion of the Arabian anticyclone (indicated in Fig. 4) juxtaposed with significant positive tilt in the westerly trough whose equatorward extension penetrates well into the Indian subcontinent. In correspondence with the observed mean precipitation (Fig. 3), the 850 hPa circulation fields also indicate weaker strength in the monsoon circulation by displaying stronger westerly wind component over the peninsular and central parts of the Indian subcontinent while the easterly wind component is found to be weaker over the IGP. Deepening/thinning of upper level trough occurs when the Tibetan anticyclone splits and sheds an anticyclone upstream of the trough in many of the cases.

Fig. 4
figure 4

A composite map of horizontal winds (vectors) and geopotential height (shaded units in meters) for the events described in Table 1 valid at a 200 hPa, b 500 hPa, and c 850 hPa. Tibetan anticyclone and Arabian anticyclone referenced in the study are indicated in the figure. (Source ERA-Interim reanalysis)

For example, the time evolution of circulation fields at 200 hPa and 850 hPa for two events that occurred in September 1995 and June 2013, shown in Figs. 5 and 6, exhibits the shedding process and amplification of the Tibetan anticyclone downstream of the upper level trough. With prevailing hot conditions over North Africa, the Middle East, and over the Tibetan Highlands during the warm season, there is a thickness ridge that pushes the STJ and polar jet stream in closer proximity which leads to a closer juxtaposition of the southern branch of the polar jet stream and STJ. These two jets effectively merge when there is strong baroclinic forcing between the extratropical latitudes and the Tibetan plateau. These jets converge and the support for the converged jet comes from deep cold air advection in the southern branch of the polar jet. The ridging over the hot and elevated zone extending from North Africa to western part of China may enhance extratropical wave breaking mechanisms as it serves as a block in the flow. One can say that the geography over these locations facilitates a hybrid upper jet structure. The STJ referred in the text is a synthesis of STJ and southern branch of polar jet. Hence the premise is that while the large-scale monsoon circulation is relatively not strong, interactions between rapidly advancing monsoonal flow transients and the equatorward advancing midlatitude circulation patterns in concert with the orographic ascent in the Himalayan foothill region can be reasonably attributed to the cause of flood-producing extreme precipitation events in the WEH region. The rationale for this contention is first examined through the use of large-scale anomalous circulation composites in the following.

Fig. 5
figure 5

Time evolution of 200 hPa horizontal winds (vectors; ≥20 m s−1 is shaded), geopotential height (contour interval 60 m), and air temperature (blue-dashed contour; contour interval 3 °C) valid during 14–17 June 2013 (left panel) and 31 August–3 September 1995 (right panel). (TA = Tibetan anticyclone) (source: ERA-Interim reanalysis)

Fig. 6
figure 6

Time evolution of 850 hPa horizontal winds (full barb = 5 m s−1) and mean sea level pressure (shaded; units in hPa) valid during 15–17 June 2013 (left panel) and 1–3 September 1995 (right panel) (source ERA-Interim reanalysis). The marker L indicates the monsoon low. Crossed-circle marks the location of the extreme precipitation

Daily composites of lower and upper-tropospheric anomalous circulation fields are shown in Figs. 7, 8 and 9. Day 0 corresponds to the day of the extreme precipitation which occurred over the WEH region (see also Table 1). The 200 hPa anomalous circulation fields exhibit a Rossby wave train structure with a gradual build-up of anomalous cyclonic-anticyclonic gyre pattern in the horizontal wind and mass (height) fields poleward of the WEH region, and a strong Ω-type anomalous blocking anticyclone to the west of this pattern extending from the mid-troposphere (Figs. 7, 8). The lower tropospheric anomalous circulations show a west-northwestward moving transient monsoon low 3–4 days prior to the extreme precipitation episodes (Fig. 9). Conspicuous features at the upper levels include the formation of two anomalous high patterns, the first building up north of the Caspian Sea and the other over the eastern part of China during this time (Figs. 7a–c). This setup induces a strong positively tilted cyclonic-anticyclonic pair of gyres in upper level mass field thereby enhancing the meridional winds poleward of the WEH region 1–2 days prior to heavy rains (Fig. 7d, e).

