1 Introduction

The BOB (2°N–22°N, 80°E–100°E) is a semi-enclosed tropical basin affected by tropical cyclones mainly during pre-monsoon (April–May) and post-monsoon (October–December) seasons. Besides the tropical cyclones, many depressions form during monsoon (June–September). It is well known that the cyclogenesis points are observed in the northern part during monsoon (June–September) and in the southern part during pre-monsoon and post-monsoon seasons (Fig. 1a). The northern part of the BOB (12°N–22°N, 80°E–100°E) shows the 72% cyclogenesis points (Table 1; Fig. 1a) from IMD datasets derived from Cyclone e-Atlas. The monthly variations of the occurrence of cyclogenesis show enhancement from April to August and then gradually decrease from September to March (Table 1) in the northern BOB (Fig. 1a). The southern BOB (2°N–12°N, 80°E–100°E) shows the semiannual pattern of cyclogenesis with one highest frequency during post-monsoon and another highest frequency during pre-monsoon. Generally, cyclones in the BOB basin move northwestward and strike Andhra Pradesh and Orissa coasts. However, some of them re-curved and hit Bangladesh and Myanmar coasts (Kotal et al. 2008; Webster 2008; Fritz et al. 2009). Because of its different nature in different seasons, the prediction of the cyclone movement over this basin is very difficult. Although some details about the cyclones and storm surge modeling are provided by Dube et al. (2009), but more understanding is still required for better prediction of the movement of the cyclone over this basin.

Fig. 1
figure 1

Seasonal variability of a the number of occurrence of IMD cyclogenesis in the southern BOB (2°N–12°N, 80°E–100°E) and in the northern BOB (12°N–22°N, 80°E–100°E) during 1891–2007, and b CHP in the southern BOB and in the northern BOB from SODA v2.0.4

Full size image
Table 1 List of monthly cyclogenesis points for the northern and southern boxes of the BOB based on 117 years (1891–2007) IMD data
Full size table

The tropical cyclones are generally generated in the regions where sea surface temperature (SST) is greater than 26°C (Gray 1968; Shay et al. 2000; Sadhuram et al. 2004; Lin et al. 2005, 2009). Other conditions required for their genesis are; (a) conditional instability through a deep atmospheric layer, (b) more relative humidity in lower and middle troposphere, (c) large low-level relative vorticity, (d) less vertical shear of horizontal winds (<10 m/s between the surface and upper troposphere), (e) deep thermocline (at least 20 m more than the normal value), and (g) location should be at least a few degrees away from the equator (>±2° N/S) etc. (Henderson-Sellers et al. 1998). It is to be mentioned here that only one important parameter is not responsible for the genesis of cyclone in the BOB. Oceanic eddies are also related to the cyclone formation (Lin et al. 2009). Eddies are generally warm core or cold core depending on their association with downwelling and upwelling regions, and as a result, the heat and SST distribution are different for different eddies. Although the cyclonic and anticyclonic eddies have SST > 26°C in the BOB, but all eddies do not generally support to develop the cyclone. Shay et al. (2000) and Goni and Trinanes (2003) suggested that warm core eddies or loop current in the Gulf of Mexico were important for decreasing the impact of negative feedback on the intensification of hurricanes. In general, strong wind of cyclones stirs and cools the underlying ocean water, which leads to the reduction of heat fluxes from ocean to the atmosphere (Ginis 1995; Wada and Usui 2007). But it cannot cool the underlying water when a deep warm layer associated with positive SSH anomaly. This helps tropical cyclones to maintain their intensity or intensify rapidly where SSH anomaly is relatively high, which is observed for hurricanes in the North Atlantic (Bao et al. 2000; Hong et al. 2000; Shay et al. 2000) and typhoons in the western North Pacific (Holliday and Thompson 1979; Lin et al. 2005). This process is one of the important factors that control the intensity and intensifications of tropical cyclones (Emanuel et al. 2004; DeMaria et al. 2005). Many researchers argued that when the tropical cyclone encounters or leaves an eddy then its intensity changes (Shay et al. 2000; Goni and Trinanes 2003; Lin et al. 2005, 2009). Ali et al. (2007) discussed the effect of eddies on cyclone intensity using satellite-derived SSH. It is also now realized by the research community that the integrated value of temperature from surface to subsurface layer is important rather than SST as the surface eddies are linked with the subsurface layer (Whitaker 1967; Wada 2009). Bay of Bengal is influenced by precipitation and large amount of freshwater river discharge which causes stratification and shallow mixed layer depth (MLD), which absorb huge insolation, which may cause rapid intensification of cyclones (Senguptha et al. 2007).

