Intraseasonal variability of currents along west coast of India A Thesis submitted to Goa University for the Award of the Degree of DOCTOR OF PHILOSOPHY in Marine Sciences By Amol Prakash Research Guide Shankar Doraiswamy Goa University Taleigao, Goa 2014 To amma and acha Statement As required under the University ordinance OB 9.9(vi), I state that this thesis entitled IntraseasonalvariabilityofcurrentsalongwestcoastofIndiaismyoriginalcontribution andithasnotbeensubmittedonanypreviousoccasion. The literature related to the problem investigated has been cited. Due acknowledge- mentshavebeenmadewhereverfacilitiesandsuggestionshavebeenavailedof. AMOLPRAKASH CSIR-NationalInstituteofOceanography,Goa 15December2014 Certificate ThisistocertifythatthethesisentitledIntraseasonalvariabilityofcurrentsalongwest coast of India, submitted by Amol Prakash to Goa University for the degree of Doctor of Philosophy, is based on his original studies carried out under my supervision. The thesis or any part thereof has not been previously submitted for any other degree or diploma in anyuniversityorinstitution. SHANKARDORAISWAMY CSIR-NationalInstituteofOceanography,Goa 15December2014 v Acknowledgements “Nomanisanisland,entireofitself.” JohnDonne I owe my profound gratitude to my guru, mentor, and motivator Dr. Shankar Do- raiswamy for guiding me through the journey down the path of PhD. Dr. Shankar has been extraordinarily tolerant and supportive throughout my PhD. He is a person who thrives flawlessness and sets high standards to bring the best out of his students. I had a wonderful time working with him. The knowledge and wisdom he shared with me are invaluableandwouldcontinuetobeapartofmylife. TheroadtomyPhDbeganwithDr. SatheeshShenoi,whowasmysupervisorformore than a year, and I am thankful to him for his support and guidance. Discussions with Dr. Satish Shetye were enlightening and thought-provoking, and his meticulous comments provided the crucial breakthrough for our first paper. The coastal-trapped-wave theory presented in this thesis was instigated by the critical comments of Prof. Barbara Hickey and an anonymous reviewer; inclusion of this analysis has considerably improved this thesis, and I am grateful to the reviewers. I am also thankful to Prof. Kenneth Brink and thelateDavidChapmanformakingtheircoastal-trapped-wavecodespublic. Prof. Brink, inparticular,hadtakenouthisprecioustimetohelpmeunderstandthecodes. Prof. Allan Clarke also helped me get these codes from Prof. Brink. I am in debt to Prof. Julian Mc- Crearyforhelpingmefillmyintellectualvoid. IalsothankDr. Vialardforgivingmetips on statistics. Dr. Aparna was more like a friend to me and I have greatly benefited from herconstructivecommentsandwarmencouragements. Mr. Michaelalwaysansweredmy questions, especially related to Generic Mapping Tools, with great patience. Interactions withMr. I.Suresh,Mr. SundarandMr. Ashokwereuseful. IwouldliketothanktheDirector,CSIR-NationalInstituteofOceanographyandVice Chancellor (V.C.), Goa University for providing me with all the necessary infrastruc- ture and facilities for research. I cannot express enough thanks to my Faculty Research Committee for their continued support and assistance towards my thesis. I appreciate the feedback offered by my V.C.’s nominee Dr. A. S. Unnikrishnan. He always kept me on my toes, while I presented my results. I also owe my deepest gratitude to Prof. H. B. Menon, who agreed to be my co-guide. I would like to offer my special thanks to Prof. G.N.Nayak,Prof. R.V.Pai,Prof. J.A.E.Desa,andMr. Yashwantwhohelpedme throughtheadministrationglitchesintheUniversity. TheADCPdataformyresearchworkwasmadeavailableowingtoanincredibleeffort from the members of the mooring group. I would like to show my greatest appreciation vi to Mr. Fernando and his mooring team: Yogesh, Khalap, Narayan, Prakash, Mithun, and Siddharth. I also thank the officers, crew, and seamen on the mooring cruises for providingthenecessarysupportfordeployingtheADCPs. Financialsupportforthiswork was provided by INCOIS (Ministry of Earth Sciences) and