Fig. 7
figure 7

A composite map of daily anomalous circulation fields (horizontal winds–vectors; and geopotential height–shaded; units in meters) at 200 hPa. Day 0 corresponds to the day of the extreme precipitation in the WEH region (Table 1) (source ERA-Interim daily climatology derived from the period 1979–2008)

Fig. 8
figure 8

A composite map of daily anomalous circulation fields (horizontal winds–vectors; and geopotential height–shaded; units in meters) at 500 hPa. Day 0 corresponds to the day of the extreme precipitation in the WEH region. (source: ERA-Interim daily climatology derived from the period 1979–2008)

Fig. 9
figure 9

A composite map of daily anomalous circulation fields (horizontal winds–vectors; and geopotential height–shaded; units in meters) at 850 hPa. Day 0 corresponds to the day of the extreme precipitation in the WEH region (source: ERA-Interim daily climatology derived from the period 1979–2008)

A widespread mass deficit (negative height/cyclonic vorticity anomalies) prevails in the mid-troposphere stretching from the Caspian Sea to the deserts of Mongolia in the west-east direction and from the northern Arabian Sea to southern Russia in north–south direction, respectively prior to the day of extreme precipitation (Fig. 8). Also during this time, the transient low level pressure feature moves towards northwest India and the low-level circulation fields indicate strong anomalous southwesterly flow of air from the Arabian Sea directed towards the WEH region (Fig. 9). Also, the 500 hPa geopotential height field indicated a strong equatorward protrusion of the westerly trough (Fig. 4b) aligned with the transient cyclonic vorticity at low levels (Fig. 9c–e) at the same time enhancing the meridional component of monsoon winds and moisture convergence in phase with the increasing meridional winds at the upper levels, i.e., the circulation anomalies extend to lower levels advecting moisture into the region at an angle normal to the Himalayan barrier for establishing a moisture-laden convective environment in the foothill region.

The tropical-extratropical circulation phasing across the Himalayas occurs from a day to two days prior to extreme precipitation in the WEH region. A north–south oriented cross-sectional display of anomalous wind components along the meridian 80ºE, shown in Figs. 10 and 11, noticeably indicates this phasing and enhancement of meridional winds, as well as pronounced mid-tropospheric vertical motion over the foothills of the WEH region at this time. Convergence in the foothill region is phased with enhanced convergence extending up to 400 hPa at high elevations within the plateau region and at the same time the upper level divergence is distinctly strong due to mass depletion and poleward ageostrophic motion (Rochette and Market 2006) supporting the upper level jet acceleration (Fig. 10) and its associated entrance region transverse circulation dynamics (Uccellini and Johnson 1979).

Fig. 10
figure 10

A vertical cross-section composite plot of daily anomalous zonal and meridional wind fields (units in m s−1) (zonal wind–shaded; and meridional wind–contours) along the meridian 80ºE. Day 0 corresponds to the day of the extreme precipitation in the WEH region (source ERA-Interim daily climatology derived from the period 1979–2008). J marks the upper level jet

Fig. 11
figure 11

A vertical cross-section composite plot of daily anomalous vertical motion (contour; µb s−1) and horizontal velocity divergence (shaded; s−1) averaged over the region 75°E–85°E. Day 0 corresponds to the day of the extreme precipitation in the WEH region (source: ERA-Interim daily climatology derived from the period 1979–2008)

The anomalous sea level pressure, shown in Fig. 12, indicates a persistent relative local pressure deficit centered over the west-northwestern part of the Indian subcontinent and the anomalous pressure gradient on its eastern side tends to strengthen in time and merge with the pressure deficits over the southern sectors of the upper level anticyclonic gyre prior to extreme rain activity in the foothill region. Also, these pressure deficits tend to strengthen in time in tandem with the anomalous enhancement of midlatitude mass field centered over the north of the Caspian Sea. It is apparent that persistent and stronger anomalous high at 200 hPa over eastern China dominated during the deepening of the upper level trough and the deepening also is presumably due to the dynamical instabilities arising from the shedding of upper level anticyclonic circulation due to divergent forcing at upper levels (Hsu and Plumb 2000). The Tibetan anticyclone elongates westward and sheds a large anticyclone which further enhances the already existing Ω-blocking high over the western Eurasia. Anomalous mid-level mass deficits were furthermore meridionally confined and cold air advanced from far north of 50ºN which clearly outlines the meridional interactions of monsoon and extratropical flows.

