Recent research has reported that the tropical easterly jet stream (TEJ) of the boreal summer monsoon season is weakening. The analysis herein using 60 yr (1950–2009) of data reveals that this weakening of the TEJ is due to the decreasing trend in the upper tropospheric meridional temperature gradient over the area covered by the TEJ. During this period, the upper troposphere over the equatorial Indian Ocean has warmed due to enhanced deep moist convection associated with the rapid warming of the equatorial Indian Ocean. At the same time, a cooling of the upper troposphere has taken place over the Northern Hemisphere subtropics including the Tibetan anticyclone. The simultaneous cooling of the subtropics and the equatorial heating has caused a decrease in the upper tropospheric meridional thermal gradient. The consequent reduction in the strength of the easterly thermal wind has resulted in the weakening of the TEJ.
The tropical easterly jet stream (TEJ), whose existence was established by Koteswaram (1958), the low-level jet stream (LLJ) discovered by Joseph and Raman (1966) and Findlater (1969), and the Tibetan anticyclone north of the TEJ are the important components of the Asian summer monsoon system and parts of the monsoon Hadley cell (Krishnamurti and Bhalme 1976). The TEJ has the strongest winds over the Indian Ocean, the intensity of which is found to be well correlated with the Indian summer monsoon rainfall (Pattanaik and Satyan 2000). Koteswaram (1958) showed that the TEJ is maintained by the thermal contrast between the subtropics and the equatorial Indian Ocean in the upper troposphere. Recently researchers have found a progressive decline in the intensity (strength of winds in the core of the TEJ) and areal extent of the TEJ. Sathiyamoorthy (2005) showed that the horizontal extent of the TEJ had decreased over the Atlantic Ocean and West African regions between the 1960s and the 1990s. Rao et al. (2004) reported that the core of the TEJ has been shrinking over the South Asian region. It has been hypothesized that the declining strength of the TEJ would produce conditions favorable for the formation of tropical cyclones over the Indian Ocean during the monsoon season by reducing the vertical wind shear there (Rao et al. 2008). Since the TEJ is energetically maintained by the release of the available potential energy in the Hadley and Walker circulations (Chen 1980), the interannual variation of the TEJ may be related to the interannual variability of the Walker and local Hadley circulations (Chen and van Loon 1987).
Krishnan et al. (2013) analyzed climate datasets showing that, in response to the global warming, the intensity of the boreal summer monsoon overturning circulation (monsoon Hadley cell) and the associated southwesterly monsoon flow (LLJ) have significantly weakened during the most recent 50 years. Joseph and Simon (2005a,b) have also shown the weakening trend of LLJ since 1950 by analyzing National Centers for Environmental Prediction (NCEP)–National Center for Atmospheric Research (NCAR) reanalysis data (Kalnay et al. 1996). Based on the simulations from an ultra-high-resolution global general circulation model with a spatial grid size of about 20 km and 60 vertical levels, Rajendran and Kitoh (2008) and Rajendran et al. (2012) showed that a stabilization (weakening) of the summer monsoon Hadley-type circulation in response to global warming has resulted in a weakened large-scale monsoon flow (LLJ). A recent study has suggested that the increase in convection due to the excessive warming of the sea surface temperature (SST) of the Indian Ocean is one of the reasons for the weakening of the TEJ (Joseph and Sabin 2008). Are there other factors in the weakening of the TEJ? The main objective of this work is to understand these. Section 2 gives details of the data used and section 3 presents the results of this study. A summary of the conclusions is given in section 4.
2. Data used
Upper tropospheric monthly mean wind and temperature data from 1950 to 2009 used in this study were obtained from NCEP–NCAR reanalysis (Kalnay et al. 1996). This reanalysis dataset was used in preference to other such datasets in view of the long period of six decades covered by it. These data have a spatial resolution of 2.5° latitude × 2.5° longitude. The gridded variables are strongly influenced by the available observations and are therefore the most reliable products of the reanalysis. Monthly mean SST at 1° × 1° grid resolution was obtained from the extended reconstructed SST (ERSST) dataset, version 3b (Smith et al. 2008). The equatorial Indian Ocean has high data density (ship observations) and hence this region offers better SST datasets compared to other oceanic regions (Ihara et al. 2008). To study areas of deep tropical convection, outgoing longwave radiation (OLR) datasets (Liebmann and Smith 1996) obtained from the National Oceanic and Atmospheric Administration (NOAA) at 2.5° latitude × 2.5° longitude grid resolutions are used.
