Content uploaded by KOTESWARARAO KUNDETI
Author content
All content in this area was uploaded by KOTESWARARAO KUNDETI on Oct 22, 2017
Content may be subject to copyright.
INTERNATIONAL JOURNAL OF CLIMATOLOGY
Int. J. Climatol. (2017)
Published online in Wiley Online Library
(wileyonlinelibrary.com) DOI: 10.1002/joc.5216
Investigation of temperature changes over India in association
with meteorological parameters in a warming climate
G. Purnadurga,aT.V. Lakshmi Kumar,a*K. Koteswara Rao,bM. Rajasekharc
and M.S. Narayanand
aDepartment of Physics, Atmospheric Science Research Laboratory, SRM University, Chennai, India
bSchool of Earth and Environmental sciences, Seoul National University, Seoul, South Korea
cSathish Dhawan Space Centre, ISRO, Srihari Kota, India
dDRVE, SRM University, Chennai, India
ABSTRACT: We have used 1∘×1∘resolution maximum temperature (TMAX) data sets developed by India Meteorological
Department (IMD) to examine the summer time warming over India during the period 2001– 2014 in comparison with
the period 1971–2000. The two study periods have been arrived at based on the drastic change of Moisture Index (IM)
trends over India between the two epochs. The TMAX variations over India are discussed with the corresponding changes
in Potential Evapotranspiration (PET) data of the Climate Research Unit (CRU) and Outgoing Longwave Radiation (OLR)
data of NOAA ESRL. The study shows a considerable warming over northern parts of India compared to southern parts.
Western Himalayas (WH) and Northwest (NW) regions experienced highest warming with 1.4 ∘C and 0.8 ∘C increases during
Epoch 2 (2001–2014) as compared Epoch 1 (1971 –2000) during the summer (March, April and May). Using MERRA Black
Carbon Surface Mass Concentration (BCSMC) data, we have analysed the relation of increasing BCSMC with the TMAX over
different homogeneous temperature regions of India and found that BCSMC has increased upto 1.6 times between the two
epochs. Strong linear association is found between TMAX, PET and OLR evidenced by Coherence Wavelet Spectral analysis.
It is also found that the highest warming occurred in the month of March and is 2.2 ∘C in WH and 1.4 ∘CinNWpartsof
India. We calculated mass stream function based on zonal mean meridional velocity for the two periods. In the recent periods
we observed the weakening of polar cell and northward expansion of Hadley cell. These changes may be related to warming
conditions of the atmosphere which may explain the intensication and northward expansion of the Ferrel cell with favourable
conditions during the summer season.
KEY WORDS maximum temperatures; pet; OLR; mass stream function; India
Received 5 November 2016; Revised 28 April 2017; Accepted 27 June 2017
1. Introduction
It is reported by IPCC (2013) that the change in global
mean surface temperatures will be around 0.3 ∘Cto
0.7 ∘C during 2016–2035 when compared to the period
1986–2005. Hegerl et al. (2007) reported that the global
warming from the beginning of industrial era is caused
by the anthropogenic greenhouse gas emissions. Studies
of Scott and Jones (2012) have shown, based on the
temperature records till 2010, that the global warming
is set to continue in 21st century. It is also reported that
annual surface temperatures over India have signicantly
increased during 1901–2013 (Annual Climate Summary,
2014). This increase in temperature over India is a result of
signicant occurrence of heat waves with higher frequency
and increasing duration (Rohini et al., 2016). The conse-
quent numbers of days with highest temperature, known
as heat waves, are strongly linked to ENSO (Ratnam et al.,
* Correspondence to: T.V. Lakshhmi Kumar, Department of Physics,
Atmospheric Science Research Laboratory, SRM University, Chennai
603203, India. E-mail: lkumarap@hotmail.com
2016). Many studies over India have found that there is an
increase in mean and maximum temperatures in space and
time scales (Arora et al., 2005; Srivastava et al., 2017).
These analyses have been carried out based on the station
data and also from data of homogeneous temperature
regions. The main reason for the increase of temperatures
are found to be global warming and global teleconnections
(Kothawale and Rupa Kumar, 2005; Jaswal et al., 2015).
Besides the above, the contribution of aerosols can also
be thought of. black carbon (BC) aerosols which mainly
originate from fossil fuel and incomplete combustion,
absorb the solar radiation and contribute to climate warm-
ing (Ban-Weiss et al., 2012). The atmospheric heating is
further increased by the anthropogenic emissions of BC.
Jacobson (2010) found that the increase in BC, particularly
biofuel soot gases result in cloud and atmospheric heating
due to their hygroscopic nature. Meehl et al. (2008) used
the coupled ocean-atmospheric climate model to study
the 21st century simulations and reported that the BC and
organic aerosols increase the lower tropospheric heating
over south Asian region. Hence, it is very important to see
the relation of BC and surface temperatures to understand
© 2017 Royal Meteorological Society
G. PURNADURGA et al.
the warming over a region. It is also well known that the
Hadley Cell is the principal controller of the precipitation
patterns in the tropical regions. There is an increasing
interest on the changes of width and strength of Hadley
Cell in the context of global warming. Kang and Lu (2012)
studied the expansion of Hadley Cell and reported that it
is extending polewards in all seasons. It is also reported
that the changes in meridional temperature gradient of the
troposphere suggest a total widening of ∼2∘in latitude
since 1979 (Johanson and Fu, 2009).
