Anomolous warming of Indian ocean during 2016 premonsoon
Analysis of 19 years of Sea Surface Temperature (SST) data (2000-2018) showed that 2016 premonsoon was the warmest period in the last 19 years. Anomalously warm conditions (SST anomalies of ~0.8°C) existed in the entire Indian Ocean during the March-April-May of 2016. It was further found that the anomalous warming during 2016 premonsoon was not instantaneous, and there was a gradual build-up of temperature anomalies from 2015 onwards. In this study, the processes that led to the anomalous warming are analysed and quantified following a mixed layer heat budget method. Required heat flux data sets are obtained from TropFlux project, Mixed Layer Depth (MLD) data are obtained from a 1/12° MOM model simulation, and current data are from OSCAR currents. All these data sets are then regridded into regular 1° x 1° gridding before doing the heat budget analysis.
Keywords: sea surface temperature (SST), heat budget equation, heat flux, pre-monsoon season
Indian Ocean (IO), the third largest ocean basin, is unique among other oceans in many respects. It is land locked to the north, and has a warm pool to the east unlike the western side warm pools in other oceans. Absence of steady equitorial easterlies cause an upwelling in the northern hemisphere off northeast africa and arabian peninsuala and east-west of the tip of India, similarly in the southern hemisphere along the northern edge of south-east trades. Such features of IO makes it different from other oceans and its interaction with atmosphere influence the climate on both regional and global scale (Schott et.al 2009).
Climate variability of IO range from intra-seasonal (ISO) to inter-annual and longer time scales. IO shows several kinds of active ISOs such as zonally propagating Madden-Julian oscillation (MJO) northward propagating ISOs like Monsoon Active-Break Cycle (AB Cycle), and biweekly variability in equatorial currents. Out of all these, MJO is the most common or perhaps the well-known intra-seasonal oscillation. The west-east propagation of MJO impact the rainfall over south-Asia. During winter they mostly interact with 5-10°S thermocline ridge while during summer they spin off northward propagating ISOs (AB Cycle) thereby affecting rainfall over India and various south-Asian countries. In addition to ISOs two most prominent inter-annual climate features are present in IO, namely, the El Niño–Southern Oscillation (ENSO) and Indian Ocean Dipole (IOD). ENSO has its origins in tropical Pacific and cause a change in sea surface temperature (SST) in the central and eastern tropical Pacific Ocean on periods ranging from 3 to 7 years. During this period tropical ocean warm or cool by 1 to 3°C from normal. Such oscillating warming and cooling pattern known as ENSO cycle affect the rainfall and weather across the tropics and other parts of world. In the Indian Ocean, during the warm phase of ENSO, called the El-Nino, SST is usually warmer and monsoon is weak, and vice versa during the cold phase of ENSO, called the La Nina. ENSO affects IO through teleconnections, and the peak influence is felt in the IO with a lag of one season. On the other hand, IOD is an intrinsic mode of inter-annual variability in the IO and plays an important role in regulating the weather and climate of IO. It generally occurs during the boreal summer (June, July, august - JJA) and peaks mostly during spring (September, October, November - SON). IOD varies between positive, negative and neutral phases. During the positive phase, SSTs are anomalously warmer than normal in the western equatorial Indian Ocean (WEIO), and cold in the eastern equatorial Indian Ocean (EQIO). Opposite anomaly occurs during negative IOD (Benjamin et.al). These IODs impact the weather and climate of tropical Indian Ocean (TIO) and several surrounding countries. Such anomalous warming of WEIO during positive IOD causes normal or more than normal rainfall in east Africa, India and other ad joint areas, on the other hand on the eastern side of TIO i.e.-Australia and Indonesia it causes drier than normal condition causing drought. Negative IOD plays exactly the opposite role to that of positive IOD but the impact of negative IOD is lesser than that of positive IOD.SST of the IO is one of the main controlling factors of the onset, duration and quantity of the Indian summer monsoon which occurs with regularity during June-September period. 2015-16 was a unique period for the IO, in which, effects of a strong El-Nino was felt through teleconnections and a strong positive IOD was also developed in the IO. As a result of these complex climate mode interactions, strongest warm SST anomalies were observed during the pre-monsoon period of 2016 in the Indian Ocean (figures 1, 2), with far reaching climate implications. This warm anomaly was developed in early 2015, persisted throughout the year and peaked during 2016 pre-monsoon. In this study, a comprehensive analysis of the triggering mechanisms of these warm anomalies are analysed through a heat budget analysis.
