Continental Shelf Processes Affecting the Oceanography of the South Atlantic Bight

Old Dominion University ODU Digital Commons

CCPO Publications Center for Coastal Physical Oceanography

2-1981

Continental Shelf Processes Affecting the Oceanography of the South Atlantic Bight. Progress Report, 1 June 1980-1 June 1981

Larry P. Atkinson [email protected]

Follow this and additional works at: https://digitalcommons.odu.edu/ccpo_pubs

Part of the Oceanography Commons

Original Publication Citation Atkinson, L P. (1981). Continental shelf processes affecting the oceanography of the South Atlantic Bight. Progress report, 1 June 1980-1 June 1981. United States. doi:10.2172/6549027.

This Report is brought to you for free and open access by the Center for Coastal Physical Oceanography at ODU Digital Commons. It has been accepted for inclusion in CCPO Publications by an authorized administrator of ODU Digital Commons. For more information, please contact [email protected]. PROGRESSREPORT CONTRACTDE -AS09-76EV00889

Larr y P . Atkinson -SKIO-

VOLUME1 DISCLAIMER

This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor any agency Thereof, nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof. DISCLAIMER

Portions of this document may be illegible in electronic image products. Images are produced from the best available original document. ..------DISCLAIMER------,

• This book wa1 prepared as if\ ~ 1of work spoosotetJby an ~e:ncvof the ~ite(:! Sm:es Government, · Neither t'he uni;>are.t~, Prodvct, or proc_ess d sdosed. ·tic represents that its usq would not infringe wwauily owned rifjha, Aefernnre herein to any $f.l(ltl commercial product, process, 0t service by trade name, trademark. ~nulacturer, _or o;hCfWl$8,~ not necessarily constituie Of imply hs endoriement, recommeodauon. or IOV-Or,ngbv lh,e Uni 00 States, Governmem or any agel\CVthereoL The views and O{linions of authofs itxptessed herein do nat n~ily sune or reflea th0$Cof the United States Governmentor onv uger,cythereof.

CONTINENTALSHELF PROCESSES AFFECTING THE OCEANOGRAPHYOF THE SOUTH ATLANTIC BIGHT

PROGRESSREPORT

Larry P. Atkinson Skidaway Institute of Oceanography Post Office Box 13687 Savannah, Georgia 31406

June 1, 1980 to June 1, 1981

Report Date - February 1981

Prepared for the U.S. Department of Energy . -~ Under Contract No. OE-ASO~)

DISTRHlUTIO/fOf THIS DOCUMUT IS UNUMITfl\ . VlGlJJ TABLEOF CONTENTS

Introduction ...... 1

Index Volumel: Reprints ...... 2

Index Volume2: GABEX-I• 3

Index Volume3: Technical Reports ...... 4

P11hlir:ations and Meetings . . . . . 5 Cruise Summary•• . . . . 9 Papers and Reports Published under this Contract •...... 10 Recent Results and Reports • 14

•

DtSCLAlMER TciiS !)'.JCUMENTW1\S F'RE,WlF.O f.~ MJ t.CCOUNT OF WOf?f<, S.PONSOi~EI)2.V THf. u:~irrnSTA,::~ GQ'h:0;;..;1\,ir:.NTNEITHER THE UNiT-EDST/iTES NOR. T}:f UNITEDS"i,i-,T:::., Cf:,\im,mn o::ENERGY i%f( ,'l.NYOf, HE!R£MPLOYf.f.$, MAKESANY WJ1n;,\iITV; fi

Note: Blue sheets have been inserted between articles to separate them.

GABEX-I- Preliminary Results A Spatial Look at the 20-23 April period during GABEX-I. The intrusion of Gulf Stream water across the continental shelf due to topographically-induced upwelling. Detai-led observations of a Gulf Stream frontal eddy on the Georgia continental shelf, April 1977. The relation of concentration and size distribution of suspendedparticulate matter to hydrography in Onslow Bay, North Carolina. Evidence for the deflection of the Gulf Stream at the Charleston rise. Role of Gulf Stream spin-off eddies in forming phytoplankton patches on the outer southeastern shelf. Volumeof summersubsurface intrusions into Onslow Bay, North Carolina. · Summertimeadvection of low sal intty·-su.r.face waters into Onslow Bay. Modes of Gulf Stream intrus-ion into the South Atlantic Bight shelf waters. A time~dependent model of nutrient distribution in continental shelf'waters~ Ocean chlorophyll studies from a U-2 aircraft platform. A flushing model of Onslow Bay, North Carolina, based on intru5ion volumes. Shelf flushing rates .based on the distribution of salinity and freshwater in the Georgia Bight. Pelagic tar off Georgia and Florida in relation to physical processes. Natural fluorescence as a tracer for distinguishing between piedmont and coastal plain river water in the nearshore waters of Georgia and North Carolina. INDEXVOLUME 2

First Preliminary Report of the GABEX-1Cruises: Station Logs, Standard Sections and Mapsand Time Series Transects. · INDEXVOLUME 3

Ndte: The technical reports are in order according to their tech report series nUmbers. In a case where data has been excluded, a yellow blank sheet has been inserted. The reports included are as follows:

Series # Title 80-1 Hydrographic Observatio_nsoff Savannah and Brunswick, Georgia (March, Mayand September, 1977 and January, 1978}. 79-5 Hydrographic Observations in the Georgia Bight (April 1978). 79-3 Hydrographic Observations in the G,eorgia Bight (July ·1977). 79-1 Hydrographic Observations in the Georgia Bight (December1976). 78-7 AeroSystem: Description, Operation, Data Acquisition and Processing. 78-5 Hydrographic Observations in the Georgia Bight (April 1977). 78-1 The Results of Four Oceanographic Cruises in the Georgia Bight. PUBLICATIONSAND MEETINGS

During the past year the following papers and technical/data reports were published or accepted for publication. Papers Published or in Press in Reviewed Journals or Books Atkinson, L.P. and L.J. Pietrafesa. 1980. A flushing model of Onslow Bay, North Carolina, based on intrusion volumes. J. Phys. Ocean., 10: 472-474. Atkinson, L.P., J.J. Singer and L.J. Pietrafesa. 1980. The volume of summer subsurface intrusions of Gulf Stream water into Onslow Bay, North Carolina. Deep-Sea Res., 27: 421-434. Hofmann, E.E., L.J. Pietrafesa and L.P. Atkinson. Description of a~ottom intrusion in Onslow Bay, North Carolina. Deep-Sea Res., in press. Blanton, J.O., L.P. Atkinson, L.J. Pietrafesa and T.N. Lee. The intrusion of Gulf Stream water ·across the continental shelf due to topographically induced upwelling. Deep-Sea Res., in press. Yoder, J.A., L.P. Atkinson, T.N. lee, H. Kim and C. McClain. Role of Gulf Stream frontal eddies in forming phytoplankton patches on the outer southeastern U.S. shelf. Limnol. Oceanog., in press. Cordes, C., L.P. Atkinson, R. Lee and J.O. Blanton. 1980. Pelagic tar off Georgia and Florida in relation to physical processes. Marine Pollution Bull., 11: 315-317. Hofmann, E.E., L.J. Pietrafesa, J.M. Klinck and l.P. Atkinson. 1980. A time dependent model of nutrient distribution in continental shelf waters. Ecological Modelling, 10: 193-214. Kim, H.H., C.R. McClain, L.R. Blaine, W.D. Hart, L.P. Atkinson and J.A. Yoder. 1980. Ocean chlorophyll studies from a U-2 aircraft platfonn. J. Geophys. Res., 85: 3982-3990. Paffenh6fer, G.-A., D. Deibel, L.P. Atkinson and W.M. Dunstan. 1980. The relation of concentration and size distribution of suspended particulate

matter to hydrography in Onslmv Bay, N.C., Deep-Sea Res., 27: 435-447. Singer, J.J., L.P. Atkinson and L.J. Pietrafesa. 1980. Summertims advection of low salinity surface 0ater into Onslow Bay, N.C. Estuarine and Coastal

Mar. Sci., 11: 73-82. Papers in Final Preparation Atkinson, L.P., L.J. Pietrafesa and E.E. Hofmann. On nutrient sources to I Onslow Bay, North Carolina.

Atkinson, L.P. and T.E. Targett. A synoptic observation of upwelling along

the 60 m isobath from Cape Canaveral to Cape Hatteras \vith ref~rence to

fish distribution. Papers Submitted Natural fluorescence as a tracer for distinguishing between piedn~nt and coastal plain river water in the nearshore waters of Georgia and North Carolina. Coastal Estuarine Marine Science. Technical Data Reports 1980 Singer, J.J., L.P. Atkinson and W.S. Chandler. Onslow Bay XBTData: 1975 (OBIS I, II, 111 and IV). Georgia Marine Science Center Technical Report 80-2.

1980 Singer, J.J., L.P. Atkinson, W.S. Chandler and S.S. Bishop. Hydrographic observations off Savannah and Bruns\-lick, Georgia (March, May and September

1977 and January 1978). Georgia Marine Science Center Technical Report 80-1.

1981 Lasley, S.R., L.P. Atkinson, J.J. Singer and H.S. Chandler. Hydrographic observations in the Georgia Bight (April 1979). Georgia Marine Science

Center Technical Report (in press).

1980 Atkinson, L.P., J.J. Singer ilnd \.J.S. Chandler. First preliminary report of

the Gabex-1 cruises: station logs, standard sections and maps and time series transects. Inforfiial r.eport. ';

June 1980. The nitrogen pool of a mid-shelf intrusion. O'Malley, Atkinson and Yoder. 43rd Annual Meeting of ASLO. Knoxville. June 1980. Effect of eddy-forced upwelling on nutrient and phytoplankton

distributio near the shelfbreak in the South Atlantic Bight. ..Gishop, Lasley, Atkinson and Yoder. 43rd Annual Meeting of ASLO. Knoxville. December 1980. Oceanographic climatology of the southeastern U.S. Continental Shelf Waters. Atkinson, Blanton and Chandler. AGUFall Meeting, San Francisco. December 1980. Thennal and visible color expressions of an upwelling off the Gulf Stream. Kim, McClain, Hart, Atkinson and Yoder. AGUFall Meeting, Sari Francisco.

January 1981. Nutrient sources to the southeastern United States Shelf Waters:

rivers, recycling and the Gulf Stream. \-linter Meeting of /\SLO. Seattle.

January 1981. Effect of upwelling on primary productivity of the outer south eastern shelf. Winter Meeting of /\SLO. Seattle.

Other talks not listed, have been given to regional and governmental groups.

The P.1. attended the DOEworkshop on the environmental effects of future

U I El p 1ants in Upton, New York on 22-24 January 1980. FIELDOBSERVATIONS

Cruises under this contract or cooperative1y with other contracts are summar1zedin Table 1. • I.la tf Crui\t I O•t• t,-7 Aug 197!> AO-/S-01 148(, On\lu• Bay l'\Jor1n9 (rut\t (P) 4-14 !lept 19n C • 9-1~ ii 71!(/I 11161 On\lu ■ E•r lntru\lon S\ud1 IP) 13-14 Oct 197~ Au-n-2 ,outII Sl& 1 0111 l(J ■ n•y /\oori"g CrufH (Pl ll-11 1.1,,,I 97!, AO-IS- J 9~C/~91El ~n,lo ■ bar •~ortn~ C,ut\r (r) 13· f tu J 91t, Bf· 1~-? 48, trorgl• 819~1 (hP)

2b ht, 197b Bl .If,.) 21 BC Lrorgi• bight IP) 10 HH 19/t Bf-lb-6 6UC/Sl8l tror9ta !llgnt (~) 29 Har 1911 er. n. 9 7~C/HBl C.eor9i• 81901 (r; i~ A~r 191!. 81-Jt,-1) leH (Nf') ~-b ""•11sn Bl - JL- I!, 16B( C.eor91• Bigrrt (I')

19 ,.,.,, 19/f ~r .Jt. p 1fl I (NP) iL Ko~ !~If er.n-1s Grurgt• BL/ht !•t,ortej), (f·) ) .Junr l9)b !if. )L. 7) lnl (Hf') 17 Jur,r l Si~ ~f-76-7( le\l (UP) l~-18 July 19lt ~I -JL-OU!, On\l~ ■ 6ay Stu!r (P)

7i-~j Jul1 197t llf. lb-C~t 4SCID unslo ■ 6•y SluOy (,) i& July•b Aug Bf -16-0B1 3(X:10/)llHl On1lo ■ B•1 Stujy (P) 191t 14-lb Aug 1971> 6i-lb-06i, .;?( ID Onslu ■ 8•1 S1ud1 (P) IB t.u9 19/b Bf - 7b-OS9 4Cl0 Onslo ■ 6•y Stuoy (Pl 6 Oct 1976 Bf-76-30 42(1D Grorgld Btghl, front Hun! !Pi

U Oct 1916 Bf-1!,.)? 49(10 C.eorgi• ai9h!, front Hunt l (Pl I~ Oct 1976 Bf-H--l\ HCIO C.eorgh 819hl. front lwnt ) {r) 1)-ltl Oct !Sib Bf-76-39 ll.!lCIO C.eorglo Bight, front ll"nt 4 (P) 9-l~ ll~c l9Jb Cl-n-t? He 10, 3111ar Georgia S\Qht tF1 6-9 l'lar 19]7 Bf -11- t< 3~161 C.,•0191a Blgnt If•)

6- lb Apr 1977 1.0-77-4 Hb( 10/30161 C,eor~I• Blghl (P) # 19 Apr 19]] Hf .17.7; IH81 C.eor9la l!tgnt (P) 21 Apr ISJJ lif - 77-)0 l4RS~ r,eorgia Btghl, front Hunt S (P) 76-11 l'!•y 1911 Of.]}. ~e l~CID/llBl C.eorgto 8iqhl (1 t 4-10 July 1917 Cl-71-03 b IC 10/)0lB f C.eor91a B19nt {P)

10-11 Au, 19]1 C,,,.,,91• llighl lntru\10" Stuo, (l'•llrn1101l-r) 17-IS '-u, 1911 Hf. :I-~~ C.rorglo h19ht lntru\lon Study ll 17•?1 Aug 1~71 Bf. 1/- :,: r,,•orgi• bight lntn,\lon ~tu~, Ill 7i-i) Aug 1911 Uf-11-SC C.eurgla Bight lntru1lon Stu~, IV 7L-27 1.ug 1971 llf. )}. ;,.; 2.l lHl C.eorg,a 819hl ln1,u1lon Studr V

13-14 ~ept 1971 flf-1/-Sl IUIO/l~IRl (,,.a r 9 t • b I 9 1, t ( f' ) S-9 llov 1971 c1-11.o, ~sen C.torgla Bight (lf) 13-21 J•n 19/t Cl-lH-01 'l4B(/74IB1 r.eu1·91a !light (P) 17-21 leb l97F 6f -18-07 1)(1[1 lr1t1 {IIP) 17-;J ADr l971i Cl -111-04' ~9cl0/l9Hi Leor91a 810nt (P) ~. \~ Md, l'Jit\ il-lu-11 28ClDJll~~~ Ltor910 b19rit, / ,.,nl flu, (I') ~-I?;.._,, 1971, 1.J- liJ-UI ?L R~:, L.-orgla 81901, lrnul rlu, (r) 3-10 ~ugu\t 19/t C 1- lc\.lll,tt )) )LC JD1:,?1BT L,-,.,9 t • b t ?ht ( If•)

tl-14 huv. 191b (1-78-o.~•· ~10/4lel ~CJulh Al t,nl It Ot9n1 (P) 13-19 l'\JJrlh 1579 pp. 79-01" 4)(10/lblBl ~ovlh A1lanllc 8t9hl 11 ~•,-2 Junr PP-19-0t" ~? C tt, ~outh At l on! it Right 1:1 I 5 IS ~ Ju\,- 11 /.."~. 61 - 19- )4 21l~ 1D/l'iUBl G!'Or9 I• e.19nt lnHru\lon ~,uo, (lf') 19),; Z?-?9 /..u;. l91~ H-1~-0J" ~2( 1r, ( f, i ~outn Atl•ntlt ~'t"' J-4 I.let. 19/9 Sf . 79. ~ I e1s1 ,.~orgla Et•;nt (IP 1~•17 !Jct. 1~19 Bf - 19· S~ 41B1 C.N,r~I• !ll~nt (IP 2b Ou.-J hu.-. CJ. 7

~~I~\: Bl•Blut f In, Cl•Col.,-,t,.,1 lstlln; f•l,H ■ orc; A!i•~d,,ntr ll; W•~II Jenn; p;,aJ, \ Pt lf'

·-. ___.... ·------PAPERSAND REPORTS PUBLISHED UND~R THIS CONTRACT

1975 Atkinson, L.P. and D. Wa1lace. The source of unusually low surface salinities in the Gulf Stream off Georgia. Deep-Sea Research 22: 913-916.

~ 1976 Atkinson, l.P. and J. Hall. The distribution and production of methane in the Georgia salt marsh. Estuarine and C6astal Marine Science 4: 677-681 1976 Dunstan, W.M. and L.P. Atkinson. Sources of new nitrogen for the South Atlantic Bight. pp. 69-78 in M. Hiley, ed. Proceedings of 3rd Inter national Estuarine Research Conference, Galveston, Texas. 1977 Atkinson, L.P. Modes of Gulf Stream intrusion into the South Atlantic Bight shelf waters. Geophysical Research Letters 4(12): 583-586. 1978 Atkinson, L.P., J.O. Blanton, and E. Haines. Shelf flushing rates based on the distribution of salinity and freshwater in the Georgia Bight. Coastal and Estuarine Marine Scienc~ 7(5): 464-472.

1978 Atkinson, L.P., G.-A. Paffenhofer and W.M. Dunstan. The,.chemical and biological effect of a Gulf Stream intrusion off St. Augustine, Florida. Bulleting of Marine Science 28(4): 667-679. 1978 Blanton, J.O. and L.P. Atkinson. Physical transfer processes bet.ieen Georgia tidal inlets and nearshore v,aters. pp. 515-532 in J. Wi1ey, ed. Estuarine Interactions. /\cademic Press. 1978 Pietrafesa, L.J., J.O. Blanton and L.P. Atkinson. Evidence for deflection of the Gulf Stream at the Charleston Rise. Gulf Stream 4(9): 3, 6-7. 1979 Atkinson, L.P., J.J. Singer and W.S. Chandler. The 9400 CTDSystems reliability, calibrated and data acquisition. in Proceedings of Fifth STD/Ocean Systems Conference and Workshop. Grundy Enviornmental Systems, San Diego. In Press. 1980 Atkinson, L.P., J.J. Singer and L.J. Pietrafesa. The volume of sum111er subsurface intrusions of Gulf Stream water in Onslow Bay, North Carolina. Deep-Sea Research 27(6A): 421-424.

1980 Singer, J.J., L.P. Atkinson and L.J. Pietrafesa. Summertime advection of low salinity surface w~ter into Onslow Bay, North Carolina. Estuarine and Coastal Marine Science 11: 73-82. 1980 Atkinson, L.P. and L.J. Pietrafesa. A flushing model of Onslow Bay. North Carolina, based on intrusion volumes. Journal of Physical Oceanography. 10(3): 472-474. 1930 PaffcnhBfer, G.-A., D. Deibel, L.P. Atkinson and W.M. Dunstan. The relation of concentration and size distrihution of stispended particulate matter to hydrography in Onslm-JBay, North Carolina. Deep-Sea Research 27(6A): 435-447.

ID 1980 Kim, H.H., C.R. McClain, L.R. Blaine, W.D. Hart, L.P. Atkinson and J.A. Yoder. Ocean chlorophyll studies from a U-2 aircraft platfonn. Journal of Geophysical Research 85(C7): 3982-3990. 1980 Hofmann, E.L., L.J. Pietrafesa, J.M. Klinck and L.P. Atkinson. A time dependent model of nutrient distribution in continental shelf waters. Ecological Modelling 10: 193-214.

1980 Cordes, C., L.P. Atkinson, R. Lee and J. Blanton. Pelagic tar off Georgia and Florida in telation to physical processes. Marine Pollution Bulletin 11: 315-317. .. 1980 Lee, T.N., L.P. Atkinson and R. Legeckis. Detailed observations of a Gulf Stream frontal eddy on the Georgia continental shelf, April 1977. Deep-Sea Research. In Press. 1980 Yoder, J.A., L.P. Atkinson, T.N. Lee, H. Kim and C. McClain. Role of Gulf Stream frontal eddies in fanning phytoplankton patches on the outer southeastern U.S. shelf. Limnology and Oceanography. In press.

1981 Blanton, J.O., L.P. Atkinson, L.J. Pietrafesa and T.N. L:e. The intrusion of Gulf Stream water across the continental shelf due(.to topographica11y induced upwelling. Deep-Sea Research. In press.

1981 Hofmann, E.E., L.J. Pietrafesa and L.P. Atkinson. Description of a bottom intrusion in Onslow Bay, North Carolina. Deep-Sea Research. In press.

IL» Technical Reports 1967 Stefansson, U. and L.P. Atkinson. Physical and chemical properties of the shelf and slope waters off North Carolina. Duke University Marine Laboratory Technical Report. 230 pp.

1975 Atkinson, L.P. Oceanographic observations in the Georgia Bight. R.V. EASTWARDCruise E-13-73 and E-19-73. G·eor9ia Marine Science Center Technical Report 75-6. 1976 Atkinson, L.P. Oceanograµhic observations in the Georgia Bight. R.V. EASTWARDCruise E-3-74 and E-12-74. Georgia Marine Stience Center Technical Report 76-1. 1976 Atkinson, L.P., J.J·. Singer, \LM. Dunstan and L.J. Pietrafesa. Hydrography of Onslow Bay: September 1975 (OBIS I I). Georgia Marine Science Center Technical Report 76-2. 1976 Atkinson, L.P., J.J. Singer and L.J. Pietrafesa. Onslow Bay intrusion study. Hydrographic observations during current meter StfVicing cruises in August, October and December 1975 (OBIS I, nr and IV). Georgia Marine Science Center Technical Report 76-4.

1977 Hofmann, E.E., L.J. Pietrafesa, L.P. Atkinson, G.-A. Paffenhofer a1~d h'.M. Dunstan. A mathematical model of nutrient distribution in coastal waters. Center for Marine and Coastal Studies, North Carolina State University, Technical Report No. 77-2.

1977 Pietrafesa, L.J., D.A. Brooks, R. D'Amato and L.P. Atkinson. Preliminary data report, physical/dynamical observations made in Onslow Bay; summer, fall and winter, 1975. Center for Marine and Coastal Studies, North Carolina State University, Technical Report No. 77-05. 1977 Atkinson, L.P., G.-A. Paffenliofer and l~.M. Dunstan. Hydrographic and biological observations at an anchor station off St. Augustine, Florida, 9-14 April 1975 (R.V. EASTWARDCruise E-lG-76). Georgia Marine Science Center Technical Report 77-4. 1977 Singer, J.J., L.P. Atkinson, W.S. Chandler and P.G. O'Malley. Hydrographic observations in Onslow Bay, North Carolina: July-August 1976 (OBIS V), Data Graphics. Georgia Marine Science Center Technical Report 77-6. 1978 Chandler, W.S., L.P. Atkinson, J.J. Singer, P.G. O'Malley and C.V. Baker. A. CTDSystem: Description, operation, data acquisition and processing. Georgia Marine Science Center Technical Report 78-7. 1978 O'Malley, P., L.P. Atkinson, J.J. Singer, W.S. Chandler and T.N. Lee. Hydrographic observations in the Georgia Bight (April 1977). Georgia Marine Science Center Technical Report 78-5. 1979 Deschamps, J.R., L.P. Atkinson, J.J. Singer, W.S. Chandler and T.N. Lee. Hydrographic obse,·vcttions in the Georgia Bight (December 1976). Georgia Marine Science Center Technical Report 79-1. 97 pp. 1979 Atkinson, L.P., /\.L. Edwards, J.J. Singer, ~~.S. Chandler and G.-A. Paffenhofer. Hydrographic observations in the Georgia Bight (July 1977). Georgia Marine Science Center Technical Report 79-3. 126 pp. 1979 McCarthy, J.E., L.P. Atkinson, J.J. Singer and W.S. Chandler. Hydrbgraphi• observations off Savannah, Georgia (winter/spring 1976). Georgia Marine Science Center Techndtal Report 79-2. 41 pp. 1979 Lasley, S.R., L.P. Atkinson, J.J. Singer and W.S. Chand1er. Hydrographic observations in the Geor~ia Bight (April 1978). Georgia Marine Science Center Technical Report 79-5. 1979 Kim, H.H., C.R. McClain, L.R. Blaine, W.O. Hart, L.P. Atkinson and J.A. Yoder. Ocean chlorophyll studies from .a U-2 aircraft platforn:. NASATechhical M~norandum80574. 27 pp.

1980 Singer, J.J., L.P. Atkinson and W.S. Chandler. Onslow Bay XBt Data: 1975 (OBIS I, II, III and IV). Georgia Marine Science Center Technical Report 80-2. 114 pp.

1980 Singer, J.J., L.P. /\tkinson, l~.S. Chandler and S.S. Bishc,p. Hydrographic observations off Savannah and Brunswick, Georgia (March, May and September 1977 and January 1978). Georgia Marine Science Center Technical Report 80-1. 105 pp. 1980 Atkinson9 L.P., J.J. Singer and W.S. Chandler. First preliminary report of the GABEX-1Cruises: station log, standard sections and maps and time series transects. Informal Report.