In the present study, we analyse the gridded surface
air temperature data over India to evaluate the variations
of TMAX for the two epochs 1971–2000 (Epoch 1) and
2001–2014 (Epoch 2). The main aim of this study is to
show that there is a considerable warming over India dur-
ing Epoch 2 compared to Epoch 1. The starting year of
the study period is chosen as 1971, mainly to see the
temperature rise after the rapid industrialization in India
(Kothawale et al., 2016). The two Epochs 1971– 2000 and
2001–2014 are arrived at, based on the break trend anal-
ysis of Moisture Index (IM) time series derived over India
using Thornthwaite and Mather (1955) revised water bal-
ance concept. The break trend analysis which is applied
in this study is useful to analyse the abrupt changes in IM
for the two study epochs. A drastic decline was observed
in IMover India during the summer months of March,
April and May around the year 2000 and hence the total
study period was divided into two epochs as 1971– 2000
and 2001–2014. The TMAX differences over these two
epochs are examined along with the corresponding varia-
tions of Potential Evapotranspiration and Outgoing Long-
wave Radiation (OLR) over India. The possible reasons for
the TMAX rise over different homogeneous regions along
with those over All India are discussed by associating them
with PET and OLR variations. The Black Carbon Sur-
face Mass Concentration (BCSMC) data from MERRA
has been used to examine the relation of BC with the
TMAX over India. Further, we also analysed the width of
the Hadley Cell for the above mentioned two epochs using
mass stream function (MSF).
2. Data and methodology
The main data sources for this study are: India Meteoro-
logical Department (IMD), Climate Research Unit (CRU)
and NOAA – ESRL for data on daily TMAX, monthly PET
and interpolated monthly OLR, respectively. The daily
TMAX data have been averaged over a month and is used
as monthly mean TMAX for the present study. The grid
resolution of TMAX data is 1∘×1∘covering the Indian land
mass for the period 1951–2014 (Srivastava et al., 2009).
The PET data sets are available with the 0.5∘×0.5∘grid
resolution and OLR at 2.5∘×2.5∘resolution. All the three
data sets were averaged over all the grids falling over each
of the seven temperature homogeneous regions of Indian
land mass (Figure 1, IMD – www.tropmet.res.in). Thus,
average TMAX over the seven homogenous regions and
over whole of India were computed. Monthly rainfall and
69°E
WH- Western Himalaya, NW- North-west, NC- North central
NE - Northeast, WC- West coast, EC- East coast
IP- Interior peninsula
9°N
12°N
15°N
18°N
21°N
24°N
27°N
30°N
33°N
36°N
WH
NW
NC
WC
NE
IP
EC
72°E 75°E 78°E 81°E 84°E 87°E 90°E 93°E 96°E
Figure 1. Temperature homogeneous regions of India.
PET data have been used to run the Thoronthwaite and
Mather water balance model and Humidity and Aridity
indices are thus, derived. The difference of humidity
and aridity index is known as IMby which climate of
a region is categorized as perhumid (A) (IM>100%),
Humid (B4) (100<IM>80), Humid (B3) (80<IM>60),
Humid (B2) (60<IM>40), Humid (B1) (40<IM>20),
Moist subhumid (C2) (20<IM>0), Dry subhumid (C1)
(0 <IM>−33.3), Semi arid, D (−66.7 <IM>−33.3)
and Arid, E (−100 <IM>−66.7) as per Thoronthwaite
Climate Classication (1948). The rainfall data for this
analysis have been taken from IMD with 1∘×1∘resolution
developed by Rajeevan et al. (2006) from the station data.
The seven temperature homogeneous regions of India are
(1) Western Himalayas (WH), (2) North West (NW), (3)
North Central (NC), (4) North East (NE), (5) West Coast
(WC), (6) East Coast (EC) and (7) Interior Peninsula (IP).
The difference of maximum temperatures for each of
the three summer months, March, April and May and for
average of all these three summer months, were obtained
to see the TMAX rise over India on spatial scale. Simi-
larly, the differences, as above, were obtained for PET
and OLR to understand their association with TMAX. Daily
temperatures have been calculated for different homoge-
neous regions and for All India been studied to quantify
the corresponding rise in TMAX rise over different homoge-
neous regions as well as for All India. Coherence Wavelet
Spectrum has been obtained (details are given in Section 3)
for TMAX and PET and also with OLR to examine the vari-
ation of PET and OLR with TMAX. Pearson correlation
technique is also used to understand the strength of the
relation among these variables.
The BCSMC data (in microgram per cubic meter) have
been obtained from Modern ERA Retrospective Analy-
sis for Research (MERRA) products which are developed
by Global Modelling and Assimilation Ofce Research
© 2017 Royal Meteorological Society Int. J. Climatol. (2017)
TEMPERATURES OVER INDIA IN ASSOCIATION WITH METEOROLOGICAL PARAMETERS
Site – National Aeronautics and Space Administration
(GMAO NASA). The data are available on monthly scale
with 0.5∘×0.625∘grid resolution over the globe from the
year 1980 till to date. The authors extracted the same data
for the different homogeneous regions of India as well
for the whole of India from 1980–2014. This data for the
above period has been compared with the monthly TMAX
data and a correlation analysis was carried out for the
aforementioned period.
Mass stream function (Oort and Yienger, 1996), which
is the integral the northward mass ux above a particular
pressure level has been calculated using the formula
𝜓=2Πacos 𝜓
gv∫p
0
dp
where 𝜓is Mass Stream Function, 𝜈is the
zonally-averaged meridional velocity, ais the Earth
radius, gis the gravity, Ψis the latitude and pis the
pressure level.
The MSF has been calculated for the two epochs and
studied in the context of increasing temperatures over
India. The wind data for the MSF calculation is taken from
the NCEP-NCAR Reanalysis data sets with 2.5∘×2.5∘
resolution.
3. Results and discussion
Figures 2(a) –(d) depict the break trends for the time series
of IMover India derived from the Thoronthwaite and
Mather revised water balance model for the Epoch 1
(1971–2000) and Epoch 2 (2001 – 2014) for the summer
season and for the 3 months. Due to southwest monsoon
showers over India from June, we have not included June
for this analysis.
