in the northern Indian Ocean?

This article was downloaded by: [prince prakash] On: 12 October 2013, At: 23:07 Publisher: Taylor & Francis Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Remote Sensing Letters Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/trsl20

Can oxycline depth be estimated using sea level anomaly (SLA) in the northern Indian Ocean? a

ab

Satya Prakash , Prince Prakash

& M. Ravichandran

a

a

Indian National Centre for Ocean Information Services (INCOIS), Hyderabad 500 090, India b

National Centre for Antarctic and Ocean Research (NCAOR), Goa 403 804, India

To cite this article: Satya Prakash, Prince Prakash & M. Ravichandran (2013) Can oxycline depth be estimated using sea level anomaly (SLA) in the northern Indian Ocean?, Remote Sensing Letters, 4:11, 1097-1106 To link to this article: http://dx.doi.org/10.1080/2150704X.2013.842284

PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/termsand-conditions

Remote Sensing Letters, 2013 Vol. 4, No. 11, 1097–1106, http://dx.doi.org/10.1080/2150704X.2013.842284

Can oxycline depth be estimated using sea level anomaly (SLA) in the northern Indian Ocean? SATYA PRAKASH∗ †, PRINCE PRAKASH†‡ and M. RAVICHANDRAN† †Indian National Centre for Ocean Information Services (INCOIS), Hyderabad 500 090, India ‡National Centre for Antarctic and Ocean Research (NCAOR), Goa 403 804, India

Downloaded by [prince prakash] at 23:07 12 October 2013

(Received 29 May 2013; accepted 4 September 2013) Information on depth of oxycline is critical not only for understanding magnitude and extent of the hypoxic zone but also for specifying potential fishing zones on operational basis. We analysed Argo-oxygen data from the northern Indian Ocean, along with sea level anomaly (SLA) data from altimeter, to demonstrate the correlation between depths of oxycline, thermocline and SLA. Our analysis suggests that observed variability in oxycline depth is mainly governed by physical processes such as vertical movement in the thermocline depth in the northern Indian Ocean basin. There exists strong positive correlations between depths of thermocline, oxycline and SLA. Oxycline depth and SLA are highly correlated in the Arabian Sea, but the correlation between the two is weaker in the Bay of Bengal and equatorial Indian Ocean. We propose a regression equation between SLA and oxycline depth, which may be used to estimate the depth at which water is oxygen deficient (through oxycline) in the northern Indian Ocean.

1. Introduction The northern Indian Ocean is a unique tropical basin that is landlocked at its north and is forced by seasonally reversing monsoons which make this basin a host to a wide range of biogeochemical provinces such as highly productive Arabian Sea (AS) (Wiggert et al. 2005, Prakash et al. 2008), oligotrophic equatorial Indian Ocean (EIO) (Gandhi et al. 2012) and relatively low-productive Bay of Bengal (BoB) (Kumar et al. 2004, 2007). Despite being located at similar latitudes and experiencing similar seasonal forcing, the productivity pattern in the AS and BoB are markedly different. While the former is highly productive during both the monsoons, the productivity in the latter is significantly low. High river discharge and more rainfall over the BoB make it highly stratified, restricting the supply of nutrients into the upper layer from the deep and thus limiting the productivity. Localized wind and eddies, however, contribute significantly towards biological productivity in the BoB (Vinayachandran et al. 2004, Kumar et al. 2007). One of the most intriguing aspects is that both AS and BoB have similar organic carbon flux, as shown by the time-averaged sediment trap data (Ittekkot et al. 1991), despite having large differences in column productivity. High vertical fluxes of organic matter lead to high oxygen demand for bacterial remineralization in the subsurface layer. Also, being landlocked in its north by the Eurasian land *Corresponding author. Email: [email protected] © 2013 Taylor & Francis


