Evaluation of OSCAR ocean surface current product in the tropical Indian Ocean using in situ data
Rajesh Sikhakolli1,∗,Rashmi Sharma1,Sujit Basu1,
B S Gohil1,Abhijit Sarkar1and K V S R Prasad2
1Atmospheric and Oceanic Sciences Group,Space Applications Centre(ISRO),Ahmedabad380015,India.
2Department of Meteorology and Oceanography,Andhra University,Visakhapatnam530003,India.
∗Corresponding author.e-mail:srajesh@v.in
The OSCAR(ocean surface current analysis real-time),which is a product derived from various satellite obrvations,has been evaluated in the tropical Indian Ocean(TIO)in two different ways.First,the OSCAR-derived monthly climatology has been compared with available drifter-derived climatology in the TIO.From the comparison of the two climatologies,one can infer that OSCAR product is able to capture the variabilities of the well-known surface current systems in the TIO reasonably well.Fourier analysis of the major current systems,as reproduced by OSCAR,shows that the dominant annual and miannual periodicities,known to exist in the systems,have been faithfully picked up by OSCAR. Ne
xt,the evaluation has been carried out by comparing the OSCAR currents with currents measured by moored buoys.The zonal component of OSCAR-current is in good agreement with corresponding compo-nent of buoy-obrved current with a correlation exceeding0.7,while the match between the meridional components is poorer.The locations of the peaks of the mean and eddy kinetic energies are matching in both the climatologies,although the peak in the drifter climatology is stronger than the same in the OSCAR product.Finally,an important feature of Indian Ocean circulation,namely the rever Wyrtki jet,occurring during anomalous dipole years,has been well-reproduced by OSCAR currents.
1.Introduction
Monitoring the variations of ocean surface currents is extremely important for climate studies as the currents are the major transporters of heat and salt.Surface current in the Indian Ocean displays the unique feature of annual reversal in respon to annually reversing monsoon winds(Shenoi et al. 1999).Traditionally,ocean currents have been measured by ships,moored current meters and drifting buoys.However,the situation in the Indian Ocean is wor than in the Atlantic and the Pacific becau of the relative scarcity of in situ data.Nev-ertheless,Hastenrath and Greischar(1991)stud-ied the asonal variability of surface circulation in the tropical Indian Ocean(TI
O)using long-term obrvations of surface windfield and ship drift measurements.
Later,Shenoi et al.(1999)studied the surface circulation in the TIO by deriving a monthly cli-matology of ocean surface currents from the drift-ing buoy obrvations in the TIO collected from 1976to1998.Lagerloef et al.(1999)ud satellite altimetry and vector wind data to analy surface velocity during1992–1999.In a study of somewhat similar nature,Shankar et al.(2002)studied the ocean surface currents in the Indian Ocean by com-bining geostrophic current from a surface height (SSH)measurements and Ekman current from a surface wind(SSW)field.However,neither
Keywords.Ocean surface current;remote nsing;buoy currents;drifter data;Indian Ocean circulation.
J.Earth Syst.Sci.122,No.1,February2013,pp.187–199
c Indian Academy of Sciences187
188Rajesh Sikhakolli et al.
Lagerloef et al.(1999)nor Shankar et al.(2002) could correctly reproduce the equatorial current features.
The problem of estimating satellite-derived near-surface velocity in the equatorial area was solved by Bonjean and Lagerloef(2002).They ud three surface variables,namely SSH,SSW and a sur-face temperature(SST),all available from satellite measurements,to derive the surface current prod-uct,known as ocean surface current analysis real-time(OSCAR).The method of deriving the surface current is bad on the resolution of quasi-steady quasi-linear momentum equations,neglecting local acceleration.Equatorial velocities are obtained by solving a weak formulation of the momentum equa-tions using a basis t of orthogonal polynomi-als.This formulation of Bonjean and Lagerloef (2002)differs from the one ud in Lagerloef et al. (1999)in many ways.A simplified buoyancy force, solely a function of SST,is retained in the ver-tical hydrostatic balance.This force is abnt in the earlier version(Lagerloef et al.1999).In this way,the SST enters the computation of verti-cally averaged velocity.Apart from this,the net drag force required in this computation us eddy viscosity formulation,whereas in the earlier ver-sion,a Rayleigh friction was ud.Moreover,the velocity shear is depth-dependent,which implies a higher-order turbulence closure parameterization than the earlier version.All the factors,in a combined fashion,have contributed to the superior performance of the prent algorithm for generat-ing ocean surface current.