From visual inspection of the gures we infer that IMis
distinctly different in all the individual months and also for
the season. From Figure 2(a), it can be seen that IMvaried
from - 40 to – 50 units during Epoch 1 which corresponds
to semi arid climatic nature. The period Epoch 2 has shown
the variation from - 60 to - 45 units, which also corresponds
to semi arid nature but the value of IM(−60 units) is very
close to arid category. From Figures 2(a)– (d) it is found
that there is a drastic dip in IMvalues during 1998 to 2005
and the trend is different during the two epochs. The aver-
age value of IMfor the summer season (MAM average)
for Epoch 1 is - 44 units and is - 55 units for Epoch 2. The
difference of IMis more in the month of March and is of
about −15 units. The value of IMreects the climatology
of a region and further allows us to infer the changes in
climatic indices such as rainfall and temperature. Ravin-
dranath (2011) studied the Moisture Index derived from
the Thoronthwaite Climatic Approach over tropical region
from 1901 to 2006. It is found from their study that the
period from 1970 showed negative deviations of Moisture
Index which revealed the dry weather conditions over trop-
ics. The analysis of climatology over India carried out
by Raju et al. (2013) indicated the increased regions of
semi-arid, arid and dry sub humid in different parts of India
–40
–35
–50
–45
–60
–55
–30
–65
–40
–35
–50
–45
–60
–55
–30
–20
–50
–40
–70
1970 1975 1980 1985 1990
Year
May
April
March
MAM Average
(a)
(b)
(c)
(d)
Moisture index (IM)
1995 2000 2005 2010 2015
–60
–30
–60
–55
–50
–45
–40
–35
–30
Figure 2. Break trends of Moisture Index (IM) of All India for the periods
1971– 2000 and 2001 –2014 obtained for (a) summer season (March,
April and May average), (b) March, (c) April and (d) May months.
during 1971 to 2005. They also reported that the rise in
temperatures and no noticeable trend in rainfall will lead
to more arid nature over a region. In the present study, the
difference in IMbetween the two study periods has been
veried for the level of signicance using student t test
(two tailed). It is found that the difference in IMvalue dur-
ing summer season, March, April and May are at 0.01level
of signicance.
This prompted us to further examine the changes in
surface air temperature over India on spatial and tempo-
ral scale for the aforementioned periods. We also have
seen the changes in temperatures associated with poten-
tial evapotarnspiration (PET) and OLR over India. Since
the temperature impacts the estimation of evapotranspira-
tion and it is very important parameter for agriculture, we
examined the corresponding variations in them. In the sub-
sequent sections, the spatial as well as time scale changes
of TMAX, PET and OLR have been discussed.
Figures 3(a)– (d) depict the space scale variations of
TMAX difference over India for the two epochs for the aver-
age summer season and for, March, April and May months,
respectively. It is found from the gures that the tempera-
ture rise was highest in WH region, followed by NW parts
and NE parts of India. The same has been observed dur-
ing the months of March, April and May but the incre-
ment of TMAX is pronounced during March followed by
April and May. The gures also reveal that the southern
parts of India have undergone less warming effects than
the northern parts. Some parts of Karnataka have shown
© 2017 Royal Meteorological Society Int. J. Climatol. (2017)
G. PURNADURGA et al.
MAM
9°N
12°N
15°N
18°N
21°N
24°N
27°N
30°N
33°N
36°N
(a)
69°E 72°E 75°E 78°E 81°E 84°E 87°E 90°E 93°E 96°E
March
9°N
12°N
15°N
18°N
21°N
1. 8
1. 6
1. 4
1. 2
1
0.8
0.6
0.4
0.2
0
24°N
27°N
30°N
33°N
36°N
(b)
69°E 72°E 75°E 78°E 81°E 84°E 87°E 90°E 93°E 96°E
April
9°N
12°N
15°N
18°N
21°N
24°N
27°N
30°N
33°N
36°N
(c)
69°E 72°E 75°E 78°E 81°E 84°E 87°E 90°E 93°E 96°E
May
9°N
12°N
15°N
18°N
21°N
24°N
27°N
30°N
33°N
36°N
(d)
69°E 72°E 75°E 78°E 81°E 84°E 87°E 90°E 93°E 96°E
Figure 3. Difference in TMAX on space scale over India for the periods 2001– 2014 and 1971–2000 for (a) summer season (March, April and May
average), (b) March, (c) April and (d) May. [Colour gure can be viewed at wileyonlinelibrary.com].
even a decrement of TMAX during the 2001– 2014 epoch
than in the corresponding 1971–2000 epoch. The region of
Deccan Plateau has yielded least change in temperature of
0.5 ∘C. It has been reported by Kothawale and Rupakumar
(2005) that all India mean annual temperature increased
by about 0.22 ∘C per decade during 1971 –2003. Also,
they have reported that some of the regions of India have
shown signicant warming trend during 1971– 2003. Pal
and Al-Tabbaa (2010) in their study on long term changes
of extreme temperatures over India, found that the max-
imum temperatures increased unevenly over India during
the last century. They have used the monthly maximum and
minimum temperature data of homogeneous temperature
regions provided by Indian Institute of Tropical Meteorol-
ogy, Pune, India for their analysis. They have concluded
that there are large variations in temperature changes with
highest values in WH region. The present investigation
also portrays that there is a large difference of TMAX over
WH region between the two epochs.
The study of Ganguly and Iyer (2009) on long term
variations of surface air temperatures during summer in
India showed that the air temperatures increased over
NW, NC and IP regions when compared with the EC,
WC and WH regions of India. According to their study,
the Aerosol Optical Depth (AOD) and BC optical depth
showed increasing trend over NW India. Ramachandran
and Cherian (2008) studied the seasonal variations of
aerosol characteristics from 2001 to 2005 over India using
MODIS Terra data. It is found from their study that Srina-
gar, located in WH region, has shown increased tendency
of AOD from 2001 to 2005. Also, it is reported in their
study that the states of Jammu and Kashmir and parts
of WH region have low SO2uxes. It is also found by
Nair et al. (2013) that BC concentration attains maximum
values in premonsoon season over Western Himalayas.
