Notice: Wiley Online Library will be unavailable on Saturday 27th February from 09:00-14:00 GMT / 04:00-09:00 EST / 17:00-22:00 SGT for essential maintenance. Apologies for the inconvenience.
 The observed evolution of a mini-cold pool (MCP) off the southern tip of India (STI) and its intrusion into the south central Bay of Bengal (BoB) during the summer monsoon season shows pronounced cooling episodes on intraseasonal time scale with differences in their number, intensity, duration and spatial extent. This variability that changes from year to year appears to be primarily determined by the corresponding variability in the upwelling driven by the divergence in the near-surface (Ekman + geostrophic) circulation and wind induced mixing. The signature of this cooling carried by the Summer Monsoon Current (SMC) is seen with reduction in intensity in the south central BoB limited mostly only to south of about 10°N. In the background of slow cooling caused by SMC, the cooling episodes of different amplitudes occur in the south central BoB suggesting that spatially variable wind forcing is responsible for producing these episodes simultaneously on intraseasonal time scale.
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.
 Several types of archived historic satellite and in situ measurements are utilized in this study. The TRMM TMI SST (1998–2004) data are exploited to characterize the nature of the observed intraseasonal cooling episodes south of the STI and in the south central BoB. The ship drift vectors (1854–1974), Ekman currents derived from QuikSCAT surface wind vectors (2000–2004) and the geostrophic currents derived from the T/P sea surface height anomalies (2000–2004) are utilized to characterize the SMC and the divergence in the near-surface circulation. The surface wind field is characterized with the QuikSCAT wind vectors.
3. Results and Discussion
3.1. MCP Off the Southern Tip of India
 The multi-year (1998–2004) averaged July–August SST climatology derived from TRMM TMI clearly shows the observed spatial variability around the southern peninsular India with a distinct MCP (<26°C) off the STI (Figure 1a). During the summer monsoon season these waters flow south of Sri Lanka and intrude into the south central BoB as a cold tongue. In addition, a suggestion of advection of cooler waters from the western AS into this region is also seen. Rao et al.  have suggested that the observed MCP primarily appears to be driven by the upwelling caused by the divergence (Ekman + geostrophic) in the near-surface circulation. The divergence of the near-surface circulation also shows maxima off the STI extending into the south central BoB around Sri Lanka resembling the concave shape of cool SSTs (Figure 1b). The near-surface circulation derived from ship drift vectors (Figure 1a) and geostrophic currents (Figure 1b) during the height of the summer monsoon clearly shows the signature of the SMC. The SMC also shows an anti-clockwise curvature around Sri Lanka and enters as a northeasterly current into the south central BoB suggesting its potential role in carrying these cooler waters.
3.2. Intraseasonal Variability of the MCP
 The fine resolution TMI SST data provide an unprecedented opportunity to examine the nature of the ISV of this MCP. The multi-year (1998–2004) averaged daily march of SST between 2°N and 8°N along 78°E (off the STI) clearly reveals the nature of the observed ISV during the summer monsoon season (Figure 2). At the northern end of the 78°E section (Figure 1a) the intermittent occurrence of the cooler waters (<26°C) is distinctly seen. About 5 to 6 cooling episodes typically of 1°–2°C amplitude occur from end May to October with a southward spread. This is referred as intraseasonal variability in the observed cooling. The southward spread increases with season reaching its peak by August and then recedes.
3.3. Year-to-Year Differences of the Intraseasonal Variability of the MCP
 The observed evolution of intraseasonal cold SST episodes along 78°E during the seven years 1998–2004 clearly reveals the distinct differences in their number, intensity, duration and meridional extent (Figure 3). A suggestion of southward spread is seen during all the cold SST episodes. The ISV is more (less) pronounced during 1999 and 2002 (1998 and 2003). During the years of weak ISV the SST remained relatively warmer in particular toward the equator during May–July. The probable reasons for the observed differences will be examined in the subsequent section.
3.4. Life Cycle of Cooling Episodes on Intraseasonal Time Scale
 A careful examination of the temporal evolution of SST fields revealed that the observed ISV of SST comprised of cooling episodes of variable intensity during every summer monsoon season. These episodes can be stratified into strong- (<25°C), moderate- (25°C–26°C) and weak-cooling (>26°C) categories. The daily evolution of SST, and the surface wind field during June to September for 2000–2004 is examined. The evolution of SST (surface wind field) representing typical weak- and strong-cooling episodes observed during the summer monsoon season of 2001 is shown in Figures 4a and 4c (Figures 4b and 4d). Clearly distinct differences are seen between the episodes of both the categories. During the strong-cooling episode relatively stronger winds occurred leading to enhanced divergence in the near-surface layers, upwelling and wind induced mixing. During the weak-cooling episode the surface wind field is relatively weaker. The composites of surface wind speed, divergence in the near-surface circulation during the preceding days of the coldest SST event and of the coldest SST for the weak-cooling and strong-cooling (Figure 5) episodes during June–September of 2000–2004 unambiguously show large differences in the parameter amplitudes.