The CHP, which is the heat content in the warm upper layer i.e., from surface to D 26, can be used to explain the formation of eddies and cyclogenesis points (Sarma et al. 1990) and it provides the available heat energy at a particular point in the ocean as well as cyclone potential zones in the BOB (Sadhuram et al. 2004; Sarma et al. 1990). The seasonal variations of CHPs are similar in nature for both the northern and the southern regions of the BOB (Fig. 1b). The pre-monsoon and post-monsoon peaks of cyclogenesis in the southern BOB are closely matched with the peaks of CHP. Although the seasonal variation of CHP is not exactly matching with the IMD cyclogenesis points variation for the northern BOB (Fig. 1a), but it is very interesting to note their similar characteristics. Sadhuram et al. (2004) analyzed the role of CHP in the genesis and intensification of storms and Sarma et al. (1990) analyzed the distribution of CHP in the BOB. But the detailed analysis of CHP and eddies and tracks are not analyzed yet over this basin. In order to understand the relationship of CHP with cyclogenesis points and eddies, the warm upper layer characteristic has been analyzed in this paper. The CHP and the cyclonic tracks at different seasons also analyzed in this paper.

The season-wise monthly variation of the CHP is analyzed to understand the location of cyclogenesis at the different parts of the BOB. SSH and D 26 are also investigated in the similar way to ensure the cyclogenesis points seen from CHP analysis. The eddy potential zones are depicted by 5 m depth current, CHP is correlated with 35 m depth current, and 70 m depth current is correlated with D 26. This paper is organized as follows. Section 2 summarizes the data and method used. Section 3 describes the results and discussions. Conclusions are included in Sect. 4.

2 Data and methodology

The SODA product of Carton et al. (2005) from January 2002 to December 2007 has been used for the analysis purpose. This is the ocean data assimilation with model product based on Parallel Ocean Program physics with an average 0.50° × 0.50° horizontal and 40 vertical resolution output. This recent SODA product incorporates all available in situ data in their analysis. From the year 2003, many oceanic observations available over the BOB region and as a result SODA product over this region also improved (Kumar et al. 2010). In order to study the seasonal variability, we have considered four seasons; winter (December–February), spring (March–May), summer (June–August), and fall (September–November) in our analysis and for further discussion in this paper.

D 26 (m) is calculated by interpolating the temperature vertically as

$$ D_{26} = Z_{k - 1} + (26 - T_{k - 1} )\frac{{(Z_{k} - Z_{k - 1} )}}{({T_{k} - T_{k - 1})}} $$
(1)

where T k is the temperature (degree centigrade) in the standard depth Z k of level k.

CHP (J/m2) from surface to 26°C isotherm depth (D 26) is calculated as,

$$ {\text{CHP}} = \int\limits_{0}^{{D_{26} }} {\rho C_{p} T{\text{d}}z} $$
(2)

where dz is the depth increment and T is the temperature difference between mean temperature of two consecutive levels and 26°C (Sarma et al. 1990; Sadhuram et al. 2004 ). In this work, density and specific heat of seawater at constant pressure (C p ) are derived using UNESCO formulation (Fofonoff and Millard 1983) with the help of MATLAB tool, SEAWATER.

