a volumetric temperature/salinity census for the middle atlantic bight

A Volumetric Temperature/Salinity Census for the Middle Atlantic BightAuthor(s): W. R. Wright and C. E. ParkerSource: Limnology and Oceanography, Vol. 21, No. 4 (Jul., 1976), pp. 563-571Published by: American Society of Limnology and OceanographyStable URL: http://www.jstor.org/stable/2834845 .

Accessed: 15/06/2014 08:34

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

.JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact [email protected].

.

American Society of Limnology and Oceanography is collaborating with JSTOR to digitize, preserve andextend access to Limnology and Oceanography.

http://www.jstor.org

This content downloaded from 185.44.77.28 on Sun, 15 Jun 2014 08:34:23 AMAll use subject to JSTOR Terms and Conditions

http://www.jstor.org/action/showPublisher?publisherCode=limnoc

http://www.jstor.org/stable/2834845?origin=JSTOR-pdf

http://www.jstor.org/page/info/about/policies/terms.jsp


A volumetric temperature/salinity census for the Middle Atlantic Bight'

W. R. Wright2 and C. E. Parker Woods Hole Oceanographic Institution, Woods Hole, Massachusetts 02543

Abstract Two seasonal volumetric temperature/salinity diagrams have been prepared for the

waters of the Middle Atlantic Bight from Nantucket Shoals to Cape Hatteras, to a depth of 200 m and extending as much as 130 km beyond the edge of the continental shelf. Total volume included is 23,145.6 km3, of which about half is slope water, more saline than 35%6. Most of it is in a distinctive subsurface maximum region near 130C, which is named the upper slope water thermostad. The less saline shelf water has two modes divided by a minimum near 33.6%,. The fresher mode, associated with shallow depths, is identified as coastal water; that from 33.6-35%, is called shelf edge water, and much of it is found seaward of the shelf break. There is very little seasonal change in the total volume of shelf water but its geographical distribution varies, showing the effects of spring runoff and suggesting a summer influx of slope water in the northern portion of the bight. Comparison with a similar census for the Gulf of Maine and shelf waters to the east shows some overlap but little evidence of substantial exchange.

Since its introduction in the papers of Montgomery (1958) and others, the volumetric temperature/salinity census has proved useful in describing both open ocean and coastal waters. Volumetric diagrams help to identify important water masses, their limits, and their relative abundances, to show seasonal variation, and, in some cases, to evaluate the effects of environmental influences such as advection, evaporation and precipitation, or coastal runoff.

The diagrams presented here (Fig. 1A,B) are for the winter and summer in the Mid- dle Atlantic Bight, from 35030'N-420N. The bight is one of the principal oceanographic regimes of the east coast of North America and one which is important from the point of view of shipping, fishing, de- velopment of resources, and pollution pre- vention. The purpose of the diagrams is to show the distribution of shelf water within the bight in the two seasons, with some emphasis on geographical differences within the region.

The diagrams also complement those of Barinov and Bryantsev (1972), who pre-

1 Contribution 3434 of the Woods Hole Oceano- graphic Institution. This work was supported by National Science Foundation grant GA 36499.

2 Present address: NMFS, Northeast Fish. Cen- ter, Woods Hole, Massachusetts 02543.

sented winter and summer diagrams for the waters of the Scotian Shelf, Georges Bank, and the Gulf of Maine. There is some overlap between the two studies as their coverage extends south to 40?N and west to 71?W, while ours (Fig. 2A,B) extends north to 42?N and east to 690W. The duplication amounts to about 2,000 km3.

One difference between this work and that of Barinov and Bryantsev is that they included only water inside the 200-m curve, whereas we have extended our region as far as 130 km seaward of the edge of the shelf, although we do not include any water more than 200 m deep. Our limits were chosen because it has been long known (Bigelow 1933) that in most (probably all) of the Middle Atlantic Bight, during most (probably all) seasons of the year, the main body of shelf water extends seaward well beyond the 200-m curve, and we wished to include that extension in our accounting. The lower limit of 200 m was chosen because shelf water is not found below that depth (Pollak 1947). The inshore boundaries were chosen to exclude Delaware and Chesapeake Bays, Long Island Sound, and other estuarine regions of the bight.