Fig. 12
figure 12

A composite map of daily anomalous mean sea level pressure (contour; units in hPa). Day 0 corresponds to the day of the extreme precipitation in the WEH region (source ERA-Interim daily climatology derived from the period 1979–2008)

The evolution of mid-to-upper tropospheric ageostrophic motion also corresponds with the aforementioned features by displaying a strong anomalous northwesterly component in the southwestern Himalayas slightly upstream of the base of the deepening westerly trough (not shown). The amplification of the upper level wave train produces cut-off lows flanking a massive ridge of high pressure into the upper troposphere poleward of the Tibetan Plateau that results in the amplification of the jet entrance region at 200 hPa and in its positive tilt just north-northeast of the extreme rainfall location. The 500 hPa trough also amplifies and becomes generally oriented from south-southwest to north-northeast during the same time. The divergence in the right entrance region of the upper level strengthening jet causes pressure falls at lower levels (Uccellini and Johnson 1979) which further strengthen the low-level air streams to efficiently transport the heat and moisture meridionally toward the convective region. In short, the events appear to be closely associated with a thermally-direct diabatically-forced secondary circulation (Eliassen 1959, 1962; Reiter 1963; Maddox 1979; Keyser and Johnson 1984; Wolf and Johnson 1995) within the entrance region of an intensifying upper-level jet streak that promotes the extreme precipitation while the lower branch of this circulation is enhanced by the orographic upslope lifting in a nonlinear synergistic manner.

It is seen from the upper level circulation fields that extreme precipitation events considered in this study generally begin with the association of Rossby wave mechanisms in extratropics and divergent circulation in the monsoon environment. In the following, we discuss the necessary and critical elements for the evolution of deep southward penetrating troughs into the western Himalayas which is characterized by the monsoon-extratropical interactions, and the pathways to enhancements in anomalous cyclonic vorticity in the mid-troposphere.

4 Elements for monsoon-extratropical circulation interactions

4.1 Extratropical Rossby wave breaking mechanisms

The salient large-scale features of ISM circulation features notably include the presence of upper level divergent circulation associated with Tibetan anticyclone (Ramage 1971; Hasternath 1991) and significant potential vorticity (PV) gradients on its poleward side in the middle latitudes. A non-zero mean PV gradient is necessary for the restoring mechanisms associated with Rossby waves and large-amplitude Rossby waves tend to break in the regions where the wave motions overpower the mean PV gradient, i.e., Rossby waves break in the regions of negative meridional PV gradients which facilitate strong meridional energy transfer through the propagation of the PV gradient (McIntyre and Palmer 1983; see also Postel and Hitchman 1999; Weijenborg et al. 2012; Abatzoglou and Magnusdottir 2006; Hitchman and Huesmann 2007; Homeyer and Bowman 2013). These regions are typically characterized by the overturning of isentropic potential vorticity (IPV) contours, e.g., anticyclonic Rossby wave breaking (RWB) regions correspond to the regions where high [low] IPV rolls up anticyclonically equatorward [poleward]. Thorncroft and Jones (2000) indicated that low PV outflow can facilitate anticyclonic wave breaking which is frequently associated with equatorward advancing positive PV anomalies, and this wave breaking event is often associated with blocking events in midlatitudes (Pelly and Hoskins 2003). Therefore, with more preferential occurrence of RWB situations during summertime over the central Asian region in the presence of upper level divergent outflow from quasi-steady anticyclones (Hoskins 1991), their influence needs an examination to understand their role in the extreme precipitation events considered in this study.

Examination of upper level features associated with the case events indicate there is an apparent RWB signal and this begins the process of amplifying the westerly trough towards the southwest setting its base over the northernmost states of India. The RWB signals are more pronounced in some cases than the others. The setup of RWB processes is demonstrated on the 350 K isentropic surface (in proximity to 200 hPa approximately located near the STJ) prior to a few extreme precipitation events as shown in Fig. 13 (see also Table 1). Notice that the RWB region, that is located initially around the latitude 50°N poleward of the Himalayas, extends southwestward in time and these RWB events exhibit significant thinning of upper level troughs resulting in advection of high IPV air southwest and west around the upstream anticyclones. Figure 13e displays a noticeably classic type of anticyclonic RWB setup that persisted for a week. Thus, notably evolving circulation features on θ = 350 K surface include: (1) anticyclonic RWB zone poleward of the Himalayas, (2) westward stretching (left panel of Fig. 13) of the Tibetan anticyclone and eddy shedding (right panel of Fig. 13; see also Fig. 5), (3) near-stationary Ω-type blocking ridge between 40ºE and 80ºE over the western Eurasia, and (4) strong IPV gradient located to its eastward—provides a favourable environment for significant equatorward advancement of upper level westerly troughs. That is, the life cycle of midlatitude upper level eddy circulations depend on the strong and weak RWB dynamical setup and further enhancements in the western Eurasian ridging arise from the strengths of shedded anticyclone along 40ºE and the Tibetan anticyclone.