1098

S. Prakash et al.

mass, the renewal of subsurface water in northern Indian Ocean is sluggish (Sengupta and Naqvi 1984, Sarma 2002), which coupled with the high biological demand for oxygen causes severe depletion of dissolved oxygen (DO) in the subsurface layer (∼100–1000 m; Morrison et al. 1999). The northern Indian Ocean, therefore, hosts one of the most pronounced oxygen minimum zones (OMZs) among all the oceans in the world. Presence of low oxygen water in the subsurface layer affects marine habitat and, thus, a large coastal population that depends on marine fisheries for their food and economy. It is therefore important to understand the dynamics of OMZ. Increasing number of observations of DO, such as those by Argo-oxygen floats, and a comprehensive analysis of such data will be of immense help. Argo floats have been providing high-quality data of temperature and salinity for a decade now. Addition of biosensors, such as for DO, to these floats allows real-time measurement/monitoring of biogeochemical variables and, therefore, is capable of providing new insights into subsurface processes (Gruber et al. 2010). Indian National centre for Ocean Information Services (INCOIS), a regional Argo data centre, is actively involved in international Argo programme and has deployed Argo floats equipped with oxygen sensors in the Indian Ocean. Here, we analyse DO data from five floats deployed in the AS, BoB and the EIO to understand the variability of oxycline depths in the northern Indian Ocean and its governing factors. We also aim, through our analysis, to establish a relationship which would enable us to get information on depth of oxygen-deficient water in the northern Indian Ocean basin using satellite-derived parameter such as sea level anomaly (SLA). 2. Methods The DO data used in this study are obtained from five Argo floats equipped with oxygen sensors, two floats from the AS (World Meteorological Organisation identification number or WMO IDs: 5903586 and 2900776; here, we name them as floats in the northern AS and central AS, respectively) and BoB (WMO IDs: 2901073 and 2901075; named as western and eastern BoB, respectively) and one float from EIO (WMO ID: 2901082). The trajectories of the floats are shown in figure 1. Argo data were obtained from INCOIS Regional Argo Data Centre (http://www.incois.gov.in/ Incois/argo/argo_Regional_Centre.jsp). The floats were equipped with Aanderaa oxygen optode 3830 sensor along with Seabird conductivity, temperature and depth (CTD) sensors for traditional temperature, salinity and depth measurements. The float numbers 2900776 and 2901082 drift at a parking depth of 2000 m and record profiles of the above-listed parameters at an interval of 10 days, whereas the float numbers 2901073, 2901075 and 5903586 drift at 1000 m and record profiles of the samples at an interval of 5 days. Altogether, 590 profiles (108, 56, 218, 120 and 144 profiles from floats 2900776, 5903586, 2901073, 2901075 and 2901082, respectively) were analysed. The temperature and salinity data are quality controlled based on tests prescribed by Argo data management team (ADMT). Lack of any such prescribed test, however, limits the quality control of Argo-oxygen data. Nevertheless, we have ensured the quality of DO data by visually checking for its correctness and removing bad values appearing as spikes following Martz et al. (2008). Temperature data from respective floats were used to estimate the depth of thermocline: we have used depth of 23◦ C (D23) isotherm as an indicator of thermocline for the AS and BoB following Girish Kumar et al. (2011) and 20◦ C isotherm (D20) for the EIO. There is no universal

Estimating oxycline in Indian Ocean

1099

25

20 INDIA Latitude (°N)

15

10

5


0

50

55

60

65

70

75

80

85

90

95

Longitude (°E)

Figure 1. Location of Argo floats in the northern Indian Ocean: purple, red, green, blue and black dots show the float trajectories in the northern Arabian Sea (float ID: 5903586), southern Arabian Sea (float ID: 2900776), western Bay of Bengal (float ID: 2901073), eastern Bay of Bengal (float ID: 2901075) and equatorial Indian Ocean (float ID: 2901082), respectively.

