Kerala floods in consecutive years - Its association with mesoscale cloudburst and structural changes in monsoon clouds over the west coast of India

Abstract

Kerala, located at the southwest tip of India, witnessed disastrous floods during the monsoon seasons of two consecutive years, 2018 and 2019. This paper provides a detailed analysis of these two flood events using data from multiple sources. The unusually unstable and convective nature of the 2019 event, as detectable in its higher cloud tops and evidently fuelled by anomalously warm local sea temperatures, raises concerns regarding the changing patterns of rainfall over the southern parts of the west coast of India. Specifically, our analysis reveals that the flood of 2019 in Kerala satisfies the criteria for a mesoscale cloudburst (MsCB) event, more common in the north but a very rare and never before reported phenomenon in the Kerala region. Rainfall exceeding 50 mm in 2 h has been reported from many places between 8.00 and 22.00 UTC on the August 8, 2019. Satellite-derived rainfall and cloud microphysical parameters further reveal the uniqueness of the 2019 MsCB event. If 2019 is a harbinger of how global warming may continue to affect this region, transformations of the cloud structure and the recurrence and character of intense rainfall events could pose a major threat to the highly vulnerable Western Ghats ecosystems.

Full text

Weather and Climate Extremes 33 (2021) 100339
Available online 21 June 2021
2212-0947/© 2021 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Kerala oods in consecutive years - Its association with mesoscale
cloudburst and structural changes in monsoon clouds over the west coast
of India
P. Vijaykumar
a
, S. Abhilash
a
,
b
,
*
, A.V. Sreenath
b
, U.N. Athira
b
, K. Mohanakumar
a
, B.E. Mapes
c
,
B. Chakrapani
b
, A.K. Sahai
d
, T.N. Niyas
e
, O.P. Sreejith
e
a
Advanced Centre for Atmospheric Radar Research, Cochin University of Science and Technology, India
b
Department of Atmospheric Sciences, Cochin University of Science and Technology, India
c
University of Miami, USA
d
Indian Institute of Tropical Meteorology, Pune, India
e
India Meteorological Department, India
ARTICLE INFO
Keywords:
Kerala ood
Mesoscale cloudburst
Extreme rainfall
Monsoon clouds
Deep convection
ABSTRACT
Kerala, located at the southwest tip of India, witnessed disastrous oods during the monsoon seasons of two
consecutive years, 2018 and 2019. This paper provides a detailed analysis of these two ood events using data
from multiple sources. The unusually unstable and convective nature of the 2019 event, as detectable in its
higher cloud tops and evidently fuelled by anomalously warm local sea temperatures, raises concerns regarding
the changing patterns of rainfall over the southern parts of the west coast of India. Specically, our analysis
reveals that the ood of 2019 in Kerala satises the criteria for a mesoscale cloudburst (MsCB) event, more
common in the north but a very rare and never before reported phenomenon in the Kerala region. Rainfall
exceeding 50 mm in 2 h has been reported from many places between 8.00 and 22.00 UTC on the August 8, 2019.
Satellite-derived rainfall and cloud microphysical parameters further reveal the uniqueness of the 2019 MsCB
event. If 2019 is a harbinger of how global warming may continue to affect this region, transformations of the
cloud structure and the recurrence and character of intense rainfall events could pose a major threat to the highly
vulnerable Western Ghats ecosystems.
1. Introduction
Many studies have proposed a worldwide rise in the frequency and
intensity of extreme rainfall events as a consequence of global warming
caused by the growing amounts of CO
2
in the atmosphere (Allan and
Soden, 2008.; Houghton et al., 2001; Hennessey et al., 1997.; Fowler and
Hennessy, 1995.; etc.). Observational studies show an increasing trend
in heavy to extremely heavy rainfall events over the Indian mainland in
recent decades, noticeably in the central Indian region (Roxy et al.,
2017; Goswami et al., 2006). On the contrary, Guhathakurta et al.
(2015) have reported that the frequency of heavy rainfall events is
decreasing in eastern central India and north India while increasing in
peninsular, east, and northeast India. The main rainy season for
pan-India, the southwest monsoon (June to September), is known for its
regional heterogeneity in the mechanisms of cloud formation and the
resulting precipitation patterns (Vijaykumar et al., 2017; Sub-
rahmanyam and Kumar, 2013; Zuidema, 2003; Lau and Lau, 1992;
Grossman and Garcia, 1990). While most of the clouds in the central
Indian plains grow very deep, those along the Western Ghats (WG) re-
gion and over most of southern Peninsular India are modest in depth.
During the season, the deepest clouds appear over the head Bay of
Bengal, where the formation and intensication of monsoon depressions
is a major climatological feature of the Indian summer monsoon. Rain-
fall of short duration and high intensity is common over regions like the
Indo-Gangetic Plain, the northeast mountainous Himalayas, and arid
northwest India. However, west of the WG mountains in general and
over most regions of Peninsular India in particular south of about 14
o
N
latitude, rainfall intensity is generally low to medium (Francis and
Gadgil, 2006; Tawde and Sing, 2015), although its persistence makes the
windward side of the WG a major hotspot in terms of highest daily and
* Corresponding author. Advanced Centre for Atmospheric Radar Research, Cochin University of Science and Technology, India.
E-mail address: abhimets@gmail.com (S. Abhilash).
Contents lists available at ScienceDirect
Weather and Climate Extremes
journal homepage: www.elsevier.com/locate/wace
https://doi.org/10.1016/j.wace.2021.100339
Received 19 June 2020; Received in revised form 21 February 2021; Accepted 17 June 2021