Fig. 13
figure 13

Montgomery stream function (contours; divided by acceleration due to gravity; units in meters) and potential vorticity (shaded; 1 PVU = 10−6 m2 s−1 K kg−1) on the isentropic surface θ = 350 K a few days prior to extreme precipitation events in the WEH region valid on a 26 September 1988, b 8 July 1990, c 15 August 1994, d 9 July 1986, e 2 September 1995, and f 15 June 2013 (source ERA-Interim reanalysis)

Also evident is the anomalous enhancement of negative IPV gradients, shown in Fig. 14, poleward of the westerly trough (at about 60ºE, 50ºN) a few days prior to precipitation events, the negative meridional IPV gradients poleward of the trough and enhancements of positive meridional IPV gradients in the vicinity of deepening trough along the northern flank of the anticyclone pairs provide evidence of southward advancing cyclonic vorticity streamers which leads to trough thinning supported by strong cold air advection. Further local enhancements of positive IPV gradients near the upper level jet tend to inhibit the RWB process at later times. The Ω-type blocking tends to freeze the meanders of the STJ over the Mediterranean and Asian region, thus facilitating strong advection of colder air from the northern latitudes (50°N–55°N) towards the WEH region. In some cases (e.g., Figure 8f), the RWB process is initially narrower for a few days prior to extreme rainfall; however, the deepening of the westerly trough appears to be associated with successive RWB events resulting in strong advancing pool of cold air southward from the middle latitudes—coincident with IPV building southwestwards towards the west of the Himalayas.

Fig. 14
figure 14

A composite map of daily anomalous Montgomery stream function field (contour; units in m2 s−2) and meridional IPV gradients (shaded; ×10−12 m s−1 K kg−1) on 350 K isentropic surface. Day 0 corresponds to the day of the extreme precipitation in the WEH region (source ERA-Interim reanalysis)

Thus, the large-scale forcing evolves from the RWB mechanism and apparently provides a pathway to mid-to-upper tropospheric cyclonic vorticity generation poleward of WEH region, as also seen in anomalous height deficits (Fig. 8). The generation of this cyclonic vorticity spanning from Himalayan foothills to the regions poleward in the extratropics is illustrated in Fig. 15 for the flood-induced rain episode in June 2013. Notice the tongue of high IPV air emanating from the upper level PV anomaly, usually of stratospheric origin, which corresponds to northern side of the upper level trough and these streams of high IPV air subside equatorward down to mid levels imminent to the west of extreme rain location a few days prior to the extreme precipitation event. This occurrence resembles a subtropical tropopause fold that typically reaches up to subtropical mid-troposphere along the prevailing STJ. In the mean westerly background, the PV at the base of the trough is advected eastward towards the Himalayan plateau which thereby enhances the cyclonic vorticity at mid-levels. Substantial cold air advection poleward of the WEH region further sustains the trough-thinning. As will be seen later, mid-tropospheric cyclonic vorticity is further enhanced by the interaction when it comes in phase with the lower-to-mid-tropospheric wind field associated with the northwestward propagating monsoon low located south of the WEH region.

Fig. 15
figure 15

Montgomery stream function (contours; divided by acceleration due to gravity; units in meters) and potential vorticity (shaded; units in PVU; 1 PVU = 10−6 m2 s−1 K kg−1) on the a θ = 350 K and b θ = 330 K isentropic surfaces a day prior to extreme precipitation event in the WEH region valid on 16 June 2013 (source ERA-Interim reanalysis). Also shown are the vertical cross sections of c IPV (PVU) and isobars (hPa) with isentropic ordinate, and d IPV and potential temperatures (°K) with isobaric ordinate. The cross-section extends from 20ºN, 70ºE to 60ºN, 80ºE indicated by a short-dashed line in (a). Crossed-circle marks the location of the extreme precipitation (see also Fig. 1 for the location and Table 1)

The trough thinning results in a stronger east–west height gradient upstream of the strong ridge associated with Tibetan anticyclone between 90°E and 120°E at about 30°N. The amplification of the height anomaly associated with this ridging appears to have been assisted by the latent heat release such that the divergent outflow intensifies and expands in areal coverage. The height field over the eastern Himalayan region builds and forces the flow to turn polewards more rapidly and with more intensity just east of the extreme rainfall location, i.e., RWB process associated with upper level PV anomalies tend to potentially force the mass field polewards and westwards allowing ascent and transport of low-level moist-laden tropical air imminent to the flood location. This also corroborates the discussions of Houze et al. (2011) and Martius et al. (2013) associated with an extreme flood-producing rain event occurred in northwest Pakistan.