definition of the oxycline, which is defined as maximum gradient of DO in the subsurface layer. Since the DO concentration in the subsurface water is much less in AS and BoB compared to that in EIO, its gradient is also sharper in the former than in the latter. Hence, we defined the depth of 50 µmol·kg−1 as an indicator of the oxycline depth for AS and BoB, while depth of 125 µmol·kg−1 is taken as an indicator of the same for EIO. We also analysed daily SLA, having spatial resolution of 1/3◦ , from multi-satellite sensors (ERS and JASON) provided by Aviso (http://www.aviso.oceanobs.com/en/) and extracted SLA along the float trajectory by selecting the value of the pixel which corresponded to the float location. 3. Results and discussion Argo-oxygen data from the AS and BoB show presence of a well developed and seasonally stable OMZ in subsurface layer. The EIO also holds low-oxygen water in its subsurface layer but DO concentration is much higher when compared to AS and BoB; DO concentration at a depth of 200 m in EIO is ∼75 µmol·kg−1 , whereas in AS and BoB it is less than 10 µmol·kg−1 (data from present study). Oxycline depth (marked by depth of 50 µmol·kg−1 isoline) in AS and BoB, two bio-geochemically dynamic basins, shows considerable seasonal oscillations. It varies between 60 and 120 m in AS and BoB (Sardessai et al. 2007, Prakash et al. 2012). In the EIO, the oxycline (marked by depth of 125 µmol·kg−1 isoline) is relatively deeper and oscillates between 80 and 140 m. Sardessai et al. (2007) attributed observed variabilities in oxygen concentration in BoB to physical processes such as eddies which are known to play a major role in biogeochemistry in the bay (Kumar et al. 2007). Recent observation shows that high river discharge during the summer monsoon also causes

1100

S. Prakash et al.

Depth (m) Depth (m)

0 25 50 75 100 125 150 175

Depth (m)

Depth (m)

0 25 50 75 100 125 150 175

Depth (m)


intensification of oxygen minima in coastal BoB (Sarma et al. 2013). Similarly, the AS oxygen minimum zone also shows strong seasonality in DO concentration and depth of oxycline (de Souza et al. 1996, Prakash et al. 2012, Resplandy et al. 2012). Model study shows that seasonality in oxygen concentration in AS oxygen minimum zone is primarily caused by the physical processes such as Ekman pumping (Resplandy et al. 2012). Prakash et al. (2012), using Argo-oxygen data, have also shown that westward propagating upwelling Rossby waves cause shallowing of thermocline and hence the oxycline during the early winter monsoon in the Arabian Sea. In order to understand the processes that could explain the observed seasonality in oxycline depth, we analysed the time series of thermocline, oxycline and SLA along the float trajectory. We found strong co-variability of oxycline and thermocline for all the floats analysed here, shown in figure 2. Similar correlations were not observed between

(a) D23 Oxycline

(b)

D23 Oxycline

0 (c) 25 D23 50 Oxycline 75 100 125 150 175 0 25 (d) 50 D23 75 Oxycline 100 125 150 175 0 (e) 25 D20 50 Oxycline 75 100 125 150 175 1 Jan 07 1 Jul 07 1 Jan 08 1 Jul 08 1 Jan 09 1 Jul 09 1 Jan 10 1 Jul 10 1 Jan 11 1 Jul 11 1 Jan 12 1 Jul 12 1 Jan 13 Date

Figure 2. Time series plots of D23 (D20 for equatorial Indian Ocean as a proxy for thermocline; shown as bold lines) and oxycline (shown as dashed lines) for all the floats. Plates (a), (b), (c), (d) and (e) represent central AS, northern AS, western BoB, eastern BoB and EIO, respectively. Correlation coefficient (r) between D23 (D20 for EIO) and oxycline for plates (a), (b), (c), (d) and (e) are 0.92, 0.85, 0.98, 0.91 and 0.78, respectively, and is significant at 95% confidence level.


1101

oxycline and pycnocline (except in BoB) and is evident from figure 3 which shows covariability of thermocline, pycnocline and oxycline in the central Arabian Sea, western BoB and EIO. Moreover, SLA has positive correlation with the depth of thermocline, i.e. positive (negative) SLA indicates deeper (shallower) thermocline (Yu 2003, Schott et al. 2009, Girishkumar et al. 2011). This relationship has been observed in various parts of the world ocean, for example, Girish Kumar et al. (2011) reported correlation coefficients of 0.86 and 0.75 between D23 and SLA for the southern BoB and EIO (1.5◦ S, 90◦ E), respectively. Based on the strong correlation, Girish Kumar et al. (2011) argued that SLA can be used as a proxy for analysing thermocline variabilities.