Also, they have reported that the direct and surface albedo
radiative forcing leads to considerable warming over
Himalayan region during premonsoon season. A gradual
increase in absorbing aerosols due to biomass burning and
transport activities raises the possibility that they could
have contributed to warmer temperature scenario over WH
region. Similar analysis has been carried out in the present
study using the BCSMC data obtained from the Modern
ERA Retrospective Analysis for Research (MERRA)
products developed by GMAO NASA. Different data sets
from MERRA have been used to study the aerosol charac-
teristics over India. Gaurav et al. (2015) used wind elds
at different atmospheric levels obtained from MERRA
to understand the inuence of transport on aerosols
over Indian subcontinent. Mean modelled surface ozone
© 2017 Royal Meteorological Society Int. J. Climatol. (2017)
TEMPERATURES OVER INDIA IN ASSOCIATION WITH METEOROLOGICAL PARAMETERS
(a)
(b)
Figure 4. The Seasonal variation of BCSMC (𝜇gm
−3) of MERRA over (a) Dibrugarh (2008– 2014) and (b) Chennai (2011–2014). [Colour gure
can be viewed at wileyonlinelibrary.com].
concentration from MERRA has been used to understand
the air pollution impacts on Indian agriculture (Burney and
Ramanathan, 2014). By using the satellite aerosol mea-
surements and MERRA reanalysis data, Kim et al. (2016)
found that the absorbing aerosols can impact seasonal
and interannual variations in Indian summer monsoon
season. In the present study, the authors compared the
seasonal variations of BCSMC obtained from MERRA
and Black Carbon Mass Concentration from in-situ
measurements which are reported at Dibrugarh (27.3∘N,
94.6∘E) and Chennai (12.8∘N, 80.03∘E) located in NE
and EC regions, respectively. Figures 4(a) and (b) depict
the seasonal variations of BCSMC from the year 2008
to 2014 over Dibrugarh and 2011– 2014 over Chennai,
obtained from MERRA. It can be observed that over
Dibrugarh (Figure 4(a)), the BCSMC has shown higher
values during the winter season in general (January and
February), followed by Summer (March, April and May),
Northeast monsoon (October, November and December)
and Southwest monsoon season (June, July, August and
September), respectively. Studies of Pathak et al. (2010)
reported that the Black Carbon Mass Concentration
obtained from in-situ measurements have also shown
higher values during the winter season (16.3 μgm
−3)
followed by northeast monsoon (10.9 μgm
−3), premon-
soon (7.5 μgm
−3) and monsoon (3.4 μgm
−3) seasons,
respectively. Similarly, in-situ measurements of BC over
Chennai also displayed the broad maximum of BC during
NE and winter months when compared to other seasons
(Aruna et al., 2013). From these analyse, it can be inferred
that the BCSMC from MERRA is able to show the sea-
sonal variability as depicted by the in-situ measurements
over Dibrugarh and Chennai. However, the bias of BC
mass concentration of MERRA is still to be examined
with the in-situ measurements.
In the present study, the BC data over all the
homogeneous regions have shown increment during
the period 2001–2014 when compared to 1980 – 2000.
During 2001–2014, the BC in WH and NW parts have
increased 1.6 and 1.5 times, respectively compared to
that during the period 1980–2000. The BC values for
WH and NW regions for the periods 1980–2000 and
2001–2014 during summer season are 0.12, 0.18 and
0.77, 1.17 𝜇gm
−3, respectively. It is also observed that
the other regions such as NC, NE, IP, WC, EC have
shown, respectively 1.6, 1.6, 1.3, 1.4 and 1.6 times of BC
during epoch 2 vis a vis during Epoch 1. To understand
the relation of BCSMC with TMAX, we have carried out
the correlation analysis for the months March, April and
May months of the study period. The Pearson correlation
co-efcient along with the level of signicance are given
in Table 1. Among all temperature homogeneous regions,
NE part showed the highest correlation of +0.69 with 0.01
level of signicance during the month of March compared
to all other months. The regions WH and NW parts showed
the correlations of 0.56 and 0.32 which are statistically
signicant. It is also observed that the correlation is high
in March between TMAX and BCSMC in all temperature
homogeneous regions than in other months. The correla-
tion is low in the regions of IP and EC. The average of
BCSMC for all temperature homogeneous regions con-
sidered as the All India average and the correlation in this
case is 0.52, 0.33 and 0.20 during March, April and May
month, respectively. For the whole summer season, the
correlation between BCSMC and TMAX are, respectively,
0.57, 0.28, 0.29, 0.66, and 0.33 for WH, NW, NC, NE and
WC regions with 0.01, 0.10, 0.10, 0.01 and 0.05 levels
of signicance. The correlations of EC and IP regions
are found statistically insignicant. The scatter plots
© 2017 Royal Meteorological Society Int. J. Climatol. (2017)
G. PURNADURGA et al.
Table 1. Correlation co-efcient of TMAX with BCSMC (𝜇gm
−3)
for all India and for all temperature homogeneous regions of
India* denotes 0.01 and ** denotes 0.05 levels of signicance.
Region/ March April May
Year 1980–2014 1980 –2014 1980– 2014
All India 0.52* 0.33** 0.20
WH 0.56* 0.48* 0.30
NW 0.32** 0.18 −0.03
NC 0.35** 0.27 0.06
NE 0.69* 0.51* 0.40**
WC 0.49* 0.36** 0.04
IP 0.21 0.06 −0.08
EC 0.18 0.18 0.09
for the regions with signicant correlations are given in
Figures 5(a)–(e).
The spatial scale variations of TMAX of the present
study are compared with the PET and OLR variations for
the same period to understand the proportionate relation
among them. Figures 6 and 7 depict the spatial variations
of PET and OLR. The differences in these parameters
between the two epochs for the summer season over India
are shown here. Similar features of increments have also
been observed in PET and OLR. The increment/decrement
of PET/OLR were subjected to student t- test (two tailed)
to know the level of signicance. The difference in PET
over NW, NC, NE, WC and EC are found at 0.01, 0.20,
0.01, 0.20 and 0.05 levels of signicance. The difference
in OLR during the two study periods is found at 0.01,
0.05, 0.20, 0.01, 0.01 levels of signicance over WH, NW,
NC, NE, EC regions, respectively. WH and NW parts of
India experienced 0.1–0.3 mm increase of monthly mean
PET, which gradually decreased during months of April
and May. Studies of Madhu et al. (2014) revealed that the
PET estimated from Hargreeves method over India has
shown increasing trend from 1901 to 2007 during sum-
mer time and is more prominent in NW region of India
which is shown as due to increase of summer TMAX over
that region. The OLR maps also revealed the same fea-
tures as PET. NW parts of India have shown an increase
of OLR ranging from 5 w m−2to 15 w m−2. NE parts also
showed variation of about 5 w m−2during the summer
months. The higher values of OLR over NW parts, due to
low cloud cover resulting in higher surface temperatures
(Chandrasekhara et al., 1993). It is also observed that the
southern parts of India witnessed decrease of OLR during
Epoch 2 (2001–2014) than during Epoch 1 (1971 – 2000).
