Available online at www.sciencedirect.com
ScienceDirect Procedia Engineering 116 (2015) 1072 – 1077
8th International Conference on Asian and Pacific Coasts (APAC 2015) Department of Ocean Engineering, IIT Madras, India.
Impact of SST on tropical cyclones in North Indian Ocean Maneesha Sebastiana, Dr. Manasa Ranjan Beherab* b
a PhD Student, IIT Bombay, Powai, Mumbai, 400076, India. Assistant Professor, IIT Bombay, Powai, Mumbai, 400076, India..
Abstract The impact of the cyclonic activity can be measured based on annual frequency, maximum sustained intensity, landfall, duration, surge height, accumulated cyclone energy and power dissipative index of cyclones. ‘Power dissipative index’ (PDI) is function of duration, frequency and maximum sustained intensity of cyclone (Emanuel 2005). PDI can represent the cyclonic threat better than storm frequency or intensity alone as PDI measures the cyclonic energy dissipation over the basin during a cyclonic event (Emanuel, 2005). PDI of the tropical cyclone over the North Atlantic Ocean basin was related to the climate change by correlating the PDI to sea surface temperature (SST) (Emanuel, 2005). In the present study, an attempt has been made to establish a correlation between the climate change and the cyclonic activities in the North Indian Ocean (NIO) basin. The PDI over the North Indian Ocean basin and individually over Arabian Sea and Bay of Bengal basins were calculated. The PDIs over these basins were correlated to SST obtained from the HadISST1 dataset for these basins. An increasing trend in the SST and tropical cyclone PDI was observed for the past 30 year time period in the North Atlantic Ocean basin by Emanuel (2005). A similar analysis of the cyclonic activity over the North Indian Ocean basin is carried out with the help of available best-tracks of the cyclones from Indian Meteorological Department (IMD) for a period of 23 years (1990-2013). The SST was calculated on temporal period as annual, pre-monsoon and post-monsoon period for study basins. Correlation coefficients for the SSTs over these temporal periods and the PDI of basins were obtained. It was concluded from the study that the SST was not sufficient to establish the changing climate for the cyclonic activities in NIO region for the future. © 2015 Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license © 2014The TheAuthors. Authors. Published by Elsevier B.V. (http://creativecommons.org/licenses/by-nc-nd/4.0/). under responsibility of organizing committee of APAC 2015, Department of Ocean of Engineering, Peer-review Peer- Review under responsibility of organizing committee , IIT Madras , and International Steering Committee APAC 2015 IIT Madras.
Keywords: Tropical Cyclone; Cyclonegenesis; Power Dissipative Index; Sea Surface Temperature;Climate Change
* Corresponding author. Tel.: 91 22 2576 7313; fax: 91 22 25767302 E-mail address:
[email protected]
1877-7058 © 2015 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer- Review under responsibility of organizing committee , IIT Madras , and International Steering Committee of APAC 2015
doi:10.1016/j.proeng.2015.08.346
Maneesha Sebastian and Manasa Ranjan Behera / Procedia Engineering 116 (2015) 1072 – 1077
1073
1. Introduction Tropical cyclone is one of the most potentially destructive extreme events that result in severe loss to the country in terms of life and property. The impact of these events in the changing climate scenario is of extreme importance and is a widely debated topic (IPCC 2013). Accurate prediction of cyclonic activities is necessary for the proper mitigation of the extreme event. A tropical cyclone is characterized by a low pressure centre that is associated with strong wind spiraling inwards towards the centre (IMD 2015). The strong winds of the low pressure system when travel to the shallow coastal region with sufficient energy, cause surge near its landfall which results in flooding of the coastal area. Indian Ocean basin can broadly be divided as North Indian Ocean (NIO) in the northern hemisphere and South Indian Ocean(SIO) in the southern hemisphere. The present study