other CSIR-aided projects. I thankCSIRagainforprovidingmemyfellowship. I am fortunate to be in the world of internet where accessing information is just a click away. My special thanks to the open-source community for providing me with free softwares and tools for my research. I also thank Opencourseware, universities, websites and forums who have provided high quality educational materials for free. It would be surprising to know that in spite of all the information available on the web, some works are only found in the racks of a library. I would like to thank our librarians Mr. Prabhu, Mr. Mithun, and Mr. Gauns who helped me dust the old atlases and journals (in addition tootherreferencebooks)outoftheshelf. Thiswascrucialforframingthesecondchapter ofthisthesis. IamparticularlygratefulfortheassistancegivenbytheadministrativestaffofCSIR- NIO and Goa University. I would not like to miss out on our information technology groupfortheservicestheyhaveprovided. I would like to thank my seniors Dr. Suprit, Mr. Nagesh, Dr. Syam, Dr. Manoj, Dr. Sindhu, and Ms. Thejna who took me under their wings and made me comfortable in a new lab environment. They guided me through my PhD, introduced me to their fellow- mates,andhelpedmelearnnewtoolsandtechniques. DiscussionswithmyfriendsVijith, Abhisek, Arnab, andKankan have alwaysbeen illuminating. Ms. Vidya aidedme during my final stages of my thesis writing. Ronnie and Amol helped me with the processing of ADCP data. Special thanks for the behind-the-scene support from our system adminis- trators Krupesh, Ashok, Sarvesh, Mahesh, and Eric. My deepest appreciation goes to my following friends for their continued support and encouragement: Akhil, Anil, Aravind, Betty, Byju, Charls, Chinnu, Damodar, Divya, Jineesh, Keerthi, Mahesh, Manu, Mithun, Nidheesh,Nanddeep,Praveen,Rakesh,Ravi,Remya,Renjith,Renosh,Rosaline,Saheed, Shynu,Sivakumar,Soumya,Subeesh,Sunita,Veera,Vineet,andVipin. IwouldliketopaymyhighestregardstomyAmmaandAcha. Withouttheirgoodwill and blessings, this thesis would not have materialized. The thesis is dedicated to them. Last, but not the least, my heartfelt gratitude to my sister Amiti for her encouragement andmoralsupport. vii Synopsis Ocean currents transport both mass and energy from one location to another around theworld. Thistransportmaybeinformofheat,salt,nutrients,larvae,debrisoroilspills. The large movement of heat and salt makes currents one of the most important driving forces of climate. This circulation not only stabilizes the global atmospheric circulation, butalsoregulatesthelocalweather,temperatureextremaandair-seagasexchanges. Cur- rentsalsoplayanimportantroleinthedistributionofmarinelifearoundtheworld. They bring cold nutrient-rich water from deeper depths to the surface by a process known as upwelling. Such an ecosystem can support marine fish catches that are nearly 100 times theglobalmean. Today,oceancurrentsarethoughttobethefutureforalternativeenergy. As water is almost 800 times denser than air, ocean currents can generate extremely high hydrokinetic power even at very low flow rates. Whether the application is climate, ma- rinebiodiversityorcleanenergysource,knowledgeofcurrentshasbeenimportanttothe earthscientists,andtounderstandthemonehastostartbyobservingtheoceans. Motivation JustasthecirculationintheNorthIndianOcean(NIO)reversesseasonallyinresponseto the reversing monsoon winds (see, for example, the atlas of Wyrtki [1971], or the review of Schott and McCreary [2001]), so does the current off the Indian west coast (see, for example, the review by Shetye and Gouveia [1998]), which has been called the West In- dia Coastal Current (WICC) [Shankar and Shetye, 1997]. Most of our earlier knowledge of the WICC was based on hydrographic data [Shetye et al., 1990, 1991b], ship drifts [Shetye et al., 1994], and surface drifters [Shenoi et al., 1999], which could describe, at best, the seasonal cycle of the current. The advent of satellite altimetry presented a more comprehensive basin-wide view of the circulation and led to the discovery of the