1981 Singer, J.J. R/V ISELIN Hydrographic Data. GABEX-1. 200 pp. 1981 Singer, J.J. R/V EASnJARDHydrographic Data. GABEX-1. 200 pp. RECENTRESULTS AND REPORTS In this section we will present recent results some of which are brief r~ports others pre-prints or re-prints and others technical reports or data reports. Items are separated by blue pages. Yellow pages represent large blocks of data that were not reproduced for this report. Field observations in 1980-81 included the large field experiment GABEX-1,a smaller experiment during the summer of 1979 and 1980 to acquire data relevant to the proposed GABEX-2experiment. Results of the summer 1979 cruise can be found in the proposal. Gabex-1 Preliminary Results Gabex-1 was a large scale observational effort involving three ships (!SELIN,EASTWARD, BLUE FIN), three airplanes (Coast Guard P-3, NASAU-2 and C-:130), several satellites for communication and remote sensing and a large array of current meter/temperature/pressure moorings. This particular contract

/ was responsible for much of the coordination and in particular the CTO/XBT/02 ·• nutrient measurements on the !SELIN and EASTWARD. The observational schedule of GABEX-1is given in detail in the document "First Preliminary Report of the GABEX-1Cruises; station logs, standard sec tions, maps and time series trahsects'' which can be found in the Progress Report. During the cruises the EASTWARDmapped the shelf with transhelf sections corresponding to principal current meter sections (Figure I). Meanwhile, the ISELIN repeatedly mapped events at the shelf break where springtime upwelling is usually restricted (Figure 2). For this proposal, we will discuss one facet of the results which is direct ly related to this proposal and do not require data from other contracts which has not yet been merged. We will present observations from the St. Augustine Time Series which demonstrate the rapidity with which Gulf Stream associated upwelling can change the character of adjacent shelf waters. Over a period of 13 days in April, the St. Augustine transect was occupied 11 times. We will specifically discuss the temperature, density and nitrate data which are shown in Figure 3. At the onset it must be said that throughout the observation period active upwelling occurred as indicated by the cold nutrient-rich waters over the middle IS and outer shelf and even occasionally in nearshore waters (EASTWARDsection 5 and 7). Prior to April 12, a mass of 17°C water was advected onto and across the shelf at St. Augustine. Because of the high nitrate concentrations, this event probably occurred within a few days prior to the 12th. Through the 13th (!SELIN Section 4) the situation inshore of the shelfbreak apparently changed very little. Over the upper slope, however, there was a definite doming, probably associated with frontal eddy formation. This conjecture is supported by the presence of a surface filament of relatively warm water (21°C) over the shelfbreak. During this period (12-13 April) nitrate concentrations were • uniformly high and penetrated well across the shelf (EASTWARDSection 2). Between 13 April (ISELIN Section 4) and 15-16 April, the situation changed. The surface mixed layer over the upper slope deepened, dome structure over the upper slope disappeared and the warm surface filament was missing. Isotherms over the shelfbreak and upper slope descended ca. 50 m and the 17°C water mass on the shelf was stranded at mid-shelf. Thus between 13 and 15 April, the dynamics at the shelfbreak appear to have gone through a transition from frontal eddy dominated to Gulf Stream domination. During this period the shelf was vertically stratified with strong horizontal gradients occurring only at the shelfbreak. Between 15 April and 23 April hydrographic structure changed relatively little. The lower layer shelf temperature persisted at l8°C and nitrate concentrations slowly decreased, probably due to assimilation by phytoplankton. The 18°C shelf water mass appeared to move offshore between 15 and 23 April. On 24 and 25 April, the Stream appeared to move farther west (onshore) causing upper shelf isotherms to deepen and shelfbreak bottom temperatures to rise from 13°C to 21°C. The high nitrate water mass that persisted over the shelf throughout the observations was stranded on 24 April {EASTWARDSection 15). The lower layer at mid-shelf was mixing into the surface layer, reducing surface salinities. The most dramatic change in shelf hydrography occurred between 24· April and 25-26 April. Freshly upwelled (high nitrate) water penetrated at least half way across the shelf. The 18°C isotherm ascended from 140 m to less than 40 m. A warm filament appeared (stations 403-406, EASTWARD,Section 18) over the shelf break accompanied by a dome structure both of which are indicative of frontal eddy activity. The sequence of events leading to the 25-26 April upwelling can be examined ·P in more detail via the other alongshore and transhelf measurements made during the period. On the 23rd prior to ISELIN, Section 18, we made an 60 m_isobath XBTrun from 28 to 30°N (Figure 4). The thermal structure indicated an upwelling event at ca. 29°50'N, just south of St. Augustine, with minimum temperatures of 17.9°C. A second event extending from Cape Canaveral to New Smyrna (29°N) was much larger in the alongshore and more intense with minimumobserved temperatures of 15.5°C. Immediately after the 60 m isobath run a section was run at 30°N (St. Augustine) (!SELIN, Section 18) where at 60 m the near bottom temperatures were still between 18°C and 19°C. Apparently the< 18 C water mass observed in the northern part of the 60 m run was past or had not yet reached 30°N. On the 24th, the 30°N section was occupied again (EASTWARD,Section 15) and at 60 m temperatures had increased to ca. 20°C and surface temperatures approached 25°C. Clearly at this time the Stream had moved onshore and we were observing a part

of the event previously seen at 29°10 1 N at 1400 on 22 April during the 60 m run.

This translates to a phase speed of ca. 161 cm sec- 1 • On 25-26 April the section was again repeated and 60 m bottom temperatures had decreased to< l6°C indicating the event seen during the 60 m run between' 28 and 29°N was now at 30°N. The

absence of 16°C water at 60 rn at 29.0 N was confirmed by a CTD section (EASTWARD, Settion 17) run on 25 April. At that time 60 m bottom water had increased to 17.5°, the event had clearly passed 29°N and was in the 30°N area. These observations are an excellent example of the upwelling process in the

SAB demonstrating that the events propagate north, cause significant displac~ ment of shelf waters and advect significant amounts of nutrients into shelf waters. In the next year the results of this cruise will be synthesized with other data sets to further elucidate the process of shelf water/Gulf Stream interaction. 82 81 80 79 32 r,--·------..,-::-:-:;,----·--:-, ----.------, _,.___. 32 ..- 'I . , ·' 20m, \ - 100m / I • ,, I I , .,, I I .... I I... , I ., , 't .... 31 31

.,. \

'JAX ... . ·•

30 .. 30 'I C ' \ ,-,~...'

\ I \ \ ... ' ',,( ' 29 __, 'I 29 I I I I \ I EASTh'ARD ..... I \ •\ CRUISE TRACK ' \ ' 'I ', I I I I I \ \ I 'I I ... I

~ ' I ,; ' , I I • I I ' "' \ 28------2B 82 81 80 79 Figure la. EASTWARDcruise track 82 Bl 79

32·.------.---;:-:-;----:----r------1 , 32 .,I 'I ; . 2Bm,' 100m ,' . , I . ., _,_- _,. , , , , . . ,, , , • I • • , 'I ,., I I I • .-,I ,. #4 9 1 .... I • • " t I I I.... , I , I I I \ I ,, I I ,,, I 31 I 31 .,.'.. #20 • • I • I ., ; • r . • • , '· ,__,,- . :-;,_.{,••. #3~ 8, 13 I '"'• I .., .... '\ \ ,, #19 I , , ' I I ; I I I ,I I ,I ... I , , ; : • • 112. s. 7,, 1 e. 1 2. 30 . , - . - ,. - - ,...,..r-r .... 30 I ' .. ;... ·'··I #18 15 · -· •· .... , • · · II 16 '' I I ,-' ...' ~ ...... I \ I ,, I ..... I ,_ -- I ~----..---' I ' I 'I '\ I ~ ... : /11,. 6, 11, 1 7 I , ... \ I ,...... · I I I I ' - '_,,. l I 29 • \ 29 ,--' I EASTWARD I I \ I ... 1 TRANSECTS \ , \ \ ' \ ' I ', I I I I I \ \ \ I I I I ... I .. , .,• I , I I I I ' ' 28.______...L.- __ -i._._ ' _ _LJ______J28 82 Bl BB 79 figure lb. EASTWARDsections. Section numbers indicated. Refer to Table 3 for stations, etc. 82 81 79 32,------.-----,------~------32 . I ' , ,I 100m / I I ,, ,

... I I .,'I _,.,.., ., 31 31 ' .,I ' .., I ....-., '\ ,'-\,, ' , I I r I \

.,. ' ,I 30 ..I 3{] 'I I ' \

29 29

JSEL IN C RU I 5 E TR At I<

, ., I I ... ' ' 28 ' 28 B2 Bl 80 79

Figure 2a. ISflltl cruise trod gz_,1 ~~~ 82 81 BB 79 .3~ ..------..------,----.------, --- 32 ' ' , I • 2EJm\ J IZJEJm/, • , I , ,~-: , , , . . . . , , ,' , .... ' 5 I •~ • ,, I ,., I I I i... , ,' I I I '\ I , I ....' I .,,, I 31 31 \ I \ I I I , I c" I • • • • I ...... _, . ,'. . . 1 I .. '\ , ,,.. \ , ' ..,, , · ·.: 20 I ' : . 1 1 'JAX I . . . ' '\ ··: . · 2 3 I ., , ...... f •• 24 ,' , , • • ...,'.. - 2, 3 .. 4 .. 7. 8 3fJ I ' 'I I ( ...,. 1 2 &18 .. • • ,-,,' .. . ' .... L10 \ I '' . · ·. ,., . 1 3 2- 2 ...... ~ .....'---, ____, .. ·····\9 ' '\ 1 • • • • • I \ 1 4 I '\ ' \ 15 . . . \ 29 'I 29 '' \ 'I ...... ; 6. 16 .\.:· ...'•' \ \ I ' I I I !SELIN I ' t ' SECTIONS \ '\ ' I ... I • ' f ,"' 'I I I t I ' .. 'I 2BL------'-----'-----LJ._ ' ______.28 82 Bl 80 79 Figure 2b. ISELIN sections. Section number indicated. Refer to Table l for stations. etc. , .,..,.,,,..., 31 31 ' \ , .., I ,..... -., l \ ...\ ,,' ,, ,' J I 'JAX I ,,, .. '\ .. .._ ,,' ,I I 38 \. 30 'I « ' ... ,-' ...~ ', \ ,, ..... '--- ..., ' , ...--- \ 'I I ...I I.. , '\ 29 -, '•' 29 , - _1 . 17 '\ ..... \ ' \ ' \ I I I ]SELIN 63m \ JSOBATH RUN .... '• ,,., I ' I I . ' 28.______...____ ~' ______' \ J.L ______J28

82 Bl 88 79 Figure 2c. ISELIN stations on 60 m isobath. Refer to Table 2 for stations, etc. ~~ N ~ est ~

TE11PEAATIJREC •c, Sl& .. A-l !SELIN SECTION 2 '27.ll Nll~AIE 1_..m ISELI N SECTION 2 !SELIN S.:C)lc:, 2 •Ill i2 APRIL 1981!! 12 APRIL 1960 12 A?RI L 196:: 80 180 18il 10. 39 50 , 78 90 110 130 150 10 30 ~il 70 91! I Hl 133 DISTANCEOFfSHORE (M~l OISIAN(( OffSHOR(IKMI IC JO SZ 10 <;;: 11:l I J:l OISIAt.(£ OHSHCi

:-;<:;M~~~a: ...),( ...... mmm&mmmNNl'\INNN- mm

60 60

C0 ~10.l Cl. rit r;v <.~:u TEMPERATURE< "C:J w SICl•A-1 a: EASTWARD SECTION 2 C (ASIWARO S(CIION 2 EAShA~O j,EC: lC:. ;:, 40 12-13 APRIL 1980 l 41l 12-ll APNll 1900 12•13 AP~ll lSG~

B0 180 10 31l Sil 71! 90 110 130 150 10 3a 50 70 90 110 130 ISO 10 30 5:l 711 9.l I I ;J 130 DISTANCEOFFSHORE (IOI) DISIANCEOFFSHORE

5.l

TEMPERATURE ( "CJ ISEl..lN SECTION J II/! . 12-13 APRIL 1980

10 lil 59 70 90 110 130 150 OISIAN(£orfSHOA[ (KM)

... tJ>0~~125 ?C 2ii! ;:, 11:: 60 ~:? .;:10;, Tl;l'P(RA TUR£ ( "Cl SICM"•l 0.. 1.J~rott <,...•-~; ISEI.IN SECTION • ISE~IN SCCJION • I:!! Ill !Shi:• SECTl::Jr, 4 I l APRIL 1991.! 13 APRIL !980 140 I) A?RI.. 1se:: t:\ l 4 m "'8 12 180

13 32 50 71! 9D 110 130 150 10 30 50 70 90 IHI 13.I 150 3.J Sil 70 9Z I I il 13.l OISIAN([orFSHQqf (K~) OISTAN(fOFFSHORE IKM) "' OJSTArKEorrst

Figure 3. St. Augustine time series during Gabex-1 ~q) a, UMU>CUJiCU>CLJl,,Q_.DQ,( U'1• r,t\1-mC'lm~ "'"'"'"'"'"' u .. u )C .....~ _____ .,, ~~ - II> OI CD~ m m m s m m CD m ~ .. "'"'II> ISi 1\1"'"'m .mm"'"' ;;.;;~ 20

60 5 g ,u 0il . =:Illa '-15- TE~PEil:,:ruRE C·t, SIC.MA-T ..,"- ll!TRATE r... •-1> EASTWARD.SECIION5 [ASlwARO SECIION S 0 EASTli,.ROSE::, lC:1 5 ~0 15-16 APRIL l!180 15-16 APRIL 199:l 140 IS-16 APiflL 198:l '~=. -~s 80 180 IBll Ill· Ja 59 "' 90 130 150 10 30 50 711 5i! I IC I JC 10 30 50 70 9tl 110 13.l DISTANCEQrFSHORE OISIAHC[ ClffSHOREt~Ml OISTAIICEOHSHORE (l'HI

u ,. u .. .__,.... u - LI91CUJICU MLJJr--.an.l"'\ .,...... ,.,._,.,.. ,., N .. .,, ~ ; &CG>MDIMM . :;,c:i.(~~=~ .. t'\JNNNl'V"'- N N N N N N - a, GI GI ...... ti> ..,_ ~mmmmm "' "' "' &~ ms.mrsammm mm m ---- ...... - --- "'"' ~ 4 ... ·, •• 20 '";;~::: Zll 60 ~·;22 60 63 g g ~19 ~ '\26.0 0e 16 ,;;10U =: 1::il ~~o;JS TE,..PERATURE. C 'C> "- SIC.M•-1 ..,"- ~PR.•.TE <... ~, '-~:, EASTWARDSECTION 7 0"' EASIWAROsec 11ON 7 0 EASrwAR:Ja~c rc:i , «0 17 APRIL 1900 1"112 140 17 APRIL 1990 14:! 17 •.?RIL 19R:: '25

Bil 180 10.l

10 3'3 50 70 . 90 110 130 150 ' Hl 30 50 70 90 llll 13il t5J 10 Jcl 50 70 90 110 130 DISIANC[ OfFS>lOfl( 000 01S1Ar4Cf.OffSHQRt IIIHl DISIM:C( DHSHOflE O(M)

a, CD ., ....., ,c ,. • ..•>0000< ,.,...... en.. m ...,,,._., • >< )( >C ,c IC Jlf JC ,ooooc .,, .. N .. N N- Nm w, • P'a "' - OI CD ~ N N m "'II> ISi m ...ISi ISi 151 ISi ISISISl5ZSl "'"'"'"'"'"' -- "''51Slll>151ISIISIISIS>slSISIS>__ _ • ~I j ,,"t· ~j 611 6C g ll.1 :;l~il TEMPERATURE<'C> "- 11ITRAT£C.,!'l) ·,: w ~ EASTWAROSECTION 10 0 EASTw,.RoSECTIOII 10 •• 10 •0 19-20 APRIL 1900 14:l 19-20 APRIL 198:J ~\15 !·~ \2:, ea l!l~ \25 ~! ._~_.__.___._L-..._-L_..__.._.....i..i..~-1_....._ 10 30 5ll 70 90 I 10 130 15" 10 3:J 50 71l 90 11ll 13:l DISTANCEO•FSHORE IKM> OISIAIIC( orFSHOR[ (MM)

u )C ~ :,( ..,., a, ...... ISi.. .. ~,., N N N "' ;.; N ISi ISi Ill "'ISi ISi ~ isi°' f8 ~ t i i-4'' .• . 1,.... ~1' z~ ------...122 6.l 20 i ,,,, ~8 .... NI TIIAIE 1.,t') TE"P[RATURE <'C> ~16 [•STWAqD SECTION 12 ~14 [AST~Alla SECII~~ 12 «0 -• ~PRIL 198~ 21 IDRiL. 198:: 12 B3 Ill

!i! :ii! 78 90 110 I Jil I '.;e Ill j~ ~11 7:J ;;, 11ll I JD DISTANCEOfFSHOIIE (MMJ DISH'.;~[ crrsHORE (M!4)

Figure 3. ·st. Augustine time series during Gt\BEX-1 (cont_). •.~as u ..-J>oL.l;,OU N,.-,.,.._ ~~ U>IDOOD.0 ~~ tDCD.£1aaD N~ "'~ "' ~ / . 20 ;o . . . '" ",,. . DRfl.: 198: 14

180 12 180 '21.0 ,e:: 10 .30 SIi . 71!1 90 110 l30 150 10 30 Sil, 70 90 110 I 33 150 IC 3:l sa 70 92 110 130 OISYANCEOFFSHORE (KH) 01Sln4CE OHSHOR£ ll

60 6;J g g -s ,':: 100 :1:i;: s1c .. 11-T =1r;? a.. a.. NITRATE (,>!) -IC j T['4PERATURE C "C> /18 w [ASTwAAO SECTION 15 .., I 0 0 [AS h'ARQ. SECTJOII 15 EASTWARDSECTION 15 2• APRIL 1980 I c:, I 1,0 \ I(:) 24 APRIL 198.l -IS 2• APRIL 1981! ...-16 .

181l ~14 1e;i

10 30 50 711 90 110 130 150 10 30 53 70 90 I 10 I 30 150 Ill ;;J 5:, 70 . 9.l 110 1_30 OYSTAN([OFFSHORE (~Ml OISIAN(f, DHSHORF.I~~> DISTAPlC:( o,rSHOR( (KM)

U >otl..J LI~ ,.._ \."71• Plru- LI JOC LJ U Lb< U >OCU U4...... tlC U >CU ,-.. ll7"l• ,... u:n .. ,....,..._ ..., m,-... 0,1 ~••~ Plrv- mr.zg tOJ~~ ~ & l\.H'J"'~ m eas SGISI - °' a, ...... ,...,.·~ ...... egr,r, 2il

60 .:; i lCll ~100 . n. . TEMl'EA11tuHt'. 1 •c, w IIJTilAT( (~'" ·•. 5 EASTWARDSECTION 18 C [ASTwARO SECTION 18 EAShARU SECt IC:1 18 l;; 140 25 · Zf; APR I l 19B0 25•26 APRIL 198C 140 25-2& APRIL 1980 '. s l80 180 180 \ . 20 10 ·J.J 50 70 9il IIJ 13.l 15.l 10 30 s0 10 90 110 rn, 150 30 5D 71l 90 I Ill 130 Dl~TANCEOFFSHORE C~~l 0 IST ANCE OH S ► IORE (MMI "' DISTANCEorFSHORE (KH)

Figure 3. St. Augustine time series during GABEX-1(cont.). ,.

• J I

Figure 4. Thermal structure at 60 m from Cape Canaveral (242) to St. Augustine (260). A SPATIALLOOK AT THE20-23 APRIL PERIODDURING GABEX-1

In the previous section we examined the St. Augustine Time Series and noted the presence of an upwelling event in that area. In this section we will present data showing the north/south extent of the upwelling area. Figure 1 shows bottom nitrate concentrations from an·· EASTWARDmap made on 20-23 April. Between 29 and 31°N nitrate appears to be advecting shoreward which agrees with data from the April 22 60 m isobath run (Figure 2) which placed an upwelling

event at 29°451 N (Station 258, in Figure 2). The upwelling center at 28°20' thru 28°50'N (Station 244-250 in Figure 2) was south of the area mapped by the EASTWARDon 20-23 April (Figure 1). .,

During the time period the !SELIN made several transects between 28°43 1 N

and 30°191 N. The Southern most !SELIN section (Section 16, Figure 3) shows the strong upwelling occurring in that area. This section was at the location of station 246 in the 60 m isobath run (Figure 2) which was in a definite zone of upwelling. At this location nitrate concentrations where above 15 .µM

inshore of the 60 m isobath. At the next section north (Section 15, 29°06 1 N} upwelling was much less in evidence with 20°C water restricted to the upper slope and nitrate concentrations below 0.5 µMabove 30 rn. This section agrees with the observations at 29°N in the bottom nitrate map and the distribution of upwelling centers observed in the 60 m run. Section 15 intersects the 60 m isobath run (Figure 2) at station 253 where upwelling is less prevalent. Further north at ISELIN sections 14-11 upwelling is obvious with the 18°C isotherm extending across the outer shelf and high nitrate concentrations at all depths below the shallow surface layer. The 60 m isobath run on 22 April in dicated the upwelling area was north of 29°30'N. This is further north than indicated in the sections. Tt1is is probably beca~se the sections were run on 20-22 April with the northern sections run on 20 April which was 2-3 days before the 60 misobath run. No doubt the northward advection of the upwelling center accounts for this difference. The onshore extent of the intrusion appears to extend well across the shelf and possibly to the nearshore zone (EASTWARDSection 12 (Figure 4) at 30°N (Station 260 in Figure 2)). Although nitrate concentrations are low they are normally less than 0.5 µMbut in this case concentrations above 1 µM cover 2/3 of the shelf and concentrations above 0.5 µM cover nearly all of the shelf; clearly a case of massive upwelliny. 82 81 80 79 32~-.:._------~------,------,----32

' I / I I I 0 \ ,,' I • , ,I , , • I ,'•.I I I I , I I I I I I 31 '/ 31

I I ' ' '• e .,I I ' If,,' .., I 1!

...~ / t 5 ,' I I 30 • ~: 30 -~'I . ' i : ,-,,' ' ...... ' ' \ i \ ,, ' I ,_ 1'(\.

\ I I I I 29 I BOTTOM 29 , I I I NITRATE I ' ' .. I \ I GABEX I \ \ ' ... I 20-23 APRIL \ I 'I I I 198f3 I I \ \ I I ' I I , I ,.- I I I I I ' I ' I 28 L______l______i,___'_.______,__L..-______.28 82 81 80 79

Figure 1. Near bottom nitrate concentrations during 20-23 April, 1980. 0 ~

>( -N "-1' N -,.... r... a,"" X X XX XX XX XX XX XX X XX XX > N (Y) --tj- Ln CD r--,..(X) 0) CS) _. N (Y) ...-;;t- Ln CD r--,.. (X) 01 CSl ro c:: --tj" --tj" --tj" --tj" ...-;;f" ...-;;f" ...-;;f" ~ Ln Ln Ln Ln t.n t.n Ln Ln t.n t.n (.0 tO N NNNNNNNNNNNNNNNNNN w OJ C. ro 22 u E 0 20 .....s... .t::...... ro .0 0 60 .,..1/1 E ;.,J. 0 I.O ~) :r:100 O'I t- C: o... ,,...0 ,...... ,0 X w 0 Q) I.O 0 TEMPERATURE (°C) S...N ::, +,J - 140 !SELIN SECTION 17 u (U ::, s....,.. C: +,J ..,I 22 APRIL 1980 VI Vl :::, -ro O"l::, 180 t: ~ Q,J . .c +-' t- 1/) 10 30 50 70 90 110 130 150 170 190 210 230 . N DISTANCEOFFSHORE (KM) QJ !.. ::, .,..O'I I.I.. // ~~ 01- C7'0UI-- 20 20

60 60

. ~100 ,=100 :!:, TEMPERATUR£ C "C> ... NITRATE (,JOI Cl !SELIN SECTION II ~ 140 28 APRIL 1980 ISELIN SECTION II 140 211l APRIL 1981/l 8 180 180

10 30 50 70 90 110 130 150 10 DISTANCEOFFSHORE (KM> 30 50 7B 90 I le 130 158 '-' UUJ DISTANCE IHSHORE 000 ... ,r,,a-. IS"'"°"" ~~ ('\,· "'""' re~ 20 20 l,S.:~(;3 60 60 :15

;!: 100 ~100 a. TEMPERATURE ( "C> a. Ill TRATE <~> ~ ISELIN SECTION 12 ~ 140 20 APRIL 1980 140 ISEL IN SECT ION 12 211l APRIL 1981/l

180 180

10 30 50 70 90 110 130 150 10 DISTANCE OFFSHORE (KMl 30 50 70 90 IIB 130 158 U UIJ.JU DISTANCE OFFSHORE (KM ► (7'1C5)-",J"1 U UUJU c,,151_...... , 151 --- "' t'\Jf"o.l\lr"'-4 Nm N~---

20 20 "·~•.)s. ~ •

1 60 llll ' .=100 a. TEMP[RATURE ( •o rHtRAIE <~> ~ ISELIN SECTION 13 ISELIN SECTION 13 140 140 211l-21 APRIL 19Blll 211l-2I APRIL 1980

180 1B0

10 30 50 70 90 110 130 150 10 30 50 70 90 I IB DISTANCE OFFSHORE (KM> 130 158 LJ uuuu DISTANCE Of'FSKJR( (KM) '°' Q)C7,c:s;, u uuwu .- --NN ,.... mCJ1m- N N"'NN -N f\UVNN--"'"' • •!.0.2 20 . . 1 60 60 i' " g :') e 10 .=100 ~100 a. TEMPERAJURE <"Cl a. IIITRAT( (,JO) ~ w ISEL'IN SECTION 14 Cl !SELIN SECT ION 140 21 APRIL 1980 140 .. 21 APRIL 1980

180 180

10 30 50 70 90 110 130 150 10 50 70 (KM> 30 90 110 130 150 DISTANCEOFFSHOP.E DISTANCE (HSIOIE (KIO ~ ';.':,( N NN NM '.:.'~N N ''"" N ''"'' 20 20

60 60 g g 2:100 ~100 0.. TEMPERATURE_< ·c, ....0.. UITRATE (,JO) ~ ISELIN S[CHON l!I Cl IS[LIN SECTION IS 141! 21 APRIL 19110 1,a 21 ~PRIL 1981!

180 180

10 30 SB 70 90 110 130 150 10 30 SB 70 90 (KM) Ill 13B 158 01ST Ali([ OHSH'.JRE DISIAM:E OHSIOl£ OIMl U UI ,l,...,J,.J Ullll. U UL.L..L.JLDC a:, en~ rn. a, (1\G,,-

181? l8il

10 )I! 50 72 !i3 112 130 152 Ila )il 52 70 9a Ill! na 158 OISIANC[ Q,FSHOR[ 0 ...11 0 IS TAl,( [ Cf f St(),<( (IOO

-Fiyure 3. Cross shelf sections between 28°43' and 30°19.'N. 60 ....,7- ;!:!ml o_ CL NITRATE (µM) tu TEMPER,'\ TURE <•[) w \ C.) EAST~ARO SECTION 12 0 EASr~ARO SECTION i2 21 t\PR IL 1980 1,11) 21 APRIL 198~

180 180 L_.__1-_,_-'-.-'l...--~-1..-l__...1--L..L-1-.L--L_i._j 10 30 sa 10 g~ 1 10 130 1s0 DISTANCEOFFSHORE CKM>

Figure 4. Trans-shelf section at St. Augustine. High nutrient concentrations indicate intruded water is reach ing the beach. THEINTRUSION ~F GULFSTREAM WATER ACROSS THE CONTINENfAL SHELFDUE.TO TOPOGRAPHICALLf - INDUCED UPWELLING

...

• J.O. BLANTON* L. P. ATKINSOI~*

L.J. PIETRAFESA** T.N LEE***

*Skidaway Institute of Oceanography, Savannah; Georgia 31406 **Department of Marine Science and Engineering, North Cafolina State University, Raleigh, North Carolina 27650 ***Rosenstiel School of Marine and Atmospheric Science, University of Miami, Miami Florida 33149 ABSTRACT

· Sumn'le.rbottom-temperatures along the continental shelf between·cape Hatteras and Cape Canaveral are abnormally cold in regions where isobaths· diverge. These regions are located north of capes and shoals which force : ·. the flow of shelf ·water to cha·nge vorticity whereby upwelling is induced. The Gulf Stream intrudes across bottom during summerto replace the upwel led water and accounts for the colder and more:stratified water found over· the northern Florida continental shelf and the North Carolina shelf.

