11. interstitial water studies, leg 15 - a comparison of … · 2007. 5. 17. · 11. interstitial...

11. INTERSTITIAL WATER STUDIES, LEG 15 - A COMPARISON OF THE MAJOR ELEMENT ANDCARBONATE CHEMISTRY DATA FROM SITES 147, 148, and 149 l

Douglas E. Hammond, Lamont-Doherty Geological Observatory, Columbia University Palisades, New York

ABSTRACT

Examination of analytical results reported by various investi-gators for the major elements in the pore waters collected on Leg15 reveals that [Cl~], [SO 4

= ], [Na+], [K + ], and [Mg++]determinations agree well and may be reliable to 1%, 6%, 1.5%,5%, and 2.5%, respectively. [Ca+ +], determined by titration(SIO), appears reliable to ±5%, but atomic absorption analysestend to be systematically high, a problem which becomes morepronounced with depth. A model for apparent constant estima-tion (MACE) has been developed to facilitate comparison of thecarbonate parameters which were measured ( Σ alkalinity, ΣCC»2,pH, and PcOi^• Alkalinity calculated from ΣCC"2 and PCO?laboratory measurements agrees with shipboard titrations, butcalculation of ΣCC>2 from alkalinity and pH measurementssuggests that some CO2 loss has occurred between pore waterextrusion and ΣCC>2 analysis. Punch-in pH measurements areconsistent with gas pocket PCO7 within 0.2 pH units at Site 147,but are systematically lower than pH measurements on extrudedwater at Sites 147 and 148. This may be due to liquid junctionpotentials but is more likely attributable to CO2 loss duringextrusion.

MACE is used to calculate £2, the degree of saturation ofcalcite in pore waters, and to elucidate the nature and magnitudeof chemical changes in interstitial water as temperature andpressure are changed from in situ conditions. Clay minerals areshown to act as a source of strong acids as sediment temperatureis increased. Pore water can exchange material with calcite andmay attain equilibrium in the absence of organic matter. Becauseof this, squeezing temperature and pressure should be monitored.

INTRODUCTION

The aim of the DSDP pore water probram is to useinterstitial water composition to define diagenetic mech-anisms and rates. This requires a knowledge of in situconditions, and several complications must be considered.Contamination of interstitial water with seawater mayoccur during drilling. Gas may be lost from cores with highgas pressure. Temperature and pressure during pore waterextrusion differ from in situ conditions.

The intensive geochemical sampling done on Leg 15offered an opportunity to examine the effects of thesecomplications and a chance to compare several differenttypes of measurements and estimate the confidence whichmay be placed in each. Of particular interest is thecarbonate system, the one which is quite likely to beaffected by temperature and pressure perturbations from insitu conditions.

MAJOR ELEMENT COMPOSITION

Three investigators analyzed complete profiles of porewater: Sayles et al. (WHOI) determined major and minor

Lamont-Doherty Contribution No. 1834.

constituents of samples squeezed at 4°C (abbreviated CS)and 22°C (abbreviated WS), Presley et al. (TAMU) deter-mined major and minor constituents of CS samples; andGieskes (SIO) determined calcium, magnesium, and totalalkalinity of CS and WS samples. Table 1 contains acombination of their data selected to best represent thepore water composition of each sample. Procedures used inthis selection, systematic differences, and suggested confi-dence limits (followed in parentheses by the fraction ofsamples falling within these limits) are discussed in thefollowing paragraphs. All data for [Mg++] and [Ca + + ],which are particularly important in discussing the carbonatechemistry, are plotted in Figures 1 to 12 and are tabulatedby each investigator elsewhere in this volume.

Cl~TAMU determines chloride using a titration with

Hg(N03)2 and a visual end point. WHOI does a Mohrtitration using AgNθ3 and a potentiometric end point. Thetwo sets of data generally agree within ±1% (32/47),although TAMU values are systematically about \Vi%greater at Sites 147 and 148. Since the WHOI data show asmoother profile with depth, they are selected as superior.

831

D. E. HAMMOND

TABLE 1Pore Water Composition3

Sample

147B-1-2147B-1-3147B-1-4147B-1-4147B-14

147B-2-3/4147B-2-2147B-2-6146B-4-3/4147B-6-2147B-8-2/3147B-7-4147B-9-4147-10-3/4147B-11-3147C-2-1147C-4-4147C-7-4148-1-2148-1-4148-2-1148-2-3148-3-3148-1-3148-5-2148-64148-7-3148-8-3148-9-4148-10-3

148-124148-14-3148-16-3148-18-2148-20-3

148-23-4148-26-2149-2-2149-2-5149-3-5

149-4-3149-5-3149-6-4149-7-2149-8-4

149-9-5149-10-2149-11-4149-12-5149-14-3

149-16-4149-18-3149-20-4149-23-4149-26-2149-29-3149-31-1149-33-1149-35-4149-37-3149-40-1149-41-5149-42-2147B-1-2147B-1-3

Depth

2.54.24.74.85.1

8.515.021.028.051.0

63.063.083.082.0

100.0126.0148.0176.0

3.06.0

11.013.022.031.039.0

51.059.067.079.086.0

106.0122.0141.0159.0179.0

209.0232.0

4.08.0

17.0

23.032.043.049.062.0

72.078.090.0

100.0116.0

135.0153.0173.0201.0226.0

254.0271.0289.0313.0329.0354.0369.0374.0

2.54.2

Type

cscscscscscscscscscscscscscscscscscscscscscscscscscscscscscscscscscscscscscscscscscscscscscscscscscscscscscscscscscscscscscscswsws