Figure 1 shows the climatological mean of wind (at 150 hPa, the maximum wind level of the TEJ) and temperature (200 hPa) for the period 1950–2009. The area covered by the TEJ (10°S–20°N, 60°W–180°E) is marked by the letter A. At the 150-hPa level, the TEJ has maximum wind speeds (high-speed center) over the North Indian Ocean, and wind speed decreases to the east and west (Fig. 1a). Toward the north of the TEJ is the Tibetan anticyclone (B) and the subtropical westerly jet stream (C). In Fig. 1b, the mean temperature at 200 hPa is shown and it exhibits a warmer upper troposphere over the subtropics with maximum temperature over the Tibetan anticyclone region (D) and cooler temperature over the equatorial region, particularly the Indian Ocean (E). The TEJ is maintained by the upper tropospheric thermal gradient that exists between the equatorial region and the subtropics, particularly that between the tropical Indian Ocean and the Tibetan anticyclone region. It is seen from Fig. 1c that the upper tropospheric temperature over the equatorial Indian Ocean and African continent (F) has warmed during the period 1950–2009, while in the subtropical areas marked G and H cooling has occurred at 200 hPa during the same period. Statistical analysis shows that these warming and cooling trends are significant at 95% level of confidence determined using a two-tailed Student’s t test (Fig. 1d). Because of the low Coriolis force over the TEJ region (A), a small change in the temperature gradient and the consequent change in thermal wind can cause large variation in the intensity of the upper tropospheric zonal winds (Pant and Rupa Kumar 1997).
Figure 2 depicts the mean zonal wind at 150 hPa during the monsoon season [June–September (JJAS)] for each decade of the period 1950–2009. During the three decades from 1950–59 to 1970–79, the high-speed center of the TEJ, which lies over the Indian Ocean (Arabian Sea), has progressively weakened. From the decade 1970–79 to the decade 1980–89 the winds of the high-speed center (the innermost isotach) showed very little change in strength or area covered. Again, during the three decades 1980–89 to 2000–09 the high-speed center of the TEJ progressively weakened. During the most recent decade (2000–09) the 30 m s−1 wind contour ceased to exist as part of the weakening trend. Joseph and Sabin (2008) compared the changes in the TEJ over the period 1960–99, in both NCEP–NCAR and 40-yr European Centre for Medium-Range Weather Forecasts (ECMWF) Re-Analysis (ERA-40) data. They found that TEJ has weakened in both the datasets. However, the weakening in ERA-40 data is slightly less. As pointed out by Sathiyamoorthy (2005), we also find that the TEJ has weakened considerably over the eastern Atlantic Ocean and the adjoining African continent, where the weakening is particularly large during the three decades from 1950 to 1979. Our analysis shows that this weakening is associated with the intense cooling of the subtropical upper troposphere there (see Fig. 1c). The core of the TEJ exists over the Indian Ocean region because of the large thermal gradient between the equatorial Indian Ocean and the Tibetan anticyclone region. The release of latent and sensible heat over the Tibetan Plateau during the summer monsoon season keeps the temperature of the upper troposphere there very high and this is considered to be the main cause for the existence of the core (high-speed center) of the TEJ over the Indian Ocean.
We analyzed the decadal change in the upper tropospheric (200 hPa) temperature difference between the equatorial Indian Ocean and the subtropical area around the Tibetan anticyclone (Fig. 3). For this analysis, two boxes were taken, one over the equatorial Indian Ocean (10°S–10°N, 30°–120°E) and the other in the subtropics (25°–35°N, 30°–120°E), which is representative of the Tibetan anticyclone region. Figure 3 shows that the upper tropospheric temperature over the equatorial Indian Ocean had a weak warming trend of nearly 0.5°C during the six decades from 1950 to 2009. The well-known climate shift of 1976–78 (Miller et al. 1994) is not prominently seen. However, the temperature at 200 hPa over the Tibetan Anticyclone indicates a large cooling trend of about 1°C during the period 1950–76, a sudden shift around 1976–78, and a large cooling trend again of 1°C from 1977 to 2009. The net cooling across the six decades from 1950 to 2009 is about 1°C. The reason for this cooling trend in the subtropical upper troposphere is yet to be understood. To highlight the change that has taken place across the 1976–78 climate shift, the trend line graph is separated into two panels in Figs. 3b and 3c.
The vertical shear of the meridional wind averaged over the area bounded by latitudes 10°S–10°N and longitudes 30°–120°E between the two levels 200 and 850 hPa (Hadley circulation) is shown in Fig. 3c. The effect of the subtropical cooling and the equatorial warming on the intensity of the Hadley circulation is evident from the trend analysis. During the past 60 years, the summer Hadley circulation has shown a weakening trend. It may be noted that, similar to the subtropical cooling, an abrupt change has occurred in the strength of the Hadley circulation across the 1976–78 climate shift.