°C

0

32 30 28

50

Depth (m)

24 22 20

150

18 16

200

14

(a)

12 10

250 A M J J A S O N D J F M A M J J A S O N D J F M A M J J A S O N D J F M 2007 2008 2009 2010

°C

0

32 30 28

50

Depth (m)

26 24

100

22 20 150

18 16

200

14 12

(b)

10

250 N D J F M A M J J A S O N D J F M A M J J A S O N D J F M A M J J A S O N D 2008 2009 2010 0

°C

32 30 28

50

26 Depth (m)


26 100

24

100

22 20

150

18 16

200

14

(c)

12 10

250 F MAM J J A S O N D J F MAM J J A S O N D J F MAM J J A S O N D J F MAM J J A S O N D J 2008 2009 2010 2011 Date

Figure 3. Time series plots of thermocline (red line), pycnocline (blue line) and oxycline (black line) in the central AS (a), western BoB (b) and EIO (c). The background is depth-time section of temperature from each float. Colour bar on the right is temperature in ◦ C.


1102

S. Prakash et al.

Since information on depth of thermocline can be extracted from SLA, we selected thermocline over pycnocline for our analysis. We found that correlation coefficients, r (Pearson’s product moment correlation coefficient; significant at 95% confidence level), between D23 and oxycline in the northern and central AS are 0.85 and 0.92, respectively. For the two floats that we analysed in the BoB, the correlation coefficients were significantly higher, being 0.98 and 0.91 in the western and eastern BoB, respectively. Though the correlation coefficient between D20 and oxycline was also strong in the EIO (r = 0.78), it was less than their northern counterparts. Since there is a strong co-variability between thermocline and oxycline, and in light with the argument proposed by Girishkumar et al. (2011), SLA can also be used as a proxy for studying oxycline variabilities. We, then, explored the relationship between SLA and oxycline, which would enable us to estimate the latter using the former. We found a strong correlation between SLA and oxycline depth in the BoB and AS. The scatter plots between SLA and oxycline depths, along with correlation co-efficient and slope of regression lines, are shown in figure 4. We here propose that the depth of oxycline can be estimated using the following displayed equations: Depth = (2.06 ± 0.11) × (SLA) + (95.88 ± 0.98)

(1)

Depth = (1.68 ± 0.07) × (SLA) + (74.0 ± 0.75)

(2)

and

in the AS and BoB, respectively. Correlation coefficients (r) and root mean square errors (RMSE) for equations (1) and (2) are 0.83 & 0.79 and 11.55 & 12.56, respectively. The EIO correlation between SLA and oxycline is weaker because of significant intra-seasonal variations in the wind over this zone. These large variations in wind, coupled with large-scale planetary waves and its influence on thermocline variabilities (Girishkumar et al. 2011), probably cause a weaker correlation between SLA and oxycline in this region. Nevertheless, the depth of oxycline can still be estimated in EIO from SLA using the displayed equation Depth = (1.25 ± 0.25) × (SLA) + (100.0 ± 1.43)

(3)

(r = 0.40; RMSE = 13.15) with reasonable accuracy. The given relationship explains ∼65 % (coefficient of determination R2 = 0.69 and 0.63 for AS and BoB, respectively) of the total variance in oxycline depths for these regions. It is also evident from the regression equations (1) and (2) that depth of oxycline is deeper in the AS compared to BoB (∼97 m in AS and ∼73 m in BoB), mainly due to strong stratification in the latter. Moreover, since SLA is a satellite-derived parameter, the given equations enable us to determine, through oxycline depth, the depth at which the subsurface water is oxygen deficient and, therefore, help us to understand the dynamics of oxycline variabilities in the northern Indian Ocean basin. It is evident from the observed high correlations between oxycline and thermocline that the variability in oxycline depth is governed by the variability in the thermocline depth, i.e. the shallowing of the thermocline is associated with the corresponding shallowing in the oxycline and vice versa. Though it is fairly well established that biological consumption of oxygen, coupled with sluggish renewal of subsurface waters, leads to formation/maintenance of the OMZ in the northern Indian Ocean (e.g. Naqvi et al.

