Seasonal Variability of Circulation Along the North-East Coast of India Using Princeton Ocean Model

Abstract

The coastal circulation of northern Bay of Bengal, highlighting the West Bengal-Orissa coast is subjected to seasonal reversing of monsoon circulation. Additional forcing occurs due to freshwater flux from rivers. The surface circulation patterns and SST were analysed in former studies for the whole Bay of Bengal coast, and the effect of freshwater influx on circulation patterns was also studied in detail. We have corroborated those results by studying the surface circulation patterns in the north-east coast of India, on a month-to-month basis using Princeton Ocean Model (POM) forced by wind, salinity, heat flux and temperature for the year 2001. The reversal of surface current along the coast is observed between summer and winter monsoons. The study location shows signature of upwelling during the month of July, along with low-pressure cold-core cyclonic eddy. The presence of upwelling was also confirmed in our study by comparing the result with SeaWiFS chlorophyll data. The upwelling phenomenon was investigated for the July month of the years 1998 and 1996 and it is inferred that there is no correlation of upwelling with former and subsequent El-Niño and La-Nina years.

1 Introduction

It has always been a challenge to understand and depict the ocean circulation features, especially in the case of India and the monsoons. For a semi-enclosed basin such as Bay of Bengal which comes under the tropical monsoon belt and possessing a seasonally reversing circulation, it is found very essential to have knowledge of the coastal circulation patterns. There is a decreasing gradient in the bathymetry which ranges from 4 km at 5°N to about 2 km at about 20°N. The continental shelf extends from 150 to 200 km in width at the Head Bay region. Majority studies have been carried out in the coastal and surface circulation of Bay of Bengal region using models such as ROMS, OGCM and satellite data analysis and by hydrographic studies. The more recent studies of the circulation pattern along the east coast of India were done by measuring the sea level anomalies estimated from satellite altimetry.

The coastal currents along the east coast of India are subjected to seasonal reversal so that it shows different nature in the northern and southern part along the East coast of India. The southward flow of coastal waters is found to appear first in the northern Bay of Bengal by the end of September and then it goes down gaining more strength as it reaches the Sri Lanka. The south-west monsoon and North-East monsoon winds push the surface water southwards from June to September and northwards from November to January, respectively. This is the usual trend observed along the East coast of India throughout the year. In this study, we have considered a small region, the north-east coast of India, which is subjected to both the summer and winter monsoons, so that the seasonal variability of surface circulation patterns can be identified on a local basis and the physical features followed by the monsoon winds can also be identified. Mostly, earlier studies on coastal circulation over BoB were done focussing on the region below 17°N and the localized feature such as upwelling mainly over the Vishakhapatnam coast were studied in detail. A cyclonic gyre off Visakhapatnam in the deep ocean was detected using numerical model MITgcm [1].

The model studies on the circulation over Bay of Bengal reveals a large anticyclonic gyre centred at 15°N, 86°E, fully established in February and this gyre is driven by the north-easterly wind of the winter monsoon [2]. It has been found from past observations that the southward shifting of a cyclonic gyre from the head BoB region and seemed much prominent at a depth less than 200 m for the month of July. In BoB, upwelling has been detected off Walthair in February and June [3]; it probably extends up the coast to Sagar Island off Calcutta in June–July. The observations from the coastal circulation in the northern region of BoB suggests that the dominant factor which affects the salinity and thus the stratification is the estuarine process with a predominant low salinity plume cascading as far south as 17°N [4]. The freshwater influx from the major river Subarnarekha which flows into the southern part of West Bengal coast have not been considered in our study as we need to analyse the sole effect of heat flux on the surface water circulation over the specified region.

2 Data and Methodology

2.1 Site Specification and Bathymetry

The region under study is the West Bengal-Orissa coast which extends from 20.5–22°N to 86.8–88°E. The region is exposed to tropical climate. It annually receives a moderate amount of freshwater through river input from Subarnarekha River (392 m³/s) and by heavy precipitations, especially during summer monsoon period. During the arrival of monsoons, low pressure in the BoB often leads to the occurrence of storms in the coastal areas. The region under study is shown in Fig. 1.

Fig. 1

Google Earth image of site

The bathymetry data was derived from the ETOPO1 global relief model from National Geophysical Data Center (NGDC). It is a 1 arc-minute global relief model for the earth surface which integrates land topography and bathymetry. The bathymetry data was digitalized and given as an input for our model for a depth of 200 m. The model bathymetry is shown in Fig. 2.