·,i 1

.·INTRODUCTION

Wave-like perturbations.on .the western (cyclonii) edge of the Gulf Stream are advected north~ard: by t_he·strong fl ow of th~ Gulf · · Stream along the southeastern U.S.--~ontinental shelf (the South Atiantic Bight). Certai.n. mainfestations of these waves have beeri described a~ shingles (VonArx,· Bumpu~and Richardson, 1955), meanders

. '· (Webster, 1961), and.spin-off eddies (L~e, 19i5;: ~ee a~d Mayer~ 1977). Someconfusion pres.ently exists in the scientific communi_tyas to the, • • • 1 . exact process ·meant by. the use of these terms. The.re is genera] agreement that the waves occur each 5-10 days and propagate no~thward be tween Cape_Canaveral and Cape Hatteras at about 30 km/day (Legeckis, 1975; Legeckis, 1979) under a variety of .stable and unstable configur ations. The waves induce upwelling of colder North Atlantic Central Water,hereafter called Gulf Stream Water of slightly lower salinity at the edge of the shelf (Lee, Atkinson, and Legeckis, 1980). Under certain hydtographic conditions found in summer, the upwelled water can intrude along the bottom displacing large amounts of shelf water (Blanton, 1971; Blanton and Pieterfesa, 1978; Atkinson and Pietrafesa, 1980). Intrusions of Gulf Stream Water onto the shelf occur most dramat- . ically during summermonths when the shelf water has its lowest density (Atkinson, 1977) and Gulf Stream transport is maximum(Niiler and Richardson 1973). · Consistent with high transport, the horizontal density gradient through the Gulf Stream frontal zone is highest during summer and the Gulf Stream Water at the depth of the shelf break (ca. 50m) is usually of higher density than the adjacent shelf water. Thus the Gulf .' ..,J 2

·Stfeam Water intrudes shoreward underneath the shelf water {Blanton, 1971; Atkinson, 1977) and appears to be the dominant p'.ocess that maintains the vertical density stratification of the ocean_during summeron the continental shelf of th~ South. Atlantic Bight. Detailed studies of thermal.structure, winds and currents in the vicinity of Cape Canaveral {Lemin~, 1979) suggested tha~ frequent .intrusions of colder· Gulf Stream Water ~ould otcasionally- produce a narrow zone (< 20 km) of low· · surface temperatures ~orth of Cape Canav~ral.. Anomalously cold water temperatures in this re~ion were first reported by Green (1944) during July and August, and the prevailing northea~t wind stress north of Cape Canaveral was;thought to indu~e upwelling across the continental shelf. The cold coastal temperatures extend from Cape Canavera1 north to Jacksonvi 11e in summer(Taylor and Stewart, 1959). Periods of cold water are accompanied by a lowering of sea level- and. by wind stresses most conducive to upwelling. Leming (1979) dem6nstrat~d that the changing vorticity of streamlines following the curved isobaths north of Cape Canaveral could induce upwelling on the north si~e of the Cape, which is down stream under prevailing flow conditions in summer. It is the purpose of this paper to present further evidence that Gulf Strea~ intrusions prevail and perhaps are strengthened in regions where capes and shoals change the vorticity of the flow on the shelf.· This process can account for coTd bottom temperatures and strong vertical stratification of the shelf water observed north of Cape Canaveral and in the region of the Carolina Capes (Fig. 1).

TOPOGRAPHICALLY- INDUCED UPWELLING Upwelling in the vicinity of capes and other topographic irregularities

has been studied since the 19301 s by Japanese oceanographers (see Uda, 1959 for ' 3

a list of references). Arthur (1965) noted from Reid, Roden and Wyllie (1958)

' . that there appeared to be an intensificaticin of upwelling south of capes and.

points that extend out into coastal. currents on the U:.S. West Coast. Arthur (1965) demonstrated that changes in the relative vorticity along a sti~am line could enhance upwelling south of-cape~ for southward flow: The equation for conservation of vorticity is (Arthur, ·1965)

f dw = P..L+..sv . oz Dt · · · v dv where E; = R - dn = the vorticity (E;) in local coordinates of the ho.rizontal. flow velocity (v) along a stream line of· radius of curvature (R). The term~ is the ·• shear normal to the streamline (n-direction); ·s = 2 x 10-11m-1s- 1 = change of thE'. planetary vorticity (f) with latitude; w is the vertical velocity along z {positive upward). For the estimates below, we neglect t so that the vorticity is calculated by E; = v/R. The prevailing winds in July and August along the South Atlantic Bight exert . ·a northeastward wind stress (Weber and Blanton,· 1980). This induces a northeast ward f1ow on the shelf and places the north side of capes and shoals on the downstream s~de of this flow. Leming (1979) showed that north of Cape Canaver al, the term sv and Dt;/Dt cause upwelling for northward flow. Experimental data do not presently exist to evaluate the theory quantita tively. However, some regions in the South Atlantic Bight have favorable bathy metry to induce upwe11 i ng (Fig. 1). North of Cape Canavera1 and north of the Carolina Capes, the·isobaths between 20-30 m have radii of curvature as large as 50 - 100 kmwith the small value associated with the Carolina Capes.·, The iso baths downstream (north) of the curving portions diverge for several tens of kilo meters. Since low frequency currents tend to follow isobaths, the divergent flow 4 induces upwelling to preserve continuity. Thus upwelling occurs over a large region of diverging bathymetry. Wecan estimate w near bottom associated with the curving flow and apply the estimates to the entire region of diverging bath ymetry. For these estimates (Table· 1) we assume ·v = 0.3 m/s. The assumption on. R and isobath. areas required to compute D~/Dt and volume fluxes (Table 1) at each cape can have a wide range within one-half of an order of magnitude. Thus the differences between capes are not si~nificant but-the orders of magnitude are indicative. Specifically, the upwelling velocities are slightly smaller than 11t.ypical 11 values of 10-s m/s often quoted for upwelling. The intent of Table 1 is to indicate the potential sig~ificance of topo graphically induced upwelling to the hydrography of the continental ·shelf. We will present data to support the hypothesis that cold bottom temperatures and strong vertical stratification are observed downstream of the same regions where there occurs upwelling produced by the changing vorticity of the flow. 5 Table 1. Estimates of volume fluxes due to upwelling in areas of curving isobaths north of Cape Canaveral, Florida,and Cape Lookout, Nprth Carolina. Estimates are based on the curvatures of the 30-m iso bath, V = 0.3 m/s, and Wis estimated at a depth of 30 m.

CAPECANAVERAL CAPELOOKOUT f (s-1) 6.6 X 10-5 7.7 X 10-S R (km) 100 50 OE;/Ot(S- 2} -1.8 x 10-11 -2.4 x 10-ll s-v(s- 2} 6 X 10-12 6 X 10-l2 w30 (m/s) 5 X 10-6 7 X 10-G Area of diverging isobaths (km2) 80 X 40 100 X 50

3 1. X 4 3. 5 -x 104 Volumeflux (M/s) 7 10 _... .,, 6

DATAPRESENTATION

Wesuspect_ed upwelling to occur _in. areas where the flow follows curving iso- . . baths. The areas north of Cape Canaveral, Cape Fear and Cape Lookout.are thus prime areas of interest (Fig. 1). Ou~·data files (recent data) ·and those bf the . National Oceanographic Data Center (irchived data since 1953) were used to compute surface and bottom temperatures. Three depth intervals (1-20m, 21-40m, 41-60m)·

were averaged over one-degree squares from 28° N to 36° N. The region 28° - 30° N is called North Florida shelf; region 31° - 33° N is called Georgia-South Carolina

shelf; and 34° - 35° N is c~lled Carolina Capes. The d~ta for July and August are ·• summarized for bottom temperature and surface-bottom temperature differences (Fig. 2). Coldest bottom temperatures are found between 41 - 60 min the region 29° 30° n. Bottom temperature increases rapidly with increasing latitude through 31° - 32° N. A sharp decrease is noted again between 33° - 35° N. In the mid shelf regions (21-40 m) coldest bottom temperatures are found between 30° - 31° N. Bottom temperatures near the coast (1-20 m) are also the coldest in the region

. 1 29° ~ 31° N and 34° ~ 35° N. In summary, coldest bottom temperatures are found on the north Florida shelf. W~ile the gradients ar~ most dramatic along the outer sh~lf, colder temperatures also exist in the region closest·to shore. Similar patterns are reflected in the mean differences between surface and bottom temperatures. The difference is greater ~long the n6rth Florida shelf, an effect that s~ows up even in the 1-20 m depth range, btit most strongly exists ~long the 41-60 m depth range. The difference is minimum~ff Georgia and South Carolina and increases again off the Carolina Cape region. Thus, the thermal stratification of the shelf waters is greatest on the north .Florida shelf and the Carolina Capes.

.... 7

Meansurface temperature (Fig. 3) close to shore has a· max-imumbetween 31° - 33° N, off Georgia and South Carolina. In.August, there is·a minimum region. that extends iiCross the shelf off northern Florida. . . Twoquasi .:synoptic surveys of the ~eorgi a and no.rth Florida shelf were con- . . ducted bver a ten-day period in Juli - August 197~ (Fig. 4 ~nd 5). They ill~s- trate colder bottom temperatures off north· Florida .. DownstreamfrOm NewSmyrna . t . I· Beach to SaVan~ah(Fig. 4)~ isotherms were sloped parallel to . th~ .continental . I . .. . slope. Near the she if break, 18° - 24°C isotherms shoaled. upward, then downward.

'I farther inshore. 'rhe 18° ·_ 24°C isotherms pentrated into shallower water off NewSmyrna Beach and St. Augustine than in areas farther north (Fig~; 5). These isotherms formed a bottom temperature front sttonge~t in the south ~long the 40-m isobath. The front weakened and extended farther offshore with distance northward. The same region was resurveyed about ten days later, and significant changes were found.. Off St. St. Augustine, near bottom temperatures <20°C were present (Fig; 6). Surface water temperatures had decreased slightly between 50-100 km offshore. Conditions off Brunswick had remained essentially unchang~d. Bottom temperatures from 18° - 20° (Fig. 7) south of Brunswick had moved onshore and l bottom temperatures <16°Cwere found in a small region along the outer shelf. The cold bottom water originated from the Gulf Stre~m according to its temperature salinity correlation. Other surveys have shown the presence of Gulf Stream water near bottom ex tending into shallower water on the north Florida shelf (Fig. 8). Somefeatures in bottom temperatures were commonto those described above. The bottom waters at water depths less than 30 ~ 40 m were consistantly cooler off Florida than off Georgia. _Warmerbottom temperatures off Georgia were separated from the cooler

..... 8

water farther-south by a sharp temperature gradien~. The vertical distribution of temperature and salinity (not ;hown) indicated'that this gradient sep~r~ted · Gulf Stream Wat~r from less saline modified sh~lf water...... A quasi-synopt_ic survey was conducted off Cape Lookout, North Carolina, in August 1968· (Fig. 9) ... This survey was completed in less than 48 ho~rs dur- · ing moderate northeas~ ~ind stress.· Near bottom water east· (downstream) ~f the shoal area south of the cape was i0 -2° colder than that to the west. 9

DISCUSSION

·we.hypot~esize that the ~reason the downstream side of ca~es and shoals in-· duce upwelling (Arthur, 1965) and that this up~elling induces Gulf Stream Water present at the shelf break farther offshote to intrude across 'the shelf bott6m during ·summer_months. The prevailing northeast wind stress during summer (Weber and Blanton, 1980) induces northeast flow in thi in~er and mid-shelf re~ions. Thus we expec~ intrusions to be more prevalent north of Cape Canaveral(north Florida ~helf} than off Georgia and South Carolina. ·. Staiistical data (Fig. 2) supp6rt this hypothesi·s. -Regions of_cold bottom water are found off north Florida. Similar regions off the Carolina Capes are . suggested but the manner in which the data were sorted (by latitude) may not have sufficient spatial resolution for that region since the shoals associated with the capes cut across lines.of equal latitude. The surface-bottom temperature differences are a 1so consistent w"ith the hypothesis (Fig. 2). Intrusions are the primary process responsible for main taining this difference ~Jhich is clearly gr.eater off the. no.rth Florida shelf than

. , .. ' ' ' found along th~ bottom in shallower_ depth~ off Florida than off Georgia (Fi~. 5, 7 and 8). Even an April survey depicted this situation. The temperature dif ferences would not clearly show the effect in seasons other than summer.because surface temperatures are colder off Georgia than off north Florida except in summer. However, the upwelling process for flow along curving isobaths would occur regardless of season. A_small amount of data off C~pe Lookout {Fig. 9) _suggests that upwelling occurs on the downstream side.· Evidence shows that intrusions ih Onslow Bay 10

enter th_e embaymentjust east of the Cape Fear shoals then spread throughout· the embayment(Blanton and Pietra_fesa, 1978; Atkinson ahd Pietrafesa, 1980)~ This is _co~sistent .with the.:hypothes1s that cape induced upwelli·ng induc~s , . ·the Gulf Stream Water. to intrude into the upwelling region-; in thi~ case down~_ .·, stream of .Cape Fear. The riorth Flori~a ~helf is a·r~gion of spr~ading isobath~. As a given vol

ume-·of- fl·ow p9sses· ·arou.ndCape Canaveral and begins to follow the ·isoba~h·s, the spreadi_ng flow :is_compensated by upwelling to replace the mass loss by· horizontal spreading of the streamlines. In summer, this loss i? made up by inducing the Gulf Stre~m Water to intrude across the.bottom to shallo~er water. -I~ regions •SI ·. ' where the isobaths d6 n6t spread (Georgia - South Carolina shelf), th~re is no inducement for an intrusion of this type to occur. The Georgia - South Carolina shelf has minimumdifferences in surface and bottom temperature in July and August (Fig. 2).

Wind-Induced Upwelling Alongshore coastal winds favorable to upwelling prevail in the South Atlantic Bight during the months of June, July and August (Saunders, 1977; Weber arid Blanton, 1980). · Upwelling-favorable winds were thought to cause the cold coastal tempera tures observed in northern Florida (Green, 1944; Taylor and Stewart, 1959). Up welling-favorable winds are equally persistant off Georgia and South Carolina and yet we find no evidence that intrusions occur there in summerto the same degree found off the north Florida shelf.

Wave-InducedUpwellin[ The upwelling of Gulf Stream Water is induced by waves in the Gulf Stream •' .. 11

front.· Satellite imagery shows thatthese waves are a consistent feature along the shelf break {Vukovich, Crissman, Bushnell and King, 1978; Legeckis, 1979) and appear to grow to much larger size north of the Charleston bump (Fig. 1) in the region of the Carolina tapes. They occur with a frequency of approximately 35 events/_year, i_.~., once/IO days, along the North Carolina shelf. Evidence suggests · that these frontal_ disturbances.intensify as they propagate north of Cape Canaveral , and presumably any upwelling the_y would produce would likewise intensify. One

mi~ht hypothesize that upwe11 ing of Gulf Stream Water would intensify- with 1atitude and so would 1ntr~sions~ particularly with the.aid ~f the upw~lling-favorable : winds in summer. Our .data (Fig. 2} do not supp:ort this hypothesis.·~

Speculations We·have shown d~ta ·consistent with the hypothesis that the d1i~rgenci due to spreading isobaths and related. to vorticity changes in flow following curved isobaths induces upwellin~ that in turn causes denser Gulf Stream Water to flow shoreward along the bottom. This might be viewed as a process that ~nbances the

upwelling produced at the she 1f break by waves fo. the Gulf Stream front, ard we . . pose the question of whether the upweliing induced by the ~apes might have a feedback effect.that could intensify cyclohic circulatio~ of the ffow due to these events. Consider a wave pro~agating nofthward a1on~ the outer edg~ of the north Florida shelf. Colder Gulf Stream Water passes upward and.shoreward through the wave (Fig. 10). The shoreward motion would be limited in the shelf area where no upwelling was induced by wind or topography, but the shoreward motion i_~~-, intrusion) is enhanced by the topographically-induced upwelling. Thus more water must pass upward through the wave to.support the divergence in shelf flow. Intensified upwelling caused by divergence in the overlying fluid has an 12

analogy in meteorology. Sur-face. low: pressure areas frequently intensify as they propagate and pass beneath the exi't region of .an upper level (300-500 .mb} trough in the jet stream ·(Haltiner and M·artin, 1957). ·The.low strengthens be~ cause the vertftal motion already presen~ is enhanced .. :

· We.do· not presently have ·convincing. evidence that meanders passing· along the ·northern Florida shelf are intens.ified by s:preachng isobaths. This .h~po thesis is. to be tested by future research in the South Atlantic Bight.• 13

CONCLUSIONS

Historical data show·_that July and August bottom temperatures on the con.:..·· tinental shelf off n~rth. Florida and _in the ~arolina ·Capes r~gion ih Ju)y and; August are cooler than those found farther north. je~tical stratification is also stronger in the·same areas which suggests that Gulf Stre~m Water· intrudes a~ong bottom into shall~wer water. · Shelf flow.following curved isobaths can induce upwelling (Arthur, 1965) and

· regions of curved i scibaths a re found _north of Cape ca·navera1 and in the. Carolina Capes region .. Upwelling in these regions is replaced -by Gulf StreaITLWateruplifted . . . ·• by wave-like perturbations.to· levels above the shelf break. Replacement by Gulf Stream ·wat~r requires it to intrude acrosi the bottom into the regions of curved isobaths. Intrusions off Georgia and South Carolina appear much weaker and are not indicated in the historical data. No curved isobaths exist here.over a sufficiantly large region to induce upwelling. Predominant winds in _July and August are favorable for upwelling throughout the South Atlantic Bight. If wind~induced upwelling were the sole agent respon sible for the decreased bottom temperatures, we ~ould expect to find evidence of these off Georgia and_South Carolina. It is possible that Gulf Stream Water could intrude into these regions during spring when solar radiatio~ begins to warm the cooler shelf water and produces a stratified water column. ·A density-driven flow such as the gravitational flow discussed in estuaries aided by northeastward wind stress could induce Gulf Stream Water to intrude closer to shore. The flow is obviously maintained for only one or two months after which winds and tidal mixing destroy the stratification. Off.north Florida and North Carolina, Gulf Stream intrusions induced by shelf flow divergence maintain the stratification during the summermonths with wind playing a less dominant role. ;· 14 . ·ACKNOWLEDGEMENTS.

Weare grateful to Dr. Gerold Jandwitz for stimulating discussions in the. early phases of this study .. Wethank Dr. DaviiMenzel for his helpful cormients. on the original manuscript. The following support from the U. S. Department of Energy is gratefully acknowledged: Contract No. EY-77-$~09-0125and Contract No. EY-77-S-09~0889, t~ Skidaway Institute of Oceariography; Contract NQ. EY-76-S-09-0902to No~th Carolina State University, and Contract No.. EY-76-S-05-5163to Universit,Y of ' Miami. Weacknowledge the support of Duke University through ·NSFGrant G-17669, which provided.the R/V EASTWARDfor the acquisition of data utiliz~d in·Figure 9 and B~reau of Land ManagementContratt No. 209-05 which supported the acqui sition of some of the hydrographic data.

0 . 15 LIST OF FIGURES

. Fig. 1. Location map of the South Atlantic Bight. Stippled areas mark curving isobaths which diverge for predominant north ward. shelf flow.

Fig. 2. Mean bottom·temper~tur~s and mean differerices betweeri surface •· arid bottom temp~rature fo~ histor1cal d~t~ since 1953 ii the South Atl~ntic.Bight. See teit for·details.

Fig. 3. Mean•Surface temperiture ..for hist~riial :d~t~ si~ce 1953 in the·

South Atlantic Bight. See text for details.

Fig. 4. Temperature transects on the continental shelf betw6ien NewSmyrna Beach, Florida, and Savannah, Georgia, 26-28 July 1978. (a) New Smyrna Beach; (b) St. Augustine; (c) Brunswick; (d) Savannah.

Fig. 5 shows location of transects. Stippled isotherm band is the same as that shown in Fig. 4, 5 and 6.

Fig. 5. Bottom temperature chart (0 c) for 26-28 July 1978. Transect locations of Fig. 4 are shown. Stippled isotherm band is also stippled on sections in Fig. 4.

Fig. 6 .. Temperature transects 3-6 August 1978 for (a) St. Augustine; and (b) Brunswick. Fig. 4 shows location of the transects.

Stippled isotherm band is the same as stippled bands in Figs. 4 and 5.

Fig. 7. Chart of bottom temperature (°C) as in Fig. 5 except for 3-6 August 1978. .J 16

Fig •. 8. · .More bottom temperature charts illustrating that .the front

in bottom temperature fs located .•in .shallower depths south of ' ' .Jack_so~vi11e~ Florida. ;(A) 15-23 Aprfl 1978; ·cs) 4-7 .Jul·y 1977.

.. Fig. 9. Bottom temperature chart across the shelf. . off. Cape. . . Lookout,. . North·Carolina .. Flow in ·summer is predominantly eastward and bot torn·temperatures are colder on· the down·stream side.

'' . Fig. 10. Schematic representation of a sun1mer intrusion of oceanic water.

. qnto the continental snelf. The wave along the cyclonic edge: . ' . , . . ' ofthe boundary current propagates poleward .. Oceanic water is . . . . pul l_ed upward througtl the wave trough. and sprea·ds shorewar,d a- . . ' ' ·~ ' 0 • long the ·ih~lf bottom fnto regi~~s w~ere fl~~ ciri the shelf· diverges.· A bottom temperature front marks the extent of the

intrusion. 17

LIST OF TABLES

Table l. .Estimates of volume fluxes due to upwelling in areas ·of curvirig isobaths north of Cape Canaveral, Florida and Cape lookout; North Carolina.- Estimates are based on the curvatures of the 30-m

isobath, V = 0.3 m/s and-Wis.estimated at a depth of 30 m.

REFERENCES· Arthur R. S.~ (1965} On the talc~lition of vertical ~otion·in eastern boundary currents. from determinations of ·horizontal motion. Journal . , . ' of Geophysical·Research,- 70(12}, 2799.:.2803~ . . - .. -· : Atkinson L. P. (l9t7) · Modesof Gu-lf Stream i~trusion into South Atlantic·

Bight Shelf Waters. Geophysical Research Letters, .1_(12), 583-586 .. -Atkihson L. P~J and_L. J. Pi~trafesa ·(1980) A flushing model of Onslow Bay, North Carolina based on intrusion volumes.-- }Journal 'of Physic.al ·Oceanography,(in press.)~ Blanton, J.·o. (1971) Exchange of Gulf Stream water with No~th C~rolina -~ Shelf Water in OnslowBay during stratified conditions. Deep-Sea· Research, 1.§_, 167-178. Blanton J. 0. and L. J. Pietrafesa (1978) Flushing of the continental shelf south of Cape Hatteras by the Gulf Stream. Geophysical Research Letters, §.(6), 495-498. Green C. K. (1944) ·Summerupwelling~- northeast coast of Florida. Science, 100 (2607), 546-547.

Haltiner G. J. and F. L. Martin (1957) Dynamical and physical meteorology. McGra.w-H"ill;New York. 470 pp. Lee T. N. (1975) Florida current spin-off eddies. Deep-Sea Research,

22, 753-765 Lee T. N. and D. A. Mayer (1977) Low-frequencycurrent variability and spin off eddies of-southeast Florida. Journal of Marine Research, 1§.(1), 193-220. · 19.

. Lee T. N., L. P. Atkinson. and R. Legeckjs (1979) Detailed observatfons · of a Florida cu·rrent spin-off eddy on the ·Georgia continental shelf. · (Submitted to Deep-Sea Research) . . Legeckis R•. (1975) · Application of synchronous meterological satellite data to the stu~y of time-dependent sea surface.temperature ~h~ng~s along ~he boundary of the Gulf Stream; ·Geophysical.Research Lette'rs, ·

-,.2(10), 435-438 .. Legeckis R; (1979) Satellite observations of the influence of ~ottom topo

graphy on .the $eaw~rd. deflection·of . . 'the Gulf St.reamoff Charleston~. s·outh Caro1 ina. Journa 1 of Physical Oceahography, 2_{3), 4!a3-49·7.· · Leming,_T.D. {1979) Observations of temperature, current and wind variations off the central eastern coast 6f.FJorida. NOAATechnical Memorandum NMFS-SEFC~6(U.S. Dept. of. Commerce·)..172 pp.. Niilerp, P. and W. S. Richardson (1973) S~asonal vaiiability of the Florida Current. J. Mar. Res., l:!_(3), 144-167. Reid J., G. I. Roden and J. G. Wyllie (1958) Studies of California Current Syst~m. California·Cooperative Fisheries Investigation Progress~Report,

' . ' ' July 1956-Jan1,1ary1958 .. Marine Research Commiss·ion,California, Depart- ment of Fish and Game, Sacrc1mento,California. Saunders P. M. (1977)· Wind stress on the. ocean over the eastern .continental shelf of North America. Journal of Physical Oceanography, Z.{4), 555-556. Taylor C. B. and H. B. Stewart, Jr. (1959) Summerupwelling along the·east coast of Florida. Journal of Geophysical Research 64(1), 33-40. 20

Ud~M. (1959) Oceanographic Seminars. Fisheries :Research Board of.Canada. Manuscript Report Series No. 51, 110 pp~ van Arx W. S., o: Bumpasand W! S. ~ichardsrin .{1955) On·the fine-structure. of the Gulf Stream frorit. ·Deep-Sea Research·, 146~65. Vukovich F: ~-, B. M. Crissman,.M. ~ushnell ·and W. J~ Kin~ {1978) ·Sea . . su~face· temperature ~ariabili~y a~alysi~ of potential OTEC~ites utilizing satellite. data~ Final report to the U.S. Department of Energy {Con~r~ct No. EG-77-C-05-5444).· 155 pp.· WeberA. H. and J. 0. Blanton {1980) Monthly mean wind fields for the ·• ._,___ South Atlantic Bight. ,Journal of Physical Qceanograpt\Y(in Press). Webster F. (1961) A description of Gulf Stream meanders off Onslow Bay. Deep-Sea Research,~,· 130-143. 82 81 80 79 .. 78 77 76• . · 75 ..', ·

Cape Hatteras· . \ :::.::: Cope · . Lo

Cap 34

.~ Sovann 32 .·. ...

. ..c. 0 ._ --o\....._ Q) 0 ,,._..c 30 z O(f)

_JLI- __ ._· ___ _ 29 Cope Conaverar

..... Lat. JULY 35-36 34-35 NORTH CAROLINA·. 33-34 CAPES 32-33 . GEORGIA 31-32 SOUTH CAROLINA 30-31 NORTH. 29-30 FLORIDA 28-29 1-20 21,·AO 41-60 1-20 21-40 41-60

MEAN BOTTOM TEMf:>ERATURE MEAN TEMPERATURE . (SURF- BOT. DIFFERENCE) AUGUST 35-36 34-35 _,O9r1 33-34 \~·~. ... '"°l 32-33 ~. 31-32 .. ~~~~~! w 30-31 r ~ 29-30 . .3t¥f.l..· .r '28-29 ·t 1-?Q: · 21-40 __4!-60

,.._ ......

Lat. · JULY 35-36· · 34-35 .~27· ·.. 33~34 . "-;-~ . ...._ ------.:...... ~, ...... _, ____ ---__ -- --·---·. 32-33 . - -:-- ,-- _J 31~ 32 '. ---.:__ __ 2a---- 30~31 . 29~30 - ·28-29 . " l · 1-20 · 21-40 41-60

MEANSURFACE TEMPERATURE

AUGUST 35-36 34-35 33-34 32-33

31-32 _. 28 ,,,,,------30-31 \ . 29-30 '------28-29 .~28-=== 1-20 21-40 41-60 ·•

28 16 50 26 24 14 22 12 i 20 100 18 10 16 10 150 14 -~ I 1 :r: 12 I a..I- UJ 200 Cl 10 250

300

350 4- . /.. ::· ... :· ..:::

. .

+ ...... _ c ... e...... run . Swick

. ..st. Augustine , . ....;...-B --+--- +

Bottom Temp.·(°C} GEORGIA BIGHT SURVEY 26- 28 July 1978 .

...... DISTANCEOFFSHORE ·(KM) B 20 ·40 60 80 . 100 120 140 ------:----21.s---- 2s29 ~ 20 -- -.:,.:-:.. ;: .. , ~ . J-7 ~\ ~ Brunswick · · · ''%%\X',, 26 ~ ~ 40 4 August 1978 . ·· . .· . · Temperature .. ·· . . · · ·.