Cl

20.0020.1419.9319.8519.88

19.9119.9720.0219.9119.4819.1919.0718.6418.9718.4818.2618.1517.8319.5819.4919.4719.5619.4819.5819.5719.5419.5919.5919.4919.48

19.4019.4119.2619.3819.23

19.4919.3919.3819.4519.50

19.5519.6219.6519.6319.61

19.6519.5119.2917.1419.45

19.5319.7419.8719.8519.8619.9120.0920.1820.0420.0919.5020.1120.1420.0019.83

K

8.588.697.857.417.16

7.588.956.907.538.187.928.758.327.928.698.468.746.909.97

10.019.469.479.468.458.18

7.867.676.806.906.71

5.885.985.525.375.375.625.889.469.659.75

8.958.868.328.467.92

7.517.096.985.725.845.546.766.185.945.785.605.815.885.405.234.734.884.769.979.46

CA

6.253.712.422.091.99

3.219.144.175.233.585.235.236.007.035.326.006.256.159.739.127.907.477.347.446.756.635.825.455.235.17

6.056.667.418.759.439.829.54

10.2610.5511.60

11.7612.9713.4313.0513.68

15.2516.1616.4017.3320.5522.2621.9726.1227.2727.8331.8331.7632.8629.8831.1030.5630.2231.395.863.56

Mg

51.4847.1249.0145.5843.60

47.8255.1743.4741.5930.0230.9030.2127.4029.6129.3429.4826.3325.1853.7853.6652.7952.0552.3351.6350.8250.6047.6647.2044.5845.24

38.4140.4936.6635.9534.6535.8636.7253.6353.1550.1652.3350.3949.5649.5250.06

48.9648.5248.4936.9845.1541.4841.4336.6735.3934.4433.6633.0032.2534.0233.7332.4533.8533.2446.8944.42

SO4

10.205.623.330.520.31

4.6823.630.310.721.350.930.410.101.351.662.292.080.00

27.0625.9224.3523.1121.2321.8616.3416.7612.8010.098.229.99

0.006.553.013.010.003.224.26

27.8027.6927.2725.7126.9625.4026.2326.1225.0825.9223.7321.2324.1522.1722.4823.1121.8621.7521.7522.5924.2522.3821.9621.1320.9222.2724.155.41

Alk.

14.5717.7217.6017.1315.17

17.718.18

13.7513.5010.0621.2524.6427.7628.0030.0032.8931.3623.74

2.834.003.514.435.235.026.266.477.617.667.716.72

5.274.823.673.733.633.052.362.292.382.592.442.442.092.192.06

1.791.771.711.291.771.812.203.033.212.912.923.303.002.542.462.432.272.44

14.5717.71

832

MAJOR ELEMENT AND CARBONATE CHEMISTRY DATA, SITES 147, 148, AND 149

TABLE 1 - Continued

Sample

147B-1-4147B-1-4147B-1-4147B-2-3/4147B-2-2147B-2-6147B-4-3/4147B-6-2147B-8-2/3147B-7-4147B-9-4147-10-3/4147B-11-3147C-2-1147C-4-4147C-7-4148-1-2148-1-4148-2-1148-2-3148-3-3148-4-3148-5-2148-6-4148-7-3148-8-3148-9-4148-10-3148-12-4148-14-3148-16-3148-18-2148-20-3148-23-4148-26-2149-2-2149-2-5149-3-5149-4-3149-5-3149-6-4149-7-2149-8-4149-9-5149-10-2149-11-4149-12-5149-14-3149-16-4149-18-3149-204149-234149-26-2149-29-3149-31-1149-33-1149-37-314940-1149-41-5149-42-2

Depth

4.74.85.18.5

15.0

21.028.0051.063.063.0

83.082.0

100.0126.0

148.0176.0

3.06.0

11.0

13.022.031.039.051.059.067.079.086.0

106.0

122.0141.0159.0179.0209.0

232.04.08.0

17.023.0

32.043.049.062.072.0

78.090.0

100.0116.0135.0

153.0173.0201.0226.0254.0

271.0289.0329.0354.0

369.0374.0

Type

WSwsWS

wswswswswswswswswswswswswswswswswswswswswswswswswswswswswswswswswswswswswswswswswswswswswswswswswswswswswswswswsws