Fig. 2

Model bathymetry

For understanding the circulation patterns in north-east coast of India a higher resolution model is needed. The analysis area of the model extends from 20.5–22°N to 86.8–88°E, with an approximate width of 105 km from the Orissa coast, for a depth of 200 m which includes the continental shelf region. There are 73 × 91 grid points in the horizontal plane and 20 levels in vertical direction, comprising 73 × 91 × 20 computational grid points. We have used a uniform grid having resolution of 0.016 km in x-direction and 0.015 km in y-direction.

2.2 Input Parameters

The model is forced with wind stress, derived from QSCAT monthly climatology data for the year 2001, 1998 and 1996. Other input parameters are Sea Surface Temperature (SST), Sea Surface Salinity (SSS), Heat flux and Short Wave Radiation (SWR) are taken from Wood Hole Oceanographic Institution (WHOI) OAFlux. The latent and sensible heat flux for the study area was found out using the Eqs. (1) and (2). The required data for these fluxes, humidity, wind, atmospheric temperature were downloaded for each month from (WHOI) OAFlux.

The turbulent exchange of heat between ocean and water is due to sensible heat and latent heat fluxes. The latent and sensible heat flux estimates are computed from the objectively analysed surface meteorological variables by using the COARE bulk flux algorithm 3.0.

They are commonly computed from the parameterization of the fluxes as a function of surface meteorological observables, such as wind speed, sea-air humidity and temperature gradients. These flux-related variables are obtained from three major sources.

• (1) Marine surface weather reports from voluntary observing ships (VOS)

• (2) Satellite remote sensing

• (3) Numerical weather prediction (NWP) reanalysis and operational data.

The expression for the respective latent and sensible heat fluxes are:

$$Q_{LH}=\rho{L_e}{C_e}U({q_s}-{q_a})$$

(1)

$$Q_{sh} =\rho{C_p}{C_h}U({T_s}-\theta)$$

(2)

$\rho$ is the density of air, $L_e$ is the latent heat of evaporation $C_p$ is the specific heat capacity of air at constant pressure $U$ is the wind speed relative to sea surface at a height z. ${C_e}, \, {C_h}$ are the turbulent exchange coefficients for latent and sensible heat fluxes and they are function of wind speed, height and atmospheric stability ${q_s}$ and ${q_a}$ are the surface and near surface atmospheric specific humidity, ${T_s}$ is the sea surface skin temperature and $\theta$ is the near surface air potential temperature. Table 1 shows the input parameters for the model run.

Table 1 Input parameters

3 Model Description

3.1 Princeton Ocean Model

Numerical models are used to simulate oceanic flows with realistic and useful results. The most recent models include heat fluxes through the surface, wind forcing, mesoscale eddies, realistic coasts and sea-floor features, and more than 20 levels in vertical. The model to be used in this circulation study is the Princeton Ocean Model. POM is a community general circulation numerical ocean model the can be used to simulate and predict oceanic currents, temperatures, salinities and other water properties. The code was originally developed at Princeton University [5] in collaboration with Dialysis of Princeton (H. James Herring, Richard C. Patchen). POM is widely used to describe coastal currents. It is a direct descendant of the Bryan–Cox model. It includes thermodynamic processes, turbulent mixing, Boussinesq and hydrostatic approximations. The model incorporates the Mellor–Yamada turbulence scheme developed in the early 1970s by George Mellor and Ted Yamada; this turbulence sub-model is widely used by oceanic and atmospheric models. Considering the need for a high-resolution numerical model for coastal ocean, and POM was developed with many features to handle tides and coastal circulation it was best suited for the circulation study particularly in our area of study. In POM, the vertical equations are restructured by sigma coordinate system (where σ is used to represent a scaled pressure level) and horizontal equations are recast in a curvilinear coordinate system which provides variable grid sizing. This feature made the model to better handle the coastline features, complex topographies and shallow region. It uses an ‘Arakava C’ grid differencing scheme and has a free surface and split time step. Using Mellor–Yamada turbulence scheme, it solves time dependent, 3-D equations for the conservation of mass, momentum, salt, temperature and turbulence quantities in an incompressible hydrostatic fluid.

In a semi-enclosed region like the Bay of Bengal, the typical forcing at the open boundary includes temperature and salinity, while typical free surface forcing includes the wind stress and flux of heat and freshwater. Along the lateral boundaries, rivers can be prescribed with temperature, salinity and flow rates and in our particular simulation; we have excluded the river salinity at the lateral boundary and used 20 sigma levels in vertical.