A ~--_,_.,---"""""'--~o!..---,l- __ ...... ,.,...... _27 26 ~-~ - 20 ,,-:·:·:•:?@\Wt}·•--~. 2 \ff:/ .. .-v. -:r: ca 40 0 t-- St. Augustine -· (L ~ w 5 August 1978 0 60 Temperature ·

i?:=I "it' y I :a 80 It __ ___.__ -~ -- ..,..---- h . I ..• , ,•

...... +

~- "-' ·. ., t1;]\iF*(:r~r1 .·8ruosWic1r .

. ugustine st. A . +

"D-

Bottom Temp.(°C) ' GEORGIA BIGHT SURVEY I 3-6 Aug. 78 I . i

...... ,.

. + ~-

9.) . . c5 ~ ..

\ .

. \ . . . '- + + ...

. Bottom Temp. (0 G) · . 15-23 Apr. 78 ..

Bottom Temp. (°C} i . I 4-7 July 1977 ! . CAPE LOOKOUT ~:···:; ;:~ ... ..♦• •• ~ /' .

... N·····...... \ . ..., 1/f,,.e •-· ....0•

~ -t> ~c:, -; ~.

. \ ...-

.. . . Q) C ·-Q) \... (l.) V) :::,....c: ·-o·- -0 I .Q · Ol O ·:;C: =·-V) \... u:J

I . I l . -· ....~. . .

..... lll:T/\f Ll:ll OIISl:l\\11\TIONS 01· /\ GlJLF STl!EM!

FIWNT/\L FllDY ON 'llll: CEC1HC:I/\ CONTINENTAL SHELF, APRIL 1977

Thomas N. Lee*, L;1rry P. /\tki11so11"t- and Richard LegeckislY

Dec.licatec1 to \falter O. Dtling; he h;1s gone, but his spirit lives.

Abstract. Satellite, hydrographic and moored current meter data arc used to sh0\-1 the effect of Gulf Stream fro11t,ll disturbances on lm-1 frequency current ;111d te111pcraturc variahi li ty, water t!x changc and 11utrient flux i11 the outer region of the Georgia shelf.

Pcrturli:1tions of the Culf Stre:1111 cyclonic front arc commonly observed as folded 1-1:1vcpatterns rn routine satellite-derived analyses of the c:ulf :,trc:un 1•1cstcr11 boundary bet1,een Cape Hatteras and Miami. Those Jist11rb:111ces consist of soutl11-1ard-f10\ving 1,arm filaments or of stre;imers ne;_n- surface Gulf Stream water, 15-20 m\ deep, 1-11\ich can ex.tend ~S to 40 k.111over the outer shelf around a cold upwel led core. IJ01s•nstream dimensions of the warm filaments reach 100 to 200 km in the reg ion from Jupiter, Flor.i.cla, to

Ch:irfoston, So11th Carolina, 10 to 50 kill south of Jupiter, and

200 to 300 km hct1•1ecn Charleston and Cape Hatteras. 111ese fca~

turcs arc defined a'.; cyclonic, cold-core frontal eddies due to

their fl01•1 and 1-1ater mass prupcrt ics, They appe:n to form fro.Ill

->- - University o[ ~li :1mi, l

t Ski

V National l:11viron111cnt;1l Satelli.te Scrvi~es ST~, National Oceanic and Atmospheric A

average of one GVcry two weeks, !nit with .considcraple monthly

variability. TI1ey c_an exist up to three weeks and -travel to

the north with t:he same ph.asc speed as the. waves, approximately

40 cm/ sec ..

1ne cyclonic circulation in frontal ~ddies pi-ov~des a means

for rapid shclf/G\tlf Strea1!l water cxch?trige ~- They appear to

· control the. residence. time of the outer··shelf wa_ters.; which is

. defined· _as the )11ean separation time between. eddy events, or

approximately twoweeks. Upwelling in the cold core was ... observed to extend ·into the_ euphotic zone ( 45. m dep~h) and

shoreward (3S tq 40 km) beneath the south~ard llowin~ warm

filament in a bottom intrusion layer 20 m thick. The annual

nitrogen input to the shelf ·waters by this proce·ss is estimated

a~ 55,000 tons/year, which is about a factor of two greater

than all other estimated nitrogen sources combined, and can

support an annual carbon production by pJ-lytoplankton of 32 to -2 -1 64 gCm yr with no nitrogen recycling.

INTRODllCTI ON

Sa tell i te-derivecl sea surface temperature (SST) data products routinely

distributed by the National Oceanic a11ll Atmospheric Administration (NOAJ\

NESS, 1974) and the U. S. Naval Oceanographic Office (NAVOCEANO,1975)

consistently reveal folded-wave patterns in the western boundary of the

Gulf Stream from Cape Canaycral to Cape Hatteras. These features have been investigated in the Florida Straits and termed "spin-off eddies" due_

to their cyclonic rotation, cold core and exchange of heat and salt with

2 .f

ndj ace1it she 1 f i-1atci;s (LEI:, 19 75; LEE ~111cl MJ\YER, 1977). That term:ino logy

is discontinuc·d in f,nior of a more general definition of the features as

"frontal eddies".

Edge-eddies of this type _appear in the surface waters as warm, southward-·

or1.cntc

. cores. 'llicy · were first ·measured as a, succession of overlapping thermal seg.::·

ments tcrmell "shingles" in the pioneering mapping of· the Gi.11f Stream cyclonic

front off the southeast U. S. by VON J\RX,·,13UMPUSand RICHARDSON(1955}. In.

the -Flo_ricla Straits vo_rtex diameters were found to be on the order of 10-30 km

with downstre;m a~es two to three times larger than the cross-stream (LEE,

. . ·• . 1975; LEE and ~IJ\YER, 1977) •.. 'I11ey were obsc-rved to travel northward along'

the sheifbreak at spec.els _less th:{ri the mean speed bf ·the Gulf StrcffiTl. 1hey

occurred on th·e. average ·of on·c p~~r 1,cck .i.n the Florida Straits and re·sl!lted· ·

in large amplitude, subticlal, cyclon.ic flow reversals over the shelf that

distorted the temperature and salinity fields to a depth of approximately

200 m (LEE, 1975; LEE and ~li\Yl:R, 1977). ·111c classification of these features

as eddies stems from current. meter observations· of the cyclonic flow

reversals pi·oducecl by the passage of the warni filaments that appear tq

be geostrophically coupled to the cold uplifted cores. However, these

features may never become completely

the more common eddy types, such as warm and cold core rings. The shingle

shapes of frontal eddies arc more similar to "roll vortices", which are pro

duced by wave-like ro_lling up of ,1· shear zone (ROUSE, 1963); or "wake

vortices" which are fonnccl in the wake of islands as Karman vortex streets

(WILLE, 1960). LEE (1975) was able to reproduce the observed cyclonic flow

reversals and ~ccurately preclict eddy spatial dimensions and circulaticin

using a kinematic, diagnostic mo

vortex with a uniform background current. Sailing captains were probably

3. th~ first to be aware of the _transient· southward flow generated on the

shoreward side of the Gulf Stream by these events·. STOMMEL(1965) reports

· that in 1590 · t1 sailing vessel bound from Florida to Vi_rginia had to stand

far out to sea to avoid "eddy currents.setting to the south and southwest"

(KOHL~ 1868).

Satclli te. u~·crmal imagery indicates that frontal eddies evolve from

growing wave-like meanders of the Gulf Stream cyclonic front (DE RYCKE and

. RAO, 1973; LEGECKIS, 1975; STlJ~!PF and RAO, 1975). 11ms an instability pro

cess may be :involved, due. either to the vertical shear (baroclinic instability)

or horizontal shear (bar,otropic instability) across the front (ORLANSKI,

1969; ORLANSKI and COX,. 1973; .NIILER and MYSAK, 1.971). It- has been sug

gested that atmospheric forcing call trigger n disturb;rncc in the front that

travels northward with thl! Slrcam as an unstable wave, eventua:ly evolving

II into a cyclonic eclEe-cclcly (LEE and ~!!\YER, 1977; DUING_,MOOERS and IJ:E, 1977;

LEE and BROOKS, 1979). EdLli cs can form within ·a week of meander generatio_n

and then persist for ~other one to two weeks (LEE and MAYER, 1977; LEGECKIS,

1979). 11iey nppear to ·be forming all along the Gulf Stream boundary at any

time of year, and may at times serve to dissipate kinetic energy from

the mean flow (LEE, 1975). They also serve as an effective mechanism for

exchange· of.shelf and Gulf Stream w~ters.·

Frontal eddies appear to grow to larger dimensions north of Jupiter,

Florida,· where the shelf begins to widen. TI1e size increase occurs primarily

. . in the downstream direction, resulting in elongated tongues of warm Gulf

Stream water 100 to 200 km in length. I\ second e longat:i.on is observed. north

of the "Charleston bump", a topographic anomaly of the slope extending

seaward into the Stream, where downstream dimensions can rca.ch up to· 300 .km·

(LEGECKIS, 1979). TI1c Gulf_Strcam i~ observed to have a quasi-persistent'

4 eastward di:;placcmt\nt downstream of: the "bump"; which is believed to. be the cause of the enhai1ced mca11(fors and eddies ·between· the· "bump" and c·ape llatte·r.is (Pil:Tlt/\FES(\, ATKINSON and. BLAfJTON, 1978;. BROOKS and BANE, 1978;

RANE and BltOOKS, .1979; ·LEGEC:KIS, 1979) •. LEGECKIS (1979) classified the wave-like features of the Gui£ Stream surface front pctwccn Charleston and

Cape Hatteras· into fi vc types.· '!he· type V pattern: had large amplitude . . cast-:west displacements accomp,miecl by· the wa;m. filament str~'1cture,. suggestive of eddy development. BROOKS and BANE (1980) used current meter da.ta and satellite.imagery from the outer shelf ar-ea off Onslow Bay to show that the type V filament structure was associated with cyclonic si.1bsurf~ce... current reversals similar· to that found :in the Florida Straits (LEI;: and MAYER,

1977) and off the Georgia Shelf (LEE ~rnd BROOKS, 1979). East of Cape

Hatteras, Gulf Stream meanders arc no longer restricted by a shallow shelf as along the southeast U.S. coast and the well-known warm and cold core

"rings" develop n~rth and south of the Stream, respectively (SAUNDERS, 1971;

1HOMPSONand GOTirlARDT, 1971; GOT'IHAROT, 1973; FUGLISTER, 1972, BARRETT, 1971;

PARKER, 1971).

OBSERVATIONAL ME1110I)S

Current meter, wi11cl, hyclrographic and satellite SST data were obtained

for the Georgia shelf and bordering Gulf Stream regions during April 1977

as part. of a Department of Energy supported :interclisciplin;:i.ry study of the

physical, chemical ai~d biological processes affecting the southeast U. S.

continental shelf. Current :incl temperature measurements were taken every

20 minutes with Aandcraa current meters located 17 m below the surface and

3 m above the bottom in a 7-clement box array of subsurface, taut-wire

moorings. (Fig.' 1). The array was initially

s December 7, 197(,, to /\in·il 11, 1977 ("Winter 76/77"), and then expanded to

9 moorings and rcdeploye

November 5, 1978 (".Summet 77"). Between box array deployments a 2-element an;ay (moorings E arid F) was installed for a 2.5 mont.h period from April 11,

1977, to June 30, 1977 ("Spring 77ir). lnforrnatfon regarding these m.ooring deployments is given in Table 1. Wind speed and direction time series were constructed froia 6-hoi'rrly es.timates made for the center of the box array.:

The.s·e· estima1;.es were produced· (l'ARTJ\G/\S, 1978) by determining the surface wind from the :isobai·ic c9nfiguT,ition shown in surface weather charts pre-

. . . . ~ pared. by the. ~ational Hurricane Centei·, Miami, Florida, and utilizing ..wind observations from cc>astal statio.ns., .weather buoys a.ncl ships. Effects of frfrtion on the wind speedancl direction estimates were also taken into account. llourly values of sea level height were obtaii1etl :£mm the N'ational ·

Ocean Survey, NOAA, Rockville, Maryland. Low:fre.quency (subtidal) _time

.series of all data sets were generated by smoothing the orig.inal data with a 40-hour low-pass Lanczos filter k~r11al to remove· vaTiance associated with tidal. and .inerti~l fluctuations .. Diurnal tides ate attenuated by .more 5 than 10 by the filtering operation; which restil ts .in a 4-day truncation at both the start and end of the. time serieso The filtered data were suhsarnpled every 6 hours and current and wind vcctoTs rotated clock1vise 3Q degrees into a "norm::rl" coordinate system in which the off-diagonal of the Reynolds stress.

is near zero ( F0F0N0FF, l 969). 'I11e rotated vectors were converted to cross

shelf (u+ at 120°T) and along-shelf (v+ at :~u0 T) components. 111e orientation

of the along-shelf component agrees .w·ith the alignment of local isobaths to

within 5% or l 8 dc[!rces at al.1 · cnrrent meters and should therefore be the

primary axis for low frequency flowo

6 ,I

llydrogr:1phic sections were 1i1ade at the cross-shelf current meter locations

and at a

. ·. Transect idcnt;if.ications · ·arc presented in Table 2. Salinity _and temperature

were determined with a Plcssey 94·00 C'r°r>-interfacecl to a }lewlett-Packard 9825

desk-top compt1tcr aJ~d Ke1mcdy incremenUtl recorder. · Additfonal temperature

profiles were obtained with an expendable bathythermograph system.- Wat.er . . sampl~s ~ere taken at ielecte

temperature and salinity were made underway with a Bissett-Berman 6600 T thcrmosalinograph. Phosphate, silicate an

and DUNSTJ\.N(1976)" 111e compl etc hydror: raphic clata set and discussion of

1 methods can be found in 0 Mi\LLEY, ATKINSON, SINGER, C:I-IJ\NDLERand LEE (1978) o

Sea surface temperatures (SST) were measured by a Very High Re.solution

Radiometer (VIIRR) on a polar orbiting satellite (NOAJ\-5) operated by NOAA.

The scanning radiometer. measures thennal infrared (IR) radiation emitted by

the earth and atmosphere in the 10.5 to 12.S pm band and visible radiation 0 reflected in the 0.6 -_0.7 11mspectral ban

SST changes of O.S C with a spatial resolution of l km at naclar. The

satellite data arc available over the Gulf Stream twice daily at 0200 and '

1400 GMT. According to IHU\UN (1971) clouds can completely obscure the SST

features associated with the current while atmospheric water vapor degrades

the SST gradients observccl under. cloud-free conditions. Observations of

•the Gulf Stream arc also limited by the seasonal disappearance of recognizable

SST gradients during the warmer 'mQnths of the year (usually June· through

October for the Gulf.Stream south of Cape ,llattcras). Although atmospheric

moisture limits the accuracy of satellite derived temperature measurements,

7 ..... the positi()n ··of .the western SST boundary can be established accurately in

. cls)Ud free a~·c:1s. This is- accomplished by geometrically ·correcting the VHRR

data ns

images on: a b~1thymetry map (llCIIlJPI, 1968). 111e Gulf Stream boundary can

be located within an uncertainty of· 5 'km provided landmarks-are recognizable

in the images. f\ctwccn 'J\pril 7 and 17·, 1977, ,nearly 20 IR observa_tions of the Gulf

Stream were made during unus11ally cloud free conditions. This allowed the .··

evolution of two frontal ccldi:cs t.o be descrilied in a time series,of satellite

derived SST.

. RESULTS

Synoptic Observations

J\ shipboard telefax system was used on the mooring exchange cruise of

April 1977 to receive satellite SST images from the NOAA-NESS field" station

in Miami, Floritla. \'Ji th .·the use of this information and high speed ship

board mapping of the Gulf Stream cyclon1c surface front, a shingle feature

(eddy F) was· 1ocated and inappcd on two successive occasions between J\pril.12

and 16 (f:igs" 2 anti 3)" Each map.ping took approxim:itcly 1.5 days including

hydrographic sections matte through the event. Hydrographic transects were

also made before anti after the mapping period (Table 2.).

TI1e· disturbance was :first m:ipped on J\pril 13 and 14 (Fig" 2). It

appeared as an elongated tongue of wa11n (22-23 C) Gulf Stieam water extruding

out of the front and over the shelf tmv:1nl the south :irouncl a cold core

(21 C). At this time :it w:is about 160 km long and 40 km 1vicle. When next

ohs~rvcJ about 1.5 days later (Fig. 3) it was almost 220 km Jone and 70 km

wide and maximum temperature in the warm tongue had increased to 2tl C. This

rapid growth appeared to occur by cxtcns ion of the northern portion of the

8 structure along the :front, perhaps due_to the strong horizontal shear ac_ross the Gulf Stream boundary. ·nic cool core of 21 C water had also dol1blcd in si zc. 111c center of the cool core was advected to the north at·a speed of about 30 cm/sec._

A satellite· VIIfm-rn :image taken on April 16, 1977, at 0200 GMT (orbi1;

3223 N0M-5} clearly displays this feature and its Gulf Stream. roots (Fig-. 4).

The warmest water is represented by the darkest tones of gray. Both the . · western ;rnd e~1stern. frontal boundaries of ·the Gulf Stream are clearly defined.

TI1e SST data ivere. averaged in 2 km squares in the box outlined in Fig~ 4 and are displayed as alpha-numeric ch;iracters at intervals of about one degree Celsius in Fig; S. 'l11e contours arc used to separat~ areas of similar: temperature. Cor.ip;irison of the temperatures in Fig·. S with those obtained on. the hydrognphic survey on April 14-16 (Fi_g.· 3) shows that the coldest water detected in the.core of .the featuTc measured 21 C (ship). and,20.1 +

0. S C (VIIIUl). 'I11c wan11c:;t Gulf Stream water adj a cent to this cold core was

25 C (ship) and 23.6 ~ 0,5 C_ (VIiRR). TI1is shows that the satellite measure ments were about 1 to 2 C lo1~cr than the ship measurements. Estimates of

the atmospheric water vapor correction to the IR temperatures are not

available, but this comp~rison iu\plies that -the atmosphere was relatively

dry. TI1e cross-str~;11n SS'I'clfffcrcnces in the -vicinity of eddy F are 4 C

(ship) ilncl 3 to 4 C (s;1tcllitc). Thi.s indicates th:1t the SST gradients are

represented in Fig. 4 with.in an uncertainty of 1 C. A quantitative inter

comparison of shipl>o;n:d ;ind satellite-derived SST :in this event (13ROWNand

EVANS, 1979) found a mean square difference of 1.2 C between several thousand

ohscrvation pairs, with the shiphoarcl mci1surcments being h_ig.her. The. ·

difference between the stanchrcl deviations computed from the two data sources

was only 0.1 C. TI1c general position, size·an

9 from shipboanl SST ~lata (Fig. 3) arc also in clos·e agreement with that shown by the. satellite rn. i.mage. '11ie length and width estimated from the dark filamen.t shown in the satell.ite image (Fig. 4) are 230 and 70 km,· respectively, which arc within about 4% of the shipboard estimates arid indicate that the observed spatial growth did not seriously affect the ship

,deri. ved estimates. .

A composite of hyclroiraphic sections mad~ across eddy F. from.1900 G~fr on April 14 to UIOo·on April 16 arc sliown :in Figs. 6 and 7. The northern most. section off Savaniiah on April 16 was made from east to west across moorings E and F just north of the event. (Fig. 3). The steeply slo_ping 24 ... and 25 C isotherms identify the western boundary. (colcl wal_l) of the Gulf

Stream front (Figs. 3 and 11) in a layer about· 20 m deep. n1e westward deepening isopycnals suggest southward flow within the warm tongue and throughout the water column over the outer shelf, which is in appro_:ximate

111e sections made across the cl:i.sturhance south of the Savamiah transect show the warm surface filament of 22 to 24 C water extruding from the Gulf

Stream front (Figs. 3 and 4) to extend to a depth of about 20 m. The

') westward deepening isopycnal s suggest southward flow within the wann tongue

and throughout the ivatcr collimn over the outer she.Lf, which is irt approximate

geostrophic balance with the uplifted cold core.· Northwnrd ·£low is indicated

east of the cold core. Ship drifts measured during the hyclrographic stations

support the indicated fl.01-., directions. Southward flow of about so· cm/sec was

estimated from ship drifts ;it stations within the 1varm filament and northward

flows greater than 100 cm/sec were ohservcLI for stations cast of the cold,

core. Strong upwelling of deeper Cu 1 f Stream waters occurred within the

coi

Stream source of the upwclled waters is indicated by the low temperatures,

salinities and oxygen values and high nutrient· concentrations. Near the· bottom (63 m) at the shelflffcak (68· m) in the St. /\ugustine South section,

temperature was )2.5 C, salinity 35,6 °/oo, o. 26.97, -dis.solved oxygen . . t . . . -1 · . -1 3.12 m.Q.£ and nitrate 20,5 11 mo.le 9. Similar values ~ould not be found

even at a dejith of 185 m in·. the Savannah section made north of the event.

. . Historical data from sections taken in the same region bu_t without· evidence

bf eddy structure indic_;,itc that wate:r with these characteristics c_omes from _.

· a depth of approximately 200 to 300 m. · ·n1is suggest~ that the near...:bottom

waters found along the shel.f1Hcak in eddy r could have been upwelled a verti

:cal distance of_ 140 to 240 .111. · ;I11us subsurface Gulf Stream intrusions of

this type provide a. source of nutrients to the ·outer shelf euphotic zone •.

1he Ormond tr;:insect appears to. have been ·near· the s_outhern extremity of

the disturbance, as indicated hy the horizontal temperatur_e and density dis

tributions (Figs" 3, 8 and 9). 'l11e lack of doming in the temperature and

density sections suggests that the flow has returned to a northerly direction.

Also there does not appear to be any si&mificant upwelling in the ~ertical

section data (Pigs. 6 and 7) judging from the high salinity and oxygen values

ancl low nutrient concentrationso

Cyclonic circulation ii1 eddy F is indicated by the horizontal distribution

of temperature at the surface, 16 and 30 m (Figs. 3 and 8) and the subsurface .. I topography·of the 26 c\ surface (Fig. 9)o Southward flow appears to occur in

the warn filament ancl northw.-ird flow in the Gulf Stream east of the cold core.

There .-ilso appears to be a convergence zone .-it the southern extremity of the

feature and along the Gulf Stream boundary. 111is zone can be formed by the

convergence of southerly flow in the w.-irm filament with the northward flow of

the Gulf Stream south of the disturbance which forces a flow toward the east

that merges with Gulf Stream an

been observed in the Florida Straits (LEE, 1975; LEE and MAYER, 1977).

Temporal Observations

Nearly 20 IR observations of the western SST boundary of the Gulf Stream were made between Florida and Cape Hatteras from April 7 to 17, 1977. The northward propagation of two fi-ontal eddies was observed through the region of the current meter array during this period (Figs. 10 and 11). The first event, eddy E, traveled through the area between April .8 and 15 at a rate of approxir.iately 47 cm/sec (Fit~• 10). It passed the current meter locations on the 10th and 11th, which unfortunately is when the box array was being

·removed and moorings E and 1: rcdcployecl. However, current meter data were obtained during a small ·portion of the disturbance on April 12. The. second

event, eddy F, was oh served to move north1vard through moorings E and F

around April 16 and 17 (Fig. 11) at a speed of about 42 cm/sec.

Time series of low frequency (subtidal) current, wind and temperature .,. for the "Spring 77" observation period arc shown in Figs. 12 and 13. Eddy

F is clearly identified _i11 the first week of data f:rom all _current meters

as a cyclo~i~ flow reversal coupled to a decrease in temperature (Figs. 12'

and 13, event lll). Data from these instrwnents were expanded in time over

the period from April 12 to 23 in order to highlight the· event. Hourly

vectors £rem 3·-hour lo·w-pass fil tercel time series were ·used (Fig. 14)

be.cause the 4--day truncat_ioil of the beg.inning of ·the subtidal reco~·ds over'.'"

lapped with the start of eddy F and had completely cut off eddy E.

The cyclonic current·. reversal from southerly· to northerly flow on April 12

indicates the passage of the southern porhon .of e~lcly E. Satellite IR

images showed the.warm filament to he passing tl1c ci.1Trent meters at .this

time·. Southerly flow in th~. warm filament was also indicated in hydrographic

sections (not sho1•m) taken at St. Sir.ions on J\pr:i.l 10 and at Savannah on

J\pril 11. (Table 2) by doming isotherms and isopycri"als cast of the filament.

These section data further showed a significant increase in nutrient concen

trations in the near bottom waters of the outer· shcl-f within this event'.

12 Eddy F fir.st :tnfluencccl currents at the mooring locations near the

st~irt of April 15, as seen by the low frequency cyclonic rotation sup,er-

imposed on the semf-di11rnal tidal vai-iat:i.ons. (Fig. 14). The reversal.

. . occurred at mooring F about one clay_ piior to mooring E. Current meter FT;

worked for only about 1. S clays at the start of the event. The passage of

eddy F proclucc

ancl low<;!r current meters ·at both the 45 and 75,.m locations (Figs. 12; 13

·arid 14) ~ Large temperature decreases were observed .to occur almost simul-

tai1Cously with the .flow reversals (Fig. 13). Local winds were weak and

toward the north during the period of southward flow and appear to _be un- .,. related to·thc floh! reversal.. Stronecst southward currents were_recorded

at current meter ET, where speeds reached 60 cm/sec (Fig. 14). TI1e weakest

southward currents of about 1S cm/'.;cc were measured at EB. 'J11e mean · -3 -1 ·vertical shear over the 55 111 instnonent separation was about 3.6 x 10 sec

during the period of southw,1rd flow. The temperature drop was largest near

the bottom at the shelflire,1k (EB), where temperature decl inecl about 5 C

during onshore flow which -reached a speed 9 cm/ sec. 'l11e temperature de--

crease and onshore flow near the bottom at the 45 m isobath (FB) were also

about 5 C and ~ cm/sec 'respectively. In the upper layer at the shelfbreak

(ET), temperature droppctl by about 2. 5 C during onshore flow which reached

speeds of 7 cm/sec. The occurrence of 18 C water at the 45 m isobath appeared

to lag that at the shclfln-eak by about 2.5 days. 11d.s suggests an onshore

advection velocity of. ,1ppl·oximatcly 8 cm/ sec in the lower layer between the

17 km mooring separation, \,hich agrees well with the observed flow.

Temperature and salinity sections taken through the filament (Figs. 6 and 7)

also indicate that these large temperature decre,1ses were produced by

advcction of newly upwcllecl Culf Stream water past the moorings in a bottom

layer approximately 20 m thick.

13 'lhe end of eddy F occurtccl tO\vanl the hcgi.nning of /\pril 21 when current vectors revolved hack. t01~:irll the north, thus completing a cyclonic rotati'9n.

_Near-bottom tcmpcratm;e at the shclflfre:ik increased to nearly the pre-eddy values, during offshore flow of about: 10 cm/sec, signaling the end of the

event. 'The total Ume for the disturbance to pass the mooring (eddy duration)

was approximately 7 days, from April 1~ to 21 •. The most pr:obable path of

eddy r past the moorings is approximated by the dotted lino from mooring E

in Fig. Sa. This path was estimated by· comparing the. contoured temperaturE;! . . field at the depth of current meter Ei' to measured temperature at ET during

the passage of the event (Fig. 13, event I). The arrows are estima.ted cur- ... ,. rent dircct.i.~ns necessary to procluce the obsei·ved cyclonic. flow reversal.