Weather and Climate Extremes 33 (2021) 100339
2
seasonal cumulative rainfall in the country.
During the summer monsoon season, particularly in July and August,
sudden torrential downpours occur over the southern rim of western
Himalayas, leading to ash oods and landslides over small geographic
catchment areas typically about 20–30 km
2
(Das et al., 2006; Bhan et al.,
2004). Such events are dened as ‘cloudburst’ (CB) provided they satisfy
the condition that rainfall recorded in 1 h exceeds 100 mm (as per India
Meteorological Department - IMD denition). According to Bhan et al.
(2015), these cloud burst events are mostly associated with
westward-moving cyclonic circulations in the middle troposphere
(implying a cold-core in the lower troposphere) over the Tibet-Ladakh
region during active monsoon conditions. The orographic forcing and
strong convection leading to the formation of deep cumulonimbus
clouds as high as 15 km are responsible for most of the CB events (Dimri
et al., 2017). The occurrence of CBs has also been reported over the east
coast of India. The State of Kerala (Fig. 1), located at the southwest tip of
India, experienced unprecedented intense rainy spells during the
monsoon seasons of 2018 and 2019, resulting in severe ood situations
in many areas, affecting thousands of people claiming many lives and
damaging infrastructure and agriculture. This paper investigates the
dynamic and thermodynamic features during these two ood events,
and an attempt has been made to distinguish cloud characteristics of the
2019 ood event that we identify as a mesoscale cloudburst (MsCB).
It is well understood that orography plays a prominent role in the
southwest monsoon rainfall along the WG region. When the prevailing
moist wind from the warm Arabian Sea encounters the mountain range,
its forced upward lift produces ample amounts of rainfall on the wind-
ward side of the WG (e.g., Sarkar, 1966; 1967; Oruga and Yoshizaki,
1988) or upstream (Van de Boogard, 1977; Grossman and Durran,
1984). Lightning is rare over the west coast during the southwest
monsoon season, as the tops of such orographically induced clouds
usually do not grow beyond 8 km height (Rao, 1976). Nevertheless,
some studies (e.g., Grossman and Garcia, 1990; Roca and Ramanathan,
2000) suggest that deep convective clouds are common over the west
coast of India. Convective systems with larger spatial scales have been
found to sustain longer lifetimes and grow very deep (Chen and Houze,
1997). Subrahmanyam and Kumar (2013) show that deep convective
clouds in the region are conned mostly offshore from the west coast of
India.
Francis and Gadgil (2006) have studied heavy rainfall events
occurring over the WG region. They identied two main spots where the
probability of getting intense rainfall is the maximum, one between 14◦
N and 16◦N, and the other near 19◦N (Mumbai’s latitude). They
analyzed OLR patterns associated with a large number of extreme
rainfall events that occurred during the July and August months from
1974 through 1998 and listed out four types of circulation features
responsible for such events. They are (i) Tropical Convergent Zones
(TCZ) over a large-scale zonal belt, (ii) offshore convective system, (iii)
mid-tropospheric cyclone (MTC), and (iv) offshore vortex. Tawde and
Singh (2015) analyzed over 16 years of satellite-derived rainfall data
and reported that heavy rain bouts are least observed in Kerala. On
comparing the rain-producing mechanism and cloud microphysical
properties, Kumar et al. (2014) have observed that the warm rain pro-
cess is dominant near the Western Ghats, whereas cold rain from
mixed-phase processes is predominant near the Myanmar coast, east of
the Bay of Bengal.
Our analyses reveal that during August 6–11 of 2019, the clouds
grew unusually overriding into the atmosphere, causing a torrential
downpour and consequent ash ood. We speculate that profound
instability leading to vigorous updrafts and enhanced importance of ice
processes were at play. Another intense event in 2018 may suggest that
such events could become more frequent and can pose a potential threat
to fragile landforms of these regions.
2. Data and methodology
2.1. INSAT IR brightness temperature
Infrared brightness temperature (IRBT) provides the approximate
temperature of the emitting surface. The satellite Kalpana-1 (formerly
METSAT-1) of the Indian National Satellite (INSAT) series was launched
in 2002 (Kaila et al., 2002). For the present study, we have used hourly
IRBT data for the monsoon seasons of 2013–2017 obtained from the ftp
portal of the Meteorological and Oceanographic Satellite Data Archival
Centre (MOSDAC) of the Indian Space Research Organisation (ISRO).
The original data is averaged and projected to a uniform 0.25 ×0.25◦
mesh for the region 30◦S to 30◦N and 40◦E to 110◦E.
2.2. TRMM-GPM rainfall
The Tropical Rainfall Measuring Mission (TRMM) was designed to
monitor the tropical and subtropical precipitation and energy (e.g., see
Kummerow and Barnes, 1998; Adler et al., 2000; Huffman et al., 2007,
Tao et al., 2006). The Global Precipitation Measurement (GPM) mission
is the successor of TRMM, providing precipitation measurements from
space at half-hourly temporal resolution (Hou et al., 2014). We have
utilized 0.25 ×0.25◦three hourly TRMM-GPM 3B42RT data sets
(Huffman, 2016) for the same period as that of INSAT for the analysis.
The grid points of 3B42RT are chosen to coincide precisely with INSAT
to use for the pixel to pixel comparison between rain and corresponding
cloud heights.
Our analysis pertaining to the ve year period 2013–17 is to identify
the average cloud top temperature (CTT) corresponding to the intense
rainfall events over the Indian subcontinent. The pixels that record a
rainfall value exceeding 30 mm in 3-h duration are dened as ‘extreme’.
Since we used hourly BT data and 3 hourly TRMM data, we considered
each rainfall image, three consecutive BT images centred on the rainfall
data hour, to compute the average BT. In order to exclude non-cloudy
pixels, we used a minimum value of 283 K as the threshold to mark a
pixel cloudy. The contribution by rainfall exceeding or equals 10 mm/h
to the total seasonal rainfall is dened as fractional contribution by
extreme rainfall expressed in percentages.
For analysis of ood events, GPM_3IMERGHH precipitation data
(Huffman et al., 2019) with a half hour temporal resolution and spatial
resolution of 0.1 ×0.1◦have been used. The frequency distribution of
rainfall of various intensities during the two peak ood days was
analyzed by counting the ~121 km
2
pixels that recorded rainfall at
every 10 mm/2hr interval bins between the range of 10 and 60 mm/2hr.
2.3. Cloud hydrometeor proles
The daily cloud hydrometeor proles are derived from the GPM
constellation satellites using the Goddard Proling (GPROF) algorithm
at a resolution of 0.25 ×0.25 deg (Sims and Liu, 2015). The GPM sat-
ellite carries two core precipitation instruments, GPM Microwave
Imager (GMI) and Dual-frequency Precipitation Radar (DPR). In addi-
tion to the core instruments, the passive microwave algorithm is applied
to several constellation radiometers with similar channel sets as the GMI
radiometer. The vertical proles of the hydrometeor species are derived
from lookup tables based on sensor-specic data sets, surface classi-
cation, and a-priori matching proles from different microwave sensors.
We have taken available daily hydrometeor proles over the spatial
domain close to the extreme precipitation period. The data set is ob-
tained from https://gpm1.gesdisc.eosdis.nasa.gov/data/GPM_L3/. We
have selected the swaths close to the Kerala domain on each represen-
tative day.
Cloud optical thickness is derived from visible and near-infrared
channel radiances obtained from a Moderate resolution imaging spec-
trometer (MODIS). The data sets are available from https://modis.gsfc.
nasa.gov/data/dataprod/mod06.php, and details of the cloud products
P. Vijaykumar et al.