The overall spatial analysis reveals that TMAX considerably
increased over NW and WH regions of India and is main-
tained in close association with PET and OLR.
4. Association of TMAX, PET and OLR over India
This section presents the correlation analysis of TMAX
with PET and OLR for different homogeneous tempera-
ture regions of India and also for India as a whole along
with the daily time series analysis of TMAX. Also, we
present in Figures 8(a) and (b) the coherence wavelet
spectrums (CWS) for TMAX with PET and OLR for the
period 1971–2014 for whole of India. The CWS is a very
useful technique to explore the common power among
the time series of two variables and the relative phase
in time-frequency domain (Grinsted et al., 2004). The
wavelet coherence reveals the local periodicities in a given
time series and the wavelet transform coherence can be
treated as the local correlation between two time series
in time-frequency zone. Grinsted et al. (2004) have devel-
oped this technique and applied to the time series of Arc-
tic Oscillation and the Ice extent of Baltic Sea to report
the relationship among them. Many researchers have used
this algorithm of CWS for understanding the response of
ENSO on interannual sea level variability in South China
(Rong et al., 2007), variations in drought and wet spells in
China (Su and Wang, 2007) and for studying precipitation
variability (Vargas et al., 2012).
In Figures 8(a) and (b), xaxis denotes the year and yaxis
denotes the period of the wave. The gures were drawn
with cone of inuence. The arrow marks pointing towards
right in the gure, indicate that the two parameters to be
in-phase and the arrows pointing left indicate them to be
in anti-phase. From Figures 8(a) and (b) are found that the
relation between TMAX and PET is stronger than the rela-
tion between TMAX and OLR. The common power is very
high among TMAX and PET than with OLR. The arrows
in Figure 8(a) point towards right which reveal that TMAX
and PET are in phase with each other whereas TMAX and
OLR maintained in phase when the common power is
high. When the low power happens to be between Max-
imum Temp and OLR, these parameters are in anti-phase.
It is known that increased radiative forcing increases the
surface heating and surface latent heating which leads to
increase in surface temperature and increase in evapora-
tion. This may be the reason for the strong relation of
TMAX with PET. Also, OLR is measured from the top
of the atmosphere, whereas PET is measured at surface.
This might be the reason for the anti-phase relation of
TMAX with OLR in some years. The correlation analy-
sis of TMAX with OLR and PET also revealed the strong
association among them (Table 2). The correlations are
very strong with 0.05–0.01 level of signicance over India
with values of 0.42 and 0.96 for OLR and PET, respec-
tively. Over homogeneous regions also strong correlations
of TMAX with PET and OLR are seen. Among all homoge-
neous regions, WC region has shown negative correlation
between TMAX and PET and the regions EC and IP have
shown very less and insignicant correlations with OLR.
The lowest correlation over WC region between TMAX
and PET might be due to topography and climatic con-
ditions. Figures 9(a)– (h) represent the mean daily TMAX
for the periods 1971–2000 and 2001 – 2014 during the
months of March, April and May for All India, WH, NW,
NC, NE, WC, IP and EC regions, respectively. It is clear
that the daily TMAX continuously increased from March to
May except for WC region whereas in this region, TMAX
decreased during the end of May. In general, it is also
observed that the daily TMAX for the have shown higher
© 2017 Royal Meteorological Society Int. J. Climatol. (2017)
TEMPERATURES OVER INDIA IN ASSOCIATION WITH METEOROLOGICAL PARAMETERS
(a) (b) (c)
(d) (e)
Figure 5. Scatter plots of BCSMC (𝜇gm
−3) and Maximum Temperature (∘C) for the period 1980– 2014 over (a) WH, (b) NW, (c) NC, (d) NE and
(e) WC regions of India. [Colour gure can be viewed at wileyonlinelibrary.com].
70°E
10°N
15°N
20°N
25°N
30°N
35°N
MAM
75°E 80°E 85°E
0 0.05 0.1 0.15 0.2 0.25 0.3
90°E 95°E 100°E
Figure 6. Difference in PET on space scale over India for the periods
2001– 2014 and 1971–2000 for summer season (March, April and May
average). [Colour gure can be viewed at wileyonlinelibrary.com].
values than during except for a few days in all the 3
months. The All India average TMAX during Epoch 2 is
34.7 ∘C and it is 34.1 ∘C during Epoch 1 (Table 3). Thus,
a warming of 0.6 ∘C is witnessed over India between the
two epochs. In the case of homogeneous regions, WH
has shown highest warming trend of about 1.4 ∘C during
70°E
10°N
15°N
20°N
25°N
30°N
35°N
MAM
75°E 80°E 85°E
0 5 10 15 20 25 30
90°E 95°E 100°E
Figure 7. Difference in OLR on space scale over India for the periods
2001– 2013 and 1975–2000 for summer season (March, April and May
average). [Colour gure can be viewed at wileyonlinelibrary.com].
Epoch 2 followed by NW which is of about 0.9 ∘C. The
lowest warming was observed in IP region of about 0.2 ∘C
with TMAX values of 38.9 ∘C degree and 38.7 ∘Cdegree
during Epoch 2 and Epoch 1, respectively. Remaining
regions such as NC, NE, WC and EC have shown warm-
ing of 0.4 ∘C, 0.5 ∘C, 0.4 ∘C and 0.3 ∘C, respectively. It also
seen that India as a whole experiences a warming of 0.8 ∘C,
© 2017 Royal Meteorological Society Int. J. Climatol. (2017)
G. PURNADURGA et al.
4
8
8
510
Period
15 20 25 30 35 40
51015
Year
1
0.9
0.8
0.7
0.6
0.8
1
0.4
0.2
0
0.6
0.5
0.4
0.3
0.2
0.1
0
20 25 30
4
(a)
(b)
Figure 8. Coherence Wavelet Spectrum for (a) mean annual TMAX
and PET during 1971– 2014 and (b) mean annual TMAX and OLR
during 1975– 2013 over India. [Colour gure can be viewed at
wileyonlinelibrary.com].