The region of investigation thus includes an area of about 206,400 km2 whereas the area of the shelf shallower than 200 m is about 120,000 km2. We have not tried to

LIMNOLOGY AND OCEANOGRAPHY 563 JULY 1976, V. 21(4)



564 WVright and Parker

310%o 320 330 340 350 36.0%o 224 ;< ' ' r / W , , , < ' '168336 504

300 300 22-

300 300 21-

396 396

WINTER VOLUME 20 (ki~~~~~~~~~~nm3) 14 4 74 445 1200

18 , / MIDDLE ATLANTIC BIGHT 8 4 2 ,7L / l I SURFACE TO 200m 216 132 348 17 -r,

TOTAL VOLUME= 23,145.6 km3 45 144 600

240480 36 756

12 0 8401080 2040 14-

852 162 0 4080 960 7512

5601788 1176 0241526419716 40 43968 12 -J

- / / ~~~~~~~~~~~~~~~~120 16 8 4 8 44 4 100 8 88 8 3216 109 06 19 48540

/ / 19 ~~~~~~~~~~~~6 14 4 51 6588 588 1320 75613834614065 24528

10 F Z - I11, 1 10 W /1 4 144 48 48 48 48 48 48 24 48 12 744 87620705562420012002952 912 19032

25 2 54 36 54 3 36 36 108 252 1392 240 324 912 378 364287 768 168 177 43 19926

8[~~~~~~~~~3 12 20148B 7 2 4081434 1172 8 16270123 6 12 0 X@io /(4(R100'13206 7-

6 i / / | 60~~~~~~~ 016 8 121 2 49 2 282 0 134 462 4 (q < 1536 /10176 6 480 1 / 48 [264 31 2 19443864 1728 1728 864 17520

38 4 288 1272 288 3201680 688 216 11520 4

240 768 144 1272 961320 3244984 8568

3 Lf 240Ctf, , ' ' ^ 1t, , , , , A | 240 20C

Fig. 1. Volumetric temperature/salinity diagrams for the Middle Atlantic Bight. A-Winter; B- summer.

account for the occasional detached parcels of shelf water which are found in the slope water, sometimes as far seaward as the inshore edge of the Gulf Stream. Wright (1976) has estimated that such parcels may contain 20 km3 of shelf water and perhaps four or five of them may be present in the slope water west of Nan- tucket at any time.

Method The area was divided into units of 30

minutes of latitude and 30 of longitude (quarter-degree squares) or fractions thereof. Data from Woods Hole Oceano- graphic Institution and National Oceano- graphic Data Center were searched for us-

able hydrographic stations in each unit. There are 87 winter stations and 93 summer stations. Positions are shown in Fig. 2A,B.

It was a surprise to learn that even in this extensively studied region there were areas where no suitable station could be found with close bottle spacing all the way to the bottom (or 200 m) at the appropriate time of year; as a result it was occasionally necessary to assign more than one unit area to a station or to use a station made earlier or later than the optimum time of year. For the winter diagram, February and March were considered best, but some stations were used from January (and even December), when winter cooling is not



Middle Atlantic Bight T/S census 565

300%o, 310 320 33.0 340 350 36.0%o

28 31Oat / / /_ 2 .84 696 120 7 2 4/8 7 2 48 312 672 27 26 48 48 24 12 24 4 12 12 24 24 24 336 96 684 26

12 24 24 2 12 84 96 12 216 360 9672 3 00216 54 24 24 24 336 48 / 180264 2826

24 0 24 12 60 276 84 288 480 132 24 168 ,36 132 48 180 384 162 156 36 12 552 528 1068504 48 48 240f 5682 264 348 122765 24 96 96 84 96 114204 48 8540 96336 180 114 144 5165884 144 72 48216 48 5568

23- 48 24 48 60198 78 276 60 96 72 72 90252 240120 390 432648 156312 498 360 240 96 348 216 5430 22- - -

48 2264 402 174 8 436 014 41328 221 964 357 244 4432479 2222 21 4 8 7 2 10 192 69 21 48 36 24 78018198420204300 22 892 258 66 162 06 276 12840 780 114 216 168 12 48 60 204 144 6612 20