Weather and Climate Extremes 33 (2021) 100339
3
can be obtained from Platnick et al. (2015).
2.4. Thermodynamical and dynamical elds
The daily zonal, meridional winds, temperature, humidity, and
geopotential height elds are obtained from European Centre for Me-
dium Range Weather Forecasting (ECMWF) ERA version 5 (ERA5)
reanalysis data (Hersbach and Dee, 2016). The ERA-5 data set also
provides convective available potential energy (CAPE), which is one of
the critical instability parameters governing vertically developed cloud
systems. The data set has a spatial resolution of 0.25 ×0.25 deg. We
have computed moist static energy (MSE) and vertically integrated
moisture convergence during the four consecutive days of extreme
precipitation events over Kerala in 2018 and 2019.
Vertically Integrated Moisture Flux Convergence (VIMFC) measures
the apparent dynamical sink of vapor in a region (related to precipita-
tion). At the same time, Moist Static Energy (MSE) is a measure of the
water vapor itself, at least in the tropics, where the temperature is nearly
constant across geography and time. Both are elevated during deep
convection (Szoke, 2018; Waldstreicher, 1989). We have computed
VIMFC and MSE during the extreme precipitation events over Kerala in
2018 and 2019.
VIMFC is dened in this study as the horizontal moisture ux
convergence integrated between 1000 hPa and 200 hPa.
VIMFC =1/g⎛
⎝∫
200hPa
1000hPa
∂
uq /
∂
x+
∂
vq /
∂
y⎞
⎠dp
In this equation q is the specic humidity, u and v are the zonal and
meridional components of the wind velocity respectively, p is the
pressure and g is the acceleration due to gravity.
MSE =(Sensible heat +geopotential +latent heat)
where Sensible heat =CpT; geopotential =gz; latent heat =L
v
q, with C
p
the specic heat of dry air at constant pressure (1 004 J K
−1
kg
−1
), T
temperature, z altitude, and L
v
latent heat of condensation (2.5 ×10
6
J
kg
−1
). MSE is three dimensional, while VIMFC is a horizontal map.
We have used NOAA Extended Reconstructed Sea Surface Temper-
ature (SST) V4 to compare the seasonal evolution of ocean temperature
during the years 2018 and 2019. The spatial resolution of this monthly
dataset is 2 ◦×2 ◦(Huang et al., 2014, 2015; Liu et al., 2014). Higher
resolution daily SST anomalies (relative to 2003–2014 daily clima-
tology, Chin et al. 2017) and daily-mean IMERG rainfall totals can also
be viewed https://go.nasa.gov/3pvVrNG for the 2019 event.
3. Results and discussions
This section examines the occurrence of extreme rainfall events over
the Indian region and associated cloud structure during the ve years
from 2013 to 2017. Synergic use of INSAT infrared brightness temper-
ature and TRMM-GPM data sets exposes the simultaneous evolution of
rainfall and cloud structure on a sub-daily time scale. The contribution
of extreme rainfall to the cumulative seasonal rainfall over different
regions of the Indian sub-continent is presented in Fig. 2. Mean cloud
heights (inferred from brightness temperatures) corresponding to the
extreme and modest-rate rainy pixels were used to interpret the extreme
rainfall events.
There are two major rainy areas in the Indian Summer monsoon
domain where the seasonal cumulative rainfall is more than 250 cm, one
Fig. 1. The topography and location of Kerala in the south west tip of Indian subcontinent sandwiched between the Western Ghats and the Arabian Sea.
P. Vijaykumar et al.

Weather and Climate Extremes 33 (2021) 100339
4
Fig. 2. a) Cumulative rainfall in cm for JJAS period, b) Frequency of occurrence of extreme (>10 mm/h) rainfall events in percentage, c) percentage fractional
contribution by extreme rainfall events to the total seasonal rainfall and d) cloud top temperature overhead the extreme rainfall events. The period of analysis is JJAS
during 2013–17.
Fig. 3. (a) Frequency of occurrence (in %) of very deep clouds having cloud top temperature colder than 220 K and (b) extreme rain events where 3 hourly cu-
mulative rainfall exceeds 30 mm. The period is JJAS for 2013–2017.
P. Vijaykumar et al.

Weather and Climate Extremes 33 (2021) 100339
5
along the west coast of India, west of WG, and the other along the
Myanmar coast that extends northward up to the southern ranges of
Himalayas (see Fig. 2a). The monsoon trough region over the central
Indian plains also receives widespread rainfall during the season. The
frequency of extreme rainfall events (dened as those having instanta-
neous satellite-estimated rainfall rates exceeding 10 mm/h on a 0.25◦
pixel scale) is relatively higher in northeast India and neighbouring
mountainous regions (Fig. 2b). Along the west coast of Myanmar, such
events are frequent, but not over India’s southwest coast. Over a large
area in the Gangetic planes and the west coast of India bounded nearly
between 18◦N and 21◦N latitude, such events occur more frequently as
compared to the rest of India. Fig. 2c shows that out of the total rainfall
amounts for the season, more than 35% in northeast and northwest India
and about 25–35% over central India is received from intense rainfall
events.
These results conrm previous ndings by other authors (Francis and
Gadgil, 2006; Suthinkumar et al., 2019) but for a different study period.
The average brightness temperature (BT) of cloudy pixels (dened as
~25 km scale pixels having BT <283 K), concurrent within the hour of
TRMM-derived extreme rain, for the study period is provided in Fig. 2d.
It can be seen that the clouds that produce extreme rainfall are very deep
(<210K) over the entire Bay of Bengal. Over most regions in the
Indo-Gangetic planes and northwest India, the average CTT is less than
220 K. Interestingly, the average BT of clouds along the west coast,
especially south of about 18
o
N latitude, is between 220 and 235 K,
indicating the relatively shallow tops of the clouds that produce extreme
rainfall in the region.
Fig. 3a provides the frequency of occurrence of clouds with the top
temperature lower than 220 K along the west coast of India during the
monsoon season from 2013 to 2017. It can be noticed that deep clouds
tend to occur more frequently over the ocean, away from the coast. A
larger pool of deep cloudiness is located over the southeast Arabian Sea
and a relatively smaller pool west of Mumbai. Between nearly 10◦N and
18◦N latitude, deep clouds are inhibited, where the wind is interacting
with the mountainous WG and exhibits the shape of an oblique bowl.
Fig. 3b provides the frequency of extreme rainfall events dened as in-
tensity greater than 10 mm per hour. Notice that extreme rain events are
almost absent over the ocean, despite the high frequency of occurrence
of deep clouds. Perhaps the oceanic region is contaminated by cirrus
debris originating from the convective overshoots over the WG region
and carried westward by the strong easterlies aloft during the monsoon
season. The latitudinal variation of frequency of deep clouds and
extreme rain within the 1 ×1 degree boxes marked in Fig. 3b along the
west coast of India is shown in Fig. 4. Each box straddles the local
coastline.
From Fig. 4, we can identify two regions where the probability of
extreme rainfall is maximum along the west coast of India. One between
11◦N and 14◦N and another between 18◦N and 20
o
N latitude. Using
station-recorded rainfall data for the period 1951 and 1987, Francis and
Gadgil (2006) have identied almost similar regions as hotspots of
heavy rainfall along the west coast of India, but their southern hotspot
lay closer to 15-16◦N, where the Western Ghats are closer to the coast.
We hypothesize that extreme rain events in our 11-14◦N latitudes,
where the deep cloud frequency is relatively low (Fig. 4), may be forced
ascent from low-level winds meeting orography. On the other hand, the
extreme rains between 18◦N and 20◦N are accompanied by towering
clouds, which suggests the predominant role of deep buoyant
convection.
These results for the period 2013–2017 conrm that deep clouds in
concurrence with extreme rains are absent over the WG region, espe-
cially south of 14◦N during the summer monsoon season, in agreement
with previous studies (Francis and Gadgil, 2006; Subhramaniyam and
Kumar, 2013; Kumar et al., 2014; Deshpande et al., 2018.). With this
background, we further focus on two specic 2018 and 2019 ood
events over Kerala to identify the distinct cloud characteristics and
rainfall intensities in these two extreme rainfall events.
The evolution of rainfall over Kerala during 2018 and 2019 are
distinct, even though seasonal totals came out similar. Fig. 5 shows the
cumulative area-averaged rainfall over Kerala during 2018 and 2019. In
2018, the monsoon hit the Kerala coast on May 28, whereas in 2019, the
monsoon was delayed by seven days and arrived on June 8 (IMD
Report). From the slope of the normal curve in Fig. 5, the daily rainfall
rate is maximum during June and July, with a typical rainfall rate of
above 20 mm/day. After that, the daily rainfall rate drops signicantly
in August and September. During the 2018 and 2019 monsoon seasons,
extreme rainfall events in both years occurred in August.
Four active spells characterized 2018: one in June, one in July, and
two in August. The rst spell of August 2018 lasted from 8 to 10th of
August and the next one from 12th to 18th of the same month. In 2018,
the maximum rainfall was recorded on August 15, just below 140 mm
(its map may be viewed at https://go.nasa.gov/3pCYSC8). In 2019,
three active spells occurred after onset: one each during July, August,
and September, in which the most intense occurred between the 6th and
11th of August. It is quite unusual that the State received above 150
mm/day on one day, August 8 (map at https://go.nasa.gov/3bnRF3L),
which caused ooding in many parts of the state, after a substantially
Fig. 4. Frequency of deep clouds (<220 K) and extreme rainfall (>10 mm/h) in 1 ×1 degree boxes selected along the west coast of India (as depicted in Fig. 3).
P. Vijaykumar et al.