3 Results and discussion

3.1 Seasonal variation of CHP and eddy

3.1.1 Winter season (December–February)

During winter, the currents are weak for every layer (Figs. 2a–c, 3a–c, 4a–c). Small variations of SSH are observed from the central BOB toward north and northwestward. Low SSH is observed in the southwestward and near the Sri Lankan coast. Although the Sri Lankan coast is the region of strong upwelling zone during winter season (Paul et al. 2009), CHP (>60 × 107 J/m2) is favorable for the genesis of cyclone. It is also known that the surface and subsurface layers of the southern BOB has more temperature (Kumar et al. 2010) than the northern BOB. From January to February, there is weakening of temperature and currents followed by very weak currents during the eve of spring season although the CHP (>70 × 107 J/m2) intensified in the southern side of the BOB (Fig. 3d). Kumar et al. (2010) have analyzed the subsurface layer and showed that the southwestern BOB gains heat during winter due to highest vertical advection which propagate northward and northeastward. Fig. 3a–c also shows that all eddies (Fig. 2a–c) are formed in the western BOB in agreement with above work. During the month of December, an anticyclonic eddy centered at 18°N, 88.8°E holding very small amount of CHP (~55 × 107 J/m2). The intensity of the current around the eddy is strong (~0.25 m/s). There is also a small cyclonic eddy centered approximately at 17.7°N, 85.1°E holding CHP of value approximately 25.5 × 107 J/m2. During January, eddies are intensified and form three large anticyclonic eddies (Figs. 2b, 3b, 4b) with centers are at (17.4°N, 87.5°E), (14.6°N, 82.3°E), and (11.3°N, 84°E). The CHP’s for these eddies are 27.3 × 107 J/m2, 30 × 107 J/m2, and 40.9 × 107 J/m2, respectively. At the same time, one cyclonic eddy intensified around the center at 10.8°N, 88.7°E with little higher CHP (~47.8 × 107 J/m2) values compare to others. For this cyclonic eddy, both CHP and D 26 are well correlated and can produce some depression depending on the strength of the winds. Anticyclonic eddies becoming stronger and intensified at 17.5°N, 86.9°E and 14.5°N, 82.5°E, respectively during February. The corresponding CHP for these eddies are 34 × 107 J/m2 and 43.2 × 107 J/m2, respectively. These anticyclonic eddies are also simulated with numerical model (Paul et al. 2009) with a little difference in their centers. These eddies can introduce an area of high pressure or anticyclone. The intensity of the current around the eddy is more toward the western side which develops strong western boundary currents (Shetye et al. 1993) in the northeastward (Fig. 2c) side of the BOB. There is a strong current close to the coastal regions of the southern Andhra Pradesh (12.4°N–22°N, 80°E–84.4°E) and the northern Tamil Nadu (8.5°N–13.3°N, 80°E–81°E). Another strong current near the coastal region of Orissa (17.5°N–22°N, 81°E–87°E) is also observed. During this time, another small anticyclonic eddy developed centered at 14.6°N, 86.5°E with CHP value ~39.3 × 107 J/m2. The southernmost anticyclone centered at 10.8°N, 82.2°E with CHP value ~52.2 × 107 J/m2 increasing southward (Fig. 3c–d) in the next season. Overall semiannual signal is observed for the pre-monsoon activity of cyclogenesis in the southern BOB.

Fig. 2
figure 2

Monthly climatology of the currents at 5 m depth with color-shaded SSH (m) for winter: a December, b January, and c February; spring: d March, e April, and f May; summer: g June, h July, and i August; fall: j September, k October, and l November

Full size image
Fig. 3
figure 3

Monthly climatology of the currents at 35 m depth with color-shaded CHP (×107 J/m2) for winter: a December, b January, and c February; spring: d March, e April, and f May; summer: g June, h July, and i August; fall: j September, k October, and l November