Estimating oxycline in Indian Ocean 160 140

1103

Depth = (2.06 ± 0.11) × (SLA) + (95.88 ± 0.98)

(a)

R2 = 0.69 RMSE = 11.55

Oxycline (m)

120 100 80 60 40

SLA vs oxycline Linear regression

–20

–10

160

0 SLA (cm)

10

20

Depth = (1.68 ± 0.07) × (SLA) + (74.0 ± 0.75)

30 (b)

140 R2 = 0.63 RMSE = 12.56

Oxycline (m)

120 100 80 60 40


20 –30

–20

–10

0 SLA (cm)

10

20

30

160 140

Depth = (1.25 ± 0.25) × (SLA) + (100 ± 1.43) R2 = 0.16 RMSE = 13.15

(c)

120 Oxycline (m)


20 –30

100 80 60 40 20 –10


–5

0

5 SLA (cm)

10

15

20

Figure 4. Scatter plot of oxycline depth and sea level anomaly. Plates (a), (b) and (c) represent Arabian Sea, Bay of Bengal and Equatorial Indian Ocean, respectively. R2 is coefficient of determination of the linear regression and RMSE is root mean square error.


1104

S. Prakash et al.

2006), the present analysis suggests that the vertical movement of the oxygen-deficient water is mainly governed by physical processes such as vertical movement of the thermocline. Biology, however, may be playing a dominant role in determining the absolute concentration of DO in the subsurface waters or defining the strength/intensity of the OMZ. We would not have observed such coherence between oxycline and thermocline depths if biology was responsible in modulating the oxygen gradient, at least for the upper oxycline. Various earlier researchers have also reported a dominance of physics over biology in generating/maintaining OMZ in the northern Indian Ocean, e.g. largescale planetary waves such as Rossby waves and coastal Kelvin waves cause oscillation in the thermocline (Girishkumar et al. 2011) and hence in oxycline depths as well (Prakash et al. 2012). Sardessai et al. (2007) attributed seasonality observed in OMZ of BoB to the processes such as eddies and McCreary et al. (2013) have shown, for the AS, that oceanic circulation plays a dominant role in defining the spatial distribution of OMZ. Similar co-variabilities of thermocline and oxycline have also been reported from the OMZ off Chile (Morales et al. 1999), but no relationship was proposed which could be used to estimate oxycline depth. The present study, therefore, opens new avenues of research and is of significance for operational oceanography. Presence of oxygen-deficient water has profound negative effects on marine organisms, and it severely alters their migration pattern (Naqvi et al. 2006, Seibel 2011). For example, zooplankton biomass responds adversely to vertical gradients of oxygen (Madhupratap et al. 1990). The same holds true for the fisheries, as they tend to migrate away from the oxygen-stressed conditions, evident from a significant decrease in fish landing at a coastal site in Goa, India (Naqvi et al. 2006). Therefore, information about the depth of oxycline is critical for delineating potential fishing zones. Clubbing information on depth of oxycline with satellite chlorophyll/wind data for issuing advisories on potential fishing zone will certainly be beneficial for the fishing community, as it will not only be cost-effective but will also help in proper human resource utilization. 4. Conclusion The present study of Argo-oxygen data provides new insights into the dynamics of OMZ in the northern Indian Ocean. Our analysis confirms that physical factors such as SLA, through its manifestation of thermocline, control the vertical movement of oxygen-deficient water in this part of the world ocean. We also show that oxycline variabilities are strongly correlated with the thermocline which itself is strongly correlated with the SLA. Variabilities in SLA explain up to 65% of the variability in the oxycline depth. Regression equations derived from scatter plots of SLA and oxycline may be used to estimate the depth of oxycline in the northern Indian Ocean basin. Acknowledgement The encouragement and facilities provided by the Director, INCOIS, are gratefully acknowledged. We thank International Argo Programme for collecting the data and making it freely available for researchers. We also thank various organizations who have contributed significantly towards collection of data and making it available for users. We thank two anonymous reviewers for their critical comments. This is INCOIS publication no. 154.