Weather and Climate Extremes 33 (2021) 100339
6
drier than average June and July. Supplementary Fig. S1 depicts the
percentage departure of daily rainfall over Kerala, more than 853% and
998% of normal on 15th August 2018, and August 8, 2019, respectively.
The at cumulative curve in Fig. 5 shows very little rainfall in September
2018, after the extreme rainfall event in August.
The synoptic features during the two events were distinct. The wind
at 850 hPa (top row of Fig. 6) reveals that the monsoon ow in general
and the low level Jetstream (LLJ) in particular exhibited typical active
monsoon patterns from the rst week of August in 2018, culminating in
maximum speed during the heavy rain spell of 13–17 August. The core
wind speed was above 20 m s
−1
, with speeds above 10 m s
−1
extending
from the Southern Hemisphere trades to the South China Sea. In com-
parison, the overall monsoon ow was comparatively weaker for the
August 7–11, 2019 spell (Fig. 6 bottom right), although it was almost
equally strong (near 20 m s
−1
) close to Kerala and upstream over the
northern part of the Arabian Sea.
High SST values prevailed over the North Indian Ocean (NIO) during
the monsoon season of 2019. Immediately offshore of Kerala, anomalies
of 1–2 ◦C prevailed (https://go.nasa.gov/2ZxiSLX) as part of a general
warmth of the western Indian ocean upwind of India in the monsoon
ow’s path (https://go.nasa.gov/2Nfqi4c). In contrast, August 2018 had
normal, or cool SST anomalies also intensied in a cold anomaly
immediately off the Kerala coast (https://go.nasa.gov/3blwDmt).
Fig. 7 provides the monthly SST anomalies for the years 2018 and
2019. After the onset of summer monsoon in June, the SST values over
the Arabian Sea generally drop due to strong Somali current, while in situ
rainfall together with river runoff leads to cooling of the Bay of Bengal
(BOB) surface waters (Rao, 1976; Varkey et al., 1996). The SST anomaly
over the oceanic region around the Indian subcontinent during July and
August in 2018 remained insignicant. In contrast, the SST remained
Fig. 5. Time series of daily progress of cumulative rainfall in Kerala (area shown in Fig. 1) during 2018 (Green) and 2019 (red) monsoon seasons. The black line
indicates the multi-year climatological mean cumulative rainfall. (For interpretation of the references to colour in this gure legend, the reader is referred to the Web
version of this article.)
Fig. 6. Horizontal wind pattern at 850 hPa during the active monsoon spells during 2018 (Top panels) and 2019 (Bottom panels).
P. Vijaykumar et al.

Weather and Climate Extremes 33 (2021) 100339
7
anomalously high over the entire NIO during the monsoon season of
2019. Analyzing the Latent heat ux-OLR relationship over the equa-
torial Indian Ocean, Kemball-Cook and Wang (2001) found that when
moisture ux from the ocean to the atmosphere is high, convection is
aided by building up of moist static energy. Roxy and Tanimoto (2007)
suggest that weak wind speed can lead to anomalous warming of surface
waters over the Arabian Sea during the pre-active phase of WG rain
spells followed by anomalously high wind speed in the southern Arabian
Sea. During the WG active phase, a north-south temperature gradient
develops that favours convective activity over the WG region.
Consistent with the warmer water, convective available potential
energy (CAPE) during the 2019 active spell remained very high over
most regions of the Arabian Sea south of nearly 15
o
N latitude and a
large area over the Bay of Bengal (Fig. 8, right column). In contrast,
during August 14–17, 2018, the CAPE values were low over most parts
of the Arabian Sea and near-zero at the Kerala coast where the water was
especially cold (https://go.nasa.gov/3blwDmt). Sing et al. (2014) found
that wet spells during 1980–2011 exhibit enhanced moisture
convergence and an increase in CAPE over much of the core monsoon
region while there are no signicant changes in the low level cyclonic
circulation anomalies. They suggest that these conditions support
stronger convection, especially over the southern and eastern parts of
the core monsoon region. They also observed a decline in the intensity of
dry spells after 1980, which was mostly attributed to both a weaker low
level divergence and higher CAPE values. Kumar et al. (2014) have
provided convincing reasons for less ice formation in the WG region
despite the presence of higher amounts of cloud liquid water in the lower
atmosphere. They suggest that the low values of CAPE are responsible
for the suppression of ice and graupel formation in the region. Thus, if
the CAPE values remain high enough, buoyant updrafts may ascend to
higher levels, conducive to the formation of deeper cloud systems.
The vertically integrated moisture ux convergence is analyzed
adequately (noting that it must agree with P-E since the time tendency of
vapor is small) over the Kerala region during both 2018 (August 13–17)
and 2019 (August 7–11) oods (Fig. 9). However, some spectral arte-
facts of topography and nite differencing are seen. More importantly,
Fig. 7. Monthly Sea Surface Temperature anomaly for the year 2018 (left panel) and 2019 (right panel) during June to September (top to bottom).
P. Vijaykumar et al.

Weather and Climate Extremes 33 (2021) 100339
8
Fig. 8. Spatial distribution of Convective available potential energy (CAPE) during August 14–17, 2018 (left panel) and August 7–10, 2019 (right panel) in units of
J/kg.
Fig. 9. Vertically integrated moisture ux convergence (shaded) overlaid with moisture ux transport (vectors) during the peak rainfall episodes (a) 2018 (August
13–17) and (b) 2019 (August 7–11).
P. Vijaykumar et al.