Following Moore et al. (2010), the convection is enhanced during anticyclonic RWB events on the eastern/downstream side of the deepening troughs which is reflected in strengthening of the STJ to favour jet accelerations and secondary transverse circulations about the jet. This process is inherently synergistic because latent heating during heavy precipitation strengthens the outflow aloft, pressure falls, and Himalayan orographic ascent. Therefore, in concert with low-level orographic lifting, the enhanced height gradient downstream of the trough leads to an increase in the pressure gradient force directed northwestwards into the STJ entrance region, an ageostrophic wind component directed across the jet stream (an acceleration), enhanced upper level velocity divergence accompanying jetlet formation, and strong ascent in the mid-upper troposphere (e.g., Kaplan et al. 1998).

4.2 Eddy shedding and trough thinning

The observational studies of Hsu and Plumb (2000) and Popovic and Plumb (2001) demonstrated that eddies shed from Tibetan anticyclone due to dynamical instabilities arising from the upper level divergent forcing. With RWB precedence 3–5 days prior to extreme precipitation events, it is seen that low PV air propagates around the Tibetan anticyclone which elliptically elongates westward in time in part due to β dispersion-induced westward propagation. For the case events, it is noticeable that zonal stretching and shedding of anticyclones in the tropical-subtropical environment potentially lends support in facilitating the trough-thinning process and allows the westerly trough to protrude more southward in the presence of RWB setup from the middle latitudes. The strength of Tibetan anticyclone is intrinsically tied to convective variability that arises largely from mid-tropospheric heating and divergent outflow in the upper troposphere from deep convection (Garny and Randel 2013). The westward elongation of Tibetan anticyclone and eddy shedding are responses to localized upper level divergent forcing from west-northwest propagating synoptic lows from the Bay of Bengal across the subcontinent. One can also anticipate an intense and slight northeastward shift in the Tibetan anticyclone at upper levels owing to more moisture convergence support and the multitude of convectively unstable elements at low levels in the foothill region a day prior to the extreme precipitation.

The north–south orientation of the westerly trough is controlled by the strength of the anticyclone pairs on both sides. The quasi-stationarity of these anticyclones deepens the upper level trough along with strong cold air advection from the surface through 500 hPa entering the west side of trough resulting in positively-tilted trough orientation. The strong latent heat release favours the trough deepening that usually produces a divergent outflow jet and strengthening of the downstream ridge (Stoelinga 1996). The curvature in the downstream side of the trough is connected to the intensity and northeastward shifting of the TA after shedding and provides support for stronger east–west height gradients poleward and on the eastern side of the Himalayas. This forces the air flow to turn polewards throughout most of the troposphere just east of the extreme rainfall location. Notice the building up of southeastward extending anomalous anticyclonic (detaching) circulation downstream (upstream; about 50ºE) of the trough at about 100ºE (Fig. 7). The temperature evolution on 2 PVU surface (level of dynamic tropopause; not shown) from the composites also indicates the much stronger and larger horizontal extent of anomalous warm pool downstream of the trough. The trough deepening a few days prior to the extreme precipitation appears to be associated with this anomalous warm pool thus indicating the enhanced TA after the shedding process.

It is evident that the shredded eddy induces more curvature in the trough on the upstream side (see also Fig. 5), which further enhances the anticyclonic blocking and facilitates successive RWB episodes during June 2013 event, while the September 1995 event displays a classic setup of RWB wherein strong positive-tilting of the upper level trough primarily arises from the sustenance and strength of the breaking event period. In this case, an eddy sheds to the west of the thinning westerly trough and a stronger TA to its east causing stronger STJ. Subsequently the ageostrophically-forced transverse circulations (see also Fig. 18) in its entrance region facilitated strong ascent in the foothill region and resulted in extreme daily precipitation accumulations of 350–400 mm day−1 for two consecutive days over the western Himalayas (see also Table 1). Notable for moderate RWB events (e.g., Figure 13f), successive RWB occurrences resulted in rainfall amounts greater than 300 mm day−1.