4 Model Results

4.1 Circulation Study for the Year 2001

The model clearly depicts the annual surface water circulation that alters twice a year. These surface water currents change their direction from season to season in response to the seasonal rhythm of monsoons. The surface circulation in the West Bengal-Orissa coast is characterized by the coastal current that flows northward from November to February with maximum intensity along the coast during December, and from March to September it flows southward with maximum intensity near the coast during June and July. Then in between months are the transition period from winter to the summer monsoon period (January–March) and from the summer to the winter monsoon period (September–November). The strength of EICC varies from month to month and eventually it will reverse its direction twice in a year during November and February.

During North-East monsoon (winter monsoon) the winds blowing from the north-east direction will move the surface water southwards along the coast of West Bengal. By October which is the onset of north-east monsoon, southward flow starts along the coast of west Bengal which seems to be very weak downwards.

The coastal current moving along the Orissa coast weakens and flow with no such particular direction of flow (Fig. 3a). The SST varies from 28 near the coast to 32 at a depth of 100 m. The current starts gaining more strength by the beginning of November along the coast of Orissa and the southward moving monsoon current flows further south near to the West Bengal coast and found decaying by the end of November. This drives the surface water to circulate in an anticlockwise direction as shown in (Fig. 3b). The temperature along the coast varies from 26 to 29 as it is moving away from the coast. The SST is observed to be less in December and a strong northward flow is found with intense surface currents, whole along the north-east coast of India (Fig. 3c) extending to Head Bay. It can be analysed that during north-east monsoon the coastal currents is less affected by local winds and is more forced with wind stress. A westward flow is also observed at 20.6°N with very less intensity which is the way it should be. The northward flow starts weakening during January and is observed with minimum SST for the whole year as shown in (Fig. 3d).

Fig. 3

a–d Surface circulation patterns for the month October–January, respectively

Weak currents are flowing southwards near the coast for the month of February. SST shows a uniform variation of around 26 °C except for the coastal region as shown in Fig. 4a. Most of the winds are weaker which does not show a particular direction of flow but depicts wind reversal. By March, the wind is getting more intensified and a small clockwise (eddy like formation) is found near the coast of Orissa at about 20.8°N as shown in Fig. 4b. The SST is getting higher and rose to 28 °C. The wind pattern is not clear for the centre portion of our domain where the intensity is much less. For the month of April, the southward flow is found to be more intensified below 21.4°N and a weak eastward flow is also noticed as in Fig. 4c. The SST is increased to 30 °C. Since it is a pre-monsoon period the SST should increase and as it can be observed in the Fig. 4d. It is increasing uniformly and reaching to 32 °C in May. The winds gain more strength by May and circulation pattern is clearer with an anticlockwise pattern of flow. The weak easterly flow still prevails at around 20.9°N.

Fig. 4

a–d Surface circulation patterns for the month February–May, respectively

For the month of June, (Fig 5a), it is observed that at 21.6°N, a freshwater intrusion from the Subarnarekha river changes the circulation pattern north of 21.6°N. Below this latitude coastal current flowing southward is of greater strength and further flows down along the coast of Orissa. A countercurrent flowing northward is also observed. South-west monsoon is building up along the West Bengal coast, above 21.7°N which is pushing the surface water northwards. Due to the seasonal wind change, there is clockwise circulation in the north of 21.6°N and anticlockwise in the south of it. The SST decreases to 30 °C and it found less than 28 °C near the coast. The SST reduces to 26 °C where the river inflow is found. By this time, the currents flowing away from the coast starts losing strength.

Fig. 5

a Surface circulation pattern for the month of June. b The surface circulation of coastal currents for the month of July. It clearly depicts the upwelling phenomenon. c Chlorophyll concentration along the same region (SeaWiFS chlorophyll data)

The month of July clearly depicts the phenomenon of upwelling which stretches from 21 to 22°N and found up to a depth of 30–35 m. The southward moving coastal current prevails over the Orissa coast with lesser intensity compared with the month of June. The along shore current gains more strength as it moves to the north and is directed north-eastward. This north-eastward wind stress shifts offshore and leads to Ekman transport, thereby causes upwelling. A low-pressure cold-core cyclonic eddy is observed around 21.2°N and 87.5°E. Since it is a cold-core eddy, more nutrients rich water is seen to appear there, which is confirmed by comparing the result with chlorophyll data obtained from SEAWiFS which can be seen in Fig. 5b, c. From the figure, it can be inferred that there is a high chlorophyll concentration at the centre of the eddy, which is due to upwelling. The centre of the eddy is characterized with a temperature of 26 °C. The SST went down to 22 °C at the up-welled region where the sub-surface cold water comes to the surface. To the east of eddy a strong southward flow which further moves to the right of it at around 21°N and a portion of it merges with the counter current which is flowing northward off the coast and this again merges with the eddy. Since the south-west monsoon is getting stronger, this intensifies the northward flow along the west Bengal coast. There is further decrease in SST from 31 to 29 °C at a depth of around 150 m.

