Figure 1 shows the global map of geographical locations of TCCs from 1982-2009. Blue colour dots indicate the positions of TCCs and red colour dots indicate the locations of TCCs that are grown into cyclogenesis. Though the TCCs are more or less uniformly distributed across all the oceanic basins, there are preferential locations for the TCCs that are developing into TCs. To demonstrate this aspect quantitatively, Figure 2(a&b) shows the spatial distribution of TCCs and distribution of TCCs that are developed into TCs, respectively. From figure 2(a), it can be noted that on average maximum number of TCCs is found over the regions of East Pacific (EP), North Atlantic (NA), North Indian (NI) and South Pacific (SP). But the maximum numbers of TCCs transformed into cyclonic storms are more over West Pacific (WP) and EP region, as shown in Figure 2(b). The genesis productivity (GP), which is the percentage of TCCs that develops into a TC, is estimated (Hennon et al., 2013) and shown in Figure 2(c). A higher (lower) GP indicates a more (less) favourable conditions for TCCs that develops into TCs. The high GP values are found over WP, EP, NA and South Indian Ocean (SIO) basins as seen in Figure 2(c). However, their efficiency of transformation of TCCs into TCs is high only over EP and WP regions. Though, the higher GP values are found over NA and SIO basins, the number of TCCs are less as compared to EP and WP regions. Overall, around 5.5% of TCCs are converted into TCs per year over the globe.
Figure 3 (a) shows the number of TCCs developed into TCs from 1980 to 2009 over all the oceanic basins. From this figure, it is evident that TCCs that are developed into TCs are increasing with time with a few large peaks. In order examine whether there is any preferential size of TCCs that are more efficiently developing into TCs, the size spectrum of TCCs are analysed. The size distribution of TCCs is shown in figure 3(b). The maximum number of TCCs are found to be having 100-200 km2 size. It is also noted from this figure that the number of TCCs decrease with the size. This figure thus provides the mean size distribution of TCCs observed in all the oceanic basins. Figure 4(a-d) shows the spatial distribution of TCCs of size 0-100 km2, 100-200 km2, 200-300 km2 and greater than or equal to 300 km2, respectively that are developed into TCs. From figures 3 and 4, it is clear that the dominant size of TCC is 100-200 km2 and at the same time it is the most favourable cloud cluster size to develop into a TC. The tendency of TCCs developing into TCs is significant for the sizes up to 300 km2 in EP, NA, WP, SI and SP basins. There are relatively less number of TCCs with sizes greater than 300km2 that are developing into TCs as shown in figure 4(d). Thus figures 3 and 4 provide information on TCCs size distribution and the preferential sizes that are favourable for developing into TCs, which is relatively a new aspect of the present study. Recently, Wang (2018) emphasized the importance of spatial structure of cloud cluster for the formation of tropical cyclogensis and noted that the organized cloud clusters is one of the key features for TCC to become TC. It is also to be remembered that the development of TCCs into TCs is more sensitive to environmental conditions apart from their size distribution as discussed in the section 1.
To examine whether the duration of a give TCC play a role in triggering cyclogenesis, the life time of these cloud clusters are investigated. Earlier works demonstrated that the life time of TCC plays significant role in formation of cyclogenesis. The knowledge of how long TCCs stay in the atmosphere is important information as they release a large amount of latent heat into the atmosphere. This further alters the vertical structure of atmosphere (e.g., Houze, 1982). Figure 5 shows the percentage of TCCs developed into TCs as a function of their lifetime. Around 48% of TCCs transforms into TCs within 24hrs of being identified. Furthermore, 85% of TCCs develop into TCs within 84hrs of the first identification and only 5% of TCCs develop into TCs after 84hrs. Interestingly, the occurrences of TCCs turned into TCs are increased very steep within 24hrs of first identification as seen in figure 5. It was found that the short lived TCCs develop rapidly into a TC and dissipate very quickly. In contrast, the long lived TCCs develop gradually to become a TC and decay slowly. Recently, Kern and Chen (2013) studied the differences between the developing and non-developing TCCs based on large-scale environments. They identified that the minimum life time of 8hrs is required for TCC to transform into a TC. Hennon et al. (2013) studied the lifespan of TCC and they observed that the most of the TCCs developed into TCs within 24hrs of their identification. Figure 6 shows the life time of TCCs with respect to their size for those developed into TCs. From this figure, it is clear that the cloud clusters having the size in the range of 100-200km2 with 18-42hrs life time are most probable candidate to become cyclogenesis. The number of TCCs is relatively large in the size range of 100-200km2 as compared to others. As the lifetime increases, the efficiency of transforming into TCs are declining at all the size ranges as shown in this figure. The life time of large TCC with size of the order of 600km2 is found to be 18hrs. From this figure, it is evident that for a TCC to develop into a TC, its size as well as its life time are equally important.