4.3 Ageostrophic circulations and diabatic forcing

The ERA-Interim reanalysed circulation composites from the foregoing discussion provide some confidence to the hypothesis of the large-scale connections of RWB and eddy shedding mechanisms to extreme precipitation activity in the WEH region. However, owing to the limitations of the coarser reanalyses in addressing the smaller scale motion processes, e.g., diabatically-induced circulation manifestations and orographic effects, the results from high-resolution WRF simulations of two events which produced extreme precipitation for two consecutive days are discussed here. One of the events occurred during the monsoon advancing period of 16–19 June 2013 (Ray et al. 2013; Mishra 2015) which resulted in calamitous floods and the submerging of a large region in the WEH region (see Fig. 1), and the other occurred just prior to the monsoon withdrawing phase during 3–6 September 1995. The observed and simulated precipitation amounts from these events are shown in Fig. 16, which show a good correspondence in both the spatial spread and intensity.

Fig. 16
figure 16

IMD-observed (top panel) and 6 km grid WRF-simulated (bottom panel) total precipitation accumulations (mm) during the periods (a, c) 3–6 September 1995 and (b, d) 16–19 June 2013

Figures 17, 18 show the WRF simulated 200 hPa geopotential heights and ageostrophic winds diagnosed from the total horizontal winds, and the vertical motions arising from the secondary circulation across the jet streak for the periods 3–5 September 1995 and 16–18 June 2013. These two simulated events display a similar dynamical sequence and the differences in these events come from the relative strengths and geometries of features controlling dynamical processes. The jet entrance [exit] region of an idealized jet streak is typically characterized by a transverse circulation with the ageostrophic wind directed towards the cyclonic [anticyclonic] shear of the jet (Uccellini and Kocin 1987). One can see that the entire upper level jet streak amplifies in time as leftward-directed ageostrophic flow builds over the southwestern Himalayas a few days prior to heavy rainfall (Fig. 17). Anomalous conditions from all the case events also indicated similar ageostrophic wind vectors as discussed in the previous section. The effect of upper level ageostrophy on the vertical motion in the jet entrance region through the thermally-direct transverse circulation (Fig. 18) which clearly contributes to the ascending motion needed for the development of intense rainfall and the intensity is tied to the variations of the ascending flow which is controlled by the upper level ageostrophy. The heavy rains occur when low-level orographic lifting is phased with the divergent balanced ageostrophic right quadrant of the entrance of the jet.

Fig. 17
figure 17

WRF (6 km grid) simulated 200 hPa wind speeds (shaded; m s−1), geopotential height (solid contour; meters) and ageostrophic winds (full barb = 5 m s−1) during 3–5 September 1995 (top panel) and 16–18 June 2013 (bottom panel). Circled times indicates the region extreme precipitation activity

Fig. 18
figure 18

WRF (54 km grid) simulated vertical structure of p-velocity (ω; units in µb s−1) and potential temperature (Kelvin) prior to extreme precipitation events in September 1995 (top; along DE shown in Fig. 2) and June 2013 (bottom; along AB shown in Fig. 2) cases. Also shown is the isotachs to indicate upper level jet streak. Arrows show the ascending and descending nodes of transverse circulation across the jet entrance region

The diabatic forcing is generally strong in the right entrance region of intensifying jet streaks. Prior to extreme precipitation events, a new jet streak formation is noticed at the time of phasing of circulation between the upper level trough and the low-level west-northwest migration of the monsoon low from the Bay of Bengal to north-central part of India. This is facilitated by latent heating as air is advected northeastwards and lifted over the Himalayan foothills into the deepening trough. Moist convectively unstable atmospheric conditions were more prevalent below 650 hPa in the IGP region over a two day period prior to heavy rains. The latent heat released in the right entrance region of the developing jet ahead of the deepening trough accelerates as ageostrophic flow aloft strengthens the jet, i.e., heated parcels are accelerated into the jet which continues to strengthen the entrance region. The leftward-directed ageostrophic motion induces mass divergence in the right entrance region of the strengthening jet which causes surface pressure falls in the WEH foothill region. These pressure falls further strengthen the low-level air streams from the monsoon low towards far north-western parts of India, i.e., the low-level branch of the circulation is enhanced by orographic ascent and strong vertical motion associated with the thermally-direct ageostrophic circulation within the jet entrance region.

Figure 19 shows the simulated vertical structure of horizontal winds, vertical motion, and diabatic forcing from latent heating along southwest-northeast cross sections positioned across the western Himalayas for the two case events considered here for a few days prior to heavy rains. There is a strong divergent flow aloft forced by the diabatic component resulting from the latent heat release and local enhancements of 150–200 hPa velocity divergence are seen over the southern rim of the plateau prior to heavy rains (see also Fig. 11). Also evident is a strong mid-level ascent over the plateau early on, followed by strong ascent in the foothill region once thermally-direct forcing sets up due to strong upper-level ageostrophy—which is evidenced by the amplification of the jet streak downstream of the upper level deepening trough in time. Further, at the time of phasing of monsoon and subtropical circulation fields, wind speeds south of the foothill region are amplified and one can see a confluence of the strong southerly wind component with speeds greater than 15 m s−1 prevailing over most of the troposphere and directed towards the WEH region. This strengthens the orographic uplift leading to air parcels with large ascent rates in the foothill region.