is focus on the NIO region, which has historical record of the best track datasets from 1990 provided by Indian Meteorological Department (IMD). The landmass of Indian subcontinent further divides the NIO region into two basins, Bay of Bengal (BoB) to the east and Arabian Sea (AS) to the west of Indian subcontinent. Various methodologies have been adopted worldwide to study the impact of the cyclone activities across the world. Though the spatial distribution of the cyclone activities varies in different basin, the GCMs results showed an increase in the cyclonic intensity while the frequency of the cyclones reported to be either decreased or remained unchanged (IPCC 2013). The impact of the cyclone activity is measured in terms of its frequency of occurrence, duration, maximum sustained wind intensity, landfall of the cyclone, surge height in terms of the significant wave height, rainfall associated and basin integrated quantities like Accumulated Cyclone Index (ACE) and Power Dissipative Index (PDI) (Wu et al. 2008). NOAA classifies the hurricane season of the North Atlantic Ocean (NA) basin based on the ACE (Trenberth 2005) and is calculated as the square of the maximum sustained wind speed. PDI measures destructive power of a tropical cyclone in terms of the cube of the maximum sustained wind speed achieved by a cyclone during its life time (Emanuel 2005). Studies by Camargo and Sobel (2005) showed that ACE was affected by ENSO in the North Atlantic. Emanuel (2005) showed strong correlation of PDI with SST in the North Atlantic and western North Pacific. Knutson et al. (2010) concluded that the change in tropical cyclone activities due to climate change can be detected by combinations of theoretical, observed and modelling results of the cyclone activities. Theoretically it is well established that, SST greater than 26ºC as one of the requirement for the cyclone genesis (Gray 1968). Other factors that influence cyclone genesis are Coriolis force, vorticity in the low level troposphere, relative humidity and vertical shear wind shear, large scale circulations (Gray 1968, Girishkumar & Ravichandran 2012). Coriolis force is zero at the equator, but it increases while moving towards the poles. Webster et al. (2005) studied the tropical cyclone activities like the number of cyclone, its duration and intensity with the increasing SST in all basins. The study concluded that the SST was increasing over all the basins with the highest rate in the NIO and smallest rate in the North Atlantic. From past observations of tropical activities like number and duration of occurrence of cyclones, decreasing cyclone activities in all the basins except in the North Atlantic were observed (Webster et al. 2005). Higher numbers of cyclonic activities are observed in the North Atlantic and only 5-6% of the total cyclone occurred in NIO (IMD 2015), though the SST observed were higher, leading to further investigation into relation of SST and the cyclone activities in this region. Investigation on the historical track set of cyclones from 1975 to 2004 for North Atlantic (NA), western North Pacific (WNP) and eastern North Pacific (ENP) were carried out by Wu et al. (2008). The study discussed how the changes in the individual frequency, lifetime and intensity of tropical cyclones contribute on the annual accumulated PDI. The study concluded that even the SST was increasing over all the regions, only PDI increased in NA, indicating the importance of PDI in assessing the cyclonic activity in the regions (Wu et al., 2008). Figure 1 shows a typical plot of the tracks of severe cyclonic storms that occurred in the NIO region during 1990-2013 (Cyclone eAtlas IMD). The recent statistical analysis of the past cyclone frequency showed that there is decrease in the cyclone frequency in BoB and small increase in the frequency of cyclone in the AS (Sebastian and Behera, 2015). Though frequency of the cyclones in the NIO is lesser compared to the other basins, the destructive power of the cyclones occurring in these region is higher due to its shallow bathymetry and highly populated coastal area.