LakshadweepHighandLowinthesoutheasternArabianSea(SEAS)[Bruceetal.,1994; Shankar and Shetye, 1997]; the high and low in sea level, once identified in altimeter data, were apparent in earlier hydrographic data too [Bruce et al., 1994]. A monthly climatology of the WICC based on ship drifts [Mariano et al., 1995] and a weekly cli- matology of its geostrophic component based on altimeter sea-level anomalies (SLAs) show a distinct seasonal cycle of the WICC. The WICC flows equatorward during the Indian summer monsoon (May–September) and poleward during the winter monsoon viii (November–February). Below the surface current, the hydrographic data suggested the presenceofanundercurrent[Shetyeetal.,1990,1991b]. Direct current measurements have been few in the region and have been restricted to shortdurations,typicallyamonthorless;theseshort-durationdataenabledthedescription of the major tidal constituents and high-frequency (period of the order of a few days) wind-drivencurrents[Varkey,1980;Shenoietal.,1988;UnnikrishnanandAntony,1990; Shenoi and Antony, 1991; Antony and Shenoi, 1993; Hareeshkumar and Mohankumar, 1996;DineshkumarandSrinivas, 2007;Shetyeetal.,2008],but thelatterwererestricted toafewisolatedmeasurements. Given this paucity of data that can resolve the variability at time scales other than the seasonal, it is not surprising that most of the theoretical studies of the WICC have been restricted to explaining the seasonal cycle [McCreary et al., 1993; Bruce et al., 1994; ShankarandShetye,1997;Shankaretal.,2002]andtheheatbudget[Durandetal.,2004] of the SEAS and salinity variation [Durand et al., 2007; Kurian and Vinayachandran, 2007] (the SEAS) at seasonal time scales. Hence, in order to map the variability of the currents along the Indian coast over a range of time scales, in particular the variability at time scales shorter than a season, a set of acoustic Doppler current profiler (ADCP) moorings have been deployed off the west coast of India in the last few years. The first description of the currents measured by these ADCPs was by Vialard et al. [2009], who showed that the intraseasonal variability of the WICC on the continental slope off Goa over a two-year period during 2006–2008 was much stronger than the seasonal cycle, which was very weak. In contrast, the altimeter SLAs showed comparable seasonal and intraseasonal variability. The observation was in contradiction to all the earlier evidence from ship drifts to models, which invariably showed a strong seasonal cycle along the westcoastofIndia. Today,wehavedirectcurrentmeasurementsthatspanmorethan3–4 years off the west coast of India. These observations are high-frequency (interval of 15 minutes to 1 hour), provide a vertical profile of currents, and are available both on the slope and shelf. These data allow a hitherto unprecedented analysis of the variability of theWICCoverarangeoffrequencies. Aim of the thesis The study aims at documenting quantitatively the intraseasonal variability (time scale ranging from a few days to a few months) of currents along the west coast of India using theselong-termtime-seriesmeasurements. ix A brief summary of thesis chapters Chapter 1: Introduction We begin the thesis with a short introduction, which starts by commenting on the impor- tance of ocean currents and their role in human society, followed by the motivation and objectiveofthedoctoralthesis. Chapter 2: Historical Background AbriefsurveyoftheliteratureonearlycurrentmeasurementsalongthewestcoastofIndia ispresentedinthischapter. Themeasurementsvaryfromindirectmeasurementslikeship drifts,hydrography,andsatellitealtimetrytodirectmeasurementslikecurrentmetersand ADCPs. Only a few direct current measurements were made off the west coast of India priorto2005andtheyweremostlyofshortduration,rangingfromafewhourstoatmost a fortnight. These observations permitted the analysis of only high-frequency variability like internal waves, tides, and inertial currents. The chapter also sheds light on several theoretical studies that aimed at explaining the observed seasonal cycle of