Weather and Climate Extremes 33 (2021) 100339
9
the ux vectors themselves reveal the presence of a monsoon depression
over the central Indian region is evident in both years. The ux transport
vector was almost perpendicular to the coast and WG during 2019. This
ow directly up the WG mountain slope might have contributed to the
development of deep clouds, also supported by the high value of CAPE
supported by the anomalously warm coastal waters. In other words, the
extreme rainfall event during 2019 had both dynamical and thermo-
dynamical factors in favourable alignment.
We now examine the 5-day “spells” encompassing these peak
ooding episodes in Kerala, between 13 and 17 of August 2018 and 7–11
August 2019. From here on, we refer to these two 5-day periods as 2018
and 2019 active spells, respectively. An analysis of daily rainfall during
the 2018 and 2019 active spells over Kerala is provided in supplemen-
tary Fig. S2. The most remarkable feature is the enormous downpour
recorded on the August 08, 2019. The maximum rainfall values are
located over the Kozhikkode, Malappuram, and Idukki districts of Ker-
ala, where rainfall of 260 mm per day and above is observed. Deshpande
et al. (2018) provide the statistics of “cloudbursts” and “mini cloud-
burst” events during the monsoon season in India. Cloudbursts are
short-duration violent heavy rainfall that usually occurs over a limited
area. According to Deshpande et al. (2018), there are two categories of
cloudbursts. An intense rainfall event that occurs in the Himalayan
mountain ranges can produce ash oods, landslides, and loss of human
life and properties. Such an event is category ‘a’ cloudburst (CBa)
dened irrespective of the amount of rainfall. Category ‘b’ cloudbursts
(CBb) are events where the hourly accumulated rainfall is 100 mm or
above over a smaller geographical area. They further dene a third
category of cloud bursts called mini cloudbursts (MCB) to represent
rainfall events where two-hourly accumulated rainfall is 50 mm or
above. Such events can also produce ash oods and catastrophes like
the other two categories of cloudbursts. MCB events have been found to
occur over the Western Ghats region, mostly in June.
The peak daily rainfall during the 2018 active spell occurred on 15th
August. Fig. 10 provides the 2 h of accumulated rainfall for that day. A
maximum of 25–30 mm/2 h was recorded in Ernakulum and Thrissur
districts during 16.00–18.00 UTC hours. The 2-h rainfall during other
periods of the day remained less than 20 mm. It indicates that the
rainfall of August 15, 2018 remained distributed almost equally
throughout the 24 h period, but still had the potential to produce heavy
ooding over a large geographical area, especially the low lying areas on
the banks of major rivers in Kerala. According to Mishra and Srinivasan
(2013), events like debris slides in Kedarnath on 17th June 2013 can
occur even without concurrent cloudbursts or heavy rainfall events. If
the cumulative rainfall for a given period is higher, relatively low in-
tensity rainfall can also trigger landslides and the bursting of lakes. The
2018 Kerala ood was not because of the sudden torrential downpour
Fig. 10. Two hour accumulated rainfall on August 15, 2018.
P. Vijaykumar et al.

Weather and Climate Extremes 33 (2021) 100339
10
over any smaller area but due to accumulated rainfall, as can be seen
from Fig. 5 and supplementary Fig. S1.
On the other hand, 2 h accumulated rainfall observed on August 08,
2019 provided in Fig. 11 shows that rainfall exceeding 50 mm/2 h have
occurred over a few locations in Kerala, noticeably like, Kozhikkode,
Malappuram, and Idukki during 8.00–10.00 UTC, 12.00–14.00 UTC,
16.00–18.00 UTC, 18.00–20.00 UTC, and 20.00–22.00 UTC. This
qualies the denition of MCB of Deshpande et al. (2018). However, this
event occurred over a larger area than that cloudbursts or MCB event
usually affect (typically 50–100 square km). Pixels having high rainfall
values are distributed spatially at mesoscale, and we designate this event
as a mesoscale cloudburst (MsCB). Unlike the usual Mesoscale convec-
tive systems that occur along the west coast of India (Ramage 1971;
Houze and Churchill 1987; Virts and Houze, 2016), this system pro-
duced such intense rainfall to name it a mini cloudburst. This event is
perhaps the rst of its kind in the recorded weather history of Kerala in
August during the monsoon season. Flash oods occurred following this
MsCB event in many regions of Malappuram and Kozhikode. Debris
slides and landslides were reported from several places in these two
districts. Many people and animals lost lives. A comparison of the fre-
quency distribution of different categories of intensities of recorded
rainfall during 2018 and 2019 active spells is provided in Fig. 12. A large
number of pixels recorded above 50 mm/h and between 40 and 50
mm/h rainfall in 2019. In 2018, above 40 mm/h rain pixels were absent.
During 2019, rainfall pixels with an intensity 30–40 mm/h were more
than 15 times that of 2018. A six-fold increase in 20–30 mm/h rain
pixels is recorded in 2019 compared to 2018.
The proles of moist static energy (MSE) during the peak rainfall day
of 2018 and the 2019 MsCB events, together with the hydrometeor
proles provided in Figs. 13 and 14, respectively, reveal that deep
convection coincides with high values of MSE in the lower and middle
levels of the troposphere. The MSE values south of roughly 10◦N appear
to be rather low, inhibiting the deepening of the convective system.
Interestingly, the approximate 10.7◦N latitude is a narrow channel in
the WG Mountains known as the Palghat Gap, through which the lower
level westerly wind ows uninterrupted during the monsoon seasons.
Based on observational and modelling studies, Meenu et al. (2020)
report that the 2018 heavy rain spells are characterised by mixed-phase
clouds and ice phase precipitation bands. They propose that a mesoscale
cluster with a leading convective core as deep as 16 km, windward of the
southern WG catalysed by the supply of moisture convergence from
strong westerly jet resulted in the heavy rain spells of August 2018. In
agreement with this, during both the years, an abundance of solid hy-
drometeors are present between 10◦N and 11◦N. But in 2019, another
distinct cluster appears between 11◦N and 12◦N. The cloud water
content in the northern cluster is relatively low, but large amounts of
Fig. 11. Two hour accumulated rainfall on August 8, 2019.
P. Vijaykumar et al.

Weather and Climate Extremes 33 (2021) 100339
11
rainwater mostly contributed by the melting of higher level solid hy-
drometeors are present. Huang and Cui (2015) have used WRF model
simulations to study the dominant cloud microphysical processes of the
torrential rain. They propose that the accretion of cloud water and the
melting of graupel were the main source terms that caused the rapid
growth of rainwater and enhanced surface. It indicates that a precipi-
tation mechanism similar to the one over the Myanmar coast has
resulted in a ood situation in the southwest ranges of WG during both
these years.
Therefore, it may be assumed that the high mid-level MSE values
might have supported and sustained the vertical development of clouds
in the two pockets leading to the MsCB event. The initiation mechanism
for the lifting of air at the boundary level can probably be orographic but
could have further intensied due to the availability of high MSE sus-
tained by moisture advected by the wind from the warm Arabian Sea.
Fuelled by continuous moisture supply from the lower levels, mid-level
instability must have carried enough water droplets into the higher al-
titudes that froze to form ice and snow particles and remained as huge
convective pillars for a few hours. A few days before this MsCB event, the
weak rainfall might have set the background atmosphere for this 2019
event.
Hence we infer or interpret that the ooding that occurred in Kerala
in the 2019 monsoon season was different from the 2018 Kerala ood,
despite many things in common. One of the reasons why the 2018 Kerala
ood became more widespread and severe was the consistent excess in
daily rainfall for an extended period prior to the major ood event. From
the onset of the monsoon, the rainfall was above normal, and by the end
of July, most of the reservoirs in Kerala had reached the maximum ca-
pacity level. The extended heavy spell that started in the rst week of
August set the stage for the ooding that happened from 15th-18th. The
area-averaged seasonal cumulative rainfall for Jun–Sep 2018 for Kerala
was still not the highest in the recorded history. On the other hand, the
2019 Kerala ood was associated with the MsCB event on the 8th of
August. The accumulated seasonal rainfall till the rst week of August
has little effect on this ooding. Such torrential downpours or ash
oods are not commonly seen south of 14◦N latitude over the Western
Ghats. The presence of deeper clouds more towards the east and the
resulting heavy convective rainfall is well manifested during this 2019
episode. The Cloud Ice Optical Thickness (CIOT) shown in supplemen-
tary g- S3 provide evidence for the abundance of ice formation during
the two ood episodes with noticeable high values during 2019.
The present study reveals that during the monsoon years of 2013–17,
Fig. 12. Frequency distribution of rainfall of various intensities. (a) 2018 active spell and (b) 2019 active spell.
Fig. 13. The vertical prole of moist static energy between 5 and 15
o
N (a) and the vertical distribution of different hydrometeors (b) in the atmosphere associated
with the Active Spell of 2018 (August 15, 2018).
P. Vijaykumar et al.