1074
Maneesha Sebastian and Manasa Ranjan Behera / Procedia Engineering 116 (2015) 1072 – 1077
Various studies by Miller (1958), Malkus and Riehl (1960), Emanuel (1991) modelled the tropical cyclone physically using thermodynamic approach. A widely accepted thermodynamic approach considers the cyclone as a natural Carnot engine (Emanuel 1986) that obtains its energy from the warm tropical sea. In thermodynamic approach, the maximum potential intensity of the tropical cyclone refers to the theoretical upper limit of the intensity that a tropical storm can achieve. The thermodynamic models of the tropical cyclone predict the possible position of the cyclone formation. Following similar principle, Emanuel (2005) introduced PDI as the measure of the destructive power generated by the cyclone over a year. It is a function of frequency, duration and maximum sustained intensity (Emanuel 2007) of the tropical cyclone. The present study focuses on estimation of PDI for the NIO region. The PDI was computed separately for AS and BoB basin as their dynamics are known to be different. The computed PDI of NIO region was compared with the North Atlantic basin to understand the relative energy content in the study region and corresponding cyclonic activity. Correlation of SST and PDI for the NIO region was obtained to ascertain the influence of SST variability on the cyclonic activity. Nomenclature ACE AS ASO BoB ENP HadISST1 IMD IPCC MAM NA NIO NOAA OND PDI SST WNP
Accumulated Cyclone Index Arabian Sea August-September-November Bay of Bengal Eastern North Pacific Hadley Ice and Sea Surface Temperature Dataset version 1. Indian Meteorological Department Intergovernmental Panel on Climate Change March-April-May North Atlantic Ocean North Indian Ocean National Oceanic and Atmospheric Administration October-November-December Power Dissipative Index Sea Surface Temperature Western North Pacific
Fig. 1. Track of the severe cyclonic storms in NIO during 1990-2013 (Source IMD, last access 1st July,2015)
1075
Maneesha Sebastian and Manasa Ranjan Behera / Procedia Engineering 116 (2015) 1072 – 1077
2. Data and Methodology The methodology adopted by Emanuel (2005) to compute the total dissipative power for the tropical cyclone activity was considered in the present study and was applied over the NIO region. The PDI for Arabian Sea and Bay of Bengal were computed separately. The PDI for the North Atlantic basin from 1949 to 2009 were obtained from Emanuel (2005). Emanuel (2005) computed PDI for the North Atlantic basin from the best track dataset of the Atlantic hurricane database 2 (HURDAT2) from the national hurricane centre of NOAA. The best track dataset of the NIO were provided by IMD (2015) from 1990. The PDI for the NIO region was computed using the equation (1) from the best track dataset provided by IMD for the duration from 1990 to 2013. The power dissipative index defined by (Emanuel, 2005) is given as τ
PDI V 0
3
dt
(1)
max
where, τ is lifetime of the event and Vmax is the maximum sustained wind speed at any given time during a storm event. The maximum sustained wind speed is defined as the one minute average wind speed at an altitude of 10m (NOAA). The best track dataset of the tropical cyclone contains the position as well as the maximum sustained wind reported at 6 hour intervals. PDIs of the individual storms are summed up over a year to obtain the cumulative PDI (here after referred as PDI only) for a given year. The storms with intensity of a cyclonic storm (sustained wind speed = 18 m/s) or higher was used to calculate the PDI. The other lower intensity wind velocities do not contribute to the formation of the cyclonic storms, hence not considered in computation of PDI. The SST for the NIO region was obtained from the UK Met Office Hadley Centre’s Ice and SST dataset Version 1, represented as HadISST1. The SST data is available from 1871 with a grid resolution of 1˚x1˚ (Rayner et al. 2003). However, the SST data from 1990 is used for the present study for the NIO region. The SST of NIO is obtained by averaging the SSTs over the latitude 0.5 ˚ N to 25.5˚ N and longitude 49.5˚ E to 99.5˚ E. The SST of AS is the average SSTs over the bounded