The length of the filament, determined from an ave1~age of shipboard· and

sat~lJitc SST estimates (rigs. 3 and 4) was about 225 km. 1he northward

phase speed computed from the etldy length and duration is 37 cm/seG, which

is close to the 42 cm/sec specJ previously determined from time series of

satellite IR images (Fig. 11). Either th:is ~tgrecmcnt from the two completely

separate data sources is ext'l:cmely fortuitous or the eddy length must have

remained relatively constant during the passage of the event by the moorings,

suggesting that eddy growth had stahili zed at this time.

Weekly averaged SST m:1ps of the Gulf Streain produced by the IJ. S. Naval

Oceanographic Office (NJ\VOCEJ\NO, 1975) reveal an unusually large warm fila

ment on the Goergia shelf tluring the weeks of May 12-18, 1977, and May 19-25,

1977 (Figs. !Sa and b). '111c spatial cl-Lmensi.ons estimated from these maps

were 185 x 80 km during the first week and 240 x llO km in the second week,

which indicate that the width of the distuTbance was almost the same size

as the width of the Gulf Strcmn at this location and t.imc. The curTent

14 meter records show the effect of this feature from about May 20 through

May 26, giving a dui-ation of approximately 7 days (Figs. 12 and 13, event 6).

Assuming the filament length of 240 km remai.ns relatively unchanged over

the duration ~ives a rough estimate of its northward phase speed of 40

· cm/sec. The disturbance appeared to occur at the 75 m and 45 m isobaths

almost simultaneously and produced a cyclonic current reversal in the upper

layer at the shelfbreak of + 100 cm/sec to -60 cm/sec in about one day. A

temperature drop of 7 C occurred in the bottom layer at the shel fbreak during ·

the period of onshore flow, which reached speeds of -20 cm/ sec in the lower

layer ;md ·-30 cm/sec in the upper layer.

Low Frequency Variability

Subtidal current, wine! anti coastal sea level time series from the

"Winter 76/77" and "Summer 77 11 periods are shown in Figs. 16 and 17. At

the shelfbreak (75 m isoliath) current amplitudes in the upper layer ranged

from: 40 to+ 80 cm/sec about a mean northward flow of 65 cm/sec in the

winter (Fig. 16) ar1d 55 cm/sec in the summer (Fig. 17). TI1ese energetic

fluctllations produced broad-banded spectral peaks (not shown) at periods

of 2. 5, 3.5, S and 10 days that were coherent between the cross-shelf (u),

and jlon~-shelf (v) components, with the cross-sl1elf leading by 90° in

phase, suggestive of propagating wave motion. D01•mstream current mcte·r pairs

of the winter ·experiment rcvca1 C(l 11orthw;1 rd propagation that was coherent ·

over the greatest along-shelf array separation of 40 km at periods of 3.5

and 10 clays in the up11cr layer and 10 t!ays in the lower layer. During ·the

summer coherent northward propagation of veloc:i ty an

was found over the 90 km along-shelf spacing at periods of 5 to 12 days in

the upper layer. Southward propagation was observed during the winter in the

lS lower 1ayer. a.t pcrjotls of 9 to, 12 days and tluring the summer in both upper and lo1ver l;1ycrs at pei·iods of 2 to:~ days. Temperature fluct,uations with ampl i tudcs tanging fron; :!: 2 to + 11 ,C. occurrecl througho.ut the year and were significantly ·coupled to .the velocity var:iatioi1s (Fig. 13). Largest temperature changes occurred i11 the lower layer. Current and temperature fluctuat.ions \(ere· coherent over the SS m vertical instrument separation at periods of ·3.5, 5 and 10 days during the winter and at periods of 2 to 2"5

.clays only during· the summer. Significant coherence was also found during the winter between along-shelf current fluctuations :intl t:he local· cross shelf wind at the 4 and 10 day periods .in: the upper layer and at periods near 2.5 days near the bottom. Along-shelf win

No significant coherence was :fountl between currents at the shelfbrcak and coastal sea level.

Low frequency heat and momentum fluxes at the shelfbreak were found

to be selectively grouped in the frequency b:inds of the more energetic

motions. Largest values were usually associated with fluctuations in

the S and 10 day period bands, with the Gulf Stream g:i.v:i.ng up heat and momentum to the shelf wat1)rs in the upper layer and the lower layer normally indicating an offshore he:it flux·. The offshore heat flux is believed to be produced by the coupling of lower layer onshore flow events to cold anomalies

(Fig. 13), which resul·ts in a positive u'T'. '!he seasonal average momentum

flux was founcl to he onshore in the upper layer during both winter and summeT

? -2 at about -70 and -40 cm- sec respectively. TI1e seasonal average heat flux

· in the upper layer was onshore at about -4 C cm sec-l in the winter and

-2 in the summero In the 101..-cr layer the heat flux was offshore at about

+0.5 in the winter and +2 in the summer. lbe negative momentum flux indicates

, r...... that the shelfl>renk strip off Georgia is a region of energy dissipation associated w:i.th the tr:rnsfer of energy from the mean Gulf Stream current to the fluctuations. A significant portion of the energy transfer appears to be occurring with fluctuations having periods in the 5 to 10 day period ba,id. On the continent:ll s'lopc off Onslow 13ay BROOKS and BANE (1980) recently found that. the fluctuations were supplying energy to the mean flow as was found previously by WEBSTER (1961). In the near bottom water

BROOKS and 13ANE(1980) also found an offshore heat flux of similar magnitude as found here for the Georgia shcl £break.

At mid-shelf locations (30 rn isohath) lo~ frequency along-shelf current ... fluctuations were highly coherent over the 90 km array scp?,ration distance and nearly in phase for alJ periods >2 days during the winter with very similar results for the summer season (Figs. 16 ancl 17). Current variations were al so coherent with coastal sea lcvc 1 and local wi n

of 12 to 20 hours (Figs~ 16 ,111d 17) o 11n1s at mid-shelf, s11bticlal currents are largely generated hy the.• sarric mechanism, the local winds, by producing

surface and bottom Ekman tran5ports which cause the coastal sea level to

set-up and sct-do1v11 and thereby drive a barotropic along-shelf flow over

the mid-shelf and inner-shcl f (BEARDSLEY and BUlilAN, 1974; SCOTT and CSANADY,.

1976; LEE and IrnOOKS, 197~1). ·11,c amplitude of these current variations

ranged from :!: 10 to ± 40 cm/sec ahout a 4 month mean flow in the upper

layer of 5 cm/sec toward the north in the winter and 1 cm/sec toward the

north over the sununer. 4-month period" LEE and BROOKS (1979) found no evi-

denc~ of southward-propagating continental shelf waves in the Georgia coastal

sea level data as was found south of Cape Canaveral by BROOKS and ~IOOERS (1977).

PIETRAFESA and .JANOW!TZ (1980) a 1 so rcccn t ly reported on the lack of evidence

for southward-propagating shelf waves in Onslow Bay, North Carolina. The width

(120 km) and shallo~1ess (:90% of shelf width <50 m) of the Georgia shelf was 17 believed to effectively isolate the inner region from propagating dis

turbances at the shclfhreak.

Coherence between along-shelf currents measured at mid-shelf and the

she.lfbrcak during the winter was low except for the 9 to 12 clay period range.

1his is also the only period band for which propagating wave motion was

indicated for the mid-shelf location (u and v in coherent quadrature). The

effect of Gulf Stream events appears to be more pronounced in the summer

at mid-shelf th::m during the winter. (Fig. 17). Coherence bet.ween along-

shelf currents from the mid- and outer-shelf was found to be generally higher at this time. 1his suggests that inclined isopycnal surfaces in

the summer allow Gulf Stream events to penetrate further onto the s·helf

than in the 1,intcr whc11 shelf tlens:i.ty is higher and horizontally

stratified.

DISCUSSION

Eddy Signature

Frontal eddies appear to be a common feature in the Spring 77 time

series (Figs. 12 and 13, events 1-10) ::mcl they can probably account for a

significant portion of the lo1v frcque11cy variability in the outer shelf

region. Ten cyclonic current reversals occurred at the shelfbre:1k during

. the 74-day recording period, or approximately one per 1veek on the average.

Each cyclonic 1;cversal is accomp:1niecl by a large decrease in near-bottom

temperature, which is the t)11i.cal cdd)' signature in shclfbreak current and

temperature records in the rc·gion from ~liami to Sav~mnah, where the Gulf

Stream boundary follows the shclfbrcak. A schematic characterization of

a typical frontal eddy on the Georgia shelf is shown in rig. 18. Satellite

18 IR images, hydrog'r:iphic· ;rnd current meter clat.:i indicate that southward flow occurs within the warm filament. extruding out· ~.f tl1c. Gulf Stream front toward the south (rigs. 4, 6, 12,· 14 and 15}~ Southwitrd flow extends

.to. the bottom in the outer shcl f region and is in· approximate geostrophic balance with the uplifted ~lensi ty structure of the· cold ·core (Fig. 6) .. Tii.e . . . g vertical shear estimate~! from the th9r11ial wind .eqt1

...... : ...... temperature arpcars to be produced by the nortl11v,1rd advcction of upwelled waters. in· the cold core, which extends onto the shelf beneath the warm·• fila- ment_ (Figs. 6 and 18). Upper layer temperature signatures are not as consistent and depend on the location of the ~ensor relative to the depth of the warm filament and location of the cold core. The larger current data baie from tlie Georgia shelf :i.ndic:,tes th:1t for the J 1-month period from December 1976 through October 1977 approximately m1e eddy occurred every two weeks on the average (rigs. 12, 16 and 17, eddy event lines). "ihe occurrence of these· ., disturbances a1ipears to be a random process, for at times several reversals may occur in success ion and ;1t other times they can be sepa1·ated by several weeks. Sh1ilar eddy signature~; in current mc,tcr data h.ive recently been shown to be related to the large Gulf Stre;un filament stn1ctures that form off Onslow 13;q (BROOKS and BANE, 1980)., However, measurements from the

shcl fb:reak region off_ Cape Romain (PI ETRJ\FESJ\ and JJ\NO\fl TZ, 1979), 1.,rhere the

Gulf Stream boundary is deflected seaward by the Cnarlestcin bump, indicate

that the eddy signature in currc11t and temperature records is more compli

cated than on the Ceorgia, Florida and North Carol i.na shelfs. TI1e signature

is observed to vary according "to the location of the current meter relative

19 to etldy center and 11ropagat:ion dii~ection. In the lee of the Charleston

bump eddy filament sizes call become· larger _and apparently propagated onshore

across the 11b~1mp" as well as northw,.rrd. · '11rns measurements from the shelf-·· . ' ·break at Cape Romain sometimes show·a clockwise current.rotation coupled to·

a temperature incre~ise apparently resulting from the onshore motion of a

southward f101~ing filament whose colLI core· lies so"uth. of the mooring

(PIETRAFESA and JJ\NOWI'fZ, 1979).

Gulf S lrcam Mc~mJers

,Current and temperature time· series fr<)in the Georgia outer shelf ha·ve

been co1npared to ~1yclrogr~phic data from repeated. sections made acro 3s th~ current meter locations. ai.1d into the Gu1.f StTe~m in Decemb!=)r· 1976 · (ATKlNSON.

and LEE, 1980}. · The resi.1lts of :tl1is analysis :indic,ite that the large amplitude

subtidal- current _and tempc:ratun: fluctu:1tions that occur along the shelfbreak . . . . with periods of 2 days to 2 weeks _and without tycl~nic etirrent reversals_

(Figs. 12 and 13, events 11-15) arc produced by northward-propagating~ wave

like disturbances of the _Gulf Stream cyclonic front, i.e., Gulf Stre·am

meanders. Offshore moa11dcrs :result in a dccrca!;O in ·northward current

speeds throughout the water column and upwelling of cool, de~per Gulf Stream

waters wi_th increased nutrient concentrations in the lower layer. Onshore

meanders produce down1velling and increased northward currcnt·speccls. These

onshore-offshore meanders produce coherent in-phase fluctuations of velocity

and temperature at periods of 2. 5 to 3 and 5 to 10 days throughout the year.

On many occasions offshore ·mea11clcrs produced .velocity and temperature de

creases at the shel [break, 1vhich appear to fol low northward wind events

(Figs. 12 and 13, events 12, 13 and 15), indicating that the offshore meanders

and associated upwcll ing may at t imcs be generated by offshore and onshore

Ekman transports in the surface and bottom layers, respectively.

20 Eckly rormat:i.011·

Time series of satellite-derived thermal images have clearly shown that edge-eddies fornf from growing no1~thwai,l propagating wave-like meande1·s of . . . ,' the Gulf Strcain cyclonic front (Ll:Gl3CKIS, 1975; STUMPP and RAO, 197_5). · .The folded-wave p:ittcrn, characteristic of eddy development, has been observed to occur within· two days of meander formation· (STllMPF and RAO, 1975). Wave folding appears to occur from the· shorc.:!\-lard cxtcns i.on. of. a meander producing a southward flowing warm .filament of l;ulf Stream water over the outer shelf

(Figs. 3, '1, 15 and 18). Current meter anti hydrographic observations indicate that southward currents in the warm filament associated with eddy F were in approximate gcostropld.c. balance with thq't~iplifted cold ~ore.

The above results suggest the following conceptual model for the J.ifc cycle of spin-off eddies: perturbation of the Gulf Stream cyclonic front produced by either offshore surface Ekman transport associated with north-. ward wind events or :interaction of the flow with bottom features can gener.ate an offshore meander that travels to the north as an unstable shelf wave in the upper part of the Current wlicrc the· mean current speed is greater· than the

southward phase speed of the 1,avc. 'Ille wave grows rapidly in time, possibly

feeding upon the potential energy of the mean· current (baroclinic instability)

and it may eventually evol~c into an elongated edge-eddy, which can act to

dissipate the newly acquired kinetic energy into the shelf waters. Time

sequences of satellite IR i.m;1gcs (similar to Fig. '1) between J\pril 7 to 17

reveal that the folderj-wavc patterns, which typify edge-eddy development,

begin to evolve from the onshore crests of the waves as •the crests shoal

over th9 outer shelf. Once the folding process is initiated, the shallow

(20-30 m) southward flowing filament of warm Gulf· Stream water becomes rapidly

21 elongated by the d01-mstrc!am cxte11sioi1 of ·the northern Gulf Stream connection to the filam.ent (Jiigs. i and 3). · '11le · feature l·iterally shears_ apart from the Gulf Stream and becomes stpndcd in the out,er_ shei°f; a,waiting its final di_ssipa:tion over t·he shelf. 'I11c total cycle from growth to decay may take only

2 to 3 weeks: 2 clays to.one week of wave; clcvelop1iient, about one week of-eddy_ growth and one week of dissipation. 111is process appears to be occurring about once cvpry ~ weeks. 011 the average beti"cen Miami and Cape I-lattera_s ,_ .with further amplification of the. disturbances north of the Char~estbn bump.

Eddy F was _observed to undergo a s_ignificant elongation in the along-shelf direction ov6r.the 2 day interval between measurements (Figs~ 2 and_3). · TI1e growth appeared to occur primarily by the extension of the-northern Gulf

Stream. source portion of the fil;unent, possibly due to the strong horizontal shear across the r.ui-£ Stream boundary. Tirns a·n instability process appears to be involved in the evolution o·f both meanders and.ecldicso The e~act·nature of the instability :i.s u11ccrtaino Both harotropic and haroclinic unstable waves can develop :in the Cul f Stream cycl.on:i.c front due to the intense hori zontal and vertical shears there.· NIILER ancl mS/\K (1971) theorized that the fastest growing barotropic waves would propagate to the north with wave lengths of abol1t lS0 km a·nd 1·ieriods of about 10 days. ORLANSKI (1969) obtained similar result::; for amplified baroclini.c 1-iavcs, i.e., wavelengths of 220 km with periods around 10 dfiys. D~INC, ~IOOERSanti LEE (1977) found that the most energetic current fluctuat.ions in the Florid:i Str:1i ts hacl periods of about 9 to 12 days. 111c downstream component was coherent at these periods across the entire rlorida Strajts, from the shcl f waters off ~-liarni to those off

Bimini, Bahamas, and was al ~;d coherent 1-iith the local cross-stream wind component. Current fluctu:it i ems in this period band propagated to the north with phase speeds of 2(i to 33 cm/sec and wavelengths of 200 to 450 km.

22 ...... Cohcrcnc;c in the downstream direction sharply decreased :for ,ii.stances greater than !i'.i km. It ivas suggested ( 1il\TNc;ct al., 1977) _that these results coul

ROONEY, JANOWITZ and PIETRAFESA (1978) also suggest that filaments may result from an unstable meander initiated by a wind evenL BROOKSand RAi\JE

(1980) indicate. that_ the· Charleston bump may trigger a local ii1stability, causing an intcrna 1 energy rctlis_tributi.on downstream of the 'bump through both barbclinic arid barntropic energy conversions.

Eddy Interaction with Shelf W;1tcrs

'I11e water mass properties of eddy F antl of the shelf waters arc good ... indicators of the type and d_cgrce· of shelf-Gulf Stream intE:ractions during cJdy passage." Significant amounts of shelf water within the vortex would imply int_er;1ction by cntr,1:inmc11t and mixing, whereas a paucity of shelf watGr would. imply interaction ·by displaccrncnto 'The water type "slope water" is s_eldom observed south of Cape llattcras and then only as fa:r south as Onslow

Bay (STEFJ\NSSON, ATKINSON::mcl BUMPUS, 1972) a A composite. T-:-S plot of datc1 . . from all· ~tations of the April cruise (Fig. _19) shows that shelf wate·r tends to stratify primarily_ due to salinity at this time· of year and Gulf Strea~ · waters have a tendency_ towa1:tls the11nal s.trati.f:i.cat iono 'I11e waters of the . ·

Gl!lf Stream arc fcrrtun;;1tcly qliitc constnnt in th~\ir T-S relationship and easily discernable from shelf watcT. T-S plots of stations taken _across eddy F (Fig. 20h) clearly show the characteristic Guif St:rcam T-S profiles at all stations with no tr:1cc of shelf water (salinity· <36.0 °/oo) ~ Stations taken across the outer shcl r north of the event (Fi.go 20a) show that Gulf

Stream water extends to at least the ti(, 111 isobath. 111c_subsurface salinity maximum of subtropical undcrivatcr 1vith a salinity of about 36o5 °; . is clearly 00 . vis.iblc near the 90 m cleptho The composi tc T-S plot reveals that shelf water ., never extended further" offshore than the 4 6 m .isobath. 111c absence of shelf

water over 'the' O\ltcr shelf· (4(i to 75 m_isobaths) and w:ithin the vortex.

. . . indicates that cdge.:cddics arc· cfficici1t mechanisms for rapid and persistent

shelf-Gulf Strc~m ivatcr exchange. · The repeated renewal of waters by this

. means tends to displace the Gulf Strca~ salinity front shoreward, thDs ·

creating an effective Gulf Stream front at approx:imatel)' the 45 m isobath,

whereas ·the instantaneous front (maximum hoTizontal ten1peTature and current

shti:1~·) _normal°.ly. mc:111tlcrs a.bout the 100 m isobath. 111e reason-shelf water

was not fouricl :in eddy F is that it was interacting with residual Gulf Stream

waters from prcvio,1s vo:rtices. TI1e outer shelf waters are continually Te- ... newed by this means so _that _a i-esiJence time can be define~l as the average

separation· time between spin-off ccl

Ni tr ate Flux

Upwelling in the cold core of edge-eddies transports deeper Gu,If Strea171

waters with high nutrient concentrations into the euphotic zone (<50 m) -1 along the_ outer shclfo Nitrate concentrations of 10 p moles Q, were found

at the 30 m depth.in the uplifted core of eddy F (Fig. 7). Nutrient-enriched

waters were also observed to intrude onto the .shelf in a bottom layer 20 to

30 m thick extending a distance of 35 km from the shel.fhreak beneath the

southward flowing w::mn Gulf Stream filament. Nitrate concentrations in th.c

outer shelf arc generally less than 1.0 11 moles Q.-l when eddies are not

present. ·n1e vertically a~cragcd net ni.tratc flux(u'N0 1)at the shclfbreak 3 was calculat_ed using c;urrcnt and temperature data from mooring E for the 7 day

duration of eddy r:; April lS-21 o Temperature time series from meter EB were

used to construct an ~nfcrrcd nitrate time series·, since in this area

temperature and n:itratc arc linearly related for newly upwclled waters with

24 ... temperatures -~ 20 C (S'll:FANS.SO_Nand ATKINSON, 1971; 0 1 MALLEYet al., 1978);

according to t'.hc cnipil.;ical relationship:•

] = S3.0 - 2.6•T [ NCl3 1he nitrate co11ccntration at meter ET is assumed to· be zero since temperatures

are greater· than .20 <;:, and the· meter is· loc[i.tecl in the ·e_uphotic zon~ where

planktonic uptake occurs~ Nit1~ate section data (Fig. ,7) also show near·

z·ero concentrations; at this. location. 'f11e net ·nitrate· flti..x ·.at El3 for

the_ period of eddy pa!;Snge _wu,s comput,ccl from:

- u'N0 1 = (u - ~) (N0 -_NO~) 3 3

where the overbar represents a time :ivcrage of the 7 day pcriocl·of eddy·F;

u and No arc daily ~iveragcs to suppress tidal variations; u ·and N0 are 3 3 one month averages over the J)Criod from April 16 · to May 16, which was

used to average the influence of sc/veral eddies. 111c vertically averaged

net nitrate flux was fo1md from:

u'N0 1 (72 m) * u 1 N0 1 (17 m) 3 3 2

·me negative sign indicates a 11et. flux of ni txatc to the shelf waters~ 111e 2 amount of nitrate flux through a 1 Ill column (55 m ·x 1.82 cm) at the shelf- .,,/J /sec..- . -l IP"'}/~ break 1s -18.6 11 moles sec-- " 111:is is nearly equivalent to the 4 month winter

-1 ~ average of -25 JJ moles sec onshore nitn1tc flux found for a l m~ column at

the seaward edge of the very 1~roductive Scotian shelf (S~II'n-1, 1978)" 1he

vertical eddy averaged nitrate flux was converted to a nitrogen flux QN . . -2 -] g1v1ng a value of QN = -0"2() mg N m . sec · TI1is 1vas used to find the total

25· nitrogen transported to ·1:hc shelf by the eddy, NT:

· where II is the hcieht of the shclil}reak water column (55. m), L is the

eddy length (225 km) ancl T is the eddy duration (7 days). Using these 9 values N1 = 1.94 x 10 ~N/eddy or 2128 tons N/eddy. Assuming that .on the average about one eddy occurs every two weeks, then an estimated yearly . . 4 input of nit1·ogen to the outer shelf by eddies is S.S x 10 tons N/yearo

1he. net flu_x of n_itrate at the shclfbreak from low fi-equency current: and

temperature. flu.ctuatio·ns during the winter period from. December 76 :to April . -2 . -1 77 was estimated at -8.S 11 moles m sec using the same methods as... above with 40 liLP time !>cries. If we assume this rate is constant oV'er· the year then

the annual nitrogen input through a length of shelfbreak equivalent to the . . . . 4 eddy length (225 km) is 5 ~ 1 x 10 tons N/ycaro The close agreement of these

two independent estimates over diffe1:ci11t time periods and nveragin~ lengths

suggests that the rate of 11i.ti·0Rcn transport is reasonably constant for time

periods longer than one month nnll that spin-off eddies are. a major contributor.

·to the transport •.

Impact of Eddy Nitrate. Flux on P11y_toplankton Production

· Th~ cornput:cci' net ·onshore flux .of nitrate will have an impact on phyto

plankton produ~tion that canbe estimated knowing the. .arc~ influenced by the disturbance. llydrogri1phic sections (Figs" 6 and _7) and surface temperature maps. (Figs. 2 and 3) indicate that the eddy, which was observed to be 225 km

in along:..shelf dimension, influences a region approximately .35 km irf the 9 2, · · onshore direction, an area of 7. 88 x lff m • Assuming one eddy occurs on

the• avcrag~ every two wc.9ks, or 26 events. per year, the annual direct areal . -2 -1 m.tro~cn input is -6~4 g N m yr Carbon:nitrogen r:1tios in phytoplankton

26 range from S:l to 10:1~ implying that the supplied nitrate could support an . . . . 2 . annL1a·1 c:.1rho11 p:roductic,n of 32 to C14 [: C 111- yr-l w:ith_no nit1:ogcn recycling.

· 111c estimated carbon p:roduction is based on· the computed riet flux .. from a

spring· eddy only; but the r.atc appears to be representative of the annual

input. HAINES. and DUNSTAN(] ~)75) found anannual, outer· shcl f production of

-2· -1 ·. _? .-1 132 g Cm . · yr and April production of only 18. g C m ~ yr oh an annual

basis. Sinco their samples were_ taken from a predetennined grid which did

. . . not emphasize the'shclfbrcak a':J:ca or attempt to identify or sample within

edJy-upwcll?.d waters, their values arc not rcpresc11tativc of outer shelf

carbon production. IIJ\INES (1~)74). attempted a nitrogen budget for the South.

Atlantic• night that is sche:rnt1t:ically shown in Fig. 21. Our estimat~ of Gulf

Stream induced eddy nitrate' flux has been incluclcd on this. figure for compari

son. 111c ccldy estimate of 55,000 tons N per year is a factor of 2 greater

than the combined nitrogen sources as

7 greater than HAINES (;ulf Stream estimate, which was based on diff\.1sion of

subsurface Gulf Stream waters. However at the time of the HAINES measurement

the capability to conduct satellite clircctecl field sampling had not been

developed. Clearly, the newly calculated net nitrate flux due to eddy

passage j_.s a major s·oun:e of i1cw nitrogen for 'the South Atlantic 13ight and

may control carbon product.ion in the outc1; shelf.

Slll-1~1AlffJ\ND CONCLUSIONS

\'Jc have comhi11cd a unique set of satellite II~ inwges, hydrogr::tphic

sections ::mcl current [lllcl temperature time series to conclusively show that

tl{e folded-wave patterns commonly observe cl in routi nc sat elli te-deri ved

analyses of the Gulf Stream western boundary from Cape llatteras to Miami

arc produced by ,geostroph ic, southward flowing warm filaments of near-surface

27 Gulf Stream water coupled to cold upwcllcd cores. we· define these features

as cyclonic frontal eddies partly for historical reasons, partly from their·

formation process ·(they i1ppcar to develop from growing wave_.;.likc meanders ·,

· of the Gulf Stream front), and p,u-tly from their eclcly-like cyclonic circu

lation produced by the 1~arm south1-,'ard floi•ri.ng Gulf Stream waters on the west

· side and northivdnl (;ulf Stream flow op the east side of the cold uplifted

core, w~ich is _in·~ppro~ima~ri gcostro~1ic balance wiih the cy~loni~ circula~ . . . . tion. 111ey may_never become completely

. the fin;Jl stages 9f dct'.ai 1vhc.11strandinr, over the shelf and cven·tual dissipat_ion

• is in. evidence,,

Eddy dimensions appear to have 3 ch~tracter\stic .length· scales ,·"which are

inf°lucnced by topor;raphy r.i'ncl cnn be class.Hied by 3 geographic areas along

the· southeast· Uo So shelf. Soutl1 ·of Jupitc1.:, 'Flo~i

Straitso Eddy diameters here range from about ·10 to 30 km, with the down

stream axis bci_ng 2 to 3 times larger than the cross-stream. Between Jupiter

and the Charleston bump the shelf wiJens and eclg~-ecldics take on the character

istic elongated foldcd-:-wavc patterns. 'Jhe filaments ·can extend 35 to 40 km

over the· shcl.f with downstream lengths rangii1g from 100 to 200 km. A second

elongation is obser~ccl following-~. seaward deflection of the Stream by the

Charleston bump and downstream lengths may reach up to 300 km~

'I11e e

from time series of satellite TH-images that they form fr.om growing wave-like

meanders of the Gulf ,Stream cyclonic fi·ont. It is also apparent that at

times wind events can ini t:i ate a .frontal meander, which propagates to the

north and may evolve into an edge-eddy; but whether this occurs th~·ough a

barot1"01,.ic ur IJanH.:l inie i.ll~tahi l ity process is unknown" The uncertainty

.I I

2B .. rises because the Gulf Stream front appears to satisfy the inst.ability

criteria for ~_ither process and both predict similar ,~ave properties, which

agree with the obs~rvcd propcrti~s lwa~~ l~1gths 100 _to 200 km and wa~e

· period ::: 10 days).· However, qdgc-eddies have. been· found to dissipate kinetic

energy of the mean flo~•, which points .to a barotropic instability (LEE; 1975)°.