Full size image
Fig. 4
figure 4

Monthly climatology of the currents at 70 m depth with color-shaded D 26 (m) for winter: a December, b January, and c February; spring: d March, e April, and f May; summer: g June, h July, and i August; fall: j September, k October, and l November

Full size image

3.1.2 Spring season (March–May)

The anticyclonic eddies in the northwestern BOB extend northeastward and southwestward (Fig. 2d–f) from the prior seasonal position. Shetye et al. (1993) also observed this anticyclonic eddies. During March, anticyclonic eddies intensified and take the place in the northern side with center at 17°N, 86.3°E containing the CHP value ~52.6 × 107 J/m2 and in the southern side with center at 14°N, 82.5°E containing CHP value ~70.7 × 107 J/m2. There is another anticyclonic eddy developed at the southeastern side at 13.6°N, 86.1°E with CHP 61.4 × 107 J/m2. During April, rate of intensification of the anticyclones are very slow compared to March but CHP values intensified with ~92 × 107 J/m2 at 16.3°N, 85.7°E and ~99.3 × 107 J/m2 at 13.3°N, 82.5°E (Figs. 2d, 3d). The reason can be explain by the advection terms of Kumar et al. (2010) from the subsurface response as the horizontal advection is highest at spring for the southeastern, southwestern, and northwestern while it is opposite for the northeastern BOB. Therefore, more warm water of southwestern BOB spreads slowly and trapped in the northwestern BOB. The northeastern BOB also becoming warm due to negative horizontal advection (Kumar et al. 2010) but the important mechanism of the northeastern BOB during this time enhance the upwelling due to coastal Kelvin waves as strong positive prominent vertical advection (Kumar et al. 2010) prevent the warming condition. This situation makes the sea surface more stratified and enhance downwelling Rossby waves in the western BOB (Figs. 2d–f, 3d–f, 4d–f) but at the same time, a cyclonic eddy is developed at 18.7°N, 87.7°E with CHP ~46 × 107 J/m2. The cyclonic and anticyclonic eddies discussed above are also detected by Babu et al. (2003) although there is little difference in the center locations. During May, the cyclonic eddy at 18°N, 86.8°E with CHP 58.4 × 107 J/m2 intensified but completely moved out at the end of summer. A small and weak cyclonic eddy at southwestern side at 7°N, 83.5°E with CHP ~53 × 107 J/m2 also found during this time. Sometimes this cyclonic eddies can produce the high-level depression. The anticyclonic eddies became weak as well as the western boundary current (WBC) but CHP became prominent (112 × 107–122 × 107) J/m2 for the anticyclonic eddies centered at 12.5°N, 82.1°E and 15.6°N, 85.8°E. These eddies can introduce as the cyclogenesis points which leads some sever anticyclones, depending on wind speed, holding CHP values, SSH, and D 26. During this time, currents for the layers in the southern BOB (Figs. 2f, 3f) move southeastwards and intensify during summer season. During late spring to early summer, the CHP values of the southeastern region of the basin are favorable (Fig. 3f–g) for cyclogenesis but subsurface response (Fig. 4f–g) shows that westward movement of warm water from this region can reduce its energy. These results also support the advection terms of Kumar et al. (2010) that the warm phase approaches westward and making the southern basin warmer. This situation prevents to make any strong curl at the surface layer. But because of high CHP values can produce some cyclogenesis which can work as an effective insulator along with the above anticyclonic eddies and produce mostly severe storms which can move along the line insulator (IMD cyclone tracks following northeast and northwest direction during this time, figures are not included). During May, this mechanism is prominent and then reduced slowly at the end of summer.