Cl

19.9319.7920.020.00

20.00

20.090.00

19.660.00

18.95

18.720.00

18.5318.21

18.1417.7919.7319.4219.43

19.5319.6119.6119.5919.6419.5619.5719.5019.6019.50

19.4619.5319.4819.5019.45

19.4619.4619.4519.4219.58

19.6719.6319.6019.6619.61

19.6519.4917.3619.5419.96

19.7319.8419.8319.9519.99

20.1120.1020.1419.34

20.0720.10

K

8.699.208.690.009.97

8.690.009.970.00

10.74

9.710.009.849.20

9.468.18

11.7611.2512.27

11.2510.4810.749.20

10.23

9.208.958.988.958.10

7.417.166.646.907.41

7.1610.7410.5010.259.97

10.239.719.719.468.95

7.928.437.167.677.67

7.677.926.906.647.16

6.645.625.625.376.135.62

Ca

1.981.971.720.009.25

4.090.003.580.005.23

5.610.005.856.00

5.816.159.338.097.157.217.107.335.766.015.385.124.955.025.59

5.906.507.128.789.72

9.1010.0310.2011.6111.64

12.7013.7212.9913.6514.76

16.0915.9916.3319.7021.99

21.7225.3926.9727.9229.10

19.4930.6630.3729.91

30.3729.88

Mg

44.4242.6643.19

0.0054.53

39.980.00

27.560.00

26.73

25.580.00

25.9126.73

23.6822.2149.7050.5848.5348.6349.4350.1648.2346.4743.4741.6440.9240.2235.25

35.3632.9032.8532.1032.27

33.9050.8451.7551.2150.57

48.5847.3048.5347.8346.53

45.5445.0836.7341.2339.26

40.1235.2333.3132.9932.60

32.0833.9532.6432.13

35.0934.67

SO4

2.701.350.200.00

23.94

0.410.000.310.000.31

0.100.001.871.76

1.500.62

27.3725.9224.46

23.1122.0623.2116.6518.11

11.769.998.538.957.30

4.783.432.916.450.00

3.9528.4128.2028.0027.48

24.1526.6527.1727.5825.92

25.9225.9222.1725.7122.27

23.4224.1522.0623.1121.02

23.2123.4323.6322.48

23.7322.69

Alk.

17.6017.6915.170.007.90

14.220.00

10.060.00

24.64

29.140.00

34.2332.89

30.8023.743.394.163.334.765.374.766.266.057.568.087.796.905.62

5.183.673.993.723.032.962.442.542.762.72

2.572.332.372.302.06

1.941.791.291.932.32

2.293.003.452.972.75

3.292.422.482.392.222.30

1 Averaged values in mmols/kg except Cl, which is PPT.

833

D. E. HAMMOND

[ Mg++] mmol / kg

(UJ

)

I

DE

P1

5 0

100

150

200

" HOLE

-

~—à. o

_

_ × o

× Δ

1 1 1

301 ' '

147

X

X *

X

X Δ

X Δ O

1 I ,

401 ' 1

X

I , 1

X X

Δ

ΔX°

o

_l 1 _

50ftX x Δ 0 O

×o

×o-

-

-

•

-

-

Cold Squeeze

× WHOI

o TAMU

Δ SIO

_ i I 1 • ' i

Figure 1. Site 147 magnesium profile (cold squeeze).Underlined data are considered inaccurate.

3 0

mmol / kg

40

100 -

200 -

3 0 0 -

400

6 01 1 1 1 1 1 1—

" HOLE 148

c ×Δ

>ò o

×Δ

Λ

x _

Cold Squeeze

× WHO!

o TAMU

Δ SIO

1 . 1 , 1 1 ,

1 "'

x o

1 I

Δ < * 0

MAJOR ELEMENT AND CARBONATE CHEMISTRY DATA, SITES 147, 148, and 149

[ Mg + + ] mmol / kg

30 40

100 -

UJ 200Q

1 ' 1

" HOLE

-

-

X

, 1

1 I 1 1

149

C — - Δ x

x> Δ

x o

X O Δ

o x

x o

×Δ

>ùo

X O Δ

x o

1 ' '

•

×Δ o

xo

1 , ,

X

X

Δ

| l>^ i o I I

Δ ×Δ o

ox ΔΔ 0xo

Δ»

Δ

-

_

_

-

-

Cold Squeeze

× WHOI

o TAMU

Δ SIO

1 , , , ,

Figure 5. Site 149 magnesium profile (cold squeeze).Underlined data are considered inaccurate; C =contamination with dye during drilling.

[Mg + + ] mmol/kg

40 50

200 -

1 ' 1

' HOLE

c

-

-

, , 1

149

—»- X

X

X

X Δ

ft

X Δ

x

x

X Δ

X Δ

> x

Δ

Δ

Δ

Δ

x

XΔ

×Δ

1 I 1 I

×v>Δ

-

-

-

-

Warm Squeeze

x WHOI

Δ SIO

l i i . ,

Figure 6. Site 149 magnesium profile (warm squeeze).Underlined data are considered inaccurate; C =contamination with dye during drilling.

so 4-TAMU and WHOI both determined Sθ4= microgravi-

metrically. Agreement is generally within ±0.8 mmol/kg or±6% (34/49), whichever is larger. There is a tendency forthe TAMU values to be a few percent larger than the WHOIvalues. Nine samples were analyzed by Cescon and Machi(this volume) by using a cation exchange column andBaCh titration. Their values agree with the WHOI resultswithin ±0.3 mmol/kg or ±2% (7/9), and with the TAMUdata within ±0.8 mmol/kg or ±8% (3/5). On this basis, theWHOI sulfate data is chosen as slightly more reliable thanthe TAMU data.

Na+

The TAMU and WHOI sodium analyses were both doneby atomic absorption and generally agree within ±1.5%.Since [Na+] differences between pore water and seawaterare less than the analytical uncertainties, calculations in thispaper assume that pore water has the same Na:Cl ratio asseawater.

K+

Both investigators measured potassium by atomicabsorption. The two sets of data agree within ±5% (26/38),although the TAMU values are systematically about 2%greater than the WHOI values. Neither set of data appearssuperior, so the two are averaged unless two analyses differby more than 10%, and one of these clearly departs fromthe trend with depth (3 samples). One WS analysis(149-3-5) appears anomalously low and is replaced by aninterpolated value.

Mg++

TAMU and WHOI both determined [Mg++] by atomicabsorption while SIO measured the difference between anEDTA titration of total alkali earths and an EGTA titrationof calcium only, applying a small correction for strontiumcontent. The two sets of atomic absorption data bracketthe titration data, with the TAMU results systematicallyabout 2% greater and the WHOI results systematicallyabout 1% less than the SIO results. [Mg++] listed in Table 1is an average of all analyses except values which differ fromthe average by more than 5% and show a claer deviationfrom the trend with depth (1/108 for WHOI, 3/49 forTAMU, 9/106 for SIO). Generally, each investigatorsresults fall within ±2.5% of this average (91/108 for WHOI,37/49 for TAMU, 87/106 for SIO).