They form on tl1 e average of one evc1o/ 2 wee ks and tr ave 1 to the north along

the shelfbreak ilt al,)()t~t 40 cm/sec, which is ab_ouf the same as the· northward

ph ..1sc speed of the meanders." 'J11c:i.r total life cycle, from meander to dis

·sipation over th.c shelf, usually takes place· in 1 ess · tha~· 3 weeks: approxi

mately 2 days ·to one week of meander gro1,;th, one we~k of eddy growth and

one week of decay. ·n1ey appear to form along the boundary of any major

current system and arc often observed along the outer edges of the larger

warm and cold core Gulf Stream rings, which suggests that they may play a

role in the exchange of water across ring fronts.

r:rontal eddies arc observed to.be a major component of the low fre

quency current and temperature variah:ili ty over the outer 35 to 40 km of

the Georgia shelf. during al 1 seaso11s. Ecldy signature in current and

temperature time series from the shel.fbreak·along Florida-and Georgia

consists of· a C);clonic flow. reversal thrciughout the water column coupled t'o

a large temperature decline that is more _intense near the bottomo There

is some indication that there is more ed

it will take longer time series to be conclusive. Edge-eddies provide a

mean.s of rapid water rencw;:i 1 for the; outer shelf. The frequency of vortex

passage tends to create an effective Gulf Stream front (as ,rieasurccl by

saiinit}', _36.0 °/oo or grcatqr) ~s·.to 40 k.111sliore,,ai·c.l of the; actual front

(maximum horizontal te.mper:iturc and current gradient}, which .is. normally

located seawnr

.29 ---··· ······-··•------

·'

waters can be defined as the average timc·:intcrval b.etween:edcly events, or

Upwelling in the cold core of frontal ecldies was observed to.transport.·

deeper nutrient-enriched (:t.11£ S.t:rcam waters onto_ the ·shelf in a bottom

intrusion J aye1' .20 · 111 thick licne_ath the 1'1;11111 sout}_nvar

and spring and possibly even further in the summer when the shelf waters arc

vertically st'ri1t:i..fied. '111cse nutrient-enriched waters reach :into the

euphotic zone, 1~hiclr.is approximately 4S m in this area, and thus provide

a food source for phytot1iankton •. Estima:tcs of nitrate flux indicate that

eddies transpo'rt approximately S5,000 tons of nitrogen annually to ·th~ outer

shelf, which is almost a factor of 7 g-reate'r than previous Gulf Stream·

e?timatcs and is about a factor of 2 greater than all other estimated_ shelf

nitrogen sources combined. !Ve estimate that this nitrogen input could support 2 _:111 ...·•1111t1,·.1l c.·1·1.·ho11 prot 111c t·· ion o:f .>~.,,., t o \)'t"-A g C.m - yr -l 1v1.'ti 1 no n1.·t rogcn · recyc. 1·1ng.

'111c estimated rate of nitr.ite input is only a factoi- of 2 less than _that

estimated for the: seaward edge of the very productive Scotian shelf. The

question then is: why is the Georgia shelf usually considered to have low

biological productivity? Since we feel our nitrate flux estimates are

reasonahly uccu:rate, we can thi.nk of two possible· answers: (.1) It is highly

productive, at least in the outer shelf, hut past sirn1pling strategics have

not been adcqu.1t:c to resolve such a11 evcnt-domi.natecl region. '!111.s could have

easily occurred because only recently, with the advent of routine satellite

' . coverage, have the cdcly up1vc1 ling events been obse1·vetl in any detail; (2)" 'fnc

eddy-induced upwelling events do 1iot last long enough to establish a well

structured food chain., In the 2 to 3 week duration of upwelling events there

is ample time for primary producers to respond :u1d for some of the faster

30 growing zoopJ ankto11, ho1vcvcr, the time may not he sufficient for higher trophic level organisms·.

If the ai1swer to .the ahovc question is number 1, then there should exist undiscovered fishcrie.s along the southeast U.S. outer shelf. If number 2 is the answer~ ·t.Jic11 since the nutrient food source is almost. continuous this region would seem ideal for establishing artificial reefs within the cuphotic zon·e along the shelfbreak strip.

Acknowlcdger.icnts-.;.Wc tlwnk ·Captain Hartshorne of the R/V Columbus Isclin , . . . .and Captain Bennet of the S/S Advance II and their crews for aid in th~ . field operations. Spec:ia 1 tban ks arc extended to Rich Findley anc.I·t{',leddra

Chalker, who p:rc1>arcd the ci1n·cnt meter instrumentation, and. Jim Singer,

Rill Chandler and Pat 0 1 ~1allcy, \:,,Jio matle the hydrographic observations~.

Pat· O' Malley is also· tlwnkcd for pteparing

Sherman Chiu is thanke

Dr~ P ~ McClain. and D. 01.son for their cdtiqt1cs. l.ve particularly iii sh.

to acknowledge the i1clp of Dr. Steve 13aig of NOAA/NESS, Miami, fot without

his ·imalys:is of sat·ell:itc ·imar,cs wemay never have found the eddies. We.

also thank Jere Crccn foT typi 111: and Diln ~ld11tosh and G;ny Rosicl lo fo·r

drafting.

Tilis resc~irch was supported by the DcpaTt1i1cnt of Energy under

contracts EY-7(,-S-OS-516:) and EY-7(i-S-09·-0R89 to T. N~ Lee and L. P. Atkinson,

rcspccti vc ly;

31 .. REFERENCES

... ATKiNSONL; P. and T. N. LEE (1979) Shelf brcnk upwelling induced by Gulf

Stream lllt~andcrs. (In .1frcpar.1tion): . llANEJ. M. Jr:. and D• ..A. BROOKS-{1979) Gulf Stream meanders along the

continental margj_rl° -from the: Florida Straits to Cape Hatteras. Geophysical.

Rese:"arch Letters; 6('1) ,: 280-282.

BJ\RRE'TI'j. R. · (1971) Avail3b1c. potential ericrg"y of Gulf Stream rings.

Deep·...,sc:.iR2~;i~;_irch, 18, 122l--::123L

BEARDSLEYR. C. and_B. BlfiM,\N (1971,) Circul.:1tion on the t;Jew Engl.:1n

BRAUN C. (19/1) Lir.d.ts on the nccuracy of infr.:1recl measurements of sea surface·

temperature from satellite. NOAA Tech. Memo. NESS 30, NOAA, Wasl}ington, DC,

1-28.

BROOKS_D. A. and C. N. K. MOOERS (1977) Wind. forced ·continental shelf waves

1.n· the Florida Current. · Journal of Geophysical Research, 82, 2569-2576.

BROOKS·.D. A. and J. M. BANE JR. (1978) Gulf Strerun deflection by. a bottom

feature off Ch.:irleston, · So.uth Carolina. Science, 201, 1225-1226.

BROOKS, D. and J. M. BJ\NE JR. (1980) Gulf Stream meanders: observations

of .their forn1;ition, struc1 urc, propar,.it ion, a11cl energetics from

O1arleston to Cipe ll:1ttC'ras. (Submitted to Journal of Physical Oceanography)

BROIYNO. B. and R. ll. EVANS (1979) J>rocessi 111: of NOAA VIIRR Infrared c;lata

for geophysic.11 applications. Presented at Second \forking Conference

on Oc?:-i.nographic Data Systems, \'IHOI, Scptcr.1bcr 26-28, 1978. DE HYCKE R. J. and P ~ K. RAO (1973) Eddies along a Gulf Stream boundary .

viewed from a very high _rcsolu tion. radiome.ter. Journal of Physical

Oceanocraphy, 3, 490-491.

' . oUING W.~ C. N. K •. MOOERS and L N. LEE {1977) Low frequency variability in

the FJorida Current and relations to ;itmosphcric forcing from 1972 to 19740

Jo.urn:il of Marine Research, 35 (1), 129-161.

DUNSTAN W. M. and L. P. ATKINSON (1976) Soiirces of new nitrogen for the South

Atlantic Bight. · Procc.;e

Conference; Galveston, Texas (M;_iTtin Wiley, c

FOFONOFF N. P. (1969) Spectral charat:tcristics of •inten1al ,-1aves in; the ocean.

Deep Sea Research, supplemental to Vol. 16, 59-71.

FuCLlSTER F. C.' (1972) Cyclm1·ic.: d.n!_:s formed in the Gulf Strear:1 1965-1966,

in Studies in Physical Oceanogr;.iphy, A. Cordon, ed., New York, Gordon and

Breach, 137-168.

GARDNER, l~. S •. , D. S. WYNNEand W. M•. DUNSTAN (1976) Simplified procedure for.

the manual analysis of nitrate· 1n· scawatr.:r. Mar. Chem., 4, 393-396.

GOTTHARDTG~ A. (1973). Obi.crvcd fon!1ation of a. Gl!lf Stream anJ:icyclonfc eddyo

Journal of Phy::;ical Occinoi:;raphy~ 3, 237-238...... HAINES E. n. (197_4) 1'1~ocesse~~ affecting prtllli.1ction rn. Georgia coastal waterso

PhD d:isscrtat:ion. D1ikc Un.ivcrs:it)', Durham, North Carolina.

HAINES, E. B. and IV. ~l. DUNSTAN (El/5) The distribution ;111d relation of

particul:ltc org;rnic _111:itcri:il ;rnd p1·irn:1ry productivity .in the Georgia Bi1tht,

1973-1974. Estuarine and Coastal 1,1:tr. Sci., 3, '131-4.41.

KOHL J. C. (1863) Cesc'hichte clcr Golfstrums uncl seiner Erforschung. Bremen:

II C. E. Mul1er.

LEE T~ ll. (1975) Florida c:uri'ent ~;pin-off cdd ics._ .Deep Sea Rcsc;:11:ch. 22, 753-765. LEE ·T. ll. .and· D. !I.AYER. (1977) Low fr~qucricy current val'.'iability aml spin-off

eddies· on the f;hclf of:f southeast F.lor':i.rla. Jominal · of Marine Research, 35(1),

193-220.

-LEE L H. and D~ A. !\ROOKS (1979) lnitiai observations of curr.:!nt, temperature

and coast;il sea :J.c;vcl · respb11se to i1tmosphcri-: and Gulf Str·eatn forci;-ig ·on the

Georgia shelf. . Gc~opliysical Research Letters, G( 4) ·, 321-324.

LEGECKIS R. (1975) · Applic;-it:i.ons of ~ynchrorious· meteorological s.3tellite data

to the study of time dependent sea surface temperature chan'ges along _the

bOlmdary of .the Gulf Stream. Geophysi:::al Res~arch Letters, 2, ~35.:..433_

LEGECKIS R. (1979) Sat.ellite observi1tions of the influence of botton~ topography

on ·the seaward defle_ction of .the ·Gulf Stream off Charleston, South Carolinao

Journal of Jihysic;:il Oc:canor(r:1phy, 9.(3), '183-497.

LEGECKIS R. and J. PRITCHARD (1976) Algorithm for corr cc tion of VHRR imagery

for geometric corrC'ctions due to earth curvature, earth rotation ind

spacecraft roll attitude errors. NOAA Tech. Memo. NESS 77, Washington, DC, 1-Jl.

NAVOCEANO(1975) Experimental ocean frontal nnalysis charts. UoS. Naval

Oce;:inographic Off ice, Code 3710, .197 5-1977.

NIILER P.P. and. L. A. }f!SAK {l97l.) Tiflrotropic w,Jvcs along an eastern continental

shelf. Gcophyriical F.lui

NOM-NESS (1974) Exped.;tlL~ntal C.ulf Strcan Analysis, NOJ\A-2 S.:1tellitc Thennal

Infrared (V1IRH), U.S. ll

F.O.Il. f/4, Room 3301, \~ashi.ngi-on, D.C. 20233.

O'MALLEY, P.G., L. P. ATKINSON, .J. J. SlNU:ll, \\'. S. C!IJ\NDLER an

(] 978) JlyclJ·ographi c ohserva tions in the Georgia Bight (April 1977).

Georgia Marine Science Center Technical l

jets in a baroclinic fluid. Journal of Atmospheric Scicn_cc, 26, 1216-1232. .•

ORLJ\NSKI I ... nd M. D. COX (1973) llaroclinic instability in ocean currents.

Geoph)•si cai Fluid Dynamics, 11, 297-332.

PARKER C.E. (1971) Gulf .StrcaP.J rini;s in the Sarga_sso Sea, . Deep-Sea Research~

18, 931-993 ~ .

PAR:'ACAS J 0 F. (1978) Surhc(i ,duds in the Georgia Embayment~· _University- of·

Mi.:irn:i.:, RSM.AS, 11 i.11al Report to. Sdcncc Applications, ·Inc., ·on subcontract

025'-:06. . 34 i'>P-_-

PIETR.J\FESA L...:J., L. P. ATKJ.NSON and·J. O. BLANTON (1978) .Evidence for deflection

. _of the Gulf Stream by _the CharJ.eston Rise •. Gulf Stream, IV (9) ,• 3-7 . ...' PIETRAFESA _L. J. and G. S. JI\NOl\'ITZ (1979) A note on the· identification

of a Gulf Stream Spin-off Edcly {1,om .Eu.1crian Data. Gcoph:ysical Research.

Letters, 6 (7j, 549-552.

PIETRJ\FES/\ L. J •. ;rnd r.. S. ,J/\NCJ\HT:7,(1980) Lack· of evidence of sotithe~ly propa- · gating shelf waves in Onslow Bay, N.D. Geophysical Research Letters, 7(2),

113-116.

ROONEY,·D._, G•.. JNNOIHTZ ;1nd L. PIE'l'RAFESI\ (1978) _A simple model of the··

reflection of the ·Gulf Strc,iin by the Charleston Bump. Gulf Stream Monthly,

4(11), 3-7.

ROUSE, II. (,1963) On the rol c of cdclics in fluid motion. American Scientist,

285-314.

SAUNDERS, P. M. (1971)- Anticyclonic eclclics fanned from shoreward me:mdcrs of

the Gulf Stream. Deep Sea Research, l 8, l 207-1219.

SCOTT, .J. T. ancl G. T. CSI\NADY (197(>) Nc:1rshore currc.nts off Long Isl:rnd.

Journal of Geophysical lki:i.:arch, 81, 5401-S 1l09. SMI'i11, P. C. (1978) ·Low-frequency fluxes of momentum, heat, salt and nutrients .., at the eui:c of the Scotian shelf.-· Journal of Geophysical Research, 83(C8),

4079-4096.

STEFANSSON, U., L. P. ATI~INSON·and D. r-. BUHPUS.(1971) Hyclrograph_ic prop'erties

and circuL1tjon of the North C:-nol in;, :shelf an

Research, 18, '383-1120 •.

STEFANSSON, U., and L. A1XINSON (1971) · ~utricnt-clcnsi ty relationships. in the

\-icst.crn North 1\tl.:intic between Cape Lookout and Dcrrnuda. ..Limnol ~ Occ:mogrZlphy,

16(]), 51-59.

STOMHEL, II~ M. (]9fiS) TI1e Gulf Strc;:un, University of C.1lifornia Press, 248 pp.

STRICKLAND, J.· D~ H. ;rnd T. H. PARSONS (196~~) A m;inunl of sea 1vater·~nalysis.

Bulletin Fi~hcrics Hesc:nch Bonrd of C:rnaJa. 125: 185 pp.

STUMPF~ H. G. and l'. K. RJ\O (197S) Evolution of Gulf Strc:im eddies as seen in

satellite infrared i1nagcry. Jotnn;d of Physical Oceanography., S, 388-393.

TilOMPSON, B. J. and G. A. GOTnJARDT (1971) Life cycle of a_ North Atlantic ecldr.

Transo_ American Geophysical U~1ion., .52, 241 (abstract only).

UOWPI, Eo (1968) Atlantic continental shcl f .1nd s'lope of the United States

physiography. Geolog. Survey Professional Paper S29-C, U. S. C,overnment

Prj_,nting Office:, Washington, D.C. 1-30.

VONARX, W. S., D. F. BlNPUS a-nd II'. s. Hlr.tlAIU>SON (1'.JS:;) On the fine structure

cif tl1c Gulf Strc~n front. Deep-Sc~ Rescirch, 3, 46-65.

WEBSTER, R. (1961). The effect of meanders on the kinetic energy balance of

the ·Gulf StTC;Jlllo Tellus, 13, 392-'101.

WILLE, R. (1960) Karman vortex streets. Ad\'anccs i11 Applied Mcch:mics, 4,

185-196. . rr GllRE. CJ\l'TI ONS .

Fig. :1.1 Locatio·n of current me.tcr mocirings J\-G (0); coastal sea· level ·(.A)· and \vind (@) sfations,

Fig.· 2. Surface temperatures. from T/S. mapping April 12 (092-5 hr.) to

. ··. 14 . (l O30 hr. ) , 19 77 • Circles with:

Ship track is.shown hy straight litws and stations by small ~lots.

Stations ;1t. section .cncls arc .sho1~n in bi-ackcts. -Arrows indicate· . . . current diTc~do.i1.

Fig. 3. Surfaqc tempc_raturcs from T/S mapping April. 1'\ (10:rn: hr.) to 16

. (()92]). C:ircks w:i.th dcits shcnv IHO()Tii1g locations. Ship track .

is show1i ·by straieht lines and stations .by sr)lall dots. Stations

at section ends arc shown in br,tcke·ts. Arrows indicate curr·!nt

direction.

rig. 4. Satellite Vl!lm IR image of a Culf Stream frontal eddy from

orbit 3223 NOAA-Son April 16, 1977, at 0200 GMT.

Fig. S. SST contours-· from alpha-m1mc.n·ic presentation of orhi t 3223 ··

NOM-5. VIIRR IR images averaged every 2 km.

Fig. 6 Co1:ipositc of hydrograph:ic sections through eddy F, April 14-16.

Stations arc identified on bottom profiles.

Fig. 7. Composite of nutrient sections thrnugh eddy F, l\pril 14-16.

Stations arc idcntifictf on bottom profiles.

Fig. 8a. llorizontal temperature (C) distribution at 16 m through eddy r,

April 14-16. Circles with smal 1 tlots arc mooring locations.

Dots arc for station location. Section end stati.ons arc in·

Fig. Sb·. llorizontal tcmpc-r.1turc (C) distribution at 30 m through eddy F,

April 14-16. Circles with small' dots are mooring locations. Dots

are for station locations. Section cn

Pig. 9·. Depth of the 26 at surface thrnugh eddy F, April 14-16. oots are

for station Jocation .. · Sequential pas i.ti ons of eddy E on J\pri 1 8 to lS, taken from

infrared VIIRI~ im;tges· from NOM-5. night (0200 GHT) orbits.

Circles show the current meter locations.·

Fig. 11 o Sequential positions of eddy Fon April 1S-t7, taken from

i nfrarcd V\ll~R i111,1gcs· froin N()/\J\-5 day (1400 GrIT) orbits~

Circles show the ~ocatio•ns of cun:ent meters E and r.

Fig. 12. Time series of: (>-hourly, i-otatcd qirrent and wind ve_ctors from

4()-IILP filtered records for J\pr:il 15 to Jun_c 29, 1977. ·Magnitude

and 1·ot;1b on of wind antl current _vectors are, shown by the· scale '

arrows 011- the left.. Instrument depth and wate1~ depth ar~ shown

on the 'ri.ght. VcTtical lines -arc. for· eddy ancl meander event

jdcnt:i.fication.

Fig. 13, Ti.me series of rotate\l velocity compo1\ents u (-0-0-0-), v ( I I I ) ,

cm/sec, and tcmpcrature_T (XX X), C, from 40-IILP fil-r6redrecords.

for J\pr:il lS to .June 29, 1977; Vertical lines are: for eddy-and

meander event _:itlentification.

: . . . . Fig. 14. Time series. of _I-hourly, rotated cutrcnt an4 wind vectors from'· 3-HLP

· .filtered reconls for Apt:il 12-24, 1977~ Magnitude and rotatioi1

of w:inll an

the left.

Fig. 15. Gulf Stream· l;oundar:ics off the southeast U. S. as shown by U. S.

Naval Occanograpil.ic Office i.n Experimental Ocean· Front·al Analysis

charts for ,,•eds nf .i) ~lay 12-JS, 1977 ~111db) ~lay 19-2S; arrows

ind·icatc current di rL:cti.011. Fig. 16. · Time series of· fr--ho11rJy, rotated curre.nt and w:i.nd _vectors, and

corrected sea level from Ch;rrlestou (Chso) and Daytona ·(Day..,),

40:..111:,pf:i ltcrc

level corrected for atmosphei·ic pressure ·by 1 mb pre_ssure = 1.01 cm

sea level" Magnitude ;md rotat:ion of ·vectors shmvn. by_ scale arr.ows

011 foft. Instr11i11ent depth and w,it.cr depth arc sho,;,,n on right.

Vertical lines are .for Clldy e_vent identification.

· Fig. 17. ·Time series of 6-hourly, rotated current an

corrected sea level from Cha:rleston (Chs.). and Daytona (Day.),

40-IILP filtered recc)rds for July 6, 1977 ·to October 31. Sea- level •S,..

corrected for atrnosp)1cri c pressure by l mb pressure = 1. 01 cm sea

Jevel. Magnituclc and rotation of vectors shoh'n by scale :irrows on

left. I nstru111c11t depth :111d w;i tcr depth arc shown on right. Vertical

lines arc for eddy event 'i dent i. f.ica tion.

fjg. 18. Schematic characterization of a Gulf Stream frontal eddy on the.

Georgia shelf.

Fig. 19. Composite T-S plot of all hydrographic station data taken during

April 8-16, 19_77. Station· nos. are given .beside some of the

identifiable shelf stations.

Fig. 20a. T-S clistrilmtions from st;1tions of Savannah (-1) section, April 16.

20b. T-S distributions from stations of St. Augustine North section,

April 1 S.

Fig. 21. Schematic of annual nitrogen.inputs to 'the Georgia shelf as computed

hy 11/\INES (197'1) ~ Values aTc in tons of nitrogen per year. J\rro,vs

indicate flux· direction. Out est:imate of annual eddy induced flux

is shown in parenthesis. Table 1. Relevant in~;truri1cn:t inforn1atio11 for Georgia· shelf currc~1t. meter deployments'\. ·

water instr. .. current record lengths moo;ring instro

.A ; 31° 00.S'N AT 30 17' I . I .A 80° 27~9 1 w· . AB 30 27'. - - B 171 '. ! 30° 50,8'N UT 7 'i B'· 0 8p • b2~0'W· n11 75 72 - .. - C Jl.o 11. J.'N CT ·75 17 C . 79°· 53.1 'W Cf3 75 72, D .DT 17' -- 31° 25.0'N 75 ' D I 79° 45.I, 'W DB ,75 72 E . ' ,· 31° 35.S'N ET 75 17 -' - - 1 .. r: 79° 40.2 \{ .EB 75 72 F 31° 39~6'N FT l15 17 •' - F 79° 50.9'W FB. 45 421 G 31° 4l+.,8'N GT 30 171 - G 80°, 08.5'H GB 30 27 i I Table 2. Sequence of hydrographic sections, April 8-20, 1977.

start encl

(hour/day) (hour/day) ; section EST EST section type stations ship

14.8/ 8_. 4o4/ 9 Savanri.i11 1 STD/XBT 1- 14 S/S Advance II 1.4/10 · 9.4/10 SL Simons 1 STD/XBT 21-.32 II 3.1/11 10.6/11 · Savannah 2 STD/XIlT 33- 43- II .· 15.4/12 23 ..8/12 Savannah 3 STD/XIlT 53- 65 II 17.7/13 2. l/ll1 Jackson:ville-Sou th STD/XIl'f 100-113 II 19.1/14 0.1/15 St. Augustine-South STD/XBT 174-182 - " 4.1/15 8.1/15 Ormond STD/XBT · 192-199 ... II 12.1/15 16.• 6/15 · St. Augustine-North STD/Xl3T 213-22.0 II 16.6/15 23.0/15 . .Jackso11ville-N0rt:h XRT 229-236 " . 2.7/16 5. 7 /i6 st .• Simons 2 XBT 249-258 II II 8.7/16 1404/16 Savannah 4 STD/X.BT 259-267 ,. 15.3/19 4.0/20 Savann;:1h 5 XBT A-P R/V Blue Fin Table 3. Data used in nitrate. flux cal~ulati6ns, values ar~ daily averages.

I N0 _-(72m) (N0 ) (NO )' 3 3 • . 3 . datq . u (17m) u (7 2m) . (t_l) T· (72m) ( u) -1 -1 ,Q,-1 (April ·1977) cm/sec cni/ sec cm/ sec . oc 'lJ moles R, . lJ moles ,Q, . cr.1/sec ).I moles ..

15 3.5 .. -2.-3 0.60 113.2 5.7 2.85 0.04 -2.75 16 -5.3 -8.2 -6_. 75_ 17.1 8.7 4.35 ~7. 3i" -1. 25 17 -4 .o· -:3.5 -3. TS 14.5 15.-3 7.65 .,..,.~ 31 .· z.os · 18. -4;7 -1.6 -Jol5 14.0 -16. 7 · - 8. 35 -3.70 2.75 19 3~6 '20 ,, 3.00 1'1.7 14.9 7.49 2.44 .1.89 20 ll1.o· ·8. 5 11. 25 16.5 10.2 5.10 · 10.69 -0.50 21 7.0 -1.5 . 2. 75 17 ,,8 6. s· 3, /10 2.19 -2.20

...