For the month of august, it is observed that the upwelling region is reduced to the near coast region and cold (<22 °C) is replaced by warmer water (Fig. 6a). The south-west monsoon starts retreating so that the northward flow along BoB is less intensified but still the flow prevails. By this time, the coastal current is also reduced but continues to flow in the same direction as it was in July (southward). The SST varied between 22 and 31 °C with a uniform temperature of 30 °C over a larger area. By September, the whole coast is influenced by EICC which flows southward with lesser intensity as shown in Fig. 6b. The flow is getting more intensified as it flows southward. The SST is lesser as compared to the month of august with a temperature of 29 °C over a large area offshore. South-east monsoon is completely vanished and the flow in the northern region is reversed.

Fig. 6

a, b Surface circulation patterns for the month August and September, respectively

4.2 Circulation Study for the Month of July 1998 (Strong El Nino Year)

1997–1998 is a strong El Nino year and the Indian summer monsoon rainfall was above normal during this period. In our present study, the effect of El Nino on the upwelling along the north-east coast of India is done and is inferred that there is no trace of upwelling observed during the month of July nor an eddy formation is found even though the circulation pattern remains the same (Fig. 7). The counter current which was flowing upward off the coast with much intensity in 2001 was found to decay and forms a clockwise circulation pattern near 21°N in 1998. The SST was also found to be high than the year 2001. There is no such traces of a severe cyclonic activity either. The coastal current flowing southward due to the seasonal monsoon winds (south-west monsoon) is not at all affected by the impact of El-Niño. A counter rotating flow prevails to the south of 21.3°N and the currents are found to weaken as it reaches the Orissa coast.

Fig. 7

Surface circulation pattern for July 1998

4.3 Circulation Study for the Month of July 1996 (Weak La-Niña Year)

During July 1996, the northward flow found to the north of 21.6°N was observed to flow eastward moving away from the coast. The SST is found to be decreased to 26 near to the shore areas when compared with 1998 case, and the upwelling phenomenon is also absent in this case with no such evidence of a cyclonic eddy (Fig. 8). It has been thus inferred that La-Nina brings down the SST but the replacement of sub-surface cold water with the warm surface water along the coast is not found. Instead of the westward moving current for the year 2001, in this case, to the south of 21°N an eastward flow is found prominent. The absence of the counter current flowing northwards away from the coast is another important factor which is noticed in this weak La-Nina year.

Fig. 8

Surface circulation pattern for July 1996

4.4 Sea Surface Temperature (SST) Validation for the Year 2001

The SST variation for the year 2001s being validated with Hadley Center SST data (HadISST) downloaded from APDRC site. Monthly averaged values for the specific region are taken to validate with the model result and the scatter plot is shown in Fig. 9. The correlation coefficient was calculated and found to be 0.988. The model output is showing a similar trend with the Hadley SST data, and so the difference seen can be attributed to some model errors Fig. 10. It shows two peaks for the months of May and October, and SST is found to be maximum during May with average value around 30 °C. During south-west monsoon, the SST decreases up to 28.5 °C for the month of august.

Fig. 9

Scatter plot comprising the model SST result with Hadley SST result

Fig. 10

Comparison of model SST data and Hadley SST data

5 Conclusions

The paper is a numerical simulation of the circulation in the coastal Bay of Bengal near the West Bengal coast. We have used the Princeton Ocean Model to study the current patterns of the Head Bay. The flow in this area is highly determined by a coastal current EICC. The strength of EICC varies from month to month and seasonally it reverses its direction twice in a year; during November and February. The influence of El Nino in the coastal pattern is found to be negligible.

A low-pressure cold-core cyclonic eddy is observed around 21.2°N and 87.5°E at certain period of the year. Since it is a cold-core eddy, more nutrients rich water is seen to appear there, which is confirmed by comparing the result with chlorophyll data obtained from SEAWiFS. It can be inferred that there is a high chlorophyll concentration at the centre if eddy which is due to the strong upwelling features there.

The coastal circulation of the Bay of Bengal near the Head Bay needs to be studied in more detail to discern the effects of the Hooghly and Ganga rivers on the flow. These rivers make the circulation pattern there much more complicated and interesting. Since studying circulation pattern is the first step to the determination of nutrients and phytoplankton concentrations and eventually fisheries, it is important that we study these patterns at length.