Weather and Climate Extremes 33 (2021) 100339
12
one of the intense rainfall hotspots in the WG is located between 11◦N
and 13◦N. In the recent two years (2018 & 2019), more intense rainfall
was occurring between 10◦N and 12◦N. Even though the methodology
and data sets used by Francis and Gadgil (2006) differ, their hotspot of
heavy rainfall was still northward (between 15◦N and 16◦N). Though
inconclusive, we suspect a southward extension of heavy rainfall core
along the west coast of India. A recent study by Suman and Maity (2020)
reveals more increase in extreme precipitation in terms of its magnitude
over south India compared to north and central India during 1971–2017
(base period: 1930–1970). Considering the fragile nature of Western
Ghats ecology, this situation is raising many concerns. Events like MsCB
are unprecedented in the region and can be a potential threat to highly
sensitive WG ecosystems.
4. Summary
Our analysis reveals that a mesoscale mini cloudburst event occurred
over Kerala between 8 and 22 UTC on August 8, 2019. In terms of the
number of casualties and areas affected, the 2018 Kerala ood is
considered a major ood and 2019 a minor one. However, we suggest
that the latter is to be seen as far more convincing evidence of the
regional impact of ongoing global climate change. While the 2018 ood
partly resulted from a large excess of monsoon rainfall accumulated
throughout the season up to mid-August, the rain pattern that caused the
2019 ood is different. The warm near-coast SST anomalies of 2019 are
unprecedented among all August 8 dates (https://go.nasa.gov/
3pCJiGS), and their longevity is conrmed by 2019 being an outlier
among July 9 dates as well (https://go.nasa.gov/3qGdyBV). While de-
pressions may always bring on heavy rain, the character of rain events
could become more convective if SST crosses a subtle vertical instability
threshold. Such a change in character would have implications for ash
ooding and impacts and may also be detectable by higher cloud tops, as
seen above.
To our best understanding, such MsCB events have not been recorded
in the region ever since the meteorological data collection started. One
of the hotspots of heavy rainfall in the Konkan region between 14◦N and
16◦N seems to have shifted more southward with likely fatal conse-
quences, although longer records are needed to see if this is a systematic
shift. An increase in rainfall intensity may suggest a rising probability for
landslides in the high to mid-land slopes of Western Ghats in eastern
Kerala during the monsoon seasons. The Western Ghats have been
subject to modications by human intervention, chiey for crop culti-
vation and the nature of the land favours landslides of multiple scales.
Further, our analysis reveals that the west coast of India is prone to
massive ooding both from a moderate to high intensity rain spell that
follows a prolonged wet period and also from events such as a cloudburst
that pours enormous amounts of precipitation in a very short period.
Heavy precipitation in a short duration brings runoff water beyond the
capacity of the rivers and the sloping topography from high land WG to
low laying west coast accelerates the rush of oodwater. Under normal
monsoon conditions, the water level in the rivers of Kerala during July
and August remains high. It suggests that a prolonged/intense spell of
surplus rainfall during these months that follows a normal June
monsoon has a huge potential to produce ooding near the river basins
of Kerala. In addition, as the present study highlights, mesoscale
cloudbursts occurring under favourable ocean-atmospheric conditions
may leave a vast area of the State vulnerable to ash oods and land-
slides, any time during the monsoon season. The present work draws
attention to the importance of monitoring the extreme rainfall events
associated with changing climate over the WG region. Studies using high
resolution models may improve our understanding of the changing na-
ture of the rainfall in the region and help adopt better strategies to
reduce future risks.
CRediT authorship contribution statement
P. Vijaykumar: Conceptualization, Writing – original draft. S.
Abhilash: Conceptualization, Methodology, Funding acquisition. A.V.
Sreenath: Software, Validation. U.N. Athira: Software, Validation. K.
Mohanakumar: Supervision. B.E. Mapes: Writing – review & editing.
B. Chakrapani: Writing – review & editing. A.K. Sahai: Funding
acquisition, Conceptualization. T.N. Niyas: Data curation. O.P. Sree-
jith: Resources.
Declaration of competing interest
The authors declare that they have no known competing nancial
interests or personal relationships that could have appeared to inuence
the work reported in this paper.
Fig. 14. The vertical prole of moist static energy between 5 and 15
o
N (a) and the vertical distribution of different hydrometeors (b) in the atmosphere associated
with the MsCB event of 2019 (August 8, 2019).
P. Vijaykumar et al.