0.4 ∘C and 0.4 ∘C during months March, April and May,
respectively. WH part of India has shown highest incre-
ment of TMAX during March of about 2.2 ∘C followed by
NW which is of about 1.4 ∘C between the two epochs.
Overall, it is to be noted that the daily TMAX of March
in all homogeneous regions have shown higher warming
than in any other month of summer season. The analysis
on high temperature days in India by Jaswal et al. (2015)
has clearly shown that there is an increase of respec-
tively, 3% and 5% in the north and west regions during
1969–2013. They also have shown that contrasting fea-
tures of summer high temperature days during the periods
1991–2013 and 1969 – 1990. Also, it is reported that there
is a strong link between SST of eastern pacic region and
temperatures over India (Kothawale et al., 2010). We also
nd a considerable increase of 0.3 ∘C in Nino 3 SST during
November, December than during other months between
the periods 2001–2014 and 1971 – 2000. Similar features
of increase have been observed in PET and OLR during
the months of March, April and May and for the total
season. The PET over India increased by 3% during the
Epoch 2 when compared to that during Epoch 1. OLR has
shown 11 w m−2and 10 w m−2increments over WH and
NW parts and a decrement of 4 w m−2and 8 w m−2over IP
and EC regions, respectively during the summer season.
5. Changes in meridional circulation during the two
epochs
The MSF of the mean meridional circulation is obtained
with pressure as vertical coordinate. Figures 10(a) and (b)
show the MSF for the two distinct periods 1971– 2000 and
Table 2. Correlation co-efcient of TMAX with OLR and PET
over India and other homogeneous regions (* and ** denotes 0.01
and 0.05 levels of signicance, respectively).
Region 1975–2014 1971 –2014
OLR PET
All India 0.43* 0.96*
IP −0.12 0.92*
NC 0.72* 0.97*
NE 0.26* 0.83*
NW 0.83* 0.96*
WH 0.92* 0.94*
WC 0.94* 0.70*
EC −0.42** 0.92*
2001–2014. We can observe that the contour lines of zero
value are located around 35∘N and 60∘N, respectively in
the epochs, which are separating the three cells, Hadley,
Ferrel, and Polar cells with their centres at 20∘N, 50∘Nand
70∘N. Flow circulates around positive (negative) centres
in a clockwise (anti-clockwise) sense. Thus, in the sea-
sonal mean (March, April and May), air rises just north
of the equator and sinks around 35∘N. In the Epoch 2,
the descending branch of Hadley cell was strengthening
and it has some northward expansion over the subtropics.
There is a seasonal variation in the position of the cells.
The latitudinal positions where the values of MSF are
zero in the subtropical regions shift poleward in both the
epochs, resulting from the intensied MSF in the subtrop-
ics in the recent periods. Widening of Hadley cell also
can be conrmed from this intensication of Hadley cell.
A strong Hadley cell dominates with its rising branch in
the summer hemisphere in the both the epochs. Obser-
vational analysis have demonstrated that the Hadley cells
expanded poleward in recent decades (Hu and Fu, 2007),
and various modelling studies have investigated the Hadley
cell expansion under global warming (Lu et al., 2007).
A recent modelling study (Hu et al., 2013) suggests that
increasing greenhouse gases play an important role in
causing the observed poleward expansion of the Hadley
circulation. Te Ferrel cell has maximum negative values
over 47∘–55∘N and it shows a signal of enhancement
as well as pole ward expansion compared to the epoch
1971–2000. A weakening and poleward shrinking of the
polar cell is also observed in both the epochs. The anal-
ysis shows that the Ferrel cell strongly intensied and
that intensication stretched over the entire troposphere
at around 50∘N in the recent period. This is may be
due to anomalous warm advection from tropics to higher
latitudes. These changes may be signicantly related to
warming conditions of the atmosphere during summer
season. The strength and width of the Hadley Circulation
during the two epochs experienced substantial changes
associated primarily to greenhouse warming and contribut-
ing to temperatures rises in the recent periods. Climate
models predict a weakening of the atmospheric convective
overturning in response to surface warming driven by an
increase in greenhouse gases (Knutson and Manabe, 1995;
Tanaka et al., 2004; Held and Soden, 2006).
© 2017 Royal Meteorological Society Int. J. Climatol. (2017)
TEMPERATURES OVER INDIA IN ASSOCIATION WITH METEOROLOGICAL PARAMETERS
(a)
(c)
(e)
(g)
(b)
(d)
(f)
(h)
Figure 9. Mean daily time series of TMAX for the months March, April and May over (a) All India, (b) WH, (c) NW, (d) NC, (e) NE, (f) WC, (g) IP
and (h) EC regions of India. [Colour gure can be viewed at wileyonlinelibrary.com].
Table 3. Mean TMAX in ∘C over India and homogeneous regions for the periods 1971– 2000 and 2001–2014.
Region March April May MAM
1971–2000 2001 –2014 1971– 2000 2001 –2014 1971– 2000 2001 –2014 1971– 2000 2001 –2014
All India 31.232.034.935.336.336.734.134.7
WH 16.218.422.623.526.727.621.823.2
NW 29.931.336.737.540.240.535.636.4
NC 32.433.238.438.941.241.137.337.7
NE 28.129.029.329.430.431.029.329.8
WC 34.935.336.937.436.837.136.236.6
IP 36.436.639.339.440.440.738.738.9
EC 35.035.437.237.438.538.936.937.2
6. Conclusions
Many earlier studies have pointed out, the warming over
India is continuing and is evidenced by increase in TMAX.