3 6 240 348 120 264 72432264 72 426 84432 60 30108 336 540 348 288 288468 96 96 60 324 72 5604 19- - - 18 p A 9 > 4 12 97 168 24192 14 432366252 276 432 72324 240 288 96 1204552 24 48 5580

78 60 528 468 210 132 240 72 96 6013252 330 384 168 33 192 192 246 276 3 20 768 241 6390 17 48 192 240 192 240 228 192 102 252 288 60 96432 108 312 276 168 156 354 120 306 624 864144 540 iO56 7590

1 96 144 288 84 156 9 6 16 8 132 180 30 96 16 8 19 8 3 8 38 4 186 462 70 240 405 2 1236 96 1548 300 8226 15 14 1480 21 6 216 282 348 168 22 8 66 12 0 204 204 i5 6 51 6 22 8 348 15 6 120 360 924 864 72 13716

14 VIA 31 85 7 132 252 180 96 216' 74384 156 522 366 120 288 984 624776 573 240 17304

12 / /1 2 72 366 516 618 02 756 552 630 396 792 584 652,5244167763108 36240

24 60 5 8 288 228 336 2 456 894 6 366138

SUMMER 24 144102 276 2420 120 324 132 684 480 666 6601032402368 6

2497922

-| SUMMER VOLUME 192 798 552 11104 96 96 1092 618 648 750 92411525106865, 24 24 l (km3) l 156 12 504 36112 8 288 36 0 636 516 660 732 504 21 6 1164 8310 MIDDLE ATLANTIC BIGHT 57 2 410021104 056240 168 480 14 72 48 5130

7-L SURFACE TO 200m1/ I / / 7 SURFACE TO 200m 744 144 216 96 84 120 140.4 6-TOTAL VOLUME= 23,145.6 km3 7

50Cl r a / s a 8 s | / s 1240 240 204 72 75.6 9~~~~~~~~( (O (|/ 0 $@ (b) tK tK 4) (0@ te / f (O0 8 ?(1< n)n .0 (> ( 0 ? K g

complete, and some from April when spring warming begins to show at the sea surface. Similarly, August and September were used when possible for the summer diagram, but July and October were also al- lowed. Furthermore, although an effort was made to use observations from as many different years as possible, it can be seen in Table 1 that in some areas it was necessary to rely heavily on observations from a single year (sometimes a single cruise), so that the distribution in time is not as even as hoped. An example of the bias resulting from such difficulties is dis- cussed later.

From bottom contour charts, a mean depth was picked for each area shallower than 200 m, then curves of temperature and salinity versus depth were drawn for each station. The depths of the isotherms (every 1PC) and isohalines (every 0.2%yO) were picked off and the thickness of each inter-

val was recorded. To convert to volume the depth in meters was multiplied by a unit area. Unit area actually varies with latitude, from 2,524 km2 per quarter-degree square at 350-36?N to 2,317 km2 at 41?0- 420N. We used 2,400 km2 throughout; our final area differed from the actual area by less than 0.1%. The tabulation was done by intervals of one degree of latitude, to reveal geographical differences. Finally the totals for all such intervals were added and the temperature/salinity classes containing the greatest volumes were summed to give the 50% (hatched) and 75% (heavy line) boundaries indicated on the diagrams. The volumes are given to the nearest 0.2 km3 to balance the bookkeeping, but even the first digit to the left of the decimal cannot be considered significant as 10 km3 represents a layer only 4.16 m thick in a quarter-degree square.

Total volume by 10 temperature inter-



566 Wright and Parker

42' A ~ -A<

'f87 '8.854 A8

> :'7 9 78 .~~~7 76 64 73v

.84 41

.78 *72 71

X 15 9 If !? ~~.9/ 376

8876 4 83- 8 .88 .49 .61 403'

48 47'44 43 .41 58 5 535 2 50448 * .-'f,* *4/ A42

-ftw~~~~~~~~~~3 31 .30XW!T

3390 . 3

8 3 *2 9

.29 7 .2

~~ *2~~~~~ ~38'

*12

817 8 353

5 ' ' .'. "~-> 8'9 2 59e 41?378

.3,. ,,447z .46 A5,4*'2' 9

,;7 / ' 2 fVOLUMETRIC STUDY AREA

3~ ~ ~ ~~ 6 WINTER

STATION POSITIONS

J358 76? 75' 74' 730 72? 710 70? 69*

420

- 93.