Weather and Climate Extremes 33 (2021) 100339
13
Acknowledgments
INSAT data is accessed from https://www.mosdac.gov.in of MOS-
DAC, ISRO. ERA-5 data is provided by ECMWF and can be accessed from
the web link https://www.ecmwf.int/en/forecasts/datasets/reanalysi
s-datasets/era5. TRMM-GPM data is obtained from NASA/Goddard
Space Flight Center archived at NASA GES DISC at https://pmm.nasa.
gov/data-access/downloads/gpm. Hydrometeor proles are obtained
from https://disc.gsfc.nasa.gov/datasets/GPM_3GPROFGPMGMI_DAY_
05/summary?keywords=GPROF. NOAA_ERSST_V4 data provided by
the NOAA/OAR/ESRL PSD, Boulder, Colorado, USA, from their Web site
https://www.esrl.noaa.gov/psd/. MODIS Cloud Optical Thickness is
obtained from NASA accessed through https://atmosphere-imager.gsfc.
nasa.gov/products/cloud. This work has been carried out under the
nancial support of the Ministry of Earth Sciences (MoES), Government
of India.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.wace.2021.100339.
References
Adler, R.F., Huffman, G.J., Bolvin, D.T., Curtis, S., Nelkin, E.J., 2000. Tropical rainfall
distributions determined using TRMM combined with other satellite and rain gauge
information. J. Appl. Meteorol. 39 (12), 2007–2023. https://doi.org/10.1175/1520-
0450(2001)040<2007:TRDDUT>2.0.CO;2.
Allan, R.P., Brian, J.S., 2008. Atmospheric warming and the amplication of
precipitation extremes. Science 321 (5895), 1481–1484. https://doi.org/10.1126/
science.1160787.
Bhan, S.C., Paul, S., Kharbanda, K.L., 2004. Cloudbursts in Himachal Pradesh. Mausam
55 (4), 712–713.
Bhan, S.C., Devrani, A.K., Sinha, V., 2015. An analysis of monthly rainfall and the
meteorological conditions associated with cloudburst over the dry region of Leh
(Ladakh), India. Mausam 66 (1), 107–122.
Chen, S.S., Houze, R.A., 1997. Diurnal variation and life-cycle of deep convective systems
over the tropical Pacic warm pool. Q. J. R. Meteorol. Soc. 123 (538), 357–388.
https://doi.org/10.1002/qj.49712353806.
Chin, T.M., Vazquez, J., Armstrong, E.M., 2017. A multi-scale high-resolution analysis of
global sea surface temperature. Rem. Sens. Environ. 200, 154–169. https://doi.org/
10.1016/j.rse.2017.07.029.
Das, S., Raghavendra, A., Moncrieff, M.W., 2006. Simulation of a Himalayan cloudburst
event. J. Earth Syst. Sci. 115 (3), 299–313. https://doi.org/10.1007/BF02702044.
Deshpande, N.R., Kothawale, D.R., Kumar, V., Kulkarni, J.R., 2018. Statistical
characteristics of cloud burst and mini-cloud burst events during monsoon season in
India. Int. J. Climatol. 38 (11), 4172–4188. https://doi.org/10.1002/joc.5560.
Dimri, A.P., Chevuturi, A., Niyogi, D., Thayyen, R.J., Ray, K., Tripathi, S.N., Mohanty, U.
C., 2017. Cloudbursts in Indian Himalayas: a review. Earth Sci. Rev. 168, 1–23.
https://doi.org/10.1016/j.earscirev.2017.03.006.
Fowler, A.M., Hennessy, K.J., 1995. Potential impacts of global warming on the
frequency and magnitude of heavy precipitation. Nat. Hazards 11 (3), 283–303.
https://doi.org/10.1007/BF00613411.
Francis, P.A., Gadgil, Sulochana, 2006. Intense rainfall events over the west coast of
India. Meteorol. Atmos. Phys. 94 (1–4), 27–42. https://doi.org/10.1007/s00703-
005-0167-2.
Goswami, B.N., Venugopal, V., Sengupta, D., Madusoodhanan, M.S., Xavier, P.K., 2006.
Increasing trend of extreme rain events over India in a warming environment.
Science 314 (5804), 1442–1445. https://doi.org/10.1126/science.1132027, 2006.
Grossman, Robert L., Durran, Dale R., 1984. Interaction of low-level ow with the
western Ghat Mountains and offshore convection in the summer monsoon. Mon.
Weather Rev. 112 (4), 652–672. https://doi.org/10.1175/1520-0493(1984)
112<0652:IOLLFW>2.0.CO;2.
Grossman, Robert L., Garcia, Oswaldo, 1990. The distribution of deep convection over
ocean and land during the Asian summer monsoon. J. Clim. 3 (9), 1032–1044.
https://doi.org/10.1175/1520-0442(1990)003<1032:TDODCO>2.0.CO;2.
Guhathakurta, P., Rajeevan, M., Sikka, D.R., Tyagi, A., 2015. Observed changes in
southwest monsoon rainfall over India during 1901–2011. Int. J. Climatol. 35 (8),
1881–1898. https://doi.org/10.1002/joc.4095.
Hennessey, K.J., Gregory, J.M., Mitchell, J.F.B., 1997. Changes in daily precipitation
under enhanced greenhouse conditions. Clim. Dynam. 13, 667–680. https://doi.org/
10.1007/s003820050189.
Hersbach, H., Dee, D., 2016. ERA5 reanalysis is in production. ECMWF NewsletterNo 147
(7).
Hou, A.Y., Kakar, R.K., Neeck, S., Azarbarzin, A.A., Kummerow, C.D., Kojima, M.,
Oki, R., Nakamura, K., Iguchi, T., 2014. The global precipitation measurement
mission. Bull. Am. Meteorol. Soc. 95, 701–722. https://doi.org/10.1175/BAMS-D-
13-00164.1.
Houze Jr., R.A., Churchill, D.D., 1987. Mesoscale organization and cloud microphysics in
a Bay of Bengal depression. J. Atmos. Sci. 44, 1845–1867. https://doi.org/10.1175/
1520-0469(1987)044<1845:MOACMI>2.0.CO;2.
Houghton, R.A., Hackler, J.L., Cushman, R.M., 2001. Carbon Flux to the Atmosphere
from Land-Use Changes: 1850 to 1990. Oak Ridge: Carbon Dioxide Information
Center. Environmental Sciences Division, Oak Ridge National Laboratory.
Huang, Y., Cui, X., 2015. Dominant cloud microphysical processes of a torrential rainfall
event in Sichuan, China. Adv. Atmos. Sci. 32, 389–400. https://doi.org/10.1007/
s00376-014-4066-7.
Huang, B., Thorne, P.W., Smith, T.M., Liu, W., Lawrimore, J., Banzon, V.F., Zhang, H.,
Peterson, T.C., Menne, M., 2015. Further exploring and quantifying uncertainties for
extended Reconstructed Sea surface temperature (ERSST) version 4 (v4). J. Clim. 29
(9), 3119–3142. https://doi.org/10.1175/JCLI-D-15-0430.1.
Huang, B., Banzon, V.F., Freeman, E., Lawrimore, J., Liu, W., Peterson, T.C., Smith, T.M.,
Thorne, P.W., Woodruff, S.D., Zhang, H.M., 2014. Extended Reconstructed Sea
surface temperature version 4 (ERSST.v4): Part I. Upgrades and intercomparisons.
J. Clim. 28, 911–930. https://doi.org/10.1175/JCLI-D-14-00006.1.
Huffman, G.J., 2016. The transition in multi-satellite products from TRMM to GPM
(TMPA to IMERG). available at: https://pmm.nasa.gov/sites/default/les/docume
nt_les/TMPA-to-IMERG_transition.pdf.
Huffman, G.J., Bolvin, D.T., Nelkin, E.J., Wolff, D.B., Adler, R.F., Gu, G., Hong, Y.,
Bowman, K.P., Stocker, E.F., 2007. The TRMM multisatellite precipitation analysis
(TMPA): quasi-global, multiyear, combined-sensor precipitation estimates at ne
scales. J. Hydrometeorol. 8 (1), 38–55. https://doi.org/10.1175/JHM560.1.
Huffman, G.J., Stocker, E.F., Bolvin, D.T., Nelkin, E.J., Tan, Jackson, 2019. GPM IMERG
Final Precipitation L3 Half Hourly 0.1 Degree X 0.1 Degree V06, Greenbelt, MD.
Goddard Earth Sciences Data and Information Services Center (GES DISC). https://
doi.org/10.5067/GPM/IMERG/3B-HH/06. Accessed: [01/12/2019].
Kaila, V.K., Kirankumar, A.S., Sundaramurthy, T.K., Ramakrishnan, S., Prasad, M.Y.S.,
Desai, P.S., Jayaraman, V., Manikiam, B., 2002. METSAT-a unique mission for
weather and climate, 83.9. Current Science, pp. 1081–1088.
Kemball-Cook, S., Wang, B., 2001. Equatorial waves and air–sea interaction in the boreal
summer intraseasonal oscillation. J. Clim. 14 (13), 2923–2942. https://doi.org/
10.1175/1520-0442(2001)014<2923:EWAASI>2.0.CO;2.
Kumar, S., Hazra, A., Goswami, B.N., 2014. Role of interaction between dynamics,
thermodynamics and cloud microphysics on summer monsoon precipitating clouds
over the Myanmar Coast and the Western Ghats. Clim. Dynam. 43 (3–4), 911–924.
https://doi.org/10.1007/s00382-013-1909-3.
Kummerow, C., Barnes, W., 1998. The tropical rainfall measuring mission (TRMM)
sensor package. J. Atmos. Oceanic technol. 15 (3), 809–817. https://doi.org/
10.1175/1520-0426(1998)015<0809:TTRMMT>2.0.CO;2.
Lau, K.H., Lau, N.C., 1992. The energetics and propagation dynamics of tropical
summertime synoptic-scale disturbances. Mon. Weather Rev. 120, 2523–2539.
https://doi.org/10.1175/1520-0493(1992)120<2523:TEAPDO>2.0.CO;2.
Liu, W., Huang, B., Thorne, P.W., Banzon, V.F., Zhang, H.M., Freeman, E., Lawrimore, J.,
Peterson, T.C., Smith, T.M., Woodruff, S.D., 2014. Extended Reconstructed Sea
Surface Temperature version 4 (ERSST.v4): Part II. Parametric and structural
uncertainty estimations. J. Clim. 931–951. https://doi.org/10.1175/JCLI-D-14-
00007.1.
Meenu, S., Gayatri, K., Malap, N., Murugavel, P., Samanta, S., Prabha, T.V., 2020. The
physics of extreme rainfall event: an investigation with multisatellite observations
and numerical simulations. J. Atmos. Sol. Terr. Phys. 204, 105275. https://doi.org/
10.1016/j.jastp.2020.105275.
Mishra, A., Srinivasan, J., 2013. Did a cloud burst occur in Kedarnath during 16 and 17
June 2013? Curr. Sci. 105 (10), 1351–1352.
Oruga, Y., Yoshizaki, M., 1988. Numerical study of orographic convective precipitation
over eastern Arabian Sea and the Ghat mountains during the summer monsoon.
J. Atmos. Sci. 45, 2097–2121. https://doi.org/10.1175/1520-0469(1988)
045<2097:NSOOCP>2.0.CO;2.
Platnick, S., Ackerman, S.A., King, M.D., Meyer, K., Menzel, W.P., Holz, R.E., Baum, B.A.,
Yang, P., 2015. MODIS Atmosphere L2 Cloud Product (06_L2). NASA MODIS
Adaptive Processing System, Goddard Space Flight Center.
Ramage, C.S., 1971. Monsoon Meteorology. International Geophysics Series, vol. 15.
Academic Press, p. 296.
Rao, Y.P., 1976. Southwest Monsoon, vol. 376. India Meteorological Department, New
Delhi.
Roca, R., Ramanathan, V., 2000. Scale dependence of monsoonal convective systems
over the Indian Ocean. J. Clim. 13 (7), 1286–1298. https://doi.org/10.1175/1520-
0442(2000)013<1286:SDOMCS>2.0.CO;2.
Roxy, M.K., Ghosh, S., Pathak, A., Athulya, R., Mujumdar, M., Murtugudde, R.,
Terray, P., Rajeevan, M., 2017. A threefold rise in widespread extreme rain events
over central India. Nat. Commun. 8 (1), 708. https://doi.org/10.1038/s41467-017-
00744-9.
Roxy, M., Tanimoto, Y., 2007. Role of SST over the Indian Ocean in inuencing the
intraseasonal variability of the Indian summer monsoon. J.Meteorol. Soc. Japan. 85
(3), 349–358. https://doi.org/10.2151/jmsj.85.349.
Sarkar, R.P., 1966. A dynamical model of orographic rainfall. Mon. Weather Rev. 95,
555–572. https://doi.org/10.1175/1520-0493(1966)094<0555:ADMOOR>2.3.CO;
2.
Sarkar, R.P., 1967. Some modication in a dynamical model of orographic rainfall. Mon.
Weather Rev. 95, 673–684. https://doi.org/10.1175/1520-0493(1967)095<0673:
SMIADM>2.3.CO;2.
Sims, E.M., Liu, G., 2015. A parameterization of the probability of snow–rain transition.
J. Hydrometeorol. 16, 1466–1477. https://doi.org/10.1175/JHM-D-14-0211.1.
P. Vijaykumar et al.