1105


References DE SOUSA, S.N., KUMAR, M.D., SARDESSAI, S., SARMA, V.V.S.S. and SHIRODKAR, P.V., 1996, Seasonal variability in oxygen and nutrients in the central and eastern Arabian Sea. Current Science, 71, pp. 847–851. GANDHI, N., RAMESH, R., PRAKASH, S., NORONHA, S. and ANILKUMAR, N., 2012, Primary and new production in the thermocline ridge region of the southern Indian Ocean during the summer monsoon. Journal of Marine Research, 70, 779–793. GIRISHKUMAR, M.S., RAVICHANDRAN, M., MCPHADEN, M.J. and RAO, R.R., 2011, Intraseasonal variability in barrier layer thickness in the south central Bay of Bengal. Journal of Geophysical Research. doi:10.1029/2010JC006657. GRUBER, N., DONEY, S.C., EMERSON, S.R., GILBERT, D., KOBAYASHI, T., KÖRTZINGER, A., JOHNSON, G.C., JOHNSON, K.S., RISER, S.C. and ULLOA, O., 2010, Adding oxygen to Argo: developing a global in-situ observatory for ocean deoxygenation and biogeochemistry. In Proceedings of OceanObs’09: Sustained Ocean Observations and Information for Society (Vol. 2), J. Hall, D.E. Harrison and D. Stammer (Eds.), 21–25 September 2009 (Venice: ESA Publication WPP-306). ITTEKKOT, V., NAIR, R.R., HONJO, S., RAMASWAMY, V., BARTSCH, M., MANGANINI, S. and DESAI, B.N., 1991, Enhanced particle fluxes in the Bay of Bengal induced by injection of freshwater. Nature, 351, pp. 385–387. KUMAR, S.P., NUNCIO, M., NARVEKAR, J., KUMAR, A., SARDESAI, S., DESOUZA, S.N., GAUNS, M., RAMAIAH, N. and MADHUPRATAP, M., 2004, Are Eddies nature’s trigger to enhance biological productivity in the Bay of Bengal? Geophysical Research Letters, 31, L07309. doi:10.1029/2003Gl019274. KUMAR, S.P., NUNCIO, M., RAMAIAH, N., SARDESAI, S., NARVEKAR, J., FERNANDES, V. and PAUL, J.T., 2007, Eddy-mediated biological productivity in the Bay of Bengal during fall and spring intermonsoons. Deep Sea Research I, 54, pp. 1619–1640. MADHUPRATAP, M., NAIR, S.R.S., HARIDAS, P. and PADMAVATI, G., 1990, Response of zooplankton to physical changes in the environment: coastal upwelling along central west coast of India. Journal of Coastal Research, 6, pp. 413–426. MARTZ, T.R., JOHNSON, K.S. and RISER, S.C., 2008, Ocean metabolism observed with oxygen sensor on profiling floats in the South Pacific. Limnology and Oceanography, 53, pp. 2094–2111. MCCREARY, J.P., YU, Z., HOOD, R., VINAYCHANDRAN, P.N., FURUE, R., ISHIDA, A. and RICHARDS, K., 2013, Dynamics of the Indian-Ocean oxygen minimum zones. Progress in Oceanography. doi:10.1016/j.pocean.2013.03.002. MORALES, C.E., HORMAZÁBAL, S.E. and BLANCO, J.L., 1999, Interannual variability in the mesoscale distribution of the depth of the upper boundary of the oxygen minimum layer off northern Chile (18-24S): implications for the pelagic system and biogeochemical cycling. Journal of Marine Research, 57, pp. 909–932. MORRISON, J.M., CODISPOTI, L.A., SMITH, S.L., WISHNER, K., FLAGG, C., GARDNER, W.D., GAURIN, S., NAQVI, S.W.A., MANGHNANI, V., PROSPERIE, L. and GUNDERSEN, J.S., 1999, The oxygen minimum zone in the Arabian Sea during 1995. Deep Sea Research, Part II, 46, pp. 1903–1932. NAQVI, S.W.A., NARVEKAR, P.V. and DESA, E., 2006, Coastal biogeochemical processes in the northern Indian Ocean (14,S-W). In The Sea; : Ideas and Observations on Progress in the Study of the Seas. Vol. 14: Interdisciplinary Regional Studies and Syntheses. Part A, A.R. Robinson and K.H. Brink (Eds.), pp. 723–781 (Cambridge, MA: Harvard University Press). PRAKASH, S., BALAKRISHNAN NAIR, T.M., UDAYA BHASKAR, T.V.S. and PRAKASH, P., 2012, Oxycline variability in the central Arabian Sea: an Argo-oxygen study. Journal of Sea Research. doi:10.1016/j.seares.2012.03.003.