89 '8f .88 .8~ 410

85 *84 ~~.82 '78 .74 9 49

.75 ~ 72 '5755 40

.8 5 *3 .62 59 ~ 58 56 4

0 ~~~ ~~~43 41 47 ' 444 .46 .44 '42'

40 36,/ *33

'39 38 .35 '34

~~ P~'2l ~ '27

Fi.24 2ideAtatcB98.Hav ie

225

3 .i o 4

-37S

5 VOLUIMETRIC STUDY AREA v 4 2 2 a

SUMMER 2~~~3 ~~STATION POSITIONS

J35' 760 750 740 73' 72' 71' 70' 690

Fig. 2. Middle Atlantic Bight. Heavy lines enclose region of study. Station positions are numbered and are listed by year in Table 1. A- Winter; B-summer.

vals is printed along the right si'de of each diagram, and the totals by 0.2%oo of salinity along the bottom. Density is indicated by sigma-t (o-,) lines.

Table 1. Distribution of winter and summer stations by year (see Fig. 2).

Winter Station Nos. Summer Station Nos. by Year by Year

1928 29 1932 11,22,38,39 1929 25,57 1939 92 1932 4,11,15 1940 80 1933 17 1946 91 1934 14,16,18,81 1947 89 1940 63,70,80 1948 85 1941 44 1949 64 1946 84,86 1953 2,5,6 1947 85,87 1956 60,63,65,74,82 1952 36 1957 46,83 1954 1,3,8 1958 56,69,88 1956 49 1959 55,86 1957 77 1960 59 1958 68,69,71,78 1961 21,35,36,41, 1959 5,6,7,10,48, 79,87

75 1963 62,81 1960 33,52 1964 54,57,67,75 1961 73,74 1965 8,9,18,20,26, 1962 64 28,32,90 1964 54 1966 3,4,58,84 1965 2,51 1967 14,15,19,27, 1966 40 29,30,34,50, 1967 9,19,59 51,73,77,78 1968 35,46,62,66, 1968 33,43,47,93

67,83 1969 12,16,17,23,25, 1969 28,30,45,47, 31,37,40,45,49,

61,72 52,53,61,66,70, 1970 13,42 72,76 1971 12;20,21,22,23, 1970 44

24,26,27,31,32, 1972 42,68 34,37,38,39,41, 1974 1,7,10,13,24,48, 50,53,55,56,58, 71 60,65,76,79,82

1972 43

The total volume of the region is 23,145.6 km3 including 7,785.6 km3 of water inshore of the 200-m curve and 15,360 km3 seaward of that line. By way of contrast, the volume in the region studied by Barinov and Bry- antsev (1972) is about 37,000 kM3n8 and that of the entire North Atlantic Ocean is 137,222,000 km3 (Wright and Worthington 1970).

Description of diagrams Winter-In the winter diagram (Fig.

1A) the 75% outline incorporates two principal water masses. The mode running from 30, 32.6%o, 26.0 a- to about 110, 35.0%oo, 26.8 o-, has been variously identified as shelf water or coastal water; it is the dominant water mass on the continental shelf northeast of Cape Hatteras. The 75% boundary includes 7,274.4 km3 of such water, or 31% of the total volume in the region and 62% of all the water fresher than 35%o. The large spread in temperature at a given salinity is a function both of time and geog-




graphy; a similar diagram for the region from New York to Nantucket Shoals in 1974 has the same axis of concentration but a temperature spread of only 2 or 3 degrees (Flagg and Beardsley 1975).

The salinity subtotals along the bottom show a maximum at 32.6 to 32.8%oo, a minimum at 33.6 to 33.8%o and then another increase that continues past 35%o. A similar situation occurs in the summer (Fig. IB), when the minimum is at 33.2-33.4%o and there is actually more water between 32 and 33%yo than between 33 and 34%oo.

This minimum represents the transition between the fresher water that occupies most of the interior of the shelf and that at the shelf edge which is strongly influenced by the slope water farther offshore. There is a strong depth correlation with the water fresher than 33.8%oo. It represents 88% of all the water in depths of 60 m or less. Five-sixths of it is found in water 100 m deep or less and 99.6% of it is shallower than 60 m regardless of the depth of water. Conversely only 7.5% of the water from 33.8-35.O%o is found where the water depth is 60 m or less and 87% is in water depths of 100 m or more.