Ca++

TAMU and WHOI determined [Ca++] by atomic absorp-tion and SIO by titration with EGTA. The SIO sampleswere acidified during storage, and it has been suggested(Gieskes, this volume) that CaC03 precipitation may haveoccurred in unacidified TAMU and WHOI samples frombelow 63 meters at Site 147, causing these atomicabsorption analyses to be anomalously low. Therefore, for

835

D. E. HAMMOND

mmol / kg

10 20

50 -

x 100

tLüQ

2 0 0

Δ ^

-

_

-

-

o

— X

0 X

_ O X

-

-

-

\ ×'o A b

_o. x

x

0

X

×

1

Δ X

Δ××

Δ

1

Bt• o '

o

a Δ

Δ

Δ

Δ

i . i . . . .

Δ×-_

HOLE 147-

-

-

-

Cold Squeeze

× WHOI

o TAMU

Δ SIO

. 1 1 1 , r ,

150 - ^ i

Figure 7. Site 147 calcium profile (cold squeeze).Underlined data are considered inaccurate.

[Cα++] mmol /kg

5 10

50 -

x IOO-0_UJQ

150-

200

Δ


[ C α + + ] mmol / kg

4 5 6 7 8 9 10 II 12

100 -

H 2Q_UJQ

3 0 0 -

400

1 1 1 1 1

" HOLE 148

Δ >ex

Δ x

C — Δ

-

—

Cold Squeeze

× WHOI

o TAMU

Δ SIO

, 1 , 1 ,

1 1 1 • \JL> Δ × ' x l_o_ ' 1 '

Δ xv* oΔ X _ 0 _

Δ< 0× o —

Δ × oe _o_

X

ox

Δ ox_

Δ _×_

Δ _×_

Δ _x_

Δ _ £ _ x

-

1 , 1 , 1 1 1 , 1 1 1 I

Figure 9. Site 148 calcium profile (cold squeeze). Under-lined data are considered inaccurate; C = contaminationwith fresh water during drilling.

These may be related to the thermodynamic constants

K° =aH+aCO3

z

by the expressions

flH9O?CO9= — K°

1

K; =

K 2

wherβ7/ is the activity coefficient of species i chosen so thatai = 7/td Estimation of these activity coefficients poses adifficult problem since they are functions of both ionicstrength and solution composition.

Lyman (1957) has empirically determined K^ and Kjfor solutions of artificial seawater. Data from theGEOSECS I cruise (Takahashi et al., 1970) suggest thatLyman's K^ is accurate to 2% or better and Simpson's(1970) data indicate that near 25°C, Lyman's ratio ofK[jK^ is accurate to 10%. a has been determined to betterthan 1% in salt solutions (Bohr, 1899; Harned and Davis,

1943) and in seawater (Krough, 1904; Li and Tsui, 1971).Calculations of K{ and K[ for seawater from thermo-dynamic principles by Garrels and Thompson (1962),Berner (1965), and others have been less successful.Therefore, since pore water is often similar in compositionto seawater, the approach adopted here is to adjustLyman's empirical seawater constants for the differences in[Ca+ +], [Mg+ +], [SO 4

= ], and [HCO3"] observedbetween pore water and seawater. A model for apparentconstant estimation (MACE) is discussed in the followingparagraphs.

The major difficulty in calculating the activity coeffi-cients of interest is a result of the imprecise knowledge ofthe nature and magnitude of complex ion formation inseawater. Garrels and Thompson (1962) proposed a modelfor major ion complexing in seawater which may be used asa first approximation to the real system. Using their modeland the major element composition listed in Table 1, acomputer program was written to calculate the ratios

uncomplexed fraction in seawater' uncomplexed fraction in pore water

for the [HCO3~] , [CO3 = ] , and [Ca++] in each pore water

sample. Since pore water and seawater of the samechlorinity should have nearly the same ionic strength, theonly difference between the activity coefficients of theseions in seawater and in pore water of the same chlorinityshould be due to the differences in the uncomplexedfractions, f(. Assuming 7cO2

a n c* öH2θ depend only onchlorinity and letting

7j• = activity coefficient of i in seawater

7j• = activity coefficient of i in pore water

l

> ^ Λ -(f— Aj -V]nco

7 H C O 3 /COOΛ 2

/ H C O 3 2

where the superscript S denotes an empirical seawaterconstant.

The same approach can be used to estimate the apparentsolubility product of calcite

K' = [CO3=]

and the first apparent dissociation constant of boric acid

[H3BO3 H2O]

837

D.E.HAMMOND

[Cα+ +] mmol / kg

4 5 6 7 8 9 10 II

[Cα+ +] mmol / kg

10 20

100 h

200

300

1 1

-

•

—

-

-

-

-

-

1 l ' I

Δ XΔ×

Δ ×ΔX

Δ ×

C — ΔΔ×

Warm Squeeze

×

Δ

, i

WHOI

SIO

1 . 1

' àA

I |

×xX×Δ

X

Δ

1 ' X "

HOLE

X

Δ

Δ

1 , 1

-

148-

--

X

×

1

Figure 10. Site 148 calcium profile (warm squeeze).Underlined data are considered inaccurate; C =contamination with fresh water during drilling.