3.5. Mechanisms of Cooling Episodes on Intraseasonal Time Scale
 The observed co-evolution of 5-day boxcar smoothed surface wind speed, the estimated divergence in the near-surface circulation and SST off the STI during June–September, 2002 is shown in Figure 6. The surface wind field showed pronounced ISV throughout the season. The peaks of the cooling episodes typically lag the corresponding peaks of the strong-wind and divergence in the near-surface circulation episodes by about 3 days. But during the summer monsoon season of 2003 the strong-wind episodes are fewer resulting in relatively weaker divergence in the near-surface circulation and weak- cooling episodes (figure not shown). These episodes vary from year to year. The spectral estimates of surface wind speed, divergence in the near-surface circulation and SST show dominant periods in the range of 18–25 days during the summer monsoon seasons of 2000–2004 (figure not shown). This is in agreement with a recent study of Han et al.  who have shown that the atmospheric intraseasonal oscillations can cause significant changes in SST on time scale of 10–30 days in the equatorial Indian Ocean.
3.6. Importance of Open-Ocean Upwelling
 A careful examination of the daily SST data during June–September reveals that most of the cooling episodes show their maxima away from the coast line indicating that the cooling is primarily due to upwelling of subsurface waters in response to the divergence of the near-surface circulation. The core of intense cooling located away from the coast line with a well defined identity is clearly seen (Figure 7). However, the cooling caused by coastal upwelling in certain situations cannot be ruled out. But unfortunately the satellite SST and wind data presently available from ∼50 km away from the coast line cannot resolve this issue. From the satellite winds it is equally hard to infer (extrapolate) the surface wind field within the ∼50 km from the coast line given the known non-linear variability of winds influenced by the complex coastline geometry and the land topography.
3.7. Influence of the MCP on the SST of the South Central BoB
 It would be interesting to examine the role of SMC on the intrusion of the observed intraseasonal signature of the MCP into the south central BoB. The ISV of SST during May–October, 2002 along 78°E representing the core of the MCP and along 83°E and 88°E (transects shown in Figure 1a) is shown in Figures 8a–8c. As described earlier the ISV signature is most pronounced along 78°E with its amplitude increasing with latitude. The signature of ISV in SST is also seen in the south central BoB with its amplitude decreasing with longitude. The observed slope of background cooling along 83°E during June–September, 2002 clearly highlights the important role of SMC. However, the interesting feature to be noted here is the simultaneous occurrence of the cooling episodes along all these three transects. This feature is also seen during the summer monsoon season of 2003 when the cooling episodes are less pronounced compared to those of 2002 (Figures 8d–8f). This implies that spatially variable wind forcing as seen in Figures 4b and 4d is responsible for producing the cooling episodes simultaneously in the south central BoB on the observed intraseasonal time scale. These cooling episodes superimpose on the background cooling caused by SMC in the south central BoB.
 During the summer monsoon season the observed MCP has shown pronounced ISV off the STI with its intrusion into the south central BoB. The ISV of the MCP is attributed to the corresponding ISV in the upwelling driven by the observed surface wind field and the associated divergence in the near-surface circulation. As TMI SST and QuikSCAT wind data within ∼50 km from the coast line are not available, one cannot resolve the contribution from the coastal upwelling to the observed cooling. A field experiment conducted with appropriate high resolution measurements from the coast line may provide some useful answer to this issue. A high-resolution 3-D modelling study should also be able to resolve this important issue. Pronounced cooling episodes on intraseasonal time scales are observed with differences in their number, intensity, duration and spatial extent in every monsoon season. About 5 to 6 cooling episodes typically of 1°–2°C amplitude occur from end May to October with southward spread. The evolution of MCP has shown strong-, moderate- and weak-cooling episodes. During the strong-cooling episodes relatively stronger winds occur leading to enhanced divergence in the near-surface layers, upwelling and wind induced mixing. During the weak-cooling episodes the surface wind field is relatively weaker. The observed composites of surface wind speed, divergence in the near-surface circulation and SST during strong-cooling and weak-cooling episodes unambiguously show large differences in their amplitudes. The surface wind field shows significant ISV throughout the summer monsoon season. The strong-cooling episodes show a 3-day lag with peaks in the surface winds and divergence in the near-surface circulation. But during some summer monsoon seasons (e.g., 2003) the strong-wind episodes are fewer in number resulting relatively weaker divergence in the near-surface circulation and weak-cooling episodes. The spectral estimates of surface wind speed, divergence in the near-surface circulation and SST showed dominant periods in the range of 18–25 days. The signature of this cooling carried by the SMC is seen with reduction in intensity in the south central BoB limited mostly only to south of about 10°N. In the background of slow cooling caused by SMC in the south central BoB, the cooling episodes of different amplitudes occur along all the three transects suggesting that spatially variable wind forcing is responsible for producing the cooling episodes simultaneously on the observed intraseasonal time scale. The relationship between these intraseasonal cooling episodes and the active-break cycle of the summer monsoon will be the subject of the next study.
 Highest appreciation is placed on record for the excellent compilation by several persons and organizations of all the data sets utilized in this study. Graphics are generated using Ferret. The encouragement and the facilities provided by the Directors of NPOL, INCOIS, and IITM are gratefully appreciated. The comments of two anonymous reviewers are very helpful. This research is supported through INDOMOD-SATCORE Project of Dept. of Ocean Development, Govt. of India.