circulation in the NIO, particularly the observed boundary currents off the Indian coasts, the East India CoastalCurrent(EICC)andtheWICC. Chapter 3: Slope WICC In this chapter we present data from four ADCPs deployed on the continental slope off the Indian west coast during October 2008 to October 2012. The observations show that a seasonal cycle, including an annual cycle, is present in the WICC; this seasonal cycle, which strengthens northward, shows considerable interannual variability and is not as stronglycorrelatedalongthecoastasinclimatologiesbasedonshipdriftsorthealtimeter. The existence of a strong seasonal cycle at Goa is contrary to the report of Vialard et al. [2009], who found the seasonal cycle missing in a shorter ADCP record during 2006– 2008. They attributed this absence of the seasonal cycle, which was in contrast to the strong intraseasonal variability in the ADCP current at 50 m during 2006–2008, to the westward radiation of the seasonal Rossby wave; in contrast, there was coastal trapping at intraseasonal periods less than ∼ 90 days. The ADCP data presented here show that the seasonal cycle actually strengthens poleward during 2009–2011, contradicting this hypothesis. Exceptfortheseasonalcycle,whichisweaklycoherentfromKollamtoMumbai,the WICCismostlydecorrelatedalongthecoast. ThealongshoredecorrelationoftheWICC, x which is much stronger at intraseasonal periods, are evident during the winter monsoon allalongthecoast. Themagnitudeoftheintraseasonalcurrentisstrongerthanthatofthe seasonal cycle. A similar intraseasonal peak in the spectrum at 50 m during 2006–2008 wasattributedbyVialardetal.[2009]totheMaden-JulianOscillation(MJO).TheADCP data show that intraseasonal variability is stronger during the winter monsoon, which is consistent with the MJO peak during the boreal winter in the Indian and Western Pacific Oceans[ZhangandDong,2004]. AstrikingfeatureoftheWICCisupwardphasepropagation. Theverticalpropagation of phase, particularly for the seasonal cycle, leads to an undercurrent that shallows as theseasonprogresses. Thislinkbetweentheundercurrentandverticalphasepropagation explainswhyshort-durationobservationscanleadtodifferingdepthsfortheundercurrent. Therearealsoinstanceswhenthephasepropagatesdownward. The ADCP data at the two southern mooring locations off Kollam and Kanyakumari also show fairly strong cross-shore currents, which are dominated by intraseasonal vari- ability. The magnitude of the cross-shore current is often comparable to the alongshore current at these southern locations. Altimeter data show that strong cross-shore flows are associated with eddy-like circulations. Such eddy-like circulations, embedded within the westward propagating Rossby off southwest India, have been noted earlier too [Bruce et al., 1998; Hareeshkumar et al., 2009; Rao et al., 2013]. Such circulations need not necessarily be nonlinear phenomena: at intraseasonal periods of the order of 60 days, the wavelength is short enough to produce such small-scale “eddies” [Shankar and Shetye, 1997]. Such eddy-like circulation are not as strong at Goa and Mumbai owing to the de- crease in the characteristic length scale, the Rossby radius, with increasing distance from the equator. This decrease in the cross-shore length scale leads to far weaker cross-shore flowsatthetwonorthernmoorings. Chapter 4: Shelf WICC I — Observations Unlike on the slope, the current measurements on the shelf were discontinuous. Many of these shallow-water ADCPs were lost due to trawling activities. The analyses show that the shelf currents are relatively stronger compared to the slope currents. The Kollam and Goa moorings, which had continuous data over a year, showed a strong seasonal cycle withlargehigh-frequencyvariability. Liketheslopecurrents,theshelfcurrentswerealso decorrelatedalongthecoast. Wealsocomparedasetofsix-monthshelfandslopemoorings,separatedbyashorter distance of 150–200 km, on the central west coast of India. Unlike the more widely sep- arated, long-term moorings for which the currents were decorrelated in the intraseasonal