Fig. 19
figure 19

Vertical cross sections [left (right) end of each figure indicates southwestern (northeastern) end] and displaying the temporal evolution of vertical p-velocities (ω; dashed contours; µb s−1), diabatic heating from microphysics latent heating (shaded; K hr−1) and horizontal winds (full barb = 5 m s−1) for the periods 3–5 September 1995 (top panel; along FG shown in Fig. 2) and 15–17 June 2013 (bottom panel; along AC shown in Fig. 2). Also shown is the isotachs (≥ 25 m s−1) by solid contours and J indicates upper level jet core. Big (green) arrows indicate the regions of strong ascending motion. Also shown along the cross section (indicated by a vertical line) are the closest locations where extreme precipitation was observed for these events

Furthermore, the localized enhancement of precipitation in the foothill region arises from the meso-β scale (Orlanski 1975) mass perturbations induced in the jet entrance region located in the northern flank of intensifying Tibetan anticyclone after the eddy shedding process (Fig. 17). Notice the inverted trough formation in the height field at the upper levels oriented normal to the mountain barrier with strong poleward ageostrophy prior to heavy rains while the 500 hPa trough is oriented in the southwest-northeast direction facilitating the southerly motion towards the WEH region. The synthesis of reanalysed and simulated fields clearly demonstrate that the upper level circulation fields begin with Rossby wave breaking mechanism that creates the necessary diffluence and vorticity first from upper then to mid-levels. Lagrangian backtrajectory analysis confirms this mechanism and for example, a parcel reaching 33.2ºN and 75ºE, i.e., into the trough, at 400 hPa and hourly diagnostics of parcel motion over a period of 2 days (15–17 June 2013) is shown in Fig. 20. One can see that parcels containing high PV from upper levels descend to mid-levels thereby enhancing the mid-level relative vorticity. Meridional parcel advection is clearly evident and parcels eventually reaching the STJ entrance region prior to commencement of heavy rains.

Fig. 20
figure 20

WRF-simulated parcel back trajectory (shown on the top of the figure) reaching the 400 hPa pressure level at 33.2ºN and 75ºE (into the trough; about 400 km from north of Uttarakhand). The x axis indicates time in hours, starting from 0000 UTC 15 June 2013 (0 h) and ending at 0000 UTC 17 June 2013 (48 h). Hourly parcel diagnostics shown in the figure are a the pressure (hPa) at the parcel location, b zonal wind speed (m s−1), c meridional wind speed (m s−1), d vertical p-velocity (ω; μb s−1), e potential temperature (K), f air temperature (°C), g potential vorticity (PVU), and h relative vorticity (s−1). Wider to narrower arrows indicate regions of parcel descent

5 Sequence and interaction of dynamical processes

A sequence of monsoon-extratropical interaction mechanisms result in impending flood-producing extreme precipitation episodes in the WEH region emerges from this study. This sequence of mechanisms is presented here as follows: Equatorward deepening of midlatitude westerly troughs begins with the upper level baroclinic RWB mechanism which facilitates streams of high potential vorticity air of stratospheric origin from extratropical latitudes that isentropically descends to subtropical latitudes located north of the extreme precipitation region. This descent of high-vorticity air enhances the mid-level cyclonic vorticity centered over the northwestern part of the Tibetan Plateau. The Tibetan anticyclone disperses far westward and sheds an eddy to its west and this provides support to trough thinning and cold air advection into the trough. As a consequence of the deepening of upper level troughs, new jet streaks form downstream of the trough which promote the transverse mode of ageostrophic circulation resulting in jet accelerations and strong upward motion in the jet entrance region. The upper level mass divergence in the entrance region induces surface pressure falls in the Himalayan foothill region. The meridional circulation associated with west-northwest propagating monsoon low comes in phase with the extratropical circulation which expedites the enhancements in mid-tropospheric cyclonic vorticity and organizes a confluence of strong southerly wind throughout most of the troposphere directed towards the WEH region. In concert with subsequent orographic lifting, the monsoon low aids strong moisture convergence into the more prevalent convectively unstable environment in the foothill region. The diabatic forcing from the latent heat release in the entrance region further amplifies the divergent outflow which intensifies the Tibetan anticyclone. Mesoscale ageostrophy and mass perturbations induced in the jet entrance region provide support to precipitation enhancements over a confined region in the foothill region. The strength of the westerly trough weakens in time and the upper level jet streak downstream of the trough moves eastward and consequently the Tibetan anticyclone weakens and the detached eddy re-merges with the Tibetan anticyclone with the westerly background flow in its north.