To indicate the distinction between these two subdivisions of the water on the continental shelf we suggest that the water fresher than the minimum be known as coastal water and the more saline water, from the minimum to 35%oo, as shelf edge water. We would use the general term shelf water for all the water on the shelf fresher than 35%o.

Taking 33.6%oo as the dividing line gives a coastal water volume in winter of 4,944.6 kmi3, or 42% of the shelf water total. The summer figure is 5,146.8 km3, or 46% of the shelf water total.

The mode running from 140, 35.2%o, 26.6 o-, to 50, 35.0%o, 27.6 o"- is slope water. The 75% boundary includes 10,158 km3 of slope water, which is 44% of the total volume of the region. The slope water temperature/salinity curve lies almost parallel to the curve for western North Atlantic water (Wright and Worthington 1970) but

is somewhat fresher at all temperatures. Iselin (1936) put the difference at 0.02- 0.045oo for the slope water in the offing of Chesapeake Bay; McLellan (1957) showed a difference of about 0.1%oo for the region east of 650W. Their curves are compared by Colton (1968).

The juncture of these two water masses -the region from 10?, 35%oo to 130, 35.6%o -is where the greatest volumes of water are found. The seven hatched squares contain 8,812.8 km3-38% of the total. Water of these characteristics occupies a layer as much as 100 m thick, centered at about 100-120 m, and is widespread throughout the slope water. In the winter it forms a temperature and salinity maximum between the slightly colder and fresher slope water underneath and the much colder and fresher shelf water above. In the summer it appears as a region of small temperature gradient between the seasonal thermocline and the deeper slope water.

This water represents the extent of pene- tration of seasonal influence of the atmo- sphere. It is renewed annually when au- tumn gales and cooling break down the summer thermocline. It is the slope water analog of the 18-degree water in the Sar- gasso Sea (Worthington 1959) and deserves a name of its own. Miller (1950) refers to water of these temperature/salinity characteristics as "mid-depth slope water," and Chamberlin (unpublished manuscript) calls it "the warm band." As a more spe- cific name, we suggest "upper slope water thermostad," using terminology suggested by Seitz (1967).

The water warmer than 150C in the winter was found in only two stations, both located south of 36?N and east of 750W, in the Cape Hatteras region where the Gulf Stream is found close to shore. One of the stations, station 1, Fig. 2A, was entirely warmer than 150 and more saline than 36.0%Oo in the upper 200 m and thus is clearly Gulf Stream water. The other, station 2, Fig. 2A, is both colder and fresher, running from 190, 36.18%oo to 90 3 5Yoo, and




is representative of the southwest extremity of the slope water.

The values enclosed in ellipses near the bottom of Fig. IA are all associated with a single station-Albatross IV station 66-72, (station 40, Fig. 2A) on 12 March 1966. Those which are starred are the only classes where that station overlapped other observations. Clearly if station 66-72 had not been used, the temperature/salinity plot would have been tighter. It was included partly because it was the only suitable one from the quarter-degree square in which it was located and also because it is a special case. It is representative of observations made during the U.S. Fish and Wildlife Service surveys of 1964-1966 (Colton et al. 1968) which were the three coldest years recorded at the Nantucket Weather Station from 1931 to 1969 (NOAA 1969). Station 66-72 may thus represent simply a colder than usual year. On the other hand, the temperature/salinity char- acter of the station is like that of coastal water usually found much further east, and Colton (1968) has suggested that it represents an influx from that direction.

The two hatched classes at 12?-13?, 34.4%oo-34.8%o (Fig. IA) are also anoma- lous. They are based almost entirely on observations made in 1932, which was one of the warmest years in this region (NOAA 1969; Chase personal communication) and in December of that year, before the full impact of winter cooling had developed. East Coast weather conditions for winter 1971 approached the historical means for the area. Therefore we feel that the selec- tion of the large number of stations for this year (Table 1) did not significantly bias the results.

