Indian Summer Monsoon (ISM) is mainly characterized by seasonal wind reversal in low level jet stream and tropical easterly jet (TEJ) among several other elements of monsoon systems. TEJ is observed in general between 100 and 150 hPa during June–September over the Indian region and its strength is directly related to the monsoon rainfall. In the context of changing climate, large reduction in its extent and weakening of its strength were reported. Using high resolution measurements, we report here the observation of a sharp strengthening of the TEJ during the recent warmest decade (2001–2010), reaching its 1970s value. We also show that this change is reflected in the tropical cyclone systems and finally on the precipitation patterns over the Indian region as they are interlinked. We attribute this unusual change partly to the change in the circulation due to the tropospheric warming and lower stratospheric ozone recovery.
If you can't find a tool you're looking for, please click the link at the top of the page to "Go to old article view". Alternatively, view our Knowledge Base articles for additional help. Your feedback is important to us, so please let us know if you have comments or ideas for improvement.
 Indian Summer Monsoon (ISM) is the manifestation of large meridional thermal gradient between the warm Asian continent in the north and cooler Indian Ocean in the south. Among several elements of the monsoon system [Krishnamurti and Bhalme, 1976], the tropical easterly jet (TEJ) [Koteswaram, 1958] plays a pivotal role in deciding the strength of rainfall. As per World Meteorological Organization (WMO) criteria, TEJ is defined as an intense (above 30 m/s in the upper troposphere) narrow horizontal current of air associated with strong vertical wind shear. It is mesoscale in the cross flow direction and synoptic scale in the along flow direction and less than a kilometer thick vertically. Since TEJ is synoptic in nature, changes in its strength are considered as one of the indicators of long-term changes in atmospheric general circulation [Pielke et al., 2001]. The study of these features is important as it reveals the strength and year-to-year variability of ISM [Kobayashi, 1974].
 The ISM indices based on wind shear between low level jet (LLJ) and TEJ showed significant decrease during the past few decades [Stephenson et al., 2001]. It was reported that there was a significant reduction in the spatial extent of TEJ between 1960s and 1990s [Sathiyamoorthy, 2005] and that there was also a decreasing trend in the strength of TEJ [Bansod et al., 2012] and the number of tropical convective systems (TCS) over Bay of Bengal (BoB) [Srivastava et al., 2000; Rao et al., 2004; Rao et al., 2008]. It was also reported that the strength of TEJ and monsoon rainfall is directly related [Kobayashi, 1974; Hulme and Tosdevin, 1989; Pattanaik and Satyan, 2000]. Using an analysis of 140 year historical record, Krishna Kumar et al.,  suggested that the inverse relationship between the El Niño-Southern Oscillation (ENSO) and the ISM (weak monsoon associated with warm ENSO event) has broken down in the recent decades. They further suggested that increased surface temperatures over Eurasia in winter and spring, which are a part of the mid-latitude continental warming trend, may favor enhanced land-ocean thermal gradient leading to a stronger TEJ conducive to a strong monsoon. Thus, the variability of TEJ has critical importance for monsoon rainfall.
 In the recent past, several changes have been reported in the middle atmosphere that took place in the recent decade and were attributed mostly to the change in the circulation due to the tropospheric warming and the ozone recovery [Hu et al., 2011] leading to climate change. In this report, interesting features observed in TEJ characteristics over Indian region are presented. How such changes may be projected for the future under climate change scenario is also discussed.
 High resolution observations available over Gadanki (13.5°N, 79.2°E) since 1996 using mesosphere-stratosphere-troposphere (MST) radar [Rao et al., 1995] and radiosonde have proven to be a potential source for investigating the detailed characteristics of TEJ due to their high vertical resolutions [Narayana Rao et al., 2000; Raman et al., 2009; Ratnam et al., 2011]. Note that we have used most of the data from MST radar which works on the principle of Doppler shift, and stability will not come into picture for measuring winds. It can measure the horizontal winds with an accuracy of 0.5 m/s. The radiosonde data used in the present study are the WMO certified (Meisei, Japan) receiver, unlike routine India Meteorology Department (IMD) systems over India, which provides the data based on the principle of GPS poisoning which is very accurate. It also provides data with an accuracy of 0.5 m/s. As illustrated in Figure 1, Gadanki is strategically located in the core of TEJ, thus providing valuable information on the monsoon circulation. Since TEJ is of synoptic nature and for getting reliable estimates of spatial distribution, National Center for Environmental Prediction/National Center for Atmospheric Research (NCEP/NCAR) reanalysis data [Kalnay et al., 1996] available since 1948 with 2.5 × 2.5° grid, European Center for Medium range Weather Forecasting reanalysis ERA 40 available since 1957 with 2.5 × 2.5° grid, ERA-Interim reanalysis [Simmons et al., 2007] with 1.5 × 1.5° grid available since 1979 over the Indian region (0–30°N, 30–120°E) covering Indian Ocean, Arabian Sea, and Bay of Bengal (BoB) have also been used in the present study. It is to be noted that these reanalysis data sets underestimate the TEJ strength on an average by ~5 m/s when compared to the MST radar and radiosonde data over Gadanki. Due to these differences, the onset of TEJ will be delayed and lasts for about a week in the months of June and September, respectively, over a given location by employing WMO criteria [Raman et al., 2009]. Nevertheless, several studies have proven that these data can rightly be used for investigation of the long-term trends [Sathiyamoorthy, 2005; Bansod et al., 2012]. We also make use of gridded (1° × 1°) rainfall data obtained from India Meteorology Department (IMD) [Rajeevan and Bhate, 2009] available since 1901 and the number of tropical cyclone systems (TCS) that are formed over north Indian ocean.