6 Concluding remarks

This study examines the dynamics of monsoon-extratropical interactions associated with flood-producing extreme precipitation events in the WEH foothill region during the monsoon rainy season based on analysis of gridded precipitation observations and global reanalysis circulation products. It is important to emphasize that the features associated with extreme precipitation events over the WEH foothills are rather different from the anomalous precipitation enhancement that occurs over the CEH foothills during monsoon-break situations (Vellore et al. 2014). In the latter case, the large-scale circulation is typically associated with dry air subsidence over the Indian subcontinent due to southward intruding extratropical westerly troughs into the mid-to-upper troposphere. In contrast, the extreme precipitation events over the WEH (e.g., Northwest Pakistan in July 2010 and Uttarakhand in June 2013) are accompanied by vigorous interactions of moisture-laden monsoon circulation and southward penetrating midlatitude westerly troughs into the Himalayan region.

The present investigation analyzed the evolution of monsoon and extratropical interactions based on 34 objectively identified extreme precipitation case events in the WEH region during the period 1979–2013. For these case events, the one-day accumulated precipitation over the WEH region ranged between 217 and 460 mm. A key issue addressed in this study, concerns the dynamics underlying the equatorward deepening of the midlatitude westerly trough into the WEH region. Our results provide evidence for a common large-scale circulation pattern that prevails during the evolution of these extreme precipitation events. This pattern consists of a blocking high over western Eurasia and a midlatitude westerly trough on its eastern side, together with an upper tropospheric Tibetan anticyclone in the subtropical belt.

In presence of this background pattern, Rossby wave breaking is found to occur downstream of the Eurasian blocking high about 2–3 days prior to commencement of heavy rains over the WEH region, which allows descent of high PV stratospheric air from the extratropics towards the northwestern Tibetan Plateau. Concomitant with these midlatitude circulation developments, our findings also indicate that the Tibetan anticyclone undergoes westward elongation and eddy shedding in response to localized upper-level divergent forcing from west-northwest propagating synoptic lows from the Bay of Bengal across the subcontinent. This is consistent with earlier investigations of westward shedding of eddies from an unstable and elongated Tibetan anticyclone in the summer monsoon environment. As a consequence, the Tibetan anticyclone bifurcates and facilitates southward penetration and positive tilting of the midlatitude westerly troughs leading to generation of jet streaks downstream of the trough. Strong convective motions are generated over the WEH foothills through interactions of the convectively coupled monsoon synoptic low and the deepened westerly trough. Coincident with these tropical and extra-tropical developments, the ageostrophic effects in the jet entrance region are found to generate transverse circulation across the Himalayas. Results from high-resolution numerical simulations suggest that this transverse circulation can further amplify in response to latent heat release over the WEH foothills and mesoscale orographic forcing.

While the synthesis of this study sheds light on the underlying dynamics of monsoon and extratropical coupled interactions during extreme precipitation events over the WEH region, they also offer fair guidance to impending heavy rains and disaster alertness through tracking of the following elements: (a) Midlatitude blocking pattern and Rossby wave breaking (b) West-northwest migration of synoptic monsoon low (c) Elongation of Tibetan anticyclone and eddy shedding (d) Strong convective motions over the WEH foothills. Further, this study is based on the internal dynamics associated with the extreme precipitation events and it should be noted that anomalous responses from large-scale external drivers associated with the sea state and ocean–atmosphere coupling (e.g., Krishnan et al. 2011; Mujumdar et al. 2012; Priya et al. 2015) and their impact on heavy precipitation in the Himalayas entails further investigation. Furthermore, the current state-of-the-art weather/climate models exhibit poor skill in the reproduction of localized precipitation extremes being the fact that processes involved in generation of these extremes are not yet fully represented in these modelling systems (e.g., Bader et al. 2008; Chaturvedi et al. 2012; Joseph et al. 2014). Improved representation of the aforementioned elements is crucial for better assessment of extreme precipitation events and associated hydrological stress, in particular over the geographically complex regions such as the Himalayan river basin (e.g., Hasson et al. 2013, 2014).