The main aim of the present analysis is to quantify the
differential warming trend over India during the Epoch
2 (2001–2014) compared to Epoch 1 (1971 – 2000). The
difference of TMAX over India during these two epochs
were studied in the context of PET and OLR variations
and the study has been extended also to homogeneous
temperature regions of India. TMAX has been related to
BCSMC to understand the role of aerosols in warming over
India. The results of the study show that:
(i) Warming over India during the epoch 2001–2014 was
more in northern India than in southern India when
compared to the epoch of 1971–2014.
(ii) WH and NW regions are the warmest regions with
warming of 1.4 ∘C and 0.8 ∘C, respectively during the
summer season.
(iii) The correlations are signicant between BCSMC and
TMAX over the different temperature homogeneous
regions of India except for IP and EC regions. This
revealed the linear association of BCSMC with the
rise in maximum temperatures.
(iv) A strong association is found between TMAX and, PET
when compared to TMAX and OLR.
© 2017 Royal Meteorological Society Int. J. Climatol. (2017)
G. PURNADURGA et al.
(a)
EQ
1000
900
800
700
600
500
400
300
200
100
10°N 20°N 30°N 40°N 50°N 60°N 70°N 80°N 90°N
(b)
EQ
1000
900
800
700
600
500
400
300
200
100
10°N 20°N 30°N 40°N 50°N 60°N 70°N 80°N 90°N
Figure 10. (a) The mass stream function of the mean meridional circulation in pressure coordinates, for March, April and May in the period
1971– 2000. (b) The mass stream function of the mean meridional circulation in pressure coordinates, for March, April and May in the period
2001– 2014. [Colour gure can be viewed at wileyonlinelibrary.com].
(v) Most of the warming have been observed during
March with highest values in WH region (2.2 ∘C)
followed by NW region (1.4 ∘C).
(vi) It is found that the intensication and northward
broadening of Hadley cell and northward expansion
of the Ferrel cell during the Epoch 2 is more compared
to the Epoch 1.
Acknowledgements
The authors are thankful to Department of Science and
Technology (DST) Scientic and Engineering Research
Board (SERB), Govt of India for sponsoring this work
under EMR scheme.
References
Annual Climate Summary. 2014, National Climate Centre, India
Meteorological Department, p 24. http://imdpune.gov.in/Clim_Pred_
LRF_New/Reports.html.
Arora M, Goel NK, Singh P. 2005. Evaluation of temperature trends over
India. Hydrol. Sci. J. 50(1): 81–93.
Aruna K, Lakshmi Kumar TV, Narayana Rao D, Krishna Murthy BV,
Suresh Babu S, Krishna Moorthy K. 2013. Black carbon aerosols in
a tropical semi-urban coastal environment: effects of boundary layer
dynamics and long range transport. J. Atmos. Sol. Terrest. Phys. 104:
116– 125.
Ban-Weiss GA, Cao L, Bala G. 2012. Dependence of climate forcing
and response on the altitude of black carbon aerosols. Clim. Dyn. 38:
897– 911.
Burney J, Ramanathan V. 2014. Recent climate and air pollution impacts
on Indian agriculture. Proc. Natl. Acad. Sci. 111: 16319– 16324.
Chandrasekhara RK, Kyle LH, Mecherikunnel TA. 1993. Diurnal varia-
tion of the outgoing long wave radiation: as revealed by the E.R.B.E.
Tellus A 45: 1 –14.
© 2017 Royal Meteorological Society Int. J. Climatol. (2017)
TEMPERATURES OVER INDIA IN ASSOCIATION WITH METEOROLOGICAL PARAMETERS
Ganguly DN, Iyer KN. 2009. Long term variations of surface air temper-
ature during summer in India. Int. J. Climatol. 29: 735– 746.
Gaurav G, Nanjundiah SR, Satheesh SK, Krishna Moorthy K, Kota-
marthi VR. 2015. Performance of WRF Chem over Indian region:
comparison with measurements. J. Earth Syst. Sci. 124: 875– 896.
Grinsted A, Moore JC, Jevrejeva S. 2004. Application of the cross
wavelet transform and wavelet coherence to geophysical time series.
Nonlinear Processes Geophys. 11: 561– 566.
Hegerl GC, Zwiers FW, Braconnot P, Gillett NP, Luo Y, Marengo
Orsini JA, Nicholls N, Penner JE, Stott PA. 2007. Understanding and
attributing climate change climate change 2007: the physical science
basis. In Contribution of Working Group I to the Fourth Assessment
Report of the Intergovernmental Panel on Climate Change, Solomon
S, Qin D, Manning M, Chen Z, Marquis M, Averyt KB, Tignor
M, Miller HL (eds). Cambridge University Press: Cambridge, UK,
663– 745.
Held IM, Soden BJ. 2006. Robust responses of the hydrological cycle to
global warming. J. Clim. 19: 5686– 5699.
Hu Y, Fu Q. 2007. Observed poleward expansion of the Hadley circula-
tion since 1979. Atmos. Chem. Phys. 7: 5229–5236. https://doi.org/10
.5194/acp-7- 5229-2007.
Hu YY, Tao LJ, Liu JP. 2013. Poleward expansion of the Hadley
circulation in CMIP5 simulations. Adv. Atmos. Sci. 30: 790–795.
https://doi.org/10.1007/s00376-012- 2187-4.
IPCC. 2013. Climate Change 2013:The Physical Science Basis. Con-
tribution of Working Group I to the Fifth Assessment Report of the
Intergovernmental Panel on Climate Change, Stocker FT, Qin D,
Plattner G-K, Tignor M, Allen SK, Boschung J, Nauels A, Xia
Y, Bex V, Midgley PM (eds). Cambridge University Press: Cam-
bridge, UK and New York, NY, 1535 pp. https://doi.org/10.1017/
CBO9781107415324.
Jacobson MZ. 2010. Short-term space effects of controlling fossil-fuel
soot, biofuel soot and gases, and methane on climate, Arctic ice, and
air pollution health. J. Geophys. Res. 115: D14209. https://doi.org/10
.1029/2009JD013795.
Jaswal AK, Rao PCS, Singh V. 2015. Climatology and trends of summer
high temperature days in India during 1969– 2013. J. Earth Syst. Sci.