3.1.3 Summer season (June–August)

During the eve of summer season, one cyclonic eddy centered at 18.4°N, 87°E (Paul et al. 2009) with CHP value ~57.5 × 107 J/m2 with weak circulation pattern and another cyclonic eddy centered at 7.6°N, 84.9°E with CHP ~49.4 × 107 J/m2 are observed (Figs. 2g–i, 3g–i). D 26 comes up (~30 m) in the western boundary (Fig. 4g–i) making this region a strong upwelled zone. The anticyclonic eddies of the northwestern BOB which were strong during spring season reduce their strength but still containing CHP values 101.5 × 107 J/m2 and 104.5 × 107 J/m2 at the centers 13°N, 82.5°E and 15.3°N, 85.8°E, respectively (Figs. 2g, 3g). This anticyclonic eddies completely disappeared during September. D 26 reduces for these anticyclonic eddies also (Fig. 4g). During July, the cyclonic and anticyclonic eddies have taken the forms of lower intensity. The cyclonic eddies at 18.6°N, 87.3°E and 8.8°N, 83.9°E contain the CHP’s ~42 × 107 J/m2 and ~34.3 × 107 J/m2 respectively but both the anticyclonic eddies at 13.7°N, 82.3°E and 15.8°N, 86.3°E contain the CHP ~86 × 107 J/m2 (Figs. 2h, 3h). From June to July, the WBC broken further and circulation became inward to the basin and then southward (Fig. 2h, i). During August, WBC completely reduced and makes the situation prior to the fall season. During this time, all eddies vanish although low CHP value is observed. In the subsurface layer, CHP as well as D 26 are more in the southern BOB (Figs. 3g–i, 4g–i). Only one cyclonic eddy found at 18.6°N, 88.1°E with CHP ~34.6 × 107 J/m2. The anticyclonic eddies are formed at 14.1°N, 81.9°E and 15.2°N, 85.5°E with CHP ~82.2 × 107 J/m2 and ~69.8 × 107 J/m2 respectively (Figs. 2i, 3i).

3.1.4 Fall season (September–November)

In September, only one weak anticyclonic eddy found at 13.9°N, 81.8°E (Fig. 2j) with CHP value ~87.1 × 107 J/m2 (Fig. 3j) During this time, the boundary currents flow southward which is known as East India Coastal Current (EICC) along the western boundary of the BOB basin. Three new cyclonic eddies occur at 15.6°N, 83.2°E; 9.9°N, 82.3°E; and 18.6°N, 88.5°E with corresponding CHP 40 × 107 J/m2, 41.7 × 107 J/m2, and 36.8 × 107 J/m2 respectively. During October, the third cyclonic eddy centered at 18.5°N, 88.9°E became weak but CHP value enhances to ~45.7 × 107 J/m2. The rest two cyclonic eddies are intensified more and move southward, centered at 10.3°N, 82.4°E, 4.9°N, 82.4°E with CHP 48 × 107 J/m2, 50.3 × 107 J/m2 respectively (Figs. 2k, 3k). These eddies are also simulated by Paul et al. (2009) in their modeling study. During November, eddies are completely moved out but all over the BOB CHP values vary from 40 × 107 J/m2 to 80 × 107 J/m2 (Fig. 3l) from west to east. During this time, D 26 lies between 50 and 60 m depth (Fig. 4l). A strong southward surface current is observed in the eastern coastal region of Sri Lanka (Fig. 2l). The magnitude of this current is less in the month of December as seen from the annual cycle pattern.

3.2 Eddy and cyclone potential zone

Six cyclonic and four anticyclonic prominent eddies have been observed from the monthly climatological data of SODA over the BOB region (Fig. 5). Corresponding to their lifespan, seven eddy potential zones have been identified as: Box 1 (18°N–19°N, 86.5°E–89°E); Box 2 (17°N–18°N, 86°E–89°E); Box 3 (15°N–16.5°N, 85.5°E–86.5°E); Box 4 (12.5°N–15°N, 81°E–83°E); Box 5 (10.5°N–11.5°N, 82°E–84°E); Box 6 (7°N–10.5°N, 82°E–85°E); and Box 7 (4°N–5°N, 82°E–84°E).