Objectives of the Research
Our study mainly focuses on north Indian ocean basin (NIO). We select 30°S-30°N and 40°E-120°E as the latitude/longitudianl bounds of NIO. For a detailed analysis and understanding, we divide NIO into 4-sub basins i.e-Arabian sea (AS), Central equitorial Indian Ocean (CEIO), Bay of Bengal (BOB) and southern tropical Indian ocean (STIO) (figure-2).
This study mainly focuses on the processes that contributed to the warming of NIO during 2015-16 period. During March through May, 2016, NIO was found to be warmest compared to last 19 years (2000-2018). But this peak warming of NIO during 2016 premonsoon is not an instantaneous one, rather there was a steady build up of anomalous surface temperatures from 2015. Reasons for these anomalous warming episode is analyzed in the rest of the thesis.
The data used in this study include Sea surface temperature (SST), latent heat flux (LHF), sensible heat flux (SHF), long wave radiation (LWR), and short wave radiation (SWR) from TropFlux datasets (Praveen Kumar et.al, 2012,2013) at 1°×1° spatial and monthly resolution. TropFlux data provides daily data and monthly data over the global tropical oceans (30°S-30°N) from 1979 to 3-4 months behind the present (Praveen Kumar et.al 2012).. We used a MOM model simulation mixed layer depth (MLD) data originally at 1/12° resolution since no gap-free observational record was available over 2000-2017 period. Ocean surface current analysis-real time (OSCAR;.oscar.noaa.gov) data was used for meridional and zonal currents, which provides a 5 day resolution gridded data since 1972. The OSCAR current data combines geostrophic currents from satellite-derived sea level, as well as the wind-driven (Ekman) contribution. Data used for the analysis are then re-gridded into 1°×1° grid and also all weekly and daily data were also converted into monthly data. All the computation and analysis are done using MATLAB and CDO.
A mixed layer heat budget method was followed to analyze the reasons of the anomalous warming observed during 2015-16 period. Specifically, the following heat budget equation was used to estimate various terms:
Here, T is the sea surface temperature (SST), Qs is the surface net short wave flux; Q* is the sum of latent heat, sensible heat and long wave fluxes; the f(-h) function describes the fraction of short wave penetrating below the mixed layer. ρ0is the sea water density; Cp sea water volumetric heat capacity; h is mixed layer depth and U and V are mixed layer average zonal and meridional currents respectively. The first term in the RHS represent the effect of atmospheric heat fluxes on the mixed layer, second and third terms represent the effect of zonal and meridional advection by mixed layer currents respectively and the fourth term is the residual term or the contribution of all sub-surface process. The SST tendency term is calculated from monthly TropFlux SST data. The atmospheric forcing is estimated using TropFlux fluxes and model based MLD. Surface lateral advection is estimated using Oscar currents and TropFlux SST. The sign convention followed in this thesis is that the flux warms the ocean (or going into the ocean) is positive and that leaving it (cooling the ocean) is negative.
RESULTS AND DISCUSSIONS
Evolution of Sea Surface Temperature during 2015
Figure 3: (a) to (l) corrosponds to SST evolution of NIO from January to December
Figure 3 shows the evolution of SST anomaly (SSTA) in the Indian Ocean during 2015. Except western tropical IO, warm SSTAs were observed in the IO from January 2015. The SST warmed steadily and evenly across the IO basin, and by December 2015, the whole of IO was warmer by 0.5 - 1°C. During SON period, cold SSTAs were observed along the eastern equatorial IO, suggesting conditions of a positive IOD. In the remaining sections, the contribution of each of the processes that contributed to the anomalous warm conditions of the IO during 2015 are analysed. Time series analysis is also performed to see the specific processes relevent to subdomains and also to have a meaningful segregation of seasons (northern hemisphere summer corrosponds to suthern hemisphere winter).