124(1): 1– 15.
Johanson CM, Fu Q. 2009. Hadley cell widening: model simulations
versus observations. J. Clim. 22: 2713–2725.
Kang SM, Lu J. 2012. Expansion of the Hadley cell under global
warming: winter versus summer. J. Clim. 34: 8367– 8393.
Kim MK, Lau WKM, Kim MK, Sang J, Kim YH, Lee SW. 2016. Ampli-
cation of ENSO effects on Indian summer monsoon by absorbing
aerosols. Clim. Dyn. 46: 2657– 2671.
Knutson TR, Manabe S. 1995. Time-mean response over the tropical
Pacic to increased 10 CO2in a coupled ocean atmosphere model.
J. Clim. 8: 2181– 2199.
Kothawale DR, Rupa Kumar K. 2005. On the recent changes in surface
temperature trends over India. Geophys. Res. Lett. 32: L18714. https://
doi.org/10.1029/2005GL023528.
Kothawale DR, Munot AA, Krishna Kumar K. 2010. Surface air tem-
perature variability over India during 1901– 2007, and its association
with ENSO. Clim Res. 42: 89– 104.
Kothawale DR, Deshpande NR, Rupakumar K. 2016. Long term temper-
ature trends at major, medium, small cities and hill stations in India
during the period 1901–2013. Am. J. Clim. Change 5: 383–398.
Lu J, Vecchi GA, Reichler T. 2007. Expansion of the Hadley cell under
global warming. Geophys. Res. Lett. 34: L06805. https://doi.org/10
.1029/2006GL028443.
Madhu S, Lakshmi Kumar TV, Koteswara Rao K, Barbosa H, Vizayab-
haskar V. 2014. Trend analysis of evapotranspiration and its response
to droughts over India. Theor. Appl. Climatol. 121(1): 41–51.
Meehl AG, Arblaster MJ, Collins WD. 2008. Effects of black carbon
aerosols on the Indian monsoon. J. Clim. 21: 2869– 2882.
Nair VS, Suresh Babu S, Krishnamoorthy K, Sharma AK, Marinoni
A, Ajai. 2013. Black carbon aerosols over the Himalayas: direct and
surface Albedo forcing. Tellus B 65: 19738.
Oort A, Yienger J. 1996. Observed interannual variability in the
Hadley circulation and its connection to ENSO. J. Clim. 9:
2751– 2767.
Pal I, Al-Tabbaa A. 2010. Long term changes and variability of
monthly extreme temperatures in India. Theor. Appl. Climatol. 100:
45– 56.
Pathak B, Kalita G, Bhuyan K, Bhuyan PK, Krishna Moorthy K. 2010.
Aerosol temporal characteristics and its impact on shortwave radiative
forcing at a location in the northeast of India. J. Geophys. Res. 115:
D19204. https://doi.org/10.1029/2009 JD013462.
Rajeevan M, Bhate J, Kale JD, Lal B. 2006. The high-resolution daily
gridded rainfall for the Indian region; analysis of break and active
monsoon spells. Curr. Sci. 91(3): 296 –306.
Raju BMK, Rao KV, Venkateswarlu B, Rao AVMS, Rama Rao CA,
VUM R, Bapuji Rao B, Ravi Kumar N, Dhakar R, Swapna N, Latha
P. 2013. Revisiting climate classication in India: a district level
analysis. Curr. Sci. 105: 402 –405.
Ramachandran S, Cherian R. 2008. Regional and seasonal variations
in aerosol optical characteristics and their frequency distribu-
tions over India during 2001 to 2005. J. Geophys. Res. 113:
D08207.
Ratnam JV, Behra SK, Ratna SB, Rajeevan M, Yamagata T. 2016.
Anatomy of Indian heat waves. Nat. Sci. Rep. 6: 24395. https://doi
.org/10.1038/srep24395.
Ravindranath M. 2011. Impact of climate change on the moisture regime
of tropical climate, Research Report, National Centre for Medium
Range Weather Forecasting (NCMRWF), India.
Rohini P, Rajeevan M, Srivastava AK. 2016. On the variability and
increasing trends of heat waves over India. Nat. Sci. Rep. 6: 26153.
https://doi.org/10.1038/srep26153.
Rong Z, Liu Y, Zong H, Cheng Y. 2007. Interannual sea level variability
in the South China Sea and its response to ENSO. Glob. Planet.
Change 55: 257– 272.
Scott AP, Jones SG. 2012. Observed 21st century temperatures fur-
ther constrain likely rates for future warming. Atmos. Sci. Lett. 13:
151– 156. https://doi.org/10.1002/asl.383.
Srivastava AK, Rajeevan, Kshirsagar SR. 2009. Development of a high
resolution gridded temperature data set (1969–2005) for the Indian
region. Atmos. Sci. Lett. 10(4): 249– 254.
Srivastava AK, Kothawale DR, Rajeevan M. 2017. Observed climate
variability and change over the Indian region. Springer Geology.
305– 326. https://doi.org/10.1007/978-981-10- 2531-0_2.
Su MF, Wang HJ. 2007. Relationship and its instability of ENSO -
Chinese variations in droughts and wet spells. Sci. China Ser. D Earth
Sci. 50(1): 145– 152.
Tanaka HL, Ishizaki N, Kitoh A. 2004. Trend and interannual variability
of Walker, monsoon and Hadley circulations dened by velocity
potential in the upper troposphere. Tellus A 56: 250–269.
Thornthwaite CW, Mather JR. 1955. The Water Balance. Publications in
Climatology: Centerton, NJ.
Thoronthwaite CW. 1948. An approach towards a rational classication
of climate. Geogr. Rev. 38(1): 55– 94.
Vargas R, Collins LS, Thomey ML, Johnson JE, Brown FR, Natvig
OD, Friggens TM. 2012. Precipitation variability and re inuence
the temporal dynamics of soil CO2efuxinanaridgrassland.
Glob. Change Biol. 18: 1401– 1411. https://doi.org/10.1111/j.1365-
2486.2011.02628.x.
© 2017 Royal Meteorological Society Int. J. Climatol. (2017)