1106

S. Prakash et al.

PRAKASH, S., RAMESH, R., SHESHSHAYEE, M.S., DWIVEDI, R.M. and RAMAN, M., 2008, High new production during a Noctiluca scintillans bloom in winter in the northeastern Arabian Sea. Geophysical Research Letters, 35. doi:10.1029/2008GL033819. RESPLANDY, L., LEVY, M., BOPP, L., ECHEVIN, V., POUS, S., SARMA, V.V.S.S. and KUMAR, D., 2012, Controlling factors of the oxygen balance in the Arabian Sea’s OMZ. Biogeosciences, 9, pp. 5095–5109. doi:10.5194/bg-9-5095-2012. SARDESSAI, S., RAMAIAH, N., PRASANNAKUMAR, S. and DESOUZA, S.N., 2007, Influence of environmental forcings on the seasonality of dissolved oxygen and nutrients in the Bay of Bengal. Journal of Marine Research, 65, pp. 301–316. SARMA, V.V.S.S., 2002, An evaluation of physical and biogeochemical processes regulating the oxygen minimum zone in the water column of the Bay of Bengal. Global Biogeochemical Cycles, 16, p. 1099. doi:10.1029/2002GB001920. SARMA, V.V.S.S., KRISHNA, M.S., VISWANADHAM, R., RAO, G.D., RAO, V.D., SRIDEVI, B., KUMAR, B.S.K., PRASAD, V.R., SUBBAIAH, Ch.V., ACHARYYA, T. and BANDOPADHYAY, D., 2013, Intensified oxygen minimum zone on the western shelf of Bay of Bengal during summer monsoon: influence of river discharge. Journal of Oceanography, 69. doi:10.1007/s10872-012-0156-2. SCHOTT, F., SHANG-PING XIE, A. and JULIAN P. MCCREARY JR., 2009, Indian Ocean circulation and climate variability. Reviews of Geophysics, 47, RG1002. doi:10.1029/2007RG000245. SEIBEL, B.A., 2011, Critical oxygen levels and metabolic suppression in oceanic oxygen minimum zones. Journal of Experimental Biology, 214, pp. 326–336. SENGUPTA, R. and NAQVI, S.W.A., 1984, Chemical oceanography of the Indian Ocean, north of the equator. Deep Sea Research, 31, pp. 671–706. VINAYCHANDRAN, P.N., CHAUHAN, P., MOHAN, M. and NAYAK, S., 2004, Biological response of the sea around Srilanka to summer Monsoon. Geophysical Research Letters, 31. doi:10.1029/2003GL018533. WIGGERT, J.D., HOOD, R.R., BANSE, K. and KINDLE, J.C., 2005, Monsoon driven biogeochemical processes in the Arabian Sea. Progress in Oceanography, 65, pp. 176–213. YU, L., 2003, Variability of the depth of the 20◦ C isotherm along 6◦ N in the Bay of Bengal: its response to remote and local forcing and its relation to satellite SSH variability. Deep Sea Research, Part II, 50, pp. 2285–2304. doi:10.1016/S0967-0645(03)00057-2.