Fig. 5
figure 5

Domain of the study (BOB) and important eddy potential boxes derived from SODA. Blue color indicates cyclonic eddy or eddy potential region and red color indicates anticyclonic eddy or eddy potential region. Arrows are indicating the direction of movement of the centered of eddies evaluated by the 5-m depth current structure (Fig. 2)

Full size image

The synopsis of this eddies in the BOB have been enlisted in Table 2. The box-wise analysis is presented in Fig. 6a–g, which is the detailed descriptive form of Fig. 5. Aggregate IMD cyclogenesis from 1891 to 2007 has been displayed in Fig. 7 for the better understanding of this analysis. But which parameters are important for the genesis of cyclone in the BOB is still not clear (Sadhuram et al. 2004). The analysis is totally depending with the oceanic thermal structure; however, the cyclogenesis can form due to wind speed, wind stress curl, bottom topography effect, instability conditions of the atmospheric parameters, etc. and, it may be possible to produce a good climatological situation of the cyclogenesis by considering all the conditions.

Table 2 The synopsis of the eddies encountered in the BOB (Fig. 5)
Full size table
Fig. 6
figure 6figure 6figure 6figure 6

a (i) Position of the movement of eddy along with the information on mean SSH, CHP, and D 26, and, (ii) annual mean removed monthly anomalies for SSH (top panel), D 26, and CHP (middle panel) and correlation coefficient among these parameters (bottom panel) for Box 1. b (i) Position of the movement of eddy along with the information on mean SSH, CHP, and D 26, and, (ii) annual mean removed monthly anomalies for SSH (top panel), D 26, and CHP (middle panel) and correlation coefficient among these parameters (bottom panel) for Box 2. c (i) Position of the movement of eddy along with the information on mean SSH, CHP, and D 26, and, (ii) annual mean removed monthly anomalies for SSH (top panel), D 26, and CHP (middle panel) and correlation coefficient among these parameters (bottom panel) for Box 3. d (i) Position of the movement of eddy along with the information on mean SSH, CHP, and D 26, and, (ii) annual mean removed monthly anomalies for SSH (top panel), D 26, and CHP (middle panel) and correlation coefficient among these parameters (bottom panel) for Box 4. e (i) Position of the movement of eddy along with the information on mean SSH, CHP, and D 26, and, (ii) annual mean removed monthly anomalies for SSH (top panel), D 26, and CHP (middle panel) and correlation coefficient among these parameters (bottom panel) for Box 5. f (i) Position of the movement of eddy along with the information on mean SSH, CHP, and D 26, and, (ii) annual mean removed monthly anomalies for SSH (top panel), D 26, and CHP (middle panel) and correlation coefficient among these parameters (bottom panel) for Box 6. g (i) Position of the movement of eddy along with the information on mean SSH, CHP, and D 26, and, (ii) annual mean removed monthly anomalies for SSH (top panel), D 26, and CHP (middle panel) and correlation coefficient among these parameters (bottom panel) for Box 7

Full size image
Fig. 7
figure 7

Seasonal distribution of IMD cyclogenesis points from (1891–2007). During winter season, number is less and formation region is bounded by nearly 5°N–12°N. During summer, number is more but concentrated near head of the BOB, north of 15°N. Spring and fall season shows distribution throughout the BOB with highest concentration during fall

Full size image

The Box 1 is affected by the cyclonic eddy during April–October (Fig. 6a). This region has been identified as the main region for the depression condition during this time. The direction of the movement of the eddy center has been shown in Fig. 6a(i). SSH and D 26 show the same nature of fluctuations for this box but CHP do not show the similar nature like SSH or D 26 (Fig. 6a(ii)). During summer, heavy precipitations can reduce both D 26 and CHP but in this box, CHP reduction is less because of increased solar radiation that may input the surface to subsurface temperature high. The good correlation found during April–June for these parameters. A cyclonic eddy with very weak intensity is formed during April. Thus, effect of this cyclonic eddy in the formation of cyclogenesis is found during May to June due to thermal structure. In this cyclonic eddy centered at 18°N, 86.8°E, the maximum value of CHP, SSH, and D 26 is ~58.4 × 107 J/m2, 0.4 m, and 56.7 m respectively. A very good correlation is also found between CHP and SSH during May. The centered positions are little deviate from the optimal position for the other months and constitute region can produce the region of lowest pressure is called a barometric low or depression or cyclonic storm depending on the strength of the winds.