Figure-4: (a) to (l) corrosponds to anomaly of SST tendency of NIO from January to December
Atmospheric forcing is the sum of net non-solar radiation (latent, sensible and longwave radiation) and net solar radiation that resides within the mixed layer after a fraction of it penetrate below the MLD. Figure 5 shows the anomaly of atmospheric forcing over the IO region. Late winter and early premonsoon periods are marked by less than normal flux forcing in the NIO basins, thus providing a damping effect on SST tendency. It picks up in late premonsson and monsoon periods and provide anomalous warming of the NIO MLD surface layers. During these periods, ie. from late winter through monsoon period, the SO experiences a reversal of role by atmospheric forcing. ie. it provides a warming tendency during later winter and early premonsoon and a cooling tendency until the end of summer monsoon. Anomalous warming tendency is stronger in BoB in the rest of the time, but weaker in the AS. Another region of significant flux warming is the eastern tropical Indian Ocean.
Figure 5: (a) to (l) corresponds to anomaly of Atmospheric forcing of NIO from January to December
Climatological subsurface processes, in general, contribute to cool the mixed layer trough processes like vertical upwelling and entrainment. Both processes result in the mixing of warm mixed layer waters with relatively cooler subsurface waters, thus cooling the mixed layer water. In the present analysis, contribution of subsurface processes is estimated as a residual of the contribution of other processes. One drawback of such an approach is that it results in the accumulation of errors from other terms, but provides a computationally convenient method. Figure 6 provides anomalous contribution from subsurface processes.
Hence an anomalous warm (cool) subsurface contribution suggests a decrease (increase) of the cooling from subsurface processes.
Subsurface processes contribute an anomalous warming (a decrease of climatological cooling) throughout late winter and early summer and then cools the surface layer until the end of northern monsoon in the NIO basins. This is, in fact, counter balancing the role of anomalous flux forcing in the NIO basins during this period. Warm anomalies are present in the 5-15°S region. Both warm and cold patches of anomalous subsurface contribution is present in the rest of the basins in the remaining period.
Figure 6: (a) to (l) corresponds to anomaly of Residual term of NIO from January to December
This term is evaluated but found very noisy and less significant compared to atmosphere if flux frocing and subsurface processes, hence a detailed explanation is not provided here. Variation of it is shown in fig-7.
In the following section the whole IO region is divided into 4 regions, namely AS, BoB, Central Equatorial IO and Southern Tropical IO, and timeseries analysis of the anomalous heat budget terms are presented. This gives a more succinct idea about the role and magnitude of each of the budget terms. Following de Boyer et al (2007), the AS is further divided into Eastern AS (EAS) and Western AS (WAS) considering different oceanographic conditions of the regions.
Eastern Arabian Sea
In the EAS region, two warming episodes are observed during 2005, one during premonsoon period (April-May) and another during monsoon period (July-Aug-Sept). The stronger warming of more than 0.5°C during the monsoon period persisted rest of the year and provided the background conditions for the anomalous peak warming of 2016 premonsoon period. Timeseries of anomalous budget terms suggest that the dominent terms that contribute to the anomalous SST conditions are atmospheric flux forcing and subsurface processes. Lateral advection is a weak secondary process compared to the other two terms. Climatological maximum warming occurs during March-April driven mstly by atmospheric fluxes and the secondary warming occurs during the end of summer monsoon period (Sept-Oct), which is driven again by flux forcing and a reduction in subsurface climatological cooling. Anoamlous flux forcing and subsurface terms were mostly in phase opposition during 2015 suggesting that they balance each other and the dominent term drives anomalous warming/cooling. First peak in the SST warming seen during April-May was due to anomalous warm flux forcing acting on climatological positive flux forcing. Subsurface forcing was counter acting and cooling the MLD waters, but was not enough to balance the strong warming from flux forcing. During the second warming period, ie. July-Aug-Sept, initiation of the warming was triggered by flux forcing, but at a later stage, ie, after Sept, anomalous flux forcing and subsurface terms become weak, thus allowing the warm SST to persist.