To bring out the long-term changes in the TCCs with size ranging from 100-200 km2 that are transforming into TCs, the time series is constructed over various oceanic basins as depicted in Figure 7. Overall increasing trends in developing TCCs are observed in NA, NI, SI and WP basins from 1980 to 2009 in this size ranges whereas decreasing trends are observed in SP and EP basins. Interestingly, the rate of increase is large from the year 2000 in all basins except SP and EP, wherein sharp decrease is observed. Initially, developing TCCs shows decreasing trends in the NA basin up to 1993 and then exhibit increasing trends as shown in figure 7 (a). It is interesting to note a plateau region in number of TCCs during 1992 to 1998 in the WP basin. A decreasing trend in developing TCCs is observed in the SP and EP basins. To quantify the observed long-term changes, the trend analysis is carried out and figure 8 shows the trends along with 95% confidence intervals in number of developing TCCs in various oceanic basins. From this figure it is clearly evident that number of developing TCCs are increasing in NA, WP, NI and SI basins. Though an increasing trend found in these oceanic basins, only NA and NI basins show the significant trends. If 95% confidence intervals contains zero, then the estimated trends are not significant as one cannot rule out the null hypothesis. The observed decreasing trends over EP and SP are also not significant as the confidence intervals contain zero. However, from the time series shown in figure 7(f) and (g), one can infer the decreasing trends. Further, one should have long-term measurements to arrive at any general conclusions. The present time series is limited to 1981-2009 and efforts have to be taken to extend this dataset on TCCs to arrive at any general conclusion.
As discussed in section 1, the background environmental conditions play a key role in deciding whether or not a TCC develop into a TC (Lee, 1989; McBride and Zehr, 1981; Zehr, 1992; Chan and Kwok, 1999; Fu et al., 2007 & 2012; Peng et al., 2012). There were many studies on this topic focusing on western North Pacific and North Atlantic basins. For example, Fu et al. (2012) reported that the western North Pacific region showed distinct large-scale synoptic conditions for developing TCCs as compared to the North Atlantic basin. However, large-scale synoptic conditions favourable for developing TCCs are relatively less explored in the NI basin. In this regard, in the present study, the differences in the background environmental conditions for developing and non-developing TCCs in the NI basin are analysed. The TCCs that are formed during 2005 to 2009 period are considered. Earlier, Pattanaik (2005) studied the environmental differences between the high frequency and low frequency storm activity periods in the NI ocean basin. The author found that low frequency storms are associated with relatively large vertical shear in horizontal winds and decrease in mid-level humidity compared to that of high frequency storm period. As of now, six geophysical parameters such as sea-surface temperature (SST), Coriolis parameter, low level vorticity, low level vertical shear, instability and tropospheric moistening in mid-level are identified as key parameters associated with the genesis of TCs (e.g., Teng et al., 2020). In order to understand the general background environmental conditions in NI basin, a composite of background environmental parameters are constructed. Table 2 provides the list of the cases considered for developed and non-developed TCCs in NI basin during 2005-2009. Strong low-level convergence, weaker vertical shear of horizontal winds, sufficient low-level vorticity and upper tropospheric humidity characterize the tropical cyclogenesis. Therefore, to identify the differences in background environment conditions for developing and non-developing cases, wind speed at 850hPa, divergence and vorticity at 850hPa, shear between 200 and 800hPa and relative humidity at 500hPa are considered. Further, we constructed the composite of all these parameters with respect to each storm center for all the cases listed in Table 2. The storm center location and date is taken from India Meteorological Department (IMD) as provided in Table 2. To investigate the background conditions for developing and non-developing cases, all cloud clusters are considered for the same year and season. These composite results indicate mean state of the atmosphere during the occurrence of TCCs in the NI basin.
Table 2
List of cases considered for developing and non-developing TCCs over NI basin during 2005-2009 period.