Latent heat release in the deep moist convection is an important factor in the maintenance of the upper tropospheric temperature. Over the equatorial oceanic regions, SST is one of the major factors in the generation of atmospheric convection. Figure 4 illustrates the SST change (2000–09 minus 1950–59) of the JJAS season. From the figure, it is evident that intense warming in SST has occurred over the equatorial Indian Ocean. During this period, the Indian Ocean has warmed from 0.8° to 1.2°C. This part of the Indian Ocean hosts a major convection center of the global atmosphere. Because of the prevailing high SST there, a small change in SST can have a large response in atmospheric convection (Wang et al. 2004). The SST increase and the corresponding increase in deep convection (2000–09 minus 1980–89) are shown in Figs. 4b and 4c, respectively. These difference plots confirm that convection had an increasing trend particularly over the equatorial Indian Ocean region. The shaded areas in Figs. 4d–f denote the areas with trends statistically significant at 95% confidence level. Over the equatorial Indian Ocean, the areas of intense convection are close to regions having high SST. The increased convection over the equatorial Indian Ocean has resulted in the excessive heating of the upper troposphere.
The horizontal temperature gradient that exists between the tropical and subtropical upper troposphere is directly related to the strength of the TEJ through the thermal wind relation. The decrease in the upper tropospheric temperature over the subtropics and the simultaneous increase in the upper tropospheric temperature over the Indian Ocean region reduced the meridional temperature gradient between these regions. The consequent weakening of the TEJ is evident in Fig. 2. As described earlier, the 200-hPa temperature of the equatorial Indian Ocean (E) of Fig. 1b has increased 0.5°C during the period 1950–2009. During the same period the temperature around the Tibetan anticyclone (D) at the same 200-hPa level has decreased by 1°C. Thus the meridional temperature gradient around the core region of TEJ (i.e., between E and D) has decreased about 1.5°C. In height coordinates (z), the thermal wind relation for the zonal geostrophic wind υg can be written as
where f is the Coriolis parameter, T the temperature in kelvin, and g the acceleration due to gravity. For the core region of the TEJ around the latitude 10°N, taking f as 2 × 10−5 and the mean T for the layer 300–150 hPa as 230 K, the mean thermal wind is about 1 m s−1 km−1 in the vertical. Thus, during the period 1950–2009, the thermal wind of the layer 300–150 hPa, about 5 km in thickness, has decreased by 5 m s−1. Since, the easterly thermal wind begins at around 500 hPa, the decrease in the wind at 150 hPa of the core region of TEJ as shown in Fig. 2 is mostly accounted for by the weakening of the thermal wind.
Joseph and Simon (2005a) reported a prominent weakening trend of the monsoon LLJ during the 5–6 decades after 1950. It is hypothesized that this weakening is related to the weakening of the monsoon Hadley circulation reported in the present study. Krishnamurti and Bhalme (1976) have shown that the upper limb of the monsoon Hadley circulation is the TEJ and the lower limb the LLJ, and these are two of the six semipermanent components of the boreal summer monsoon system, the others being the heat low, Mascarene high, monsoon trough, and Tibetan anticyclone. According to them, a change in any of these six components changes the intensity of the Hadley circulation. A schematic of the changes in the upper troposphere leading to the weakening of the TEJ, the monsoon Hadley cell, and the LLJ is given in Fig. 5.
The present analysis using 60 years of data shows that upper tropospheric temperature over the Tibetan anticyclone region has cooled while that over the equatorial Indian Ocean has warmed during the recent decade. Further analysis shows that these warming and cooling trends are statistically significant at the 95% level of confidence. The warming of the upper troposphere over the equatorial Indian Ocean is associated with the enhanced deep convection there caused by the rapid warming of the Indian Ocean SST. These warming and cooling trends have resulted in the decreasing trend of the upper tropospheric meridional temperature gradient. These changes have caused a reduction in the strength of the easterly thermal wind at the core region of the TEJ, resulting in the weakening of the TEJ. The effect of the 1976–78 “climate shift” is clearly seen in the cooling trend of the subtropics and in the consequent strength of the TEJ and the monsoon Hadley cell. The weakening of the Hadley cell is also responsible for the weakening of the monsoon LLJ.
This work was done as part of the project under the acronym Indo-Mareclim, which is funded by the European Union under the project EU-FP7 “Indo-European research facilities for studies on marine ecosystem and climate in India.” The Research Council of Norway through the Project India-Clim led by Ola M. Johannessen also funds this work.