A geographical pattern in the distribution of shelf water is seen in the upper half of Fig. 3, which divides the region at the 39th parallel. There are 10,788 km3 in the region north of 39?N and 12,357.6 km3 south of that latitude. As might be expected, the warmer portions are southern and the colder portions northern, while the

30 31 32 33 34 35 36 %o2 I 2500

0 l_ONLY SOUTH OF 39~N~ *= ONLY NORTH OF 20

o xx J 3000

_ o 0 ooooooooo: xxx 251

W/TE :: * 0 0 0 *.. 2 0000 0 ~~~~~~~0 15

I0 0 X 0

= 00000 5 000

3 0 0000000 0 0 0 0 0 0 0 3 3O/ . 3. D ion 0 00 T ce by l 0 O

wnEr ando summe. Oecilsarfor obsra

tions made SOnl sOuth of =oNL sol crl r

reen obsevaton mad onl n ?orth of tha lati

~~x =xxxLTTUE

som but no al of the one-dere inevl of

30 31 32 33 34 35 36%o 300

laitud inlue Distiuin tergon./ lse b aiue

wanter ind thmer 50 e and75%moes isrenfrbervall

fiounsmade througout the regi9?N; son.d ci

There is a minimumbetween0 60 and 8

iesnth tempervatursae totals alngrho thet rigti eaitdges ofFi.ma.ke Thisss occlursbue dthero rsidenbt wnoterlo wther onte-shelfe southral of

latituNe isnrarelyucolderethan 80 wi tha of

wae nthe New Egand Shelfde is generally40t

60,oand theoug6out toh80water region. t nr

rhegio is fon principall iewen the limited are ofe sharpegradirentoas atlthe ede ofithe

reiB)coerst grnerater rangte ofhbot salinity

6,and thempraur th ? aten the wntertdiaram

reio is found 0000 prnial in -h limte are of sharp gradent atthegeofth shelf.0

Summer~~~~ Th sume diara (ig 1B)~~~~0 coer a grae rag of boh 2aint and temperature~~~~~ tha th witrdiga




and the water masses are not so clearly identifiable by temperature/salinity characteristics, except that the slope water mode remains strong, with a cold water extension at 35%O.

There is no water colder than 50C in the summer diagram, as opposed to a lower limit of 20 in winter. Even 50 water is rare in summer and in fact was observed only at one station, located at 38?40'N, 73014'W at a depth of 95 m on 7 July 1961. There is a diffuse mode which parallels the winter shelf water mode and is the slightly warmed remnant of winter water described by Bigelow (1933) and by Ketchum and Corwin (1964). If we consider "winter water'3 to be all that which is colder than 100 and fresher than 35%o, its volume is 2,532.6 km3, about a third of the volume found within the same limits in winter. Seasonal warming occurs to nearly the same temperature (28-290) at all salinities, but in the lower half of Fig. 3 it can be seen that the water warmer than 250 is found almost exclusively south of 390N.

The effects of spring runoff are reflected in the left side of the summer diagram (Fig. iB), which includes three times as much water between 31 and 32%o as does the winter diagram, and 90 km3 of water fresher than 31yoo which does not appear at all in the winter. (Water fresher than 31%o can, of course, be found in winter in coastal waters, such as Long Island Sound, and close to the mouths of estuaries, but such regions are excluded from this study.) The freshening occurs at all latitudes, but as Fig. 3 shows, there is a geographical distinction between the temperatures at which it takes place. The slightly saltier water found in the summer diagram (>36.4%oo between 18 and 240) is southern too, all from one September station at 36038'N, 74002'W, more than 50 km seaward of the 200-m curve.

Seasonal effects One reason for undertaking the census

was to ascertain what seasonal changes, if any, occur in the volume of shelf water in

the Middle Atlantic Bight, particularly as fluctuations in the position and shape of the shelf water/slope water boundary south of Cape Cod have been noted (Wright 1976). For that purpose 350oo has been taken as the dividing line between shelf water and slope water, except that cold slope water lying underneath the upper slope water thermostad may drop below that salinity and is excluded from the shelf water totals. The criterion seems reasonable because temperature is clearly not a reliable index in the upper 200 m and oceanic water in the western North Atlantic Ocean at those depths is all more saline than 35%o except in the Labrador Sea.