Developing | Non-developing |
S.No. | Date | Longitude (oE) | Latitude (oN) | Date | Longitude (oE) | Latitude (oN) |
1 | 13-Dec-2005 | 90.75 | 4.75 | 9-Dec-2005 | 99 | 0 |
2 | 26-Jun-2005 | 89.5 | 22.25 | 20-Jun-2005 | 85 | 14 |
3 | 26-Nov-2005 | 93.25 | 9.25 | 15-Nov-2005 | 89 | 3 |
4 | 25-Oct-2005 | 84.75 | 11.75 | 29-Oct-2005 | 96 | 1 |
5 | 11-Sep-2005 | 86.25 | 19.5 | 18-Sep-2005 | 96 | 9 |
6 | 26-Oct-2006 | 78.75 | 7.5 | 7-Oct-2006 | 81 | 14 |
7 | 9-Nov-2007 | 94 | 10 | 3-Nov-2007 | 99 | 5 |
8 | 15-Jun-2008 | 89.5 | 21.5 | 21-Jun-2008 | 96 | 10 |
9 | 16-Oct-2008 | 64.75 | 8.25 | 22-Oct-2008 | 70 | 11 |
10 | 14-Sep-2008 | 89.5 | 19.25 | 22-Sep-2008 | 84 | 17 |
11 | 18-Jul-2009 | 89.25 | 20.5 | 28-Jul-2009 | 86 | 19 |
12 | 20-Jun-2009 | 70.25 | 16 | 25-Jun-2009 | 88 | 13 |
13 | 7-Nov-2009 | 76 | 9 | 10-Nov-2009 | 75 | 5 |
Figure 9(a and b) shows the composite of wind speed at 850 hPa for developing and non-developing TCCs. It can be noted that there is a pre-existence of low-level vortex for cyclogenesis cases as shown in figure 9(a), while it was not observed during non-developing cases as shown in figure 9(b). These figures are composite of 13 cases and thus assume their importance as only coherent structures will be retained while averaging. This figure thus emphatically shows that all the developing cases have strong low level vortex. Figures 9(c, e & g) depict the magnitude of the environmental divergence, vorticity and vertical shear of horizontal winds for developing cases. Strong low-level convergence, high vorticity and weak vertical shear are prevailed for developing cases around 500km from the disturbance center. From figures 9(g) and (h), it can be noted that the low-level vertical shear is weak in developing cases compared to non-developing cases, which is one of the necessary conditions for a TCC to sustain and develop into a TC at later stages. TCCs having the strong low-level vertical shear are less effective in cyclonic angular momentum (Holland, 1983). In non-developing cases, there is no low-level convergence and strong vorticity as shown in figures 9(d) and (f). Significant differences are noted in the vorticity between the developing and non-developing cases both in the pattern and magnitude. Gray (1968) discussed the importance of strong low-level vorticity and weak vertical shear of horizontal wind in determining the regions of high probability of tropical cyclogenesis. Earlier studies (e.g., Lee and Gray, 1984; Lee, 1986) reported that there are low-level large-scale wind surges which cause the vorticity build-up around the vicinity of the pre-cyclogenesis. Figures 9(i and j) shows the relative humidity composites at 500hPa for developing and non-developing cases. The maximum mid-level relative humidity is found to be closer to the center of the disturbance and the distribution is more symmetric with respect to the center, which is not seen in non-developing cases. The upper level moistening is one of the key components for a developing TCC (Sadler, 1976; Emanuel, 1994; DeMaria et al., 2001). Overall, the relative humidity in the developing composite is around 10-20% higher than that in the non-developing TCCs. Fu et al. (2012) also reported similar results in western North Pacific basin.
The latent heating (LH) in the TCCs also plays a pivotal role in the developing TCCs. The LH profiles obtained from TRMM measurements are analysed to investigate the differences between developing and non-developing TCCs. Figure 10 shows the composite distribution of LH in developing cases and non-developing cases in the NI basin. From this figure, it is evident that the larger LH values are found in developing cases within 500km radius form the disturbance center compared to non-developing cases. The net warming is due to the resultant of condensation, freezing and deposition of water molecules. Though there is relatively large LH at some pockets in non-developing cases, these regions are away from the center of disturbance as shown in figure 10 (b). Further, to investigate the vertical structure of LH inside the cloud clusters for developing and non-developing cases, the mean vertical profile of LH within a radius of 500km from disturbance center is estimated as shown in Figure 11. The mean latent heating is 1.12 K/hr for developing (red color line) and 0.97 K/hr for non-developing (blue color line) TCCs. Thus on average, LH in developing TCCs is 0.15 K/hr larger than that in non-developing TCCs. It is known that the larger LH release aids in stronger updrafts which leads more intense convection for developing TCCs. Collimore (2018) reported the difference in thermal structure for developing and non-developing cases. The author observed warmer temperature in the troposphere for the developing cases indicating the larger LH release. Thus from figure 9 and 10, it can be inferred that strong low-level vorticity, weak wind shear, mid-troposphere moistening and large LH are the key factors for TCCs to develop into a TC in the NI. These results are drawn from composite of thirteen developing and non-developing cases.