Validation and error analysis of the OSCAR pro-duct was carried out in the tropical Pacific Ocean (Jo
hnson et al.2007).The OSCAR product is, however,a global product.Thus there is a pressing need to validate this product in the other basins of the world ocean, e.g.,in the Indian Ocean. The prent study is motivated by this need.In the prent study,monthly climatology of OSCAR ocean surface currents in the TIO has been gen-erated and this has been compared with climatol-ogy of surface currents bad on drifter data.The OSCAR product has also been validated by com-paring it with available moored current meter data in the TIO.
Since the prent paper is basically about surface currents from various data sources and their inter-comparison,it would be beneficial for the read-ers to have a look at the major surface currents in the Indian Ocean(Shenoi et al.1999;Schott and McCreary2001;Shankar et al.2002).Accord-ingly,infigure1,we provide such a schematic.Fig-ure1(a)shows the major currents during winter, while the major currents during summer are shown infigure1(b).In winter,the surface current pattern is similar to the pattern prevailing in the two other major ocean basins,namely the Pacific and the
Atlantic.The South Equatorial Current(SEC),the
Equatorial Current(EC),the Equatorial Counter
Current(ECC)and the Northeast Monsoon Cur-
rent(NMC)are shown infigure1(a).Associ-
ated boundary currents such as the East African
Coastal Current(EACC),the East India Coastal
Current(EICC)and the West India Coastal Cur-
rent(WICC)have also been depicted.During sum-
mer,the westward NMC is replaced by eastward
轻松搞笑小说flowing Southwest Monsoon Current(SMC).The
EICC and WICC directions are reverd,while
the Somali Current(SC)startsflowing northward,
instead of southward.There is also another major
current system in the Indian Ocean,which is the
strong eastwardflowing jet known as the Wyrtki
Jet(WJ),which appears during the transition peri-
ods,April–May and October–November(Wyrtki
1973).Sincefigure1is a description of major sum-
mer and winter currents,the WJ has not been
中国感恩节shown infigure1.However,the reprentation of
WJ in OSCAR product has been described at an
appropriate place.Infigure1,we also show the
locations of the RAMA moored buoys ud in this
study by asterisk symbols.
2.Data
2.1OSCAR current data
The OSCAR data ud in the prent study span
the period from1993to2008.The data are avail-
able at5-day intervals and at1◦×1◦spatial resolution for the TIO and are obtained from
aa.gov.The datats are mainly built
from the satellite data that have been collected
since October1992during ongoing missions.SSH
电影奖项有哪些结果咧is gridded from TOPEX/Poidon data for the
雨后的诗句period October1992–June2002and from Jason-1
for the period July2002to the prent(Johnson
et al.2007).SSW data are provided by the varia-
tionally analyd Special Sensor Microwave Imager
(SSM/I)winds(Atlas et al.1996)from October
1991to September2001and the QuikSCAT grid-
ded winds from the Center for Ocean Atmosphere
Prediction Studies(August2001till prent).The
SST data ud in generating the OSCAR cur-
rents are Reynolds and Smith’s version2SST,a
product blended from the satellite and in situ data
(Reynolds et al.2002).
2.2Drifter buoy data
Under the auspices of the drifting buoy pro-
gramme,the National Institute of Oceanography
(NIO),Goa,is maintaining and archiving the
satellite-tracked drifting buoy data for the TIO.
Evaluation of OSCAR currents in the Indian Ocean
189
Figure1.Schematic of major surface currents in the Indian Ocean during(a)winter and(b)summer.The asterisks show the moored buoy locations ud for comparison.
Recently,a monthly climatology of ocean surface
currents has been developed from this data on a
2◦×2◦grid from the drifting buoy obrvations in the TIO using data collected from1985to2006
(www.nio).Shenoi et al.(1999)have described
the methodology ud to develop this climatology.
In the prent study,this drifter climatology has
been ud to validate the important large-scale fea-
tures of the TIO circulation that are reflected in
the OSCAR product.
2.3Moored current meter data
A key element in the Indian Ocean Obrving System(IndOOS)is the basin-scale array of moored buoys,known as,rearch moored array for African-Asian-Australian monsoon analysis and prediction(RAMA).It is designed speci-fically to study large-scale ocean–atmosphere interaction,mixed-layer dynamics and ocean cir-culation related to the monsoons on intraasonal
190Rajesh Sikhakolli et al.
to decadal time-scales.The details of RAMA buoy network and their implementation plan and scien-tific objectives have been prented recently in a paper by McPhaden et al.(2009).In the prent study,the time ries data of ocean currents from buoys of this network in the Indian Ocean are ud.