Assuming that the complexing of [B(OH)4 ] is similar to[HCO3~], these constants become

Expressions to compute apparent constants for seawaterare listed in Table 2a. The temperature and chloronitydependence for K^, A 2̂, and Kß is from a least squares fitto Lyman's (1957) data, α, Ks , and the pressure depen-

sp

dence of the apparent constants are taken from relationsgiven by Li et al. (1969) which were based on experimentaldata from Bohr (1899) and Harned and Davis (1943) asrecalculated by Buch (1951), from Maclntyre (1965), andfrom Disteche and Disteche (1967), respectively.

The effect of MACE is illustrated in Table 2b. The ratioof pore water constants to seawater constants is shown forseveral varieties of pore water (19%o chlorinity, 25°C),including the extremes of chemistry exhibited by samplesconsidered in this paper. The only compositional changes towhich the apparent constants show marked sensitivity arevariations in [Mg+ +]. Site 149, where [Mg++] decreases bya factor of 2, should provide a test for the validity of thismodel. This will be discussed in a later section.

ALKALINITY

In these cores, the ionic species which can contribute tototal alkalinity determinations are [H 3 Si0 4 ~], [PO 4~],[HP0 4=], [ H 2 P O 4 - ] , [B(0H) 4 -], [ H C 0 3 - ] , and[CO3~]. Table 3a indicates the abundance of each speciesat pH 7.0 and 8.0 as a fraction of the total inorganic

iooh

xa. 200üJQ

300 h

1 ' ' ' l× 'oc

oxè>ΔαΔ

c —

-

Cold Squeeze

× WHOI

0 TAMU

Δ SIO

, , , . 1 1

(

3X

ox

a & xoc

- Δ X

1 '

ΔO X

2 Δ

S>

A —

1

X

X

Δ

A

1 ' "1

HOLE

Δ 2. x.

0^

Δ 0_

• Δ £

Δ £

A- -Δ 0

Δ

0 *

1

149 "

-

X

X

Δ

X

2 ×

£ ×

Figure 11. Site 149 calcium profile (cold squeeze).Underlined data are considered inaccurate; C =contamination with dye during drilling; A = calciumtitration result considered low since it is accompanied bya complementary magnesium anomaly.

elemental abundance, calculated from the apparent dis-sociation constants for seawater listed in Table 3b. Analkalinity titration converts all of these to undissociatedspecies, except phosphate anions, which become primarily[H2PO4—]. Therefore, for pore water of pU 8, alkalinitycould be expressed as

= O.O4× 1.2 × ΣP + O.IÓX Σ B + I .OÓXΣC

and for pore water of pH 7,

Σ A = 0.83 XΣP + 0.02 x Σ B + 0.92 x ΣC

The silica, phosphate (Gieskes, this volume), and borate(Presley et al., this volume) data may be compared to thealkalinity data (Table 1). The above equations clearlyindicate that, in the observed pH range (Table 5), the firsttwo species contribute 0 to 0.5% and borate contributes 0.5to 1.0%, while carbonate species contribute 99 to 100% ofthe total.

Gieskes (this volume), using a modified Gran titration,measured alkalinity on shipboard. His results (Table 1) areplotted in Figures 13 to 15. He estimates their accuracy as±0.5%. To check their reliability, they can be compared(Table 4) with the Σ A calculated from the Σ C O 2 and P c θ 2results of Takahashi and Prince (Queens College, thisvolume), Σ B (TAMU), and ΣSI and Σ P (SIO). The mediandifference between calculated and measured alkalinity is 1%and the two generally agree within ±6% (13/18), abouttwice the sum of each investigators estimated uncertainties.

838


[Co+ +] mmol / kg

20 30

100-

200-

300 -

x

ΔX×Δ×

C — -

-

•

Worm Squeeze

× WHOI

Δ SIO

1 '

*Δ ×Δ X

x

1

Δ X

A

1

Δ

Δ ×

—- Δ

X

Δ

A —

1 1

HOLE

X

Δ ×

x

Δ 21

Δ X

Δ

A - Δ

Δ —x_

i 1

149 "

-

-

X

X

400

Figure 12. Site 149 calcium profile (warm squeeze).Underlined data are considered inaccurate; C =contamination with dye during drilling; A = calciumtitration result considered too low since it isaccompanied by a complementary magnesium anomaly.

The directly measured values are probably more reliable.Alkalinity titrations on stored samples (WHOI) scatterbadly about the SIO and QC results.

At Site 147 and the upper 70 meters at Site 148, the CSand WS alkalinities lie within about 4% of each other, withno apparent systematic trends. Below this depth at Site148, however, and at Site 149, some interesting tempera-ture effects appear, as noted by Gieskes (this volume). Inthe lower portion at Site 148, the WS alkalinity is about 0.2meq/kg greater than the CS alkalinity. At Site 149 the samephenomenon is observed down to about 240 meters, wherea transition occurs and the WS alkalinity is about 0.1meq/kg greater than the CS alkalinity. The cause of thesedifferences is not immediately obvious, but several mech-anisms can be proposed.

Alkalinity of the aqueous phase is a conservativeproperty and can be changed only by a transfer of ionsbetween sediment and pore water. Increases with warmingmight occur by:

1) release of H 3 Si0 4 ~, PO 4~, B(OH)4~, or OH" fromadsorption sites on clay minerals;

2) adsorption of calcium by clay minerals, causingdissolution of CaCO3 in order for the solution to return tosaturation;

3) release of H+ by clay minerals followed by solutionof carbonate minerals.

Decreases with warming may occur by:1) release of H+ by clay minerals and no

solution;2) precipitation of CaCO3.

These mechanisms can be distinguished by considering theeffect of temperature on both alkalinity and pH.