The Box 2 is affected by the anticyclonic eddy during December through March (Fig. 6b(i)). But during this time, there is not good correlation exists between the parameters (Fig. 6b(ii)) indicating it cannot create any cyclogenesis. During March, the above eddy intensified more (Fig. 6b(i)) and moves southward and then stays in Box 3 for the rest of months. During May through June, CHP is high and correlations between the parameters are also high suitable for cyclogenesis. The Box 3 is also affected by the prior anticyclonic eddy of Box 2 during April through August (Fig. 6c(i)). CHP, SSH, and D 26 values are prominent during this time. The correlation coefficients among the parameters are prominent during April through July (Fig. 6c(ii)), which is suitable for the genesis of cyclone.

The Box 4 is affected by another anticyclonic eddy during January to September (Fig. 6d(i)). This eddy start developing in January then it moves southward. The maximum value of CHP associated with that anticyclonic eddy is found to be ~112.4 × 107 J/m2 during May. The center of this eddy is at 12.5°N, 82.1°E with SSH and D 26 values ~0.64 m and ~94.8 m respectively and with good correlation with CHP. After May, it moves slowly northeastward and moves out in September (Fig. 6d(i)). Figure 6d(ii) shows that although there are good correlations among the parameters from March to November but CHP, SSH, and D 26 are favorable for cyclogenesis only during April to May. Thus, during April–June, this box introduces an area of high pressure; the region of this high pressure is called a barometric high or anticyclone.

An anticyclonic eddy is developed in Box 5 during January to February. It moves slowly southwestward and then moves out during the rest of the months (Fig. 6e(i)). The CHP value is less and not favorable for cyclogenesis but good correlation among the parameters is found. Good correlation among the parameters also exists during April to July (Fig. 6e(ii)). It can be interpreted that the Boxes 3, 4, and 5 act as an effective insulator to each other and combined region is the suitable zone for cyclogenesis or intensity change of tropical cyclone during April to July. Pattern follows like the monsoon activity (annual cycle) of cyclogenesis in the northern BOB.

In Box 6, a cyclonic eddy is observed during May to October (Fig. 6f(i)). It moves from south to north. A good correlation among the parameters during its lifetime is observed but prominent CHP found only in spring season (Fig. 6f(ii)). Thus, this region is favorable for cyclogenesis during spring season. Another cyclonic eddy is seen in Box 7 during October to November (Fig. 6g(i)). A cyclonic eddy starts developing in September and become prominent in October. CHP being less this region is not favorable for cyclogenesis although very good correlations are found among the parameters during October to November (Fig. 6g(ii)).

3.3 CHP and cyclone tracks

Spatial distributions of CHP along with few important cyclone tracks for pre-monsoon, monsoon, and post-monsoon seasons are shown in Fig. (8). During pre-monsoon time, the southeastern side of BOB and the northern side of Sri Lanka along the eastern coast of India show larger CHP compared to other regions (Fig. 8a). These two regions are connected by the regions where CHP is more than 80 × 107 J/m2. The cyclone formed during May 8, 2003, at the location where CHP is larger (>80 × 107 J/m2). Because of available energy for intensification, it gets intensified and as a result, coriolis force increases and it slowly move northward. As it enter the region of high CHP (85°E, 12°N) compared to its previous location of the path, it gets intensified further and because of coriolis force it bent right side of its motion and move eastward, where CHP is less and so it slowly moves toward Myanmar coast (Fig. 8a). Similarly the cyclone MALA formed on April 24, 2006, at the location where CHP is larger and moves slowly northward but as it reaches 13°N, its intensification decreases as the region is having less CHP (~50 × 107 J/m2) and it moves toward Myanmar coast (Fig. 8a). During the monsoon season, cyclones (June 30–July 3, 2006, and September 28–September 30, 2006) move almost eastward and formed in the region where CHP is less (Fig. 8b). So during monsoon, the movement of the cyclone cannot be explained by the CHP distribution. But during the post-monsoon season, the CHP distribution shows the maximum CHP in the eastern BOB including the Andaman Sea (Fig. 8c). The cyclone which formed on November 22, 2002, moves northward as the formation region is bounded by minimum CHP in its western side and maximum in the eastern side. In the further north (14°N), it passes through a region where CHP is larger in all around and as a result, it gets intensified and bent in the right direction of its movement. It reaches 16N on November 28, 2002, and moves toward Myanmar coast (Fig. 8c).