Western Arabian Sea
The warming anomaly progressively increased in the WAS region from March onwards and peaked during Sept-Oct 2015. SST warming during March to May was triggered by excess atmospheric fluxes, and the peak in SST anomaly during Sept-Oct 2015 was due to a reduction in the climatological subsurface cooling. Figure 9c suggests that there is a phase reversal of the role of atmospheric forcing and subsurface cooling, as anomalous atmospheric forcing contributed to the initial warming and later the warm conditions were maintained and intesified by a reduction in subsurface cooling. Lateral advection is a weak secondary process compared to the other two forcing terms.
Figure 9: Time series plot for western arabian sea, a)SST anomaly, b) climatology of tendency term, c)anomaly of tendency term for 2015
Bay of Bengal
In 2015, El-Nino event was accompanied with a positive IOD phenomena, which made western side of IO i.e. AS region warmer than that of BOB region from August onwards. Positive anomaly in fig-10a showed an increase in SST but compared to AS region, this rise in SST was marginal. SST raised to its maximum during the end of the year 2015 and this warming prevailed fairly till the pre-monsoon period of 2016.
In BOB region, anomalous warming is mostly driven by atmospheric forcing and sub-surface cooling. As seen in figure 10c, there was a clear role reversal of contributions from subsurface processes and atmospheric forcing. A reduction in subsurface cooling initiated the initial warming of the Bay, but after May 2015, the warming was driven by anomalous surface fluxes. At the end of 2015, both flux forcing and subsurface process contributed positively to reach warmest SST in the Bay.
Figure 10: Time series plot for bay of bengal a)SST anomaly, b) climatology of tendency term, c) anomaly of tendency term for 2015
Central equatorial Indian ocean
This region is bounded by 5oS-5oN latitude and 50oE-100oE longitude. In this region there was continuous increase in SST during 2015 except during July where there was a slight decrease in SST. Climatological atmospheric forcing plays a dominant role except during monsoon when there is a slight decrease in atmospheric forcing. Sub-surface cooling tend to balance the heat flux term during most of the year. Compared to NIO basins, climatological advection plays a larger role, it cools during beginning of year and warms during summer monsoon time. Until the end of pre-monsoon, the initial warming seen in this basin was due to the anomalous positive contributions from flux forcing and subsurface processes which acted in seesaw. During the summer monsoon period and later into early winter, anomalous contributions from flux forcing and subsurface processes were to warm the mixed layer. This anomalous warming was balanced with a cooling effect from lateral advection. It is to be noted that the lateral advection cooling was not sufficient to balance the combined warming effects of flux forcing and subsurface processes, hence resulted in a warming episode that started from beginning of summer monsoon.
Figure 11: Time series plot for central equitorial indian ocean, a) SST anomaly, b) climatology of tendency term, c)anomaly of tendency term for 2015
Figure 12: Time series plot for southern equatorial IO, a) SST anomaly, b) climatology of tendency term, c) anomaly of tendency term for 2015
Southern tropical Indian ocean
Southern tropical Indian Ocean (STIO) is characterized from 10oS-25oS latitude to 55oE-100oE longitude. Warm SSTs were maintained in this basin almost throughout the year 2015. Climatological flux forcing follows the annual cycle of sun, resulting in a cooling during northern summer and warming during northern winter. Climatological advection is mostly warming the mixed layer waters and comparable in magnitude compared to the other two terms of heat budget. Climatological subsurface cooling is maximum during northern winter and it reduces during northern summer. Anomalous contributions of flux forcing and subsurface processes always counterbalanced during 2015, but the positive effect always dominated resulting in a uniform and persistent warming of STIO.