Examination of the effects of the increase in squeezingtemperature on water composition reveals that ZP showslittle change (Gieskes, this volume), ESi generally increasesby 0.05 to 0.10 mmol/kg (Gieskes, this volume), and Σ Bmay increase by about 30% or about 0.1 mmol/kg(Manheim et al., this volume). It is possible that[H 3 Si0 4 ~], [B(OH)4~], or more negatively chargedspecies of these ions are adsorbed on clay minerals andreleased on warming, consequently increasing ZA. Manheimet al. (this volume) have noted a temperature effect forcations in the pore waters of these cores; divalent cationsdecrease and singly charged cations increase on warming.No temperature effect, however, is apparent for the majoranions, chloride and sulfate. If [B(OH)4~] and [H 3Si0 4~]should be released at the pH observed in these pore waters,they would react with [H+] and consequently increase thepH.

There is a tendency for [Ca++] to be absorbed by clayminerals and decrease by a few percent as temperature isincreased. If this effect is balanced by a release of Na+ andK+, and CaCO3 dissolves until saturation equilibrium isregained, alkalinity increases must also be accompanied bypH increases.

Since clay minerals release the univalent cations K+ andNa+, it seems likely that they may also release H+. Thereaction

+ +CaCO3 + H

+ Ca

could then occur, increasing alkalinity. If the carbonatephase interacts with the aqueous phase in this manner, analkalinity increase on warming must be accompanied by apH decrease sufficiently large to drive this reaction to theright, despite the decrease in calcite solubility.

If clay minerals release [H+] and the carbonate phase isinert, alkalinity should decrease and pH should alsodecrease. If clay minerals do not affect the alkalinity butthe carbonate phase is reactive, precipitation of CaCθ3should occur, and again both alkalinity and pH shoulddecrease. Thus, the alkalinity and pH can be used togetherto determine the nature of the influence of the clay phaseand the responsiveness of the carbonate phase.

In 30/45 samples, alkalinity increases with temperature.As pointed out earlier, these appear to be real variations,showing depth dependence and increases often several timeslarger than the estimated analytical error. These increasesare usually (21/30) accompanied by a pH decrease (to bediscussed more completely in the following section) andonly a few (4/30) show a pH increase. This stronglysupports the hypothesis that, with an increase in tempera-ture, clay minerals release [H+] ions (or possibly absorb[OH"], and H2O [OH~] + [H

+]) causing carbonate todissolve and alkalinity to increase. Thus, both clay andcarbonate phases are capable of responding rapidly totemperature and chemical changes in their environment.

At Site 149, a change in lithology from calcareous clayto calcareous radiolarian ooze occurs between 195 and 240meters. This transition is reflected in the temperature effecton alkalinity. As the amount of clay drastically decreases,

839

D. E. HAMMOND

TABLE 2aApparent Constants for Seawater

(1) pKLλ = 6.339 - 7.78 × 10"3 X T + 7.1 X 10'5T2 - 0.01 X Cl

0'2 '5(2) pKL2 = 9.775 - 1.09 X 10'2 X T - 3.57 X 10'5 X T 2 - 0.02 X Cl

(3) pKLß = 9.255 - 1.06 X 10~2 X T + 3.57 X 10"5 X T 2 - 0.016 X Cl

(4) K*L = (0.69 - 6.3 X 10"3 X T) X 10"6 X Cl/19.0sp

K( (P)/Kt (1 bar) = Exp [-ΔV. X (P - 1)/R (T + 2731)]

aQ = 770. - 29.5 X T + 0.685 X T2 - 7.5 X 10~3 X T 3

a = α Q X 1O"7/(1.O13 X Exp [2.303 X (0.0806 - 7.4 X 10"4 X T) X Cl/20] )

mol/1-mbars

T = temperature (°C)

Cl = chlorinity (%o)

P = pressure (bars)

Vj = -19.0 cc/mole

V2 = -10.7 cc/mole

V3 = -23.1 cc/mole

ΔV4 = -29.7 cc/mole

R = 83.14 cc-bars/mol-degree

the alkalinity response to warming changes from an increaseto a decrease of about 0.05 meq/kg, suggesting thatprecipitation of carbonate occurs. Since pH does not alwaysshow a corresponding decrease, some of the carbonateprecipitation may be countered by the solution of anions ofacids which are weaker than carbonic acid.

To estimate the magnitude of each of these processes,the pH (Table 5) and alkalinity (Table 1) data for eachsample were used to calculate Σ C O 2 The difference inΣ C O 2 between CS and WS samples is attributed to solutionor precipitation of CaCO3. The amount of strong acid orweak base required to account for the alkalinity changewhich cannot be attributed to carbonate solution orprecipitation was then calculated. The resulting trends arelisted in Table 6. In all but two regions, release of [H+] byclay minerals and solution of carbonate is the dominanteffect. In these two regions, carbonate precipitation andsolution of anions of weak acids is apparent. At Site 149,this correlates with the disappearance of clay minerals, butat Site 148, the reason is unexplained.

pH

Three sets of shipboard pH measurements were made oneach sample, two by Gieskes on the water samples extrudedat 4°C (CS) and 22°C (WS) and one by Broecker byinserting electrodes directly into the sediment (PI). A glasscapillary electrode in series with surface seawater and asaturated calomel electrode was used for the first two setsof measurements, and an ORION R model 90-02 doublejunction reference electrode and a Beckman 40471 glasselectrode were used for the third set. Standardization wasmade with Beckman buffers of pH 6.86, 7.41, and 4.01.The accuracy of the CS and WS samples is estimated at

±0.05 pH units and that of the PI samples at ±0.08 pHunits. The measurement temperature was usually 26 to30°C for the pore water measurements and 18 to 25°C forthe punch-in measurements. This data is tabulated inTable 5.