Weather and Climate Extremes 33 (2021) 100339
14
Sing, Deepti, Michael, Tsiang, Bala, Rajaratnam, Diffenbaugh S, Noah, 2014. Observed
changes in extreme wet and dry spells during the South Asian summer monsoon
season. nature climate change 4, 456–461. https://doi.org/10.1038/nclimate2208.
Subrahmanyam, K., Kumar, K.K., 2013. CloudSat observations of cloud-type distribution
over the Indian summer monsoon region. Ann. Geophys. 31, 1155–1162. https://
doi.org/10.5194/angeo-31-1155-2013.
Suman, M., Maity, R., 2020. Southward shift of precipitation extremes over south Asia:
evidences from CORDEX data. Sci. Rep. 10, 6452. https://doi.org/10.1038/s41598-
020-63571-x.
Suthinkumar, P.S., Babu, C.A., Varikoden, H., 2019. Spatial distribution of extreme
rainfall events during 2017 southwest monsoon over Indian subcontinent. Pure Appl.
Geophys. 176, 5431–5443. https://doi.org/10.1007/s00024-019-02282-5, 1-13.
Szoeke De, S.P., 2018. Variations of the moist static energy budget of the tropical Indian
Ocean atmospheric boundary layer. J. Atmos. Sci. 75 (5), 1545–1551.
Tao, W.K., Smith, E.A., Adler, R.F., Haddad, Z.S., Hou, A.Y., Iguchi, T., Kakar, R.,
Krishnamurti, T.N., Kummerow, C.D., Lang, S., Meneghini, R., Nakamura, K.,
Nakazawa, T., Okamoto, K., Olson, W.S., Satoh, S., Shige, S., Simpson, J.,
Takayabu, Y., Tripoli, G.J., Yang, S., 2006. Retrieval of latent heating from TRMM
measurements. Bull. Am. Meteorol. Soc. 87 (11), 1555–1572. https://doi.org/
10.1175/BAMS-87-11-1555.
Tawde, S.A., Singh, C., 2015. Investigation of orographic features inuencing spatial
distribution of rainfall over the Western Ghats of India using satellite data. Int. J.
Climatol. 35 (9), 2280–2293. https://doi.org/10.1002/joc.4146.
Van de Boogaard, H., 1977. The mean circulation of the tropical and subtropical
atmosphere-July. NCAR Tech. Note, NCAR/TN-118+STR. National Center for
Atmospheric Research 48. https://doi.org/10.5065/D6H41PCJ.
Varkey, M.J., Murty, V.S.N., Suryanarayana, A., 1996. Physical Oceanography of the Bay
of Bengal and Andaman Sea. Oceanography and Marine Biology, vol. 34. George
Allen and Unwin., London, UK, pp. 1–70.
Vijaykumar, P., Abhilash, S., Santhosh, K.R., Mapes, B.E., Kumar, C.S., Hu, I.K., 2017.
Distribution of cloudiness and categorization of rainfall types based on INSAT IR
brightness temperatures over Indian subcontinent and adjoining oceanic region
during south west monsoon season. J. Atmos. Sol. Terr. Phys. 16 (1), 76–82. https://
doi.org/10.1016/j.jastp.2017.06.017.
Virts S, Katrina, Houze Jr A, Robert, 2016. Seasonal and Intraseasonal Variability of
Mesoscale Convective Systems over the South Asian Monsoon Region. Journal of
Atmospheric Sciences 73 (12), 4753–4774. https://doi.org/10.1175/JAS-D-16-
0022.1.
Waldstreicher, J.S., 1989. A guide to utilizing moisture FluxConvergence as a predictor
of convection. Natl. Weather Digest 14 (4), 20–35.
Zuidema, P., 2003. Convective clouds over the Bay of Bengal. Mon. Weather Rev. 131
(5), 780–798. https://doi.org/10.1175/1520-0493(2003)131<0780:CCOTBO>2.0.
CO;2.
P. Vijaykumar et al.