SST evolution during 2016 pre-monsoon season
Warming that started in 2015 prevailed in the Indian Ocean till pre-monsoon season of 2016. During 2016 pre-monsoon, the entire Indian Ocean was warmer by 1.5° C than normal. By the beginning of March, the whole IO gets warm evenly, and such warming prevails throughout the whole pre-monsoon period in the entire Indian Ocean.
Figure 13: (a) to (l) corresponds to SST evolution of NIO from January to june 2016
After the onset of summer monsoon whole NIO showed a decreasing tendency of SST which is shown in later sections. In most of the regions after summer monsoon, SST starts decreasing and showed a slight positive or a negative value. In the beginning of 2016 SST was passively modulated by 2015-16 El-Nino event, and the warming prevailed till January -April throughout the basin. However, during boreal spring due to evolution of a strong negative IOD event the SST of WIO decreased by 0.5° C than normal. The northwesterly wind anomaly due to negative IOD along the coast of Sumatra reduced the local evaporation and cause a warmer EIO. However, 2016 pre-monsoon was the warmest among the last 19 years. In the following sections like the previous section, we will discuss various budget equation parameters and its impact on the warming.
Figure 14: (a) to (l) corresponds to SST tendency of NIO from January to june 2016
As per figure 13, warm SST anomalies persisted throughout the IO from January to June 2016. Figure 14 shows the corresponding SST tendencies, calculated by estimating the rate of SST change in forward difference method. Thus the negative values seen in January 2016 (blue shade) suggests that SST anomaly was warm (figure 13a), but SST decreased compared to its February values. This way, the maximum SST warming happened during March 2016 (see figures 13c and 14b), and that is the recorded warmest pre-monsoon SSTs in the IO during the last 19 years of analysis period. After that, slowly, warm SST anomalies decreased as seen in the negative SST tendency values in figure 14. In the following sections, the processes that contributed to the maximum pre-monsoon warming in the IO will be discussed.
Figure 15: (a) to (l) corresponds to atmospheric forcing anomaly of NIO from January to june 2016
Atmospheric Forcing and Sub-Surface process
In BOB and AS region, from January to March, atmospheric forcing acts as a negative contributor for the increase in SST, and warming is driven by a reduction in sub-surface cooling (seen as positive values in figure 16). The exception is the eastern part of AS where flux forcing increased the SST anomalies during March and subsurface cooling tried to reduce the warming effect. Lateral advection is very weak compared to the other process, hence not shown here
The area averaged flux forcing is positive in the central equatorial IO and Southern tropical IO, suggesting an enhancement of SST warm anomalies. Subsurface cooling provided necessary damping effect for the flux induced warming. Here also, lateral advection was found to be a minor process in the SST anomaly formation.
During April SST evolution is solely driven by atmospheric forcing only and sub-surface cooling tends to balance the warming. By the end of pre-monsoon season there occurred an increase in flux forcing which acts as a driving force for SST evolution and due to reduced upwelling along coastal regions, sub-surface process act as balancing factor for SST warming in NIO basins.
Figure 16: (a) to (l) corresponds to subsurface process anomaly of NIO from January to june 2016
Eastern Arabian Sea
In EAS region SST peaked in March 2016 (figure 17a). Heat budget analysis suggests that the process that contributed to the strong anomalous warming during 2016 pre-monsoon is due to the large reduction in cooling from subsurface processes. Flux forcing was weak, and provided a cooling effect from top, but the stronger reduction of subsurface cooling resulted in a net warming of the mixed layer. Lateral advection was found to be very weak.