To make all sets of measurements comparable, thealkalinity and pH at the measurement temperature wereused to calculate Σ C O 2 , and, making the assumption thatΣ C O 2 and alkalinity are independent of temperature, thesetwo quantities were then used to calculate the pH at 25°C.The results of this calculation are also listed in Table 5 andare plotted in Figures 16 to 18. The magnitude of pHtemperature dependence can be estimated from Figure 19in which pH is plotted as a function of temperature forpore waters with pH's of 7.0 and 8.0 at 25°C. Thiscalculation was made for waters ranging in compositionfrom seawater to varieties having the compositionalextremes listed in Table 2b The resulting plots areindistinguishable in Figure 19. Thus, pH variation withtemperature is primarily a function of pH and the tempera-ture dependence ofK±, A"2, taken here to be that measuredby Lyman (1957). Solution and precipitation of calcitewould change the slope of this curve slightly, but notintroduce a serious error in the temperature correction.

When all three sets of measurements are recalculated to25°C, removing the temperature effect, it is clear fromTable 5 that systematic differences exist. In most regions PI< WS < CS. The median differences between PI and WSmeasurements, and between WS and CS measurements, aresummarized in Table 7. The differences appear to be real.Individual measurements usually fall within 0.08 pH unitsof the median differences (46/58 for CS-WS, 29/41 forWS-PI). As suggested earlier in this paper, the discrepancy

840


TABLE 2b

Adjustment of Constants from Seawater to Pore Water forWater of 19% at 25° C

Pore water with:

Seawater composition

Seawater comp., no [SO^~]

Seawater comp., 0.5 X [ M g + + ] ^ ^

Seawater comp., 0.5 X [Ca++]sw

Seawater comp., 2 X [Ca 4 " 1 "]^^

Seawater comp., 10 X [ H C O 3 ~ ] S

‰r t I If*-'

Λ I /TV |

1.000

1.026

- 0.913

0.983

1.034

w 0.988

‰r 1 I IS*-*

2 2

1.000

1.066

0.722

0.982

1.034

0.967

1.000

1.026

0.913

0.983

1.034

0.988

1.000

1.005

0.667

0.967

1.066

1.017

between CS and WS waters may be due to release of H+

ions from clay minerals as the squeezing temperature isincreased. Again, this effect disappears as the clay fractiondecreases in the lower region at Site 149.

The differences between PI and WS pH are moredifficult to explain, since the PI measurement temperatureis close to 22°C, the warm squeezing temperature. Onepossibility is that calcite may dissolve in extruded watersamples due to the pressure increase. The terminal squeez-ing pressure was typically about 4500 kg total load, orabout 70 atm. If half this value is chosen to represent theaverage WS equilibration pressure between water andcalcite, calculation of the Σ C O 2 and alkalinity increasesexpected from the pressure increase suggests that the WSpH should be 0.02 units greater than PI pH. Thus, only asmall portion of the difference can be attributed tosqueezing pressure.

At Site 147, the large difference may be attributed toCO2 loss during squeezing from the gassy core, after the PImeasurement was made. P c o 2 °f

t n e s e samples was about0.1 atm (Hammond et al., LDGO, this volume) and even abrief exposure of the samples to the atmosphere could havehad a marked effect on pH. At Site 148, the twomeasurements are closer but still significantly different.Gieskes (personal communication) has suggested that adifference of 0.1 pH units may be due to the difference inthe liquid junction potential of the electrodes at the PImeasurement temperature of 19°C and the WS measure-ment temperature of 27°C. This could also account for thediscrepancy in the upper region at Site 149 but wouldincrease the difference in the middle region. To summarize,PI pH is consistent with WSpH within about 0.2 units. Theestimated measurement errors could account for half ofthis, but a systematic difference of about 0.1 units remainsunexplained. This might be an artifact of liquid junctionpotentials, or perhaps a CO2 loss between the two sets ofmeasurements.

One aim of the geochemistry leg was to establish areliability for the pH measurements. Except at Site 147, allthree sets of measurements are consistent to 0.2 units. Twoother estimates of pH can be made, one from Σ C O 2 andpCO2 determinations (QC, Table 4) made on some CSsamples, and another from pCO2 measurements on gaspockets at Site 147. The pH values calculated (Table 5)from the first are about 0.5 pH units greater than thosemeasured at Site 147 (Figure 16) and about 0.25 pH units

greater than those measured at Site 148 (Figure 17). Thisdiscrepancy has been attributed earlier in this paper to aCO2 loss from QC samples of about 6%, but does not ruleout the argument that all the pH measurements could begrossly in error, and, consequently, the calculation ofΣ C O 2 from alkalinity and pH is in error. The latterexplanation is unlikely because the second estimate (Table5, Figure 16) of pH, made from alkalinity (SIO) and thepCO2 of the gas pockets from Site 147 (LDGO), isconsistent with PI results. Although some scatter exists,they show no systematic variation from the PI pH and alsoshow similar systematic differences when compared to WSvalues, supporting the CO2 loss hypothesis. Another argu-ment for accepting the pH measurements as more reliablethan those calculated from ECO2 and pCO2 is that theformer indicate a degree of CaCO3 saturation at squeezingconditions which is generally between 0.6 and 2.4, whilethe QC data yields far greater values. This will be discussedin the following section. Thus, the WS and CS measure-ments appear to be consistent within the estimated error of0.05 units in each, if the influences of clay minerals aretaken into account. PI measurements are consistent withgas pocket pCO2 within 0.2 units, although they showsystematic differences from WS and CS measurements,some of which might be due to liquid junction potentials.Undoubtedly, some of the scatter with depth among thethree sets of data is due to analytical error, but it is likelythat part of this may be real, since two or three types ofmeasurements often show similar anomalies at the samelevel.