.... 82° 81° 80° 79° 78° 77° ~~~=i::;:-f;:-~~~:S~:SE:~E:~~~~~3::::~¢~g~~~~~~~~:=E$=S::::::~==';::E:==::::::~33' 0

, .. :- ,• . ,··.·· ...·· .. :. ...·•'

32°

E .. 0 0 co-·· Q 13/ 31° 0·

E 0 E 0 0 If) 0 ~ ·-· 30° 50

E ·:x: 1- 100 a. o Current Meter ~ 150 29°

·----.. 2000 50 100 150 Cape Canaveral DISTANCE (km)

1 kilometers .-·••' ~~ O . 25 50

~:?""-~l:..!. 1-LJ-L.L:t.=L:..~ J..:!.:~~~i:.:!~\rr:..us-::-~:.-.1.::.L·1_:"c":i: T~J:j :!=1'2::i.::!:"1'.=.:::i,:-.c:..;-~.,_.:_: -,-r-ccc---r::J::! 28° 82° 81° 80° 79° 78° 77°

Fig. 1. Location of current meter moorings A-G (0); coastal sea level (A); and wind stations (e). 82° 81° 80°

I I /. I I 31" l I I ~ I 0 I . C\/ I .,I I .... I I

[100] ~ \ ~ • ' --;--~---::-::--.: •SOUTH I I 25' I I I 21 I I I t I \ \ \ . \ I . I I \ E I ' \ - I lo \ 29° I ,· IQ C I I lt'l ~ I· I 0 \ · E o ., I I u> ' E g, 1 ' \ 0 _, \ \· it) I I \ \ . I I I \ CAPE CANAVERAL I I I I ~ __ _.....__.___._L_..1__11 I) . I J__ .._~ _ _.__....__...._--' 82° Bl" . 80° .. Figo 2. Surface temperatures from T/S mapping April 12 (09~5 hro) to 14 (1030 hr.),. 1977. Circles with

Fig. 3. Surface temperatures from T/S mapping April 14 (1030 hr.) to Hi (0~)21)" C:ircles with dots show mooring locations. Ship track is shown by straight lines and stations by small dots. Stations at section cntls arc shown in brackets. Arrows indicate current direction" Fie , ~ \J lf St at 0200 (;~rr. - _23 NOJ\!'\-S. . on J\pril ream16' 1977 , 32N

i : l,; ~ ~; ·.. : .- ~ :~ . : : : ~-

30N

N5"- 3Z'.l3 /(, IIPfllL 7 7 ri g . s. SST cont01t1·s from ;!lpha-11u111cric pre sentati on of orbit 3223 NOAA-5 ~ IRR IR imige~ averaged every 2 km. "' .,, ~ v• or ,., l~ ~ ~. ~ -----· . ~---··--- ··.;;>~ tti'• • 'fl.

1 ,',11 ,"i.1\,4.N,'.'111 ~,4,t , .....,,, '.:i' I I ~·1 It, .tfW/1 111··, I I 1 H\ .J'.-11 I 'I(\ .

·10 30 ·.JO g 70 JA[KSONl' I I I C. N J: 90 l[Hl'[f,'A //JR/ (l. I HI 15 Af'f,'Jt. 1977 '"0 130 '·• 150 - J"/ll IUfl --L• . .1_.__1 ..~. J _. . .1 __._t .. - , .• .1 10 31-1 '.,11 711 'Jll 1111 I :in l'.,,1 0 I 5 l Al/[I ()I I 9<11111 (KM)

~ ~ l': ~ ~ ~ ~ ~ ~ ;;; N N ~ HI .' • '----.... - • - • ~ •'-..•,.: 10 1 3H 30 -~~} 50 ~rn 511 g 71l 7(1 g 70 ..... ~ - \-\~...... _z,.• 90 911 - ST. AUGUSTINE. N r-,.:.,._, 911 ST. AUGUST/NC. N ;.: T[Hl'lNA TUh'[ /'\ ~ ;: a. '·11 SALINITY SICHA-T 110 n. I HI 0. 110 .., J5 lif'R/L 1977 ,,, - 0 130 \,. CJ /5 APF.'/I. lf/77 15 APRIi. 1977 13(1 ~ 130 - 150 - 150 - 150 170 1711 . 170 190 190 ~-'~~L.....-L.. L.,...J...... -L. 190 _:._i_'--L~L~ ~L.. L~L. L,__L,.__.._ ...... ~~~.L--~~ J 0 31l 50 70 \lll I Iii I 3ll I SU ID 30 50 711 90 I Ill I 30 I 50 IO 30 50 70 90 I 10 I 30 I 50 01 ST MICE OFFSHOh'( (K•O OISlANC( OFFSHORE DISTANCE OFFSHORE CK~>

Ul -'--'j~,;'I~:•-•...~~- ,.~ )ll 10 30 - -~((/ ,, 30 30 50 IJj 511 ~:?rri{i 5n - ;- 70 'N,, ~ "/H ~ 711 90 ST.AUGUSTIN[. 5 9(1 SI. AIIU/5 T /NE. S :--··n.1> SI . .<.<- T .,, ,,,(l. IHI o. I HI a. IHI /4 Af'R/l 1977 ',, ,.. 1' .1/'h'//. /977 ,, Al'.RIL 1977 0 130 n 1·111 "'0 1311 150 151! 15U 17H - \ " 1·rn 17[1 190 • I !HI I ~10 :JJ...J_, __l ,_I,.! .l ,_J ,.I I.,, l , I , I ... l .I , I.-. L I -l- I -·-· I _.__i_,_l _L..._j_ ... Ill 3(l '.,(l .,., !HI 11n I :ia 1',11 1u :i.i ~'" ·;n ~m 110 1.m •~jn I ;J :m '.,II 70 mi I I '1 I 3(l I '.ill 0 IS 1 ANt-1 (JI r 9

190 _1___.___1__._.J. ....__J_ __.____1 ~--1 _..,_ I ._J_ 11J'r1 IY0 l O 30 5'1 /U tia I 1~1 I Ji, I~" DISlANC[ OI I '.,Hllh'l

Fig. 6. C:ompositc o F hydronr.:1ph'i.c sect.ions through eddy F, . April 14-16. Stations arc i.clentificcl on bottom profiles. "' ii: "' ~ ~ ~ le ~ ,:; Ill •1 30 2 511 ~ 70 ;:: 90 ;: 90 a. I 10 11. llll - ·= ..., 16 APRIL 1977 Cl 130 Cl 130 150 I Sil - · 150 170 · 171l .. · 170 190 190 iY•1..J__._..1_,__1_,,_J_4-1_ ...__ J._. __j_ .. 1- ..,__L...... L-,.....1-,_J_. __~~-~ 10 30 · s~, ·;o .t.H1 110 I 3il I ~n Ill 30 50 711 !J;J l 10 \ 31J 150 0.1SI AI-/Cf. on ;,HIJl

..,.. UI VI ..... ~ ~ ~ ~ .· --··-·---"---··•--'-•··* --- · 111 Jll ~-~------✓z;(~ . 50 '/0 - :-----,:.· . ~- ," s1 .. ,ui:uSIINI .. N -- .. ~~ i !lll 1 \Ill NJIRATl <,_,l>IJ. JJ1fOS."J11U a. llfl n llll 1.i~' w. 15 APRJI.. 1977 u 13~- t: IJ(, _. 15 t.PNIL /91/7 . \'JII_ 170 1711 - l!l0 __.___i_. _J--' ..J.~-1 J_ _, __L...__l l~fl- ::_1:_,_.L.,._L..,.L. ... L:., 1 .. ~L., .. 1'1 30 'Sil 70 911 I I U 133 I~:, 111 Jll '.,II "JC ~;, I Ill I Jll I ~[I D 151 MICE orrSHORE O,M) 0 IS I At~C( orf' SHORE (K~}

10 30 - 50 g 70 :z: 90 ST. AU(.IJST/N[. 5 ·n:110 NI TRAT[ !1,/>IJ w 14 APRIi. 1977 0 130 150 170 .170

190 .__.~.,,..__~_,_i__, 1--.-L---:J. 1911 1.~L...._i_.__i__._ ... 30 Sil 711 90 . I I U I 30 I Sil 10 3ll 511 70 90 I IO 130 150 DISTANCE urrsHOR[ CKM) OISIANC[ OHSHOR£ (KMl

"' !'; ·.~ ~-~ ~ ~ '" ~ ______._·--- ···-----· Ill i"o 10•~- .·. 3H - -...... • --,- ,I : ]ll 30. - ~II - ~" 5(1 '\ ] ·111 ·10 '] ~ 71l ~ OR.'<()N{)/l(M!I \\, (If,•.~, 1'-0 /!f.lCII ~ \ ,. 911 J !m J: n~ - NJ/RAIi /.,HI \,. SIL /C.• 1£ (,,1MJ !> n 1111 . 7:, a. 1111 f!i JI~ Ill . 1 ..., '· ~ / ) AJ-'f,'/L /!Pl c.:, /) -•PRIL 1977 t:J 13,~ - C.l I J:l l'.lll 150 - ,.,,, 11,,, I ·"1 1111 1·111 '(j~\ 19\l 1\111 : L..._J.~..J~ _1_, ·-' • 1_.,_J ... , __ J., .:.1 .. ~-J ...• _I ...__l..,J_,_ J_,_I_._J_ _!_,.1._ I _.._t__..__ 1" 30 ~)~~ 7;, 1m l l n I J;, · I ~10 I~ 311 Sll ·111 \Ill I I ,I I 3:, I ~ti .Ill 31l 50 711 911 ! Ill 13:l 15.J OISll.tl(f or•~Hfjf.lf (KM) 0 I~ 1 Wll orr 9illR( '"" DISlANC( OffS•cCR~ <~M)

Fig. 7. Cornposi tc of nutrient sections through ccldy F, April 1,J-IC1. St~t:ions arc i'c!cntif:iccl on bottom profil cs.

.... E

o. QI [25 ~ • I t ... / 23 ✓- . /. I . I . [249].

QI 0 ., .- 31 &. [258] 31°

,p 6 f [229] I I [236] J • I I B [213]· (220] 0 30 30 ,1 _..-[174]:

0 ·.O O 0 •

0 20 40 60-KM t!ft::=..~ .;.;mes= 0 10 20 30 NM

80° 79°

Fig. 8a. · llorizontal tcmpcr.:iturc (C) distribution at 16 m through edcl)'· F, J\pri1 l.'l-16 •. Circles with small 'dots arc mooring locations. Dots arc for station location.· Section end stations ;ire i. 11 brackets...... ·\

81° 80° 79° . 0 3 2 r---~----,,---~-r------.----.---,----r-~-----.---

--·

I . I [258]. ....

t229] [236]

0 30 [213] . [220] 30

[174] [182]

---..,,o_ g I :199] 0 [192] " " _ 2325 ~~~~..j,OKM 21 () 10 20 30 N M ·

80°. 79°

Pig. 8b. Ilorizontal temperature (C)

' . ·N

• : 0 .... -- •

C 0 0 31 e • 31° · ...

\· 0 30 I : . Depth of _2~

•·

i 0 20 40 60KM .· ..8'7~i

29 ---~-.....___.....__...... ,._-'-- _ _J._ _ __.___ _ _J.__,--J.. _ __,1,._~ _ ___,J 81° 80° 79°

Pig. 9. Depth of ·the 16 ot surface ·thrm1gh eddy F, April 14-16. · Dots arc for _stat ion lo cat ion. .. . '

33° ..

4/13-=_..,,·_

4/10 ...... ··········· ...... ····•··········· , ...... ·········...... , ... ' ...... 31° ...... ····· ...... 31° ...... ·······• • • • • I ..• • ··········...... ·········· .. , ...... 4/8 . . ········...... ··········...... ······ . ...'' ...... ··········...... ·········...... ···········...... ······..········ ·······...... 30°

29° 29°

28° 80° 79° 78°

Fig._ 10. Sequential positions of cdcly E on J\pril 8 to 15, taken from :infrared VI!RR images from NOJ\A-5 night (0200 G~!T) ,orbits. Cirdcs show the current meter locations. . .

_J 0 0 co ; 30°

29° 29

Pig. 11. Sequential positions of c

,,·, /42m 45IT1

17m 75m

72m 75m

16 21 26 1 6 11 16 21 26 31 5 10 15 20 25

r-APRIL-----MAY-----t----JUNE-. ----1 1977

rig. 12. ·rime series ·or (,-hourly, rotated -current and wind vectors .· from 40-111.P fi.ltcrcJ.rccords· for J\pril lS to June 29, 1977. Magnitude and rut.ation of ,-.,:incl and current vectors arc. shown by the scale arrows.on the i~it. · Ins~nimcnt depth and wat~r depth arc shown ori the right. Vertical l incs. arc for eddy and me;rnclcr event identification.

... 80 28 40 24 FT ·u,v 0 20 T -40 16

-80 18 80 28

40 24 FB O , u,v 20T .-40 16 -80 12 80 28

40 24 ET O u,v 20T -40 16 -80 12 'I j 80 i 28 i ! •, 40 I 24 'EB i 0 ~~), 20T u;v I ! I -40 i 16 -80 12 6 11 16 21 26 31 5 10 15 20 25 --MAY----1---JUNE--- 1977

Fig. 13. Time scr i.cs of rota U.:ll ve loci t.)' compo11cn ts u (-0-0--0--), v ( I I I ) Cl'l/ sec, and t(.:lllpcraturc T ( X XX), C, frOl'l 40-lll.P fi l tcrctl rcc.orcls for April 15 to .J1.111e:?9, 1077. VcrLic:-11 lines arc for cchly and meander event identification. .. . ·'

_:~1~~,7~~~-·/f'j;t ...

.tddy E-...:~. I

12 13 14 15 16 17 18 19 20 21 22 .23 24 APRIL 1977

.. Fig. 14. Tjmc series of. 1-hourly, rot;:itcd cti:rrcnt and wind vectors from 3-IILP filtcrcdrcconls for Apri.1 12-14, 1977. Magnitude and rot at ion of w:incl and current vectors jre ·shown by the sc:1 le· arrows on the l cft. ·/' ~,; I•~ ~ I -., ' ,, \ 1- t I '' I ' ' ...... - . ','',, \ ' ',,,;· \ '•' '--~ ' \

' , ~' , .\->' \ \. ', \ '' ' , ._.j::::::::::·••, .,' ~, . \ / c::,' ',· ' '.\~, ,-.,c::, ' ', '\,'""' ' ...... ____ ....;.... __- ' -.. ooi- ~· 0 0 0 ro 0

0 0 ... 0 tO 0 tO r<> r<> C\I

/ "' i t ' ...._!: ... I ~->- .... J\ '\!:;,_, I

' .... I ~, I , .4 \--....J ' ,_ ... ______,,,,_-' i§ ' '-wOO~ .._ ~_,:-i~

.·.·,·,•. ·:,:,•,•:····· ',' ;:,:,.f

0 0 0 tO 0 I() r<> r<> C\I

Fig. 15. Gulf Strc(tm boundaries off the southeast U.S. as shown by U.S. Naval Oceanographic Office in Experimental Ocean Frontal Analysis charts for Wl'cks of a) ~lay 12-18, 1977, anJ h) May 19-;:s; arrows indicate current dir<'ction. "' 45m

r---,-----·-1------1 ·- ·--1·-···- -r ·-;--· -1-~---,·--··-r---···r----,-----,----, 14 24 .3 13 2 3 2 12 2 2 .4 14 24 · 3 13 1----oEC---1----JAN ·------1--- ·-·FEB·. MAR --!--APR I . 1976 . . 1977

Fig. 16. Time series of 6-liourly, rot;itcd current and wind vectors, and corrected sea level from Charleston (Chs.) and Daytona (Day.), 40-IILP filtered records for llcccmhcr 14, 1976, to April 6. Sea level corrected for atmospheric pressure by 1 mb pre-ssurc = I "Ol cm sea 1eve1. ~!agnit.rnlc and rotation of vccto,r~_shown by scale arrows . on left. Jnstnuncnt depth anti water

E E E E .E .E ,-...11.0- !J') ,.__Ill) ,-<".),.... 10 ,- '¢ "-11.0,- ,.... v-,-.:.I,-... ,- ,-... ,-...10,- M

II) () 0, ') N I t:'d0) T·· q ~· w >- f ~ a: 0) w I?, ) ~ ,- co ~ 1 'd' 0 ...... ····=··~.~········· ,- u1- 0) 0 ---·---J -: .~~· -1 ~ V en .··················.~•······Ire J·...... ------·--·-·-·-·-·~~~-,~+------·- C\I v CN 1, 0, ·. a, ~ . ,-. 1 ,- ·-·-·- ~ -·-·-:.· - . ·--·- .. - . ·- - -·- .. 5: ,-~ w (/) '-- ,-. • ·~r·. -~...... ---·-- . -•~' •.: -.:iI 0) )I. 0) . ·- ·._:_.:;.:.:~ ) -~I .. • .or; . -·-··· '¢ 1 '¢ ·!, ...- :>. ,-:.:.:-.:...1-·;;- ,..._I' ' ~ - - • -✓ 1---:- ,.~ 0 0, ·-·-·-·-·----·-·-·-·'-··-· g ---·-·----·-·--·-..._,_ --- - -·-----.. -.---·- ---·-.:J_-_;;- --·-·-·-_·--r..:.::=:--·-·-·-·-· M ,... lO N 0 VI~~ ,\\;i,,: C\I t; - : ~1•1lO .,, '1.- ~ ~:--;JP .I.O,... :;, •l ' w ' ,- ---::-~ ~ . -~-~~ (!) -~ ~ Q ~ ~~ ~~~:yt-:~ ~ L lO ~ ~ 0~~~~i A·-·-·-·-·-·-·-·-·-·-P-·-·-t M . ~ . . ~-. cc. . (__~1 n·,/ ,,,\4,# __,_·-..f~~l ·---·•·-·-·-·-·-·-·- - -·- N ..-· 7 ·t- ~ ,- ~I N,- ~ ·,:_it,..-:_...... rt ·i .. l C\I ~ 11· ,,....(/)'j"··1, .--"·-~· - . - <.O I ...,_, ...._ ---~ => I • ,- E - , ' -- -,-':_.. I- . ,- -, I " _.. . zEu .~ !j $?"-1 , 1 ,.::-::d~·J } I I 'O I •..., ~-i -~:-·1 . . . a: I.{) • I • ~ I :" ~i~ ,----1 ...:- co ~ ' l -' ·=::..:.--- ::> :1 O lO O O lO O ~ ◄ £. 1- 1- 1- I- I- U 4 I- N ,..: ,..:N ,.! r! ~7 c, u. w a co z>'\...o

Fig. 17. T:irne scr:ics.uf (1-hourly_, rot;it:cd current and 1,ind vectors, and cc,Trcctccl sc;1 level from ClinrJcston (Chs.) ;incl Ibytona (Day.), 40-!IL!' fjltl'n·d records for ,Tuly 6, ]977 to October ."i]. Sea level corrected for ~.lmosphcric pressure hr 1 rnh pressure = l.01 cm sea level. Ma1:nituclc ancl rot;1t':icrn· of vcct:01·'.; shoh'll br sc(llc ,llTOh'S on left. Instru111c11t depth ;111d \,•;1tcr depth ·;ire sho1m on right. Verb cal l:incs arc for t~ddy C:VC'lll idC'11t:if1c;it'inn.

......

·. 0 . EB EB EB ..

CJ) ffi 100 1- w 20 ~ oo

150 ()

if!l:i ..:\:!\!l\:!l!l\:l! pwellin .•.•,,·,· .... ·.

50 100 150 KILOMETERS

Pig. 18. Schematic characterization of a Gulf Stream fro11tal eddy on the r.corgia shelf. • -I . :. '

25 w 0::: :::J 20 I- < 0::: w CL w~ I- 15

29 30 31 32 33 34 35 36 37 SALINITY

F:ig. 19. Composite T-S plot of al! ,hydrograph.ic st:1tion data taken dur:ing April 8-16, 1977. Statioi1 numbers are given beside some of the iil,~nt:if:i;:1hlc shelf stations. .. . . ·,.

35 36 34 35 36 SALINITY SALINITY a b

f-ig. 20a. T-S clistrilH1t.ions from statio11s of Sav:111n;ih (4) section, April 16.

Fig. 20h. T-S clistribt1tio11s from stations of St. Augustine North section, April lS. .. . '

NITROGEN INPUTS TO THE SAB

75001 N/yr ... 12600! N/yr , ______::u_____ --r ___ _

Fig. 21. Schematic of ;111n11al nitn>t:~n inputs to the Georgia shelf as computed hy 11/\TNF.S (El7:1). Values arc :in tons of nitroeP.n per year, /\rrow·, indic.:1tti'flux d1rcct·ion. Our estimate of annuai cclcly induced flux is shown .in p:1rcnthesis. l

... NATURALFLUORESCENCE .AS A TRACER;r,'OR DISTINGUISHING

BETWEENPIEDMONT AND: COASTAL PLAIN RIVER WATERJ~N THE

-NEARSHOREWATERSOF GEORGIAAND NORTH CAROLINA

Joan D. Wiiley

Marine Science Program and Department of Chemistry

P. 0. Box 3725

University of North Carolina at 1.vilmington

Wilmington, North Carolina 28406

and

Larry P. Atldnson

Skidaway Institute of Oceanography P. 0. Box 13687 .. Savannah, Georgia 31406 2

Keywords:... tracers; fluorescence; silica concentrations; salinity variations; humit: substances; coastal waters; Georgia; North Carolina .

Natural fluorescence ,.:dissolved silica and salinity were investigated ~s possible tracers to dis'l,lnguish between piedmont river and coastal plain river waters discharged into the ocean off North Carolina and Georgia. In the Georgia study area, dissolv~d- silica was not suitable for use as a tracer because silica concentrations were.variable and did not mix conservatively with seawater. In the. North Carolina study area, dissolved silica concentrations exhibited too much short term variability for tracer use. In_botli areas, natural fluorescence was· a suitable tracer. Additional investigations relevant to tracer application were made of the method for determining natural fluorescence; these include dependence on temperature of analysis; pH dependence, sample storage effect,

sensitivity, correlation with:total organic carbon, and possible interferences

from chlorophyll-a, lignin sulfonates, petroleum, and iron. .. . .• .

. Introduction· .

. . . The pu:r:pose_ of thi~ study was to investigate whether dissolv~d silica and

...... • ... :- natural fluorescence·can be used.to di-stinguish between coastal plain and piedmont

rive; waters where they flow irito_ the nearshore. areas of Georgia ·and North Caroline .. ... Rivers· which originate in the piedmont area of the southeast.ern United ·st~te·s· (the· . . ·...... ea.stern margin 'af the: 'Appalachian Mountains) flow d~wn steeper .g,.rad~ents a:nd so

flow fa_ster and ·tend to have ·higher su,spended sediment loads and :may therefore · have higher concentrations of dissolved si_lica than coastal plain rivers. Coastal plain rivers ori°ginate:.- in swamps and marshes ·.on·the flat coastal plain, are

slower fiowing ri vcrs f:Ll,ldtend to have higheJ:' organic l!latter cont.ent than· pied..mont rivers. Because of this, coastal plain rivers should have higher natural fluorescence than piedmont rivers·.

Development of paramet·ers which can be used as chemical tracers should allow better definition of circulation patterns in the nearshore area. This is of particular importance in industriaiized areas like Savannah, Georgia, and

Wilmington, North Carolina. Di_ssolved silica and natural fluorescence -are of

interest because they have the possibil~ty of distinguishing between water from closely spaced individual rivers. This may allow better resolution on a ''small·

sea.le of vo.tcr ::iources wher':'! o~ht:r. Lnwers are inadequate. lt'or e,ca.niple, stable isotopes (oxygeri-18 and deute;ium) can be used to distinguish between water from . . the South Atlantic Bight, the Mid-Atlantic Bight and the Gulf of.Maine, but because these tracers depend on climate (temperature), precise resolution is not possible (Torgerson, 1979). Natural fluorescence and dissolved silica may hel:)? to d:istinguish waters from different sources which experience the same climate ......

A chemica,l tracer is defined as any chemical parametE:r that can be used

to _follow the· movemepts and mixing of water from different :sources. Dissolved

silica has been' used· as a tracer in·'\:>ther areas ·( for example, Gardner,· 1977; ·•, ·and Hunt, 1978), ho~.ever, its use is limited because o.f b~ological .utilization

~ . of_. aissolved silica _(Fanning and Pilson, 1973a; Va~ Bennekom, et. al.;. _1974·;•

anti H1,mt, 1978) ~. Natural fluoresceric~ ,·· whi~h 1n fresh wate~ i~ thou.g_ht to...... , result from the presence of '.,hurnic ·~om!)ounds· .(Kalle, 1963)'~ has been· '\J.Sed as a·

t·ro.c·er lJ;·.K_alle· .( 1~49) i~- the Gulfs o·f Bothnia and Finland~ b~Duursma ·.. l 196~)

in the Dutch·Wa9-d~n Sea, and by Otto and Zimmerman a~d Ronimets . ·(1967) . {1974) I I • f I ! in the North Sea. In each case,. natural ;fluo.rescence and salinity were nega.; i i t~vely corr~~ated; indi"cating conservative n:i,ixing. . ). Several criteria must be _satisfied if a pa.r~eter is used as ·a chemical

.tracer (Zimmerman and Rommets, 1974). The· concentration of the· tr~cer must .

differ by at least a factor of two between ~he end member~; the c0ncentration

of the tracer must remain constant in the end members over the time period of

interest; and the tracer must mix co;nservatively-with seawater. These criteria

were evaluated in the prese~t -study fo:r both dissolved silica and natural

fluorescence in two areas, northeastern Georgia and southeastern. No.rth_ Carolina

(Figure 1). The rivers studied in GP.oreia are th_c S~va1mah Rive't, ·which is a

piedmont river, and the Ogeechee plus Canooche River which are coastal plain j. rivers. The Canooche:River drains into the Ogeechee River; both the Qgeechee and Savannah rivers discharge directly into the ocean near Savannah,. . Georgia. ,j ! In North Carolina, the Cape Fear River, wh~ch is a piedmont river, ~nd the

Northeast Cape Fear River and the Blac~ plus South Rivers which are coastal

plain rivers were investigated. The South River drains into the Black River;

the other three rivers 0flow into the Cape Fear River Estuary near Wilmington,

Nor\h Carolina. sc t:·n r·nn·wK-:..

Sampling methods

Sampling in the Georgia study area was conducted on May 16 and 17, .1979. ·

All samples were collected from a boa£" in or near the estuarine ·r;gion; up to·4 km offshore. The intention was toobtain samples over a wide salinity .... rang.e. This was achieved by measuring the salinity on board using a refracto- meter, and ob.tainj_ng more· precise salihi ties iri• the laboratory .• Samples were also collected for analysis of dissolved silica and natural fluorescence.

The North Carolina: study area was sampled during·March, May arid November of 1979, and during March and May of 1980. River water samples from up to

Bo bu west ~f the ocea;n were· collected from bridges or docks, and seawater samples were collected from a boat 20 km east of Wrightsville Beach, North

Carolina. All samples were analyzed for salinity, dissolved silica and natural fluorescence. The intention of this sampling was to define the tracer concentrations in the rivers, and to evaluate temporal variability.

Analytical methods

Dissolved silica was measured using the method of Strickland and Parsons

(1912) as modified by Fanning and Pilson ( 1973b). Samples were frozen soon after collection and were stored frozen until analyzed.

Natural fluorescence was measured using the method of Kalle (1963). · A standard_ solution of 0.1 mg/l qui~ine bisulfate_ in O.OlN H S04 (0.236 µM 2 quinine) was set to read 73 mF'l relative to the O .OlN' H blank solution. Since 2so14 samples were JJH:!al:luredrelal.;ive· Lo this standEi.rd setting, i'luoresc~nce· rea~ings are not in absolute units (Duursma and H6mrnets; 1961)~ The excitation wavelength of

365 nm, and emission wavelength of t1&q ·run (Kalle, 1963), corresponded with the maxima for these river water samples. A Turner 430 spectrofluoro~eter was used

... 6 . t

to measure.fluorescence; this is a monochromator and not a filter instrwnent

so. sensitivity is reduced ·but· specificity is improved., On this instrument

calibration ·req_uired a scale expansion of X 10. To prevent intern.al quenching

· ( self absorption by solutions which contain high concentrations of fl_uorescing

m~(erial) samples whi~h read greater·than 100 mFl were diluted prior- to

analysis (Kalle, 1963, Dutirsma and Ro~ets, 1961). Another re~son for diluting

is that fluorescence is line_ar with respect. to concentration only when a .small

fraction.of the in9ident light is absorbed (St. John, 1913). The standard . . deviation of. the quinine solution (73 mFl) based on 16 ·determinations was 1.8, ·

corresponding to a relati.;,,.e error of. 2 .4%. Two' No:tth Carol.inasamples ( 33 j E!,nd

71.3 mFl) analyzed 9 and 21 times over a one week period had relative er:i;'ors

of·4.5 and 2.6% respectively. All samples in the Georgia-study were analyzed

within two.days of collection; North Carolina samples were analyzed within 24

hour·s of collection.