The winter volume of shelf water, then, is 11,472 km3 and the summer volume is 11,175 km3. The difference is 297 km3, less than 3%, so that for practical pur- poses the shelf water volume of the bight can be considered constant. This result was somewhat surprising because we had thought that the seasonal variation in runoff would create more shelf water in summer, and because the shelf water/slope water boundary extends farther seaward at the surface in the summer while remaining nearly stationary at the bottom (Wright 1976). To produce a seasonal water bud- get, accounting for runoff evaporation, precipitation, advection, detachment of shelf water parcels, entrainment by the Gulf Stream, and small-scale exchange across the shelf water/slope water boundary is beyond the scope of this report.

Nearly half of the shelf water was found in stations beyond the edge of the continental shelf in water deeper than 200 m. The volumes are 5,308.8 km3 in winter (46.5% of the shelf water total) and 5,154.0 km3 in summer (45.8%). This "bulge" is distributed pretty evenly north and south of 390N, which suggests a continuity in flow along the shelf water/slope water boundary. Other seasonal changes in shelf water volumes are shown in Table 2, which gives the values by salinity interval of loo, and in Fig. 4, which gives the difference between summer and winter volumes, by




1000 SUMMER GA/IN

800 _

600 -

400 , ' 33-34%o

/'32-33%. 200 - -; - 33 ><, -32%=

KM3 <- -30- 31%~ -200

-400 - 34-35%.

-600 -

-800 SUMMER LOSS

-1000_

48l42? 39-40? 38-39? 37-38? 35"/2-37?N LA TI TUDE

Fig. 4. Change from winter to summer, by latitude, of shelf water volume in each 1%o interval of salinity.

latitude, for the water fresher than 35%oo. Although the total volume of shelf water remains nearly constant, there is a summer increase in the fresher coastal water and a corresponding decrease in the more saline shelf edge water (Table 2). Note in Fig. 4 that south of 39?N there is a decrease in the amount of 34-35%o water and an increase for all other salinity values, presum- ably the effect of spring runoff as mentioned above. North of 39?N there is some indication of coastal freshening in salinities less than 32%o but otherwise the situation is just opposite of that in the south: the 34-35%oo values increase in the summer and those from 32-34%o are decreased. This is further evidence of increased summer mixing with slope water at the edge of the shelf.

It is not surprising that the seasonal effects of runoff are greater in the south: Bue (1970) gives a mean freshwater flow into the Middle Atlantic Bight of 157 km3 yr-1 but 65% of it is south of 390N, where the Delaware and Chesapeake Bays empty, and much of the 35% released in the north is undoubtedly carried to the south close to shore by the prevailing southwesterly drift.

Comparison with the Georges Bank- Scotian Shelf region

The mean temperature in the shelf water in the Middle Atlantic Bight increases from

Table 2. Seasonal changes in volume by salinity interval.

Salinity range Winter Summer Change (00?/ ) vol (km3) vol (km3) (km3)

30-31 - 90.0 +90.0 31-32 260.4 799.2 +538.8 32-33 2,433.6 2,863.2 +429.6 33-34 3,345.6 2,754.0 -591.6 34-35 5,696.4 4,668.6 -1,027.8 35-36 11,095.2 11,308.2 +213.0 >36 314.4 662.4 +348.0

7.710 in winter to 14.600C in summer. The mean salinity decreases from 33.78%oo in winter to 33.565oo in summer. These values can be compared with the means of Bari- nov and Bryantsev (1972) for the shelf water of Georges Bank, the Gulf of Maine, and the Scotian Shelf: a temperature increase from 2.710 to 6.680 and a salinity increase from 32.66%o in winter to 32.87%oo in summer. The slight salinity increase is ap- parent principally in the deeper, saltier waters of the region; Barinov and Bryant- sev suggest that it is due to Gulf Stream influence.

The bight is both warmer and saltier than the shelf water to the east, in both seasons, as expected. Nevertheless, the 50% outlines have some overlap in both winter and summer. In the winter the two modes lie along the same axis; in summer the overlap is only in the salinity range of 32-33%oo. We noted also that the anoma- lously cold station in the Middle Atlantic Bight winter mentioned above has the same T/S pattern as the mean of the Barinov and Bryantsev diagram, another indication of its possible eastern origin.