3.Results
3.1Comparison with climatology
Since there are very few moored buoys with current measurements in the TIO,a basin scale
comparison
县治Figure 2.Monthly means of ocean surface current (ms −1)from OSCAR (left panel)and drifter data (right panel)for 4chon months.Colour shading shows the magnitudes while the arrows (of fixed length)show the directions.
Evaluation of OSCAR currents in the Indian Ocean
191
of OSCAR product with in situ data is not really feasible.Hence,we took the alternative route of comparing the OSCAR-derived monthly climatol-ogy with drifter-derived climatology.Accordingly,a new monthly climatology was derived in the TIO from 16years (1993–2008)of OSCAR cur-rents.In figure 2,we show the monthly clima-tologies of surface currents from the OSCAR and drifter data.However,we show only four repren-tative months instead of all the 12months,becau in the four months,the major current systems are well-reflected in both the climatologies.From figure 2,it is apparent that the OSCAR climatol-ogy is able to capture the large-scale ocean circu-lation features in the TIO reasonably well.Thus,during the northeast monsoon ason (reprented here by the typical month of February),the charac-teristic circulation features such as the westward-flowing WMC north of equator as well as the poleward-flowing WICC are well-reprented in the OS
CAR climatology.The anti-cyclonic (clockwi)surface circulation in the Bay of Bengal (BoB),obrved in the drifter climatology,is also well-captured by the OSCAR climatology.Moreover,
the other known features such as the ECC and SEC,which are reflected in the drifter climatology,are equally well-visible in the OSCAR climatology.During the monsoon transition months of May and October,the strong eastward flowing WJ with cur-rent speed of about 0.6ms −1is vividly noticeable in the OSCAR climatology.During the peak south-west monsoon month of July,the strong northeast-ward SC can be clearly obrved in the OSCAR climatology.However,the intensity of SC is some-what underestimated.During this ,dur-ing July,the eastward flowing SMC is obrved just north of the equator.
In figure 3,we show the monthly difference between OSCAR and drifter climatologies for the same 4months.Here,only the difference between the magnitudes is shown.During all the 4months,the differences are less than 0.2ms −1in most parts of the study area.The difference exceeds this value only at isolated places and in some months.Thus,large difference between the two magni-tudes is noticed near the Somali coast during May,July and October.In the equatorial Indian Ocean (between 60◦and 80◦E),there are large
differences
Figure 3.Difference between OSCAR and drifter-derived climatologies in terms of current speed (ms −1)for the same 4months as in figure 2.
五大运动192Rajesh Sikhakolli et al.
in May and October.The differences are more pronounced in October than in May.There are also large differences in the SMC regime east of Sri Lanka.
美白减肥
3.2Fourier analysis of the major current systems In the previous ction,we have compared the OSCAR climatology with drifter climatology and have shown that the major current systems of the TIO are well-captured by the OSCAR prod-uct.Another way of qualitatively evaluating the OSCAR product is to e whether the major peri-odicities known to exist in the major current sys-tems are well-captured by the OSCAR product.Accordingly,in figure 4,we show the results from Fourier analysis carried out on the time ries of the current speed for some of the current systems.In this figure,EC signifies the equatorial current and MC denotes the monsoon current.The name MC is just a generic name.During winter,the cor-responding current is known as NMC (schemati-cally shown in figure 1a),while during summer,the
current is known as SMC (schematically shown in figure 1b).Except for EC,the annual periodicity is d
ominant in each of the current systems,which was only to be expected.In the EC regime,the mi-annual periodicity is also quite strong.This miannual periodicity owes its origin to the pres-ence of strong eastward jets (the Wyrtki jets)twice a year (April/May and September/October)and has been noted by many rearchers.We mention here just the two well-known theoretical studies (Jenn 1993;Han et al.1999).It is indeed grati-fying to e that this well-known mi-annual peri-odicity is prent in the EC computed by OSCAR algorithm.
3.3Seasonal march of surface currents So far,we have shown the comparison of OSCAR and drifter climatologies for four reprentative months.We have also found that the major peri-odicities of the known important current systems are well-captured by the OSCAR product.Never-theless,it is still instructive to investigate
whether
Figure 4.Fourier analysis of the major current systems in the TIO.The dashed curves reprent the 99%confidence level.