Western Arabian Sea
Western Arabian Sea region also shows the same trend of temperature increase, as that of EAS. In WAS region also SST increases anomalously from the beginning of 2016, and peaked in March. During January – February reduction in sub-surface cooling act as the driving mechanism for anomalous increase in SST, during this time increase in SST was seen to be counterbalancing by the atmospheric forcing. At the beginning of pre-monsoon, the decrease in SST tendency is seen to be driven by both sub-surface cooling and decreased atmospheric forcing. By the mid of pre-monsoon (April) reduced sub-surface cooling and atmospheric forcing almost balance each other thereby causing SST tendency value nearly equal to zero. In the beginning of May increased upwelling and increased solar penetration increases the sub-surface temperature more than the surface temperature. During this period atmospheric process act as a strong contributor and sub-surface heating act as a balancing factor. However, winter cooling is solely driven by atmospheric forcing only.
Bay of Bengal
During 2016 pre-monsoon the whole IO basin was warmer compared to other years. In BOB basin there is an increasing trend of SST from the beginning itself and in this region the trend persists for a longer period compared to that of EAS and WAS basin. Like other two basin it also showed a maximum value in March around 1o C than normal. In BOB basin after the beginning of monsoon also there is an increasing trend of temperature till the end of post-monsoon season. During winter season also there is a positive SST anomaly only but the value decreases to almost 0.2 C.
In this region warming at the beginning is driven by reduced sub-surface cooling which is counterbalanced by atmospheric forcing. In the beginning of pre-monsoon season decreased SST tendency occurred mostly due to combined effect of sub-surface cooling and atmospheric forcing. During mid of pre-monsoon season atmospheric forcing reaches to its maximum value of around 2.3 C/month and the warming occurs due to atmospheric forcing only. This warming was balanced by sub-surface cooling in this period. After May onwards SST tendency is mostly driven by heat flux only and throughout the year sub-surface process plays a major role in balancing the SST tendency. During mid of monsoon period cold water advection in this region comes into frame as a driving mechanism for decreasing trend of SST tendency. In winter atmospheric forcing and sub-surface cooling balance each other and decreasing tendency of SST is very small and such small decrease is driven by cold water advection and sub-surface cooling.
Central Equatorial Indian Ocean
CEIO region shows increasing tendency in SST from January to May, maximum increase in SST occurs during beginning of pre-monsoon season. After the beginning of monsoon SST decreases by -0.5°C during November. In this region decreasing SST tendency during January is due to sub-surface cooling. Such decreasing tendency in SST is balanced by heat flux and warm water advection in this region. In this region SST tendency is mostly driven either by sub-surface warming or sub-surface cooling. Atmospheric forcing and advection balance the warming or cooling of SST driven by sub-surface process. During May large negative SST tendency seems to be due to combined effect of both sub-surface cooling and negative atmospheric forcing. Like BOB in this region also atmospheric forcing reaches to maximum value of around 0.3 C/month than normal during April. In CEIO region advection plays a major role after the onset of summer monsoon. During august cold-water advection in this region justifies the negative SST tendency value during this period. In winter increase in SST tendency is the combined effect of advection and atmospheric forcing which counterbalanced by sub-surface cooling around -0.5o C/month than normal.
Figure 20: Time series plot for Central equitorial IO, a)SST anomaly, b) climatology of tendency term, c)anomaly of tendency term for 2016
Southern tropical Indian Ocean
This region shows a warming tendency in temperature till the beginning of monsoon. There occurs maximum raise in SST during April, 2016 around 1 o C warmer than normal. After the onset of summer monsoon, it shows a completely decreasing tendency in SST afterwards. SST anomaly has a minimum value during December around -0.5o C colder than normal. SST tendency has a maximum during February, there is an increasing tendency from January – May after may there is a decreasing pattern of SST tendency can be observed, such a decreasing tendency indicates a decreasing SST in STIO region after mid of spring season. Such anomalous increase in SST in STIO is primly supported by two major factors namely-sub-surface process and atmospheric forcing, advection plays a very minor role and in most of the region it seems to get balanced by any of the two major contributors. In the beginning pre-monsoon period atmospheric forcing act as the primary contributor to SST increase. During this period sub-surface cooling act as a negative feedback system there by reducing the increase in SST. During this time warm water advection in this region also contribute to increase in SST.