region 0.5˚ N to 25.5˚ N latitude and 49.5˚E to 79.5˚E longitude. The SSTs averaged over the region bounded by 0.5˚ N to 25.5˚ N latitude and 79.5˚E to 99.5˚ E longitude represented the SST of BoB. The SST for the post-monsoon season (October-November-December) denoted as OND and pre monsoon season (March-April-May) represented as MAM was also obtained from the HadISST1 dataset for NIO, AS and BoB. The analysis of best tracks datasets of IMD indicates that the cyclones attained severe cyclonic storm or higher during the pre and post monsoon season. Hence, the cyclones during the pre and post-monsoon seasons were only considered for further analysis. 3. Results In the present study, the available potential energy for cyclone formation in the NIO region was obtained by calculating the PDI of the cyclonic events occurred during 1990 to 2013. Similarly, the corresponding PDI of AS and BoB basin were also calculated The SST over these basins were obtained from HadISST1 dataset and used for obtaining the correlation with corresponding PDI. The SSTs were obtained for different temporal spans such as annual, pre-monsoon and post-monsoon. The computed SSTs were used to find their correlation with the PDIs for different basins. The correlation coefficients of the following three cases were calculated for NIO, BoB and AS basins and are summarized in the table 1. A positive correlation was observed with annual SST and PDI for all the three time periods in the NIO region with a maximum correlation coefficient of 0.64 during pre-monsoon. The post-monsoon SST was poorly correlated to PDI with a correlation coefficient of 0.12. In contrary, it is observed from the IMD cyclone track datasets that higher number of severe cyclonic storms occurred in the post monsoon season than in the premonsoon season. Thus, based on the above results, it could be concluded that SST alone is not the driving parameter for the climatic variability of the cyclones in NIO region
1076
Maneesha Sebastian and Manasa Ranjan Behera / Procedia Engineering 116 (2015) 1072 – 1077
Table 1. Correlation Coefficients between SST and PDI for NIO, BoB and AS basins. Temporal span
Annual
Correlation coefficient between SST and PDI NIO
BoB
AS
0.41
0.26
0.52
Pre-monsoon
0.64
0.39
0.73
Post-monsoon
0.12
-0.18
-0.02
a
b
Fig. 2. (a) PDI computed by Emanuel and average SST during ASO the from HadISST1 dataset for North Atlantic basin during 19902009(Emanuel 2005); (b) PDI for the North Indian Ocean basin and Annual SST and OND and MAM season SST from HadISST1 dataset for the NIO region during 1990-2013.
The hurricane season, considered in Emanuel (2005) study, in the North Atlantic Ocean basin starts from August to October (represented as ASO). The PDI and the SST during 1990 to 2009 for the hurricane season calculated by Emanuel (2005) is reproduced in the Fig. 2. (a). Figure 2. (b) represent the variation of PDI, Annual SST, post-monsoon (OND) SST and pre-monsoon (MAM) SST for the NIO region. The PDI values in the NA basin in general are higher than the NIO region, indicating higher cyclonic activities in the NA basin. However, a sudden rise in the PDI in the year 2010 is observed in the NIO region. a
b
Fig. 3. (a) PDI for Arabian Sea and Annual Average SST and Average SST during OND and MAM season from HadISST1 during 1990-2013 in Arabian Sea (b) PDI for Bay of Bengal basin and Annual Average SST and Average SST during OND and MAM season from HadISST1 during 1990-2013 in Bay of Bengal
Maneesha Sebastian and Manasa Ranjan Behera / Procedia Engineering 116 (2015) 1072 – 1077
1077