Electrode standardization of pH measurements was made relative to five

buffer solutions with pH values between 4.00 and 9.18. pH adjustments were

made when necessary by adding microl.iter quantities of 2N NaOH or 2N HCl.

Field determinations -of pH on North Carolina river water samples were made

llR i, ng shnrt. rHne;A pH JY\[lPr ,

The concentration of iron was adjusted in some solutions (Figure 9) by

adding 20 microliter ~liq_uots of 500 ppm iron stock solution (Fec1 in dilute' 3 HCl) to a 5 ml srunple. The pH was adjusted after iron was added .. Iron

concentrations were calculated from the runount of iron added.

Total organic carbon was analyzed by the method of MacKinnon (1978).

Hwnic acid was extracted from a sediment sample from the Cape Fear River

Estuary using the method of Cheshire,· et. al. ( 1977) . The sampl·e was not ·7 desalted or purified.

Results

Dissolved silica

.... Dissolved silica in both study areas was too variable to be used as a t-racer (Table 1). Dissolved silica in the Savannah River mixed conse~vatively with seawater (r = - 0 .. 718 and P. = 0.001 for 30-·data pairs) but:that from the_

Ogeechee River did not Cr = 0.. 420 and p = 0 .05 fo:i;- 28 data pairs) (Figures 2 anc;l 3).· The positive correlation coefficient for the Ogeechee River indicates that dissolved silica is increasing with salinity. In the North Carolina study area, dissolved silica differed significantly between the coastal plain and pi!;:!dmont rivers, as in the Georgia area, ·however, during a short term variability study conducted for ten days in March of 1980 dissolved. silica in the Cape Fear

River varied by more than a factor of two between morning and afternoon samplings on-the same day. Since the Cape Fear River is the major river in the area, this variability limits the usefulness of dissolved silica as a tracer. Coastal plain and piedmont rivers had large differences in silica concentrations, so use as a tracer during the winter months when biological activity is low may be· possible.

Natural fluorescence

Natural fluorescence in both the Georgia r'ivers mixed conservatively with seawater (Figures 4 and 5) and the values for the two rivers differed by slightly more than the necessary factor of two making it a potential tracer. The corre lation coefficient between salinity and natural fluorescence was - 0.983 for·30 data pairs (P > .001) in the Savannah River and - 0.996 for 28 data pairs·

(P-> .001) in the Ogeechee River. Conservative behavior suggests that ~ime . .

8 variability was not a problem during the two day sampling time. According to

Loder and Reichard (1980) any variation. in the river end member will giye

. . : apparent·non~conservative mixing behqVivr. In the North Caroli~~ study area, the difference betweer,i the.natural fluorescence of coastal plain and piedmont ri~ers was a_lso ·great enough to be useful as a tracer (Table l), and qhort .• term variability in each river was small. Seawater from both study areas also : had very low fluorescen_ce values relative to the river waters. ''Natural . . fluorescence should therefore be a useful tracer to distinguish behreen coastal

,plain and piedmont.river waters.

Bea.ause natural fluorescence can be used as a tracer in these study· areas and perhaps also tn o"ther areas, other investigations were conducte·d to evalu_ate the method for tracer applications. '11he objective was to identify potential problems with the method before undertaking a more extensive use of fluorescence as-. a tracer. The following possibilities were considered:

1. •remperature a.ependence: The natural fluorescence of two samples and of

the q_uinine standard (Figures 6 and 7) depended on the temperature at

which the analyses were made. Fortunately, the maximum sensitivity

in both cases occurred between 21 and 25°c. St. John (1973) attributes

this effect at temperatures above room temperature to i.Pcreasing ·

efficiency of nonradiative deactivation (collisional q_uenching and

vibrational deactivation) with increasing temperature. A competing

process of delayed fluorescence becomes more important with decreasing

tanperature which may explain non-linear behavior. This temperature

effect means that samples and standards should be near room temperature

in order to achieve maximum-sensitivity and precision,.and all standards

and samples should be at the -same temperature to give comparable results. . .

2. pH dependeri.2e: Black and Christman (1963) found higher fluorescence

in a sample of Suwannee River (Georgia) water at pH 11.0 than at

-pH 5,,5, which may reflect cYianges in the protonation of nuiiiic ·acids.

Dependence on pH was observed for a Black River (North Carolina) .... sample, however,. insignificant.effects·were observed in a humic ~cid.

solution which contained approximately· 80 mg/£ humic ·a~ids (Figure ·8) .

In an_ earlier investigation, an excitation wavelength 340 rather . . . . of. . than_ 365 run was used and a pH effect similar to that observed for the

river water sc!:ITlplewas observed for the same hwnic acid sample. This

·suggests _that a pH effect may be observed with a·filter fluorimeter

whi_ch allows a broader range of energy to irradiate the sample. . 340 nm

is close to the excitation wavelength maximum for thi_s humic acid

sample, whereas for the river water samples the excitation wavelength

maximum is 365 nm. Although pH effects could be significant ih some

areas, in the 62 North Carolina samples pH did not vary significantly

from 6.6. pH was not monitored in the Georgia samples, _however, the

expected range of between 6 and 8 should cause a negligible ~ffect

relative to the chang_es _caused by mixi_ng with seawater. 'l

3. Sample ·storage Effe~t_: 'l'he natural .fluorescence of fou:i; -s:amples increased

betw.een 5 and 25%with storage for two weeks (frozen or refrigerated and I' "in glass or plastic bottles). Deionized water stored in similar con~

ditions showed no fluorescence increase, so the increase is probably

not a container effect. Samples had a constant. fluorescence vaiue for

, at least a month after this in:ttial increase. Martin and Pie~ce (1971)

also observed an increase in the concentration of humic acids after

samples were stored frozen for two months, however, the 'increase was

! I l t ..,

close to the precision .of the analysis.

4. Sensitivity: _North Carolina river wat.er was diluted with.deionized . ... water, artificial seawater arid natural seawater to determine the ... lower detection limit. One part river water could. be detect.ed in 200 parts of natural seawater or in 500 parts of. artificial seawater

or deionized water. Sc~le expansion b.f 300x was requiied for ·this

reading to be obtained.. The detecti9n limit for a tracer application. . . is probably not instrumental, but rather would be set.by the background

Raman scattering of water (·St. John, 1973) and tl).e background n~tural

fluore::icence of seawater (Duursma, 1974).

5. Filtration Effect: F'lltration of samples through a 0.45 micrometer

pore size Millipore filter resulted in ah increase of between 1 and 7%

fluorescence for sj.x North Carolina. samples. Siul:e no major changes

were observed, filtration is not necessary·.

6. Correlation with Total Organic Carbon: No correlation was observed

between natural fluorescence (100, 78, 36, and 56 mFl) andtotal

organic car·uon (6.5~ 13.5, 5.0, and 6.0 mgC/!l total organic carbon,

<.J respectively) for four Cape Fear River· samples. This suggests that .. ·thi:: fluorescing material was a major part of the organic material

present.

7. Possible Interferences:

a. Chlorophyll-a: J\. solution of 1.0 mg/!l pure chlorophyll-a

from Sigm~ Chemical, and characterized by the spectrophoto

metric method of Strickland and Parsons (1972) was prepared

in 90% water - 10% acetone. 'rhis ·solution gave no .. . .

fluores<'ence.reading at the·wavelengths appropriate for measure

ment of nat~ra1· fluorescence l365 and lr60. ~ ·compared with the

436 nm and greater than ..-6,6O-,nm maxima for chlorophyll-a)_~ -Acidi

fi_ed chlorophyll-a (pheophytin) also gave no signal at the natural .... tluorescence.wavelength settings. The chlorophyll-a solution,

when added to a river wate~ sample~ caused no fluorescence decrease . ot~er than by simple d~lution of t~e river water, S(? absorption

b;v c~lorophyll-a is _also neglig-i ble. Duursma. ( 1965) reported t·hat

natural fluorescence did not car.relate with chlorophyll, organic

carbon, or organic. nitrogen in :t;he l\)'orth Sea, which also suggests.

~o an~lytical interferences from chior.ophyll or total qrganic

carbon.

b. Lignin sulf.onates: Lignin .suifonate in water can be measured

analytically by fluorescence (Christman and Minear, 1967), From

the spectra provided,-ttese compounds should not fluoresce

significantly at the wavelengths used in the ·natural fluorescence

-measurements. Kraft mill e~fluent; unlike effluent f~om sulfite

_processing, is- not significantly fluorescent. {St. John, .1973) .. • • • ' • • • <) l'feither kind of· effluent should therefore af.fect anaiy.sis.. of ·natural fluorescence.

· c. Petroleum.: A mixture of approximately 80 mg/9., of 30 weight

lubricati'ng oil in deionized water had a fluorescence of between· · ·

150 a11d 300 mFl relative to the 73 mFl for the quinfne" standard.··

Simila1· high ·values \.Tere obtained for a mixed gasoline :oil ( 50 :1)

sample and. a suspension. of mixep. gasoline and oil in wat·er. In . eacp. some bil was clen.rly. visible as a surface film -or as a

.·\,\·~.'. . · ,.• :·f,' ...,,,:?f' ....

-.. ,,--. ~·· j . 12

sheen. 011 the glassware. 'l'hese re·sults indica;te that natural

fluorescence as a tracer is limited to are~s which do not h~ve a

high input of oi~. •....

d •. Iron: Iron in mg/Q,_·amounts can cau.se low results in the analysis .... of humic ac:j,ds by spectrophotometric or- fluorimetric methqds .. (Brun and Milburn, .1977}, ·'11his prestlJ!lably results fr6m complex

ation of iron by humic.compounds~ This was s~bstari~iated in this study for ·two Noi·th . Carolina river water SaJl'\ples · and f.or. a.humic

_acid sarnp],e (Figures 9 and 10). ~fowever, varia~ion in iron con~

centrati9n.should.not cause a problem in these North Carolina rivers

with respect to using natural fluorescence as a tracer .becau~e the.

niaximwn iron concentration observed during a 1,5 year monthly

sampling program was 0.94 mg/i lWilder and Slack, 1971). Average

values reported in the Wilder and Slack (1971) study for the Cape

Fear, Black and Northeast Cape Fear Rivers are 0.06, 0.21, and 0.25

mg/i,, low enough to prevent· -significant variations in fluorescence.

Brun and Milburn (1977), in analyzing for humic acids, eliminate

j iron interference by adding a tartrate-citrate complexing agent

to complex the iron before analysis. • The advantages of using natural fluorescence as a tracer, even with the above complications, a.re that it is a rapid and' inexpensive analysis which has good sensitivity and high specificity. Most compounds are not f.luorescent.

The probability oi' having more than one compound which both have the·same excitation and emission spectra is low compared with a spectrophotometric method in which only one wavelength is·used. Fluorometry has the additional a.dvantage of ncale expansion OVl:!r i,cveral. orders of 1nagnitude compared with a 13·

usually much ·smaller range for spectrophotometry. ·-·

Conclusions . . ..

Natural fluorescence and salinity can distinguish between-coastal plain

ancr-·•piedmont rivers as tpey mix with seawater in southeastern North Carolin~.

and~in northeastern Georgia. Using the.matrix ~ethod of Zimmerman and Rommets . (1974), the relative contribution from each source· can qe calculated. Natural

fluorescence may also be applicable as a tracer in other areas if the .limi- . ·

tations· of the method are appreciated .. This method of tracing individual river

waters m~y prove:to be most useful when used in ~onjunction with other tracers.

For example, deuterium or oxygen-18 could be used for large scale information

"(Torgerson, 1979), with natural fluorescence providing better resolution on

a small scale.

Acknowledgements

The authors would like to thank D. Menzel for critical comment~.· This

research was· supported by ·nepar:tment . of Energy Contract EY-:-76-S-09_-::~ to _

L. P. Atkinson and_by a North Carolina .University· Marine Science Council.grant

0 to· J. D. Wiliey .. .. .

. . ' : J

References.

Bla_ck, A. P. and Christman,. R. F. 1963 Characteristics .cif colored surface

wat-ers· .. Journal of the American, Water Works Association 53: 0 753-770.

Brun, G. L. and Milburn, D. ~- D. 1977 Automated fluorometric determination of

.,. humic substances in natural water·$. Analyti~a:1 Letters .10(14): · 1?09-.l?J.,9.

Cheshire, M. V., Berrow, M. L., Goodman:, B. A.,. and Mundie, .C. M. 1977 Metal : distribution and nature of some C\.1., Mn and V complexes_in l'l'umic.and fulvic

- • < • acid-· fractions of soil matter. Geochimica et Cosmochimica Acta 41: 1134-1138.

Christman, R. F·. and Minear, R. A. 1967 Fluorometric Detection of Lignin

sul4:',::mates .. 'rhe Trend in Engineering· 19(1).: 3-7.

Duursma, E. K,. 19E>l Dissolved organic carbon, nitrogen and phosphorous in the

sea. Netherlands Journal of Sea Research l: l-il18.

Duursma, E. K. 1965 The dissolved organic const:i.t_uents of sea water. In

Chemical Oceanography, Volume 1 (Riley, J. P. and Skirrow, G. , eds·.)

Academic Press, New York, pp. 433-477.

Duursma, E. K. 1974 'l'he fluorescence of dissolved organic matter in the sea.

In Optical Aspects of Oceanog:r·aphy (Jerlov, N. G. and Steeman-Neilsen, E.,

eds.) Academic Press, London, pp. 237-256.

Duu~·sma, E. K. and Rommets, J. W. 1961 Interpretation mathematicute -de la

fluorescence des eaux douces, saumatres et marines. Netherlands Journal

of Sea Research l (3): 391-405.

Fahning, K. A. and Pilson, M. E. 1973a The lack of inorganic removal of dissolved

silica during river-ocean mixing. Geochimica et Cosmochimica Acta 37:

24(),5-2415.

Fanning, K. A. and Pilson, M. E .. 1973b On the spectrophotometric detennination

·- of dissolved silica in natural waters. Analytical Chemistry 45: 136-139. prn ... __.,_,,..,,, • • . . •- b""-•l'•-•· .. ••••?...... n

·. 15

_Gardner, D. 1977 Natri~nts as tracers of water mass structure in the coastal

upwelling off northwest Africa. In A Voyage of Discovery (M. Angel,- ed.)

Pergamon Press, Oxford, Englarfd /·pp. 305-326.

Hunt, D. J. E. _1978 Studies of the mixing of coastal waters in Liverpool Bay .... using dissolved silica as a tracer. Water Research 11(5): 465-470.

K~lle, K. 1949 Fluoreszenz und Gelb_stoff -im Bottnischen und Finnischen Meerbusen. : Deutsche.Hydrographische Zeitschrift 2 (4): 9-124.

K~lle, K. 1963 Uber das verhalten und die herkunft der un den Gewassern uncl in

der atmospha.re vorhandenen himmelblauen' Fluor....,eszenz. Deutsche Hydro

g~aphischc Zeitschrift 16(4): 153-166.

Loder,· T. C-~-and.Rei.chard, R. P. 1980 The dynamics of conservative· mixing-in

estuaries. Estuaries, in press.

MacKinnon, M. D. 1978 A dry oxidation-method for the analysis of the TOC in

seawater. Marine Chemistry 7: 17-37,

Martin, D. F. and Pierce, R.H. 1971 A convenient method of analysis of hwnic

acid in fresh wa~er .• Environmental Letters 1(1): 49-52.

Otto, L. 1967 Investigations on optical properties and water masses of the

southern North Sea·. Netherlands Journal of Sea Resear'cl1 3(.4): 532-551. ·

St. John, P. A'. 1973 Lumines.cence techniques in water analysis ... Iri. 1,,'Jater and

Water Pollution Handbook, Vol. 4 (L. L. Ciaccio, ed.) Marcel Dekker, Inc.,

New York, pp. 1~87-1751.

Strickland, J. D. H. and Parson, T. R. 1972 A Practical Handbook of Seawater

Analysis. Fisheries Research Board of ·canada Bulleti_n 167, 310 p.

Torger,on, T. 1979 Isotopic composition of river runoff on the U.· S. east coast:.

Evaluation of stable isotope versus salinity plots for co~stal water mass

identification. ,Journal of Geophysical Research 84(c7): 3773-3775. 16 . i

Van:··Bennekom,, A. J.,.Krijgsman,-Vari Hartingsveld, E., Van Der V~er,_G: C. ,M.,

and ~V.an Voorst, H. F. J. 1974 Th_f-:'~asonal cycl_es of rea.cti ye. ~tligate· an~

_suspended diatoms in the Dutch Wadden Sea. Netherlands Journal of Sea

.... Research 8(2-3): 174-207,

Wilder, H. B. and· Slack L. J. 19'71 Summary of data on chemical quality of 1 . streams in North Carolina. -U . S. Geological Survey Water-~upply Paper

1895-B, 236 p.

Zimmerman, J. T. Ii'. and Ronunets, J. W. 1974 Natural fluorescence as a tracer ..

in the Dutch Wadden Sea and the adjacent.North Sea. Netherlands Journal

of Sea Research 8{2~3): 117-125.

I I I. Table 1. Averag~ _v.a:J_nes(X) · for natural flu_or.escence (mFl) and for dissolved silica (.µM) based on n samples for the various water types listed~ The symbol rf indicates the standard deviation; The Georgia coastal water salinity was 30 0 100 ; the North Carolina seawater salinity was·35 o/o"o· .. Natural -Dissolved Fl\.toresc·ence, ·silica mFl µM

. n X d n X d Sav.annah River 2 55 0 2 49 18

Ogeechee River 2 114 1 2 12 2

Georgi~ Coastal Water .3. 2 4 ~ 7 5

Cape Fear River 19 37 4 19 42. 26

Black River 11 71 8 10 5 4

Northeast Cape Fear River 19 75 .11 15 9 9

North Carolina Seawater 2 1 1 6 l.'O 0.4 .:, ..

.. . ·.·17

Figu:-e captions

Figure 1. Study area, including the fall line( ______) which separates the piedmont· from the coastal plain regi:qtr .-. The locations of the r.i v-ers investi gated are shown.

Fig~re 2. Dissolved silica, µM, .plotted versus salinity, parts per thousand,. . . for water samples taken from the Savannah River Estuary. The line drawn is the : least squares regression line.

Figure 3; Dissolved silica, µM, plotted v~rsus salinity, parts.per thousand, fcir water samples· taken from the Ogeechee River Estuary. The line drawn is the least squares regression line.

Figure 4. Natural fluorescence, mFl, plotted versus salinity, parts per thousand, for water samples· collected from the Savannah River Estuary. The line drawn is the least squares regression line.

'figure 5. Natural fluorescence, mFl, plotted versus salinity, parts per thousand, for water samples collected from the Ogeechee River Estuary. The line drawn is the least squares regression line.

Figure 6. Natural fluorescence, mF'l, plotted versus temperature of analysis,

0 _C, for a 0.1 mg/9,, solution of quinine bisulfate in 0.01 N sulfuric acid., i 1 Figure 7. Natural fluorescence, mF'l, plotted versus temperature of analysis,'· . ., l °C, for two North Carolina river samples ( indicated by X and D ) .

Figure '8. Natural fluorescence, mJi'l, plotted versus solution pH for an 80 mg/'x, humic acid sample (6) and for one North Carolina river water sample (X).

Figure 9. Natural fluorescence, mFl, plotted versus concentration of iron, parts per million, for a humic acid sample (80 ppm humic acid). The solution pH was 8.2 ! 0.2, and the initial iron concentration w~s less.than 1 mg/'x,.

A' ·-cloudy red precipitate became apparent after 60 mg/ 9, of iron had been added to the solution. lT

1. Figure 10. Natural :i'luorescence, ~Fl, plott.ed _ve·rsus ·the concentratio·n of ·iron,

parts per million, in two.North Carolina river water. samples lindicated by X and . ~~ so~ution pH.-in each cas~ ~~~~ -~,9 ± ~ .3, and the initi~: ~-~nc~ntr~tion of ....iron was less than 1 ppm. In one sample {X), a cloudy red precipitate ' app~a.red after addition of 20 ppm Fe; in. the other sample ('Q) this precip~tate : was obse~yed after addition of .only tl ppm Fe.

..'

'• j . ·' I ...

, ,,'s.c. / :

,,.,.. Ga.,,.,, / · _... ,.. .. ✓

0 50 100 ~ z:.::::a Kilomelors

0 ...... I 00 .--:-----,------,------, 60.----.----:-,----.

~ ·. . :l . + ·cu •· ~- of• -~ '!O ◄· .. ... •· •- . "O •· . Q) •·

"Eio •· "' ,. ., .... Q·-~ .·· + •

0 '---..,....;..--'----'-"- a 10 20 JO Salinity ,ppt. Salinity, ppt

------:------==-(" •~:·------~.? 125 I I . ; f: . . I : :;: 100 I E . ' ' ' '

·,·•.· .. ·•

80 ---~r----.------, I \ (

LL- .. E 60 ,I ... ·-·--·-.i...____ _ ~O 0 ~o J IJ 20 .30 ...., Salinity ,ppt ~=· ~ . ~--... ,,., 0

__;_,.( ___,... __ 20 .30 Salinity. ·ppt . ,..

1l j 'I l

.•

.aor---;---.--.----.---..:... · 8 0 ------..----.-----,.-----, I ! LL i· ·E u:: I 75 : ~ Q) ., 0 0 ~ 70 i 0 ·. (i O 00. I 0 (I) c· 0 ., ! . i .U: 65. .a 0 u.. l i so--~-__. __ _,__ _.__ __. I 60'----'---'----'-----''----' 0 10 20 Jo ~o 50 0 10 20 30 i+O 50 Temperature,·c Temperature, °C l

.. 00

X u.. ·E X " (I) '70 X 0 C X . X - (I) 0 en (I) X '- X 0 60 ,. .. ~ .. u. "x" .. A .. 50 2 If 6 8 10 pH !O ...

80 T - .. LL 60 E .-l

Q) 6 (J Q) " (J C If 0 "' Q) 6,. C:. (J ... Q) (/) ",. (J Q) (/) " Q) ... A 0 20 ...... :::, A 0 4 " .2 LL I.!.. "J 'I . ' "' "' . ' 0 0 so 100 150 200 Iron Concentration, ppm F_e Iron concentration, ppm Fe

.. C' (\I ("') r-,. 01 ,, 0 0 C) ='. r:r <;; 0 w ..... ,__!- -:-·- . l/') :,;t- 00 a? 0 o .... __J -...,..... - ...... Cl.. 0 . . co Cl) >-O ),-•' ·100 u.;a:::: ·wN t- _; 0 ·-:....;o o.....;o .:::,__;' ◊ ..J H !""-- . ...JHW _;-:.::o ~ a.. 3: 0:: -~ 75 •. . ',. ◊ ~- -1> -"V) 50

~ ~- -:-.. '-"'!• •.\l'I ◊ c:S ');) o·'------....!.------'-----'------' 0 10 20 30

,_ I I ( . I I. f r f r r f I l l r ( ,HIRE ~?MP~TER INSTl:',UM \TION LIMITED,(. .. • :· .# • ,,, .. ---•·,-·,: -·:•1r·· . . . - ..,: ,+ .~ , ..

,-,. ('I') 0) 1./") (\I 0 .. ('I') ..... - 0 :r- ~ co co -.._, - ._.- 0 >-,.co 60 w C\f !-....J 0 0 _J' _J )-i r--- 0...'.3::0 ~ :::, + .. +. CtS qo T (.) + +

Cl) + + '"C Cl) > -0 20 (/) (/) .;. ·- + 0 +

0 0 10 20 30 Salinity, 1)fJt ¾o I.D O'l ,...... N ,.... O'l !I C]) co I.I) 1/) (") 0 0 0 C' t;_; 0 er ...... I- ..... l/) :::r :::r ... co 0 a:l 0 ...... '-' :l.... '-' 0 (f) a:l >-C >-' 80 l!J ct: w C\J ~_;o I- _J 0 o_;u O.....i' _j ~ L:....: _;Hr Cl..3'.il:: Q.. 3: 0

◊

Lt0

0'------'------_.______:.,...._ 0 10 ·20 30 .J(AC /',,,,; - ~ I tr'"' / .f:ft(,.J " 0 'l,

,...._ r"\ (I') (I') ..... ::::- :::r ::· C' C .. . (") - C -::r c::>- a:> - ._, -'-' - - >- 125 LL !- - Q_ .....J~ + CL 3: '-- I \ 100 ~ _, + ~. 75

so -

...__ l...i. 0 ,______j______.1_ ___ ~ 0 10 30 ,-, r-.. ~- .o 0 l/) .. r- :::- r-- 0 .. C .... 0 - 01- r-·C1l .....- r--...... '---' ~ 0 CX) 80 -wr->-' I-_; N !-- - o_;, □- ~~ .;.JH...O CL:;: Cl. 3:o l(' ~ ~' -'\ 75 00 "\ o' 0 00 0 0 \ ..i. OJ 0 ...... 0 0000 00· -\ 0 Q ''.l 70 0 ~ -~ 0 \__;

~ ~ --..,,. / 65 ./ ~- ~ l--- l - 60 0 10 20 30 ~o 50 0 .,!I ,bJ. -

.....N ,...... ('I? C'\J II LI) ...... a lO w ...... I- ..... I- co 0 ..... _J '-' a... 80 Cf) >-a. w 0:: ~ !-_JO O_JU _JHW -~ a...3: a:: X X .,.j X X X ...... ~ X X 70 xx ...... X '\. CD ...... '< X □ □ X ~ 0- 0 X ~, □□ □ X \J □ V) 60 □ ~ 0

~"' □ :5 □

( 1 ---..=::...:....- 50 0 10 20 30 · 50 lf) O') ,..... C\J (\J I.{) !I :::r· 0- 0 O') • c::i LL..i C' !- C'? !- c::i c::i 0 ,_, ...... , ...J - CL C' co Cl) .'>- '-. 80 >-o :.,J ,_, w a::: ...JO '-...JO _; '-. 6 _; u --1 '.'- ...JHW "3: 0 .__ X CL~cr.

~ ·,.L._'· ,,, X X ~ 70 X

X --.. X

""°'-.J ~ X ~ '-' X V, 60 t:,. <',._, 6 t:,. I:>.· ~:s' xx LI t:,. .u_-- .50 2 lj- 6 8 10 "

\C co N ,-,. ,-,. l/) l/) Ii l/") 01 en :::· C) 0 .. 0 ,.._ ct- r- LL ..... ~ ..... :::t· co i- .. r- 0 -~ r- ..... _...... _, a.... '-' Cf) >-a 80 w a:::: ~....JO 0--iU .....JHLJ..; CL3C::::

'-'

1..,,\.;'

~) "-. 20 ~ .0. :,., 3 6. \j_-- A .o. O:L-----'-....--__J,_---1---...... L..-~------O (0 0

r- cr: N r--. CTl !I r-- 0 0 tr.; ~ .... I-- CY) !- co 0 ..... _j '-' CL

(/) 8Q,----,---~----r---~---, >-o •..J..Ja::: 1-- _; 0 Q_jU _JI-! w 0.... ~ a:::

X --- X

X '7 X

•../1 V V X X 7 '7 X X X X X X

0 ------'------'-----'----'------,----' r 0