3 General Characteristics of TEJ
 Long-term data sets available from Indian MST radar and high resolution data from radiosonde revealed quite different features in the TEJ characteristics during active and break phases. Higher wind reversal height, jet core altitude, and shears 1 km above and below the jet core during the active phase was noticed although they differ from year-to-year. During the break phase, the jet speed was slightly greater with greater jet width. Shear below (13–14 km) the jet core is greater in the active phase but shear above (17–18 km) the jet core is greater in the break phase [Raman et al., 2009]. Large sub-daily changes in the TEJ characteristics are also observed [Ratnam et al., 2011] in addition to day-to-day changes. These changes are mainly attributed to the motion of TEJ core due to change in the thermal contrast between land and ocean, large updrafts and downdrafts, and strong wave activity. Interestingly, TEJ peak altitude is seen above the cold point tropopause altitude for about 42% of the days suggesting that it is not a tropospheric feature alone [Ratnam et al., 2011], and any change related to stratosphere can influence the TEJ.
3.1 Long-Term Trends in TEJ Spatial Extent in the Last 60 Years
 The climatological mean (1951–1960) spatial distribution of zonal wind observed during the peak monsoon months (July and August) using NCEP/NCAR reanalysis data at 100 hPa shown in Figure 1a illustrates the spatial distribution of TEJ. Spatial extent of the core of TEJ has been derived from the climatology of the 100 hPa zonal wind crossing 30 m/s by following WMO criteria. During 1950s (1951–1960), the spatial extent of TEJ has extended from African to Atlantic regions. Long-term spatial distribution from NCEP-reanalysis data set showed large reduction of about 40% in the spatial extent by the end of the last decade (1991–2000) as also reported by Sathiyamoorthy . In the recent decade (2001–2010), the core of TEJ is restricted to the region between Arabian Sea and the Bay of Bengal (BoB) (Figure 1b). Earlier studies have shown that this effect will be seen on the prolonged drought conditions, particularly over the African region. Earlier with large database from reanalysis data sets, it was shown that the strength of the TEJ decreased considerably over the African region at 200 hPa but not over the Indian monsoon region [Sathiyamoorthy, 2005]. However, present results show that not only there is a reduction in the spatial extent but also in the strength of the TEJ. Figures 1 c and 1d show the difference in the intensity of TEJ between 1950s and 1990s and 1950s and 2000s, respectively. Large reduction of about 6 m/s in the strength is noticed in the TEJ core region and to the north and south, it is lower by about 4 m/s and 8 m/s, respectively, by the end of the last decade (1991–2000). The strength seems to decrease further by the end of the recent decade (2001–2010) (Figure 1d).
3.2 Long-Term Trends in TEJ Strength in the Last 60 Years
 Interannual variability of TEJ strength observed by various reanalysis data sets in the core region over the grid box 10°N–20°N, 60°E–100°E at 100 hPa during July and August shown in Figure 2 reveals a decreasing trend in the strength by ~1.4 ± 0.013 m/s per decade between 1948 and 2000 which is quite consistent with that reported recently [Bansod et al., 2012]. However, in the recent decade, the observations from MST radar and radiosonde (at Gadanki) reveal an interesting feature of increasing trend of 1.2 ± 0.012 m/s/yr in the TEJ strength. This magnitude is two times greater than the accuracy of wind measurements. Although radiosonde data from the IMD also show this feature at some stations in the monsoon region, they are not considered here as in many years radiosondes did not reach the altitude of 100 hPa, particularly in the recent decade. Interestingly, this feature is not noticed clearly in the reanalysis data though some signal of increasing trend is seen in the recent years particularly in the NECP/NCAR reanalysis data. This delayed response in the reanalysis data sets could be mainly due to the underestimation of mean winds as reported in Raman et al.  or partly due to selection of larger grid in these data sets.
3.3 Relation Between TEJ, Rainfall, and the TCS
 As the TEJ strength is directly related to the amount of rainfall and number of TCS, it will be interesting to examine this relation for the long-term trends, particularly in the recent decade. For this 1° × 1° gridded rainfall data obtained from IMD [Rajeevan and Bhate, 2009] available since 1901 is utilized. A correlation analysis has been performed between central India (16.5°N–26.5°N; 74.5°E–86.5°E) rainfall and TEJ strength separately for June, July, August, and September (JJAS) months and also together. The correlation between rainfall and TEJ strength during JJAS months between 1949 and 1978 is observed to be −0.54 which decreased to −0.25 between 1979 and 2008. However, when we separated the recent decade, the correlation increased to −0.51.