Figure 21: Time series plot for south tropical IO, a)SST anomaly, b) climatology of tendency term, c)anomaly of tendency term for 2016
Since April onwards SST tendency decrease seems to be due to the combined effect of both atmospheric forcing and sub-surface cooling and sub-surface cooling gets balanced by warm water advection in this region, thereby causing atmospheric forcing a major contributor for SST tendency. By winter sub-surface warming in this region is mostly due to negative IOD event that contribute to anomalous warming of ocean surface mostly in the South-eastern side of IO (fig-not shown). During winter negative atmospheric forcing anomaly compensated by reduction in sub-surface cooling causing an increase in SST as a consequence of negative IOD event.
During 2016 boreal summer and fall, a strong IOD event occurred which increased the SST of EIO compared to WIO. North-westerly wind anomaly off the coast of Sumatra reduced the local evaporation and increased the temperature of EIO and south-east IO region compared to WIO region. 2015 was a year of positive IOD which caused AS region to be warmer than that of BOB region. During 2015 El-Nino and positive IOD caused the WIO basin to be warmer than that of EIO basin, such warming of 2015 prevails till 2016 pre-monsoon season as an impact of that 2015-16 El-Nino event after 2016 pre-monsoon evolution of negative IOD cause a shift in evenly distributed warming to east and south-east IO region. Moreover 2015 act as a build-up year or triggering year for 2016 pre-monsoon season anomalous warming. In most of the region atmospheric forcing and sub-surface cooling contribute to the warming or cooling of IO. Advection term was a very negligible contributor except during some time and some region.
Such warming and cooling of the two year had its own pros and cons like- during positive IOD event Indonesia and Australia were seem to be drier year than normal and there happened drought. Whereas during negative IOD event there occurred East-Africa drought condition. However, such an increase in SST of IO due to El-Nino and periodical positive and negative IOD cause a reduced rainfall in India mostly during 2015 monsoon season which is already discussed in the previous section.
Schott, F. A.,Xie, S.‐P., andMcCreary, J. P.(2009),Indian Ocean circulation and climate variability,Rev. Geophys.,47, RG1002, doi:.
Ng, B., Cai, W., Cowan, T., & Bi, D. (2018). Influence of internal climate variability on Indian Ocean Dipole properties.Scientific reports,8.
Kumar, B. P., Vialard, J., Lengaigne, M., Murty, V. S. N., & Mcphaden, M. J. (2012). TropFlux: air-sea fluxes for the global tropical oceans—description and evaluation.Climate Dynamics,38(7-8), 1521-1543.
Kumar, B. P., Vialard, J., Lengaigne, M., Murty, V. S. N., Mcphaden, M. J., Cronin, M. F., ... & Reddy, K. G. (2013). TropFlux wind stresses over the tropical oceans: evaluation and comparison with other products.Climate dynamics,40(7-8), 2049-2071.
I, Smita Panda, would like to acknowledge Dr. B. Praveen Kumar, for his advice, supervision, guidance and encouragement during the tenure of the project work without which the task would not have been completed. Whenever I showed sign of weakness, I always found his benign presence, guidance and support available to me all the time.
No words would sufficiently express my sincere gratitude to my professors Dr. Jagabandhu Panda, Dr. Krishna Kishore Osuri, Dr. Bhishma Tyagi and Dr. Naresh Krishna Vissa for their constant encouragement and moral support.
I owe special thanks to my friends Sunanda Narayan, Rony, Rajesh, Satish, Vaishnavi and Trishnita for their encouragement and moral support and for standing by my side during my ups and downs.
Last, but not the least, I shall ever remain obliged to the Almighty for my good health and my parents Mrs. Sandhya Rani Panda and Mr. Prasanta Kumar Panda, who were constant source of inspiration for me in completing this work. Their encouragement, moral support and unconditional love, helped me to overcome occasional moment of distress.