Fig. 8
figure 8

Spatial distribution of CHP (107 J/m2) along with a cyclone tracks (May 8–May 19, 2003; MALA: April 24–April 29, 2006) for pre-monsoon period, b cyclone tracks (June 30–July 3, 2006; September 28–September 30, 2006) for monsoon period, and c cyclone track (November 22–November 28, 2002) for post-monsoon period

Full size image

4 Conclusions

The role of CHP in the genesis and intensification of storms is analyzed by Sadhuram et al. (2004), and Sarma et al. (1990) analyzed the distribution of CHP in the BOB. In this paper, the detailed analysis of CHP and eddies and tracks are analyzed over this basin. The relationship of CHP with cyclogenesis points and eddies, the warm upper layer characteristic has been investigated. The CHP and the cyclonic tracks at different seasons are also analyzed in this paper.

The seasonal variation of the CHP along with SSH and D 26 shows that the region with CHP greater than 70 × 107 J/m2 as well as good correlations (>0.9) with SSH and D 26 can form an area cyclogenesis. For the cyclonic eddies corresponding CHP greater than 50 × 107 J/m2 and good correlation (>0.9) with SSH and D 26 can form as high-level depression. The cyclogenesis zones from our analysis are consistent with the IMD cyclogenesis points. The oceanic parameters analyzed here show annual pattern with sharp peak phase during monsoon seasons for the northern BOB. The semiannual pattern of cyclogenesis is observed in the southern BOB with one highest peak during post-monsoon and second highest peak during pre-monsoon periods. The overall findings are that during winter through spring, the southern BOB has the more warm water than the northern BOB and supports the pre-monsoon activity of cyclogenesis in the southern BOB. The southeastern BOB has the highest warm phase with prominent D 26 (Fig. 4) peak during May to July and second prominent peak during October to November which lead the pre-monsoon and the post-monsoon activity of formation of the cyclogenesis. During summer, the entire BOB basin shows CHP between 60 × 107 J/m2 and 100 × 107 J/m2. The corresponding D 26 lies between 80 and 90 m depth except western coastal areas. The SSH is consistent with CHP variation. All eddies found to be the lowest in intensity or collapse prior to the fall season. During fall season, the entire BOB stratified by warm water (CHP > 60 × 107 J/m2) except northern coastal areas. Some cyclonic eddies intensified and may produce depression from any portion of the BOB basin. It is found in our study that eddies are the one of the cause of cyclogenesis. The seven boxes have been identified as eddy potential zone over the BOB. The Boxes 3, 4, and 5 act as an effective insulator to each other and combined region follows the monsoon activity of cyclogenesis in the northern BOB. It is observed in this study that CHP distribution is useful to explain the movement of the cyclone tracks during pre-monsoon and post-monsoon seasons. During monsoon season, the cyclone tracks cannot be explained from only CHP distribution. However, the cyclogenesis can form due to wind speed, wind stress curl, bottom topography effect, atmospheric instability condition, etc. and need further coupled modeling study which may be suitable for delineating the effective cyclogenesis regions of the BOB and the potential cyclone tracks.