 Large amount of rainfall during the monsoon season is observed along the tracks of the westward moving depressions or low pressure systems over BoB [Rao et al., 2004]. During the last 110 year period from 1891–2010 as reported by IMD, the total number of TCS that are formed over north Indian ocean are essentially dominated by the number of cyclonic systems that form over the BoB out of which monsoon depressions are the prominent ones in the ISM season.
 The anomaly in the TCS calculated over a mean of 62 years, i.e., 1948–2010 that dominated over BoB shown in Figure 3 reveals negative anomaly up to 1980 followed by positive anomaly with a peak in the year 1998. Quite an opposite feature is observed in the zonal wind anomaly at 100 hPa calculated over 10°N–20°N and 40°E–100°E grid over a mean of 62 years, i.e., 1948–2010 for the months of July and August with positive anomaly prevailing up to 1980 followed by negative anomaly with peak observed in 2000 in NCEP data. ERA-Interim data also showed similar features. It is quite interesting to see the decrease in the trend of positive anomaly and increase in the trend of TEJ anomaly in the recent decade. The increase in the trend of TEJ anomaly is quite prominent in the data of the Indian MST radar and radiosonde. Since the reanalysis data set underestimates the TEJ strength, it is projected that in the coming decade the trend in both the TEJ and TCS anomalies may be reversed leading to the situation of 1970s even in the reanalysis data sets.
4 Summary and Discussion
 We showed strong evidence for the reversal of decreasing trend in the TEJ strength in the recent past decade of 2001–2010 from that prevailed till the year 2000 using high resolution MST radar and radiosonde observations. This feature is not reflected in the routine IMD radiosonde observations probably due to course resolution or non-availability of data up to 100 hPa. It is quite natural to expect that length of the data being considered from MST radar and radiosonde is small. However, note that this data is available from 1996 which is 5 years before the increasing trend started. Further, very good consistency can be noticed among different data sets during the overlapping periods. All these made us to report the interesting fact that TEJ intensity is no longer decreasing but started increasing in the recent decade. We attribute the strengthening of the TEJ in the recent decade to the changing climate. Several reports reveal that there is increase in the sea surface temperatures (SST) as a result of global warming in the Indian Ocean in the recent times. Since the TEJ is resultant of thermal gradient between Indian Ocean and Tibetan high, it is quite natural to expect change in the relation between the two.
 Naturally, an increase in the SST is expected to lead to more energy that would be available for tropical convection. This should therefore provide more energy to the TCS to develop. However, the formation and intensification of TCS do not depend only on the thermodynamic conditions such as ocean temperature or moisture. Rather, dynamic conditions such as shears in the horizontal winds like TEJ are also important. As mentioned earlier, dynamical changes in the background wind play an important role in determining the development of TCS. Thus, a close (negative) relationship between the TEJ and TCS anomaly is noticed (Figure 3). This anomaly is expected to revert back to its original values (that prevailed in the 1970s) in the coming decade owing to the strengthening of the TEJ shown in the present study. These changes can be partly attributed to increase in the tropospheric temperatures due to global warming. However, the role of ozone recovery in the lower stratosphere cannot be ruled out, though large debate exists whether circulation shifts are due to changing SST or ozone recovery.
 Large ozone recovery, due to Montreal protocols in the mid to high latitude that is being reported, is expected to change the circulation in the tropical latitudes [Hu et al., 2011] including ISM. Since the ozone layer is an important component in determining stratosphere and troposphere-surface energy balance, the recovery of stratospheric ozone may have significant impact on troposphere-surface climate [Hu et al., 2011]. In this context, it is noteworthy to recall that about 42% of the time TEJ is seen above the cold point tropopause [Ratnam et al., 2011]. Thus, any change in the ozone will reflect on the TEJ characteristics. At this stage, though it may look somewhat speculative, large evidences lead to this conclusion.
 For better understanding of these changes, the relation between SST, upper tropospheric circulation due to thermal contrast between the troposphere over Indian Ocean and Tibetan high leading to TEJ, its relation to TCS and finally their role on ISM rainfall are to be tested using the general circulation models to predict the future state of ISM circulation. Whether such changes are reflected in the troposphere meridional thermal gradient in the observations is to be tested and how these are observed in the model simulations need to be explored by introducing the changes in the SST and ozone recovery in the changing climate.
 We thank Y. Durga Santhi for helping us in the data analysis. This work is carried out under CAWSES Indian Phase II program theme 1 fully sponsored by Department of Space, Government of India. We thank various reanalysis centers for allowing us to access the data sets through their ftp sites.
 The Editor thanks two anonymous reviewers for their assistance in evaluating this paper.