Conclusions Despite uncertainties arising from irreg-

ular space and time distribution of data and those inherent in the method, some conclusions can be drawn from this volumetric temperature/salinity tabulation for the Middle Atlantic Bight:

1. The volume of shelf water (defined as being fresher than 35%yo) is virtually un- changed, winter to summer.

2. Nearly half of the shelf water in the




bight is located in surface layers well beyond the edge of the continental shelf.

3. A prominent and permanent feature of the region is the upper slope water thermostad, analogous to the 180 water in the Sargasso Sea.

4. Seasonal effects in the New England portion of the bight differ from those farther south. North of 39?N there are indi- cations of more near-bottom intrusion of slope water in summer whereas south of that latitude the effects of runoff are more pronounced.

5. Although there are clear similarities between the waters of the bight and those on the shelf to the east of Nantucket Shoals, there is little overlap of the 50% modes for the two regions and little evidence that water from the east reaches the bight unmodified.

References

BARINOV, A. A., AND V. A. BRYANTSEV. 1972. A volumetric statistical T/S analysis of the Nova Scotia Shelf and Georges Bank water masses. Int. Comm. N. Atl. Fish. Res. Bull. 9: 21-25.

BIGELOW, H. B. 1933. Studies of the waters on the continental shelf, Cape Cod to Chesa- peake Bay. 1. The cycle of temperature. Pap. Phys. Oceanogr. Meteorol. 2(4): 135 p.

BuE, C. D. 1970. Streamflow from the United States into the Atlantic Ocean during 1931- 60. U.S. Geol. Surv. Water Supply Pap. 1899-1. 36 p.

COLTON, J. B., JR. 1968. Recent trends in subsurface temperatures in the Gulf of Maine and contiguous waters. J. Fish. Res. Bd. Can. 25: 2427-2437.

H R. R. MARAK, S. R. NICKERSON, AND R. R. STODDARD. 1968. Physical, chemical and biological observations on the continental

shelf, Nova Scotia to Long Island, 1964-1966. U.S. Fish Wildl. Serv. Data Rep. 23. 190 p.

FLAGG, C. N., AND R. C. BEARDSLEY. 1975. The 1974 MIT New England shelf dynamics ex- periment data report. Part 1. Hydrography. Mass. Inst. Technol. Data Rep. 75-1. 87 p.

ISELIN, C. O'D. 1936. A study of the circula- tion of the western North Atlantic. Pap. Phys. Oceanogr. Meteorol. 4(4): 101 p.

KETCHUM, B. H., AND N. CORWIN. 1964. The persistence of "winter" water on the continental shelf south of Long Island, New York. Limnol. Oceanogr. 9: 467-475.

McLELLAN, H. J. 1957. On the distinctness and origin of the slope water off the Scotian Shelf and its easterly flow south of the Grand Banks. J. Fish. Res. Bd. Can. 14: 213-239.

MILLER, A. R. 1950. A study of the mixing processes over the edge of the continental shelf. J. Mar. Res. 9: 145-159.

MONTGOMERY, R. B. 1958. Water characteristics of Atlantic Ocean and world ocean. Deep-Sea Res. 5: 134-148.

NATIONAL OCEANIC AND ATMOSPHERIC ADMINIS- TRATION. 1969. Local climatological data. Nantucket Mass. Annual summary with com- parative data.

POLLAK, M. J. 1947. Hydrography of the western Atlantic; hydrography of the subsurface slope water. Woods Hole Oceanogr. Inst. Rep. 5: 1-15.

SEIrz, R. C. 1967. Thermostad, the antonym of thermocline. J. Mar. Res. 25: 203.

WORTHINGTON, L. V. 1959. The 180 water in the Sargasso Sea. Deep-Sea Res. 5: 297- 305.

WRIGHT, W. R. 1976. The limits of shelf water south of Cape Cod, 1941 to 1972. J. Mar. Res. 34: 1-14.

, AND L. V. WORTHINGTON. 1970. The water masses of the North Atlantic Ocean: A volumetric census of temperature and salinity. Serial atlas of the marine environ- ment, folio 19. Am. Geogr. Soc.

Submitted: 29 September 1975 Accepted: 6 February 1976



a volumetric temperature/salinity census for the middle atlantic bight

Documents