Figure 3. (a) and (b) show the variation of PDI and SST in AS and BoB basins. It can be concluded from Fig. 3. cyclonic activities were highly active in both BoB and AS basins in the year 2010 which resulted in the higher cyclonic activities in NIO region. Further observations conclude that the SST during the OND season in the both AS and BoB basin reduced while the annual SST and the MAM SST increased in the 2010. It is observed from Fig.3. (a) that the cyclonic activities in the AS was almost nil except in 1994, 1996, 1998, 2004, 2007 and 2010. Thus, it can be concluded from the PDI values that the cyclonic activities are moderate in AS basin. 4. Conclusion The correlation of SST with PDI for NIO region as well as individually for AS and BoB basins were calculated considering the cyclone best track datasets of IMD from 1990 to 2013. For a better understanding, the present results were compared with the SST and PDI for the North Atlantic basin given by Emanuel (2005). The post monsoon SST were poorly correlated with the PDIs in the NIO contradicting the correlation results on the North Atlantic basin (Emanuel 2005) that has strong correlation with the SST of the hurricane season. The cyclonic activities are lower in NIO region than the NA basin that can also be seen in the power dissipative index as the PDI in the NIO is lesser than in the Atlantic basin. The poor correlation between SST and the cyclone energy calculated in terms of PDI proves that SST is not the only driving parameter that influence the cyclonic activity in the NIO region. Thus, it can be concluded that the influence of the changing climate on the cyclonic activities are region specific and the influencing parameters affecting the cyclone genesis could be obtained with detailed investigation of physical parameters like vertical shear, relative humidity, sea level pressure, Coriolis force, effects of interseasonal oscillations in NIO region. References Camargo, S. J., & Sobel, A. H. 2005. Western North Pacific tropical cyclone intensity and ENSO. Journal of Climate, 18(15), 2996-3006. Emanuel, K. A.,1987. The dependence of hurricane intensity on climate. Nature, 326(6112), 483-485. Emanuel, K. A.,1986. An air-sea interaction theory for tropical cyclones. Part I: Steady-state maintenance. Journal of the Atmospheric Sciences, 43(6), 585-605. Emanuel, K. A.,1991. A scheme for representing cumulus convection in large-scale models. Journal of the Atmospheric Sciences, 48(21), 23132329. Emanuel, K. A.,2005. Increasing destructiveness of tropical cyclones over the past 30 years. Nature, 436(7051), 686-688. IPCC 2013: Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change [Stocker, T.F., D. Qin, G.-K. Plattner, M. Tignor, S.K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex and P.M. Midgley (eds.)]. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, 1535 pp, Girishkumar, M. S., & Ravichandran, M. 2012. The influences of ENSO on tropical cyclone activity in the Bay of Bengal during October– December. Journal of Geophysical Research: Oceans (1978–2012), 117(C2). Gray, W. M. (1968). Global View of the Origin of Tropical Disturbances and Storms. Monthly Weather Review, 96(10), 669–700. Rayner, N. A. D.E Parker, E. B Horton, C. K Folland, L.V.A. And D.P.R., 2003. Global analyses of sea surface temperature, sea ice, and night marine air temperature since the late nineteenth century. Journal of Geophysical Research, 108(D14), p.4407. Knutson, T. R., McBride, J. L., Chan, J., Emanuel, K., Holland, G., Landsea, C., Sugi, M. 2010. Tropical cyclones and climate change. Nature Geoscience, 3(3), 157–163. doi:10.1038/ngeo779 Sebastian, M. and Behera, M. R., (2015). Climate change and its correlation with the frequency and intensity variability of cyclones in the Indian Ocean, 3rd IMA International Conference on Flood Risk, Swansea University, Wales, UK, 30-31 March. Miller, B. I. 1958. On the maximum intensity of hurricanes. Journal of Meteorology, 15(2), 184-195. Malkus, J. S., & Riehl, H. (1960). On the Dynamics and Energy Transformations in SteadyState Hurricanes. Tellus, 12(1), 1-20. Webster, P. J., G. J. Holland, J. A. Curry, and Chang, H.-R. 2005. Changes in tropical cyclone number, duration, and intensity in a warming environment, Science, 309(5742), 1844–1846. Trenberth, K. E., Fasullo, J. & Smith, L, 2005. Trends and variability in column-integrated atmospheric water vapor. Clim. Dynam. 24, 741– 758 Wu, L., Wang, B., & Braun, S. A. 2008. Implications of tropical cyclone power dissipation index. International Journal of Climatology, 28(6), 727–731. doi:10.1002/joc.1573 Web Reference Emanuel, K.A Website (http://eaps4.mit.edu/faculty/Emanuel/) last accessed on March, 2015. Cyclone eAtlas IMD (http://www.rmcchennaieatlas.tn.nic.in/) last accessed on 1st July, 2015.