DEGREE OF SATURATION OF CaCO3

Calcite is present at all levels in each drill hole. Perhaps,as suggested by Gieskes (this volume), the most instructivetest of the consistency of the analytical data and the modelfor apparent constant estimation (MACE) is to calculate thedegree of saturation of calcite,

for each sample at the sampling conditions. Unfortunately,the picture is clouded by several problems. It has beenshown earlier in this paper that carbonate solution andprecipitation can occur on short time scales. It has not beenshown, however, that the kinetics of these reactions aresufficiently rapid for equilibrium to be attained, or, more

841

D. E. HAMMOND

TABLE 3aDissociated Fraction of Weak Acids in Seawater of

19.4% Chlorinity at 25° C

^ S K ^ - / Σ S I

H 2 P O 4 ~ / Σ P

HPO 4 ~/ΣP

P O 4= / Σ P

B(OH) 4~/ΣB

H C O 3 ~ / Σ C O 2

C O 3= / Σ C O 2

pH = 7.0

0.004

0.101

0.876

0.023

0.020

0.902

0.008

pU = 8.0

0.040

0.009

0.787

0.205

0.163

0.917

0.072

TABLE 3bDissociation Constants for Seawater of 19.4% Chlorinity at

25°C,HA* - H ÷ + A~*- 1

Acid

H4S1O4

H3PO4

H2PO4~

HPO4=

H3BO3 H2O

H2CO3

HCO3-

pKacid

9.4

1.61

6.06

8.56

8.71

6.00

9.10

Source

Sillen, 1966

Kester and Pytkowicz,



Lyman, 1957

Lyman, 1957

Lyman, 1957

1967

1967

1967

importantly, whether or not the reactive phase is calcite.Dolomite rhombs appear at Sites 147 and 148 (lithologicdescriptions, this volume). If the pore water was saturatedwith calcite under in situ conditions, carbonate precipita-tion should have occurred as the cores were raised from thedepths. As has been pointed out by Weyl (1957) andothers, precipitation on a calcite substrate is likely to resultin the construction of a magnesium-rich coating whichinhibits further equilibration of solid and aqueous phases.

With these -problems in mind, the sensitivity of thecalculation to input parameters can be examined.

[CO3: [HC03

and

Δ[HCO 3~] AaH+

[HCO3-] " α H +

The ratio K^jK^ will increase up to 10% with depth as thefraction of free [HCO3~] increases, primarily due to thedecrease in [Mg+ +]. Assuming this ratio is correct for

seawater, taking 10% as an upper limit for any error MACEmight introduce into the calculation or 12, and taking[HCO3~] Σ alkalinity, the error in £2 can be estimatedat about ±30% (Table 8). If MACE overestimates the effectsof [Mg++] decrease, it will be difficult to detect, but anunderestimate greater than a factor of 2 could introduce asignificant systematic trend in 12 with depth.

12 was calculated for CS and WS samples, using the porewater composition listed in Table 1, the pH in Table 5, thesqueezing temperature, and estimated average squeezingpressure of 35 atm for Sites 147 and 148 and 23 atm forSite 149 (L. Waterman, personal communication). Calcula-tions of 12 for PI measurements were made at 1 bar and themeasurement temperature, using water composition inter-polated from CS and WS analyses. The results of thesecalculations are plotted in Figures 20 to 22. 12 is approxi-mately constant with depth for each sample type in eachsite, but CS results are systematically more undersaturatedthan WS and PI values, and each site appears to have adistinct 12 (Table 9).

Figure 23 schematically illustrates the pressure andtemperature history of each sample. Dashed lines indicatechanges which should cause calcite precipitation, solid linesthose which should cause solution. To quantitativelyestimate the magnitude of these changes, a pressurereduction of 50 atm should cause the precipitation of 0.1mmol CaCC^/l and warming from 4°C to 22°C shouldcause an equal amount of solution. If precipitation kineticsare too slow to allow equilibrium to be attained, PI samplesshould be supersaturated, with those from Site 149 beingfarthest from equilibrium. The results, despite their scatter,indicate that PI measurements are consistent with thehypothesis that pore water equilibrates rapidly with calcite.

Cs and WS samples do not suggest this as clearly. Severalpuzzling trends exist. 12 for CS samples tends to be smallerthan 12 for WS samples, and at Site 149, CS samples appearundersaturated with calcite. Since calcite appeared to bedissolving as temperature was increased, 12 for WS samplesshould be close to or less than 1.0. The third problem isthat the core expected to be farthest from equilibriumwhen sampled, Site 149, has pore waters closest tosaturation. As suggested earlier in this paper, CO2 loss fromSite 147 pore water samples during squeezing may haveincreased the pH, and consequently 12. Since P c θ 2increases with temperature, greater gas loss would beexpected from WS samples, explaining the higher WS 12.The same argument could be applied to Site 148, where thePCO2 is roughly 0.01 atm. Only a small amount of CO2loss (1-2%) would be required to change the pH by 0.15units and bring 12 close to 1.0. At Site 149 the PCO2 °fsamples is roughly half that of Site 148 samples andconsequently less CO2 loss might be expected. WS 12 isclose to 1.0 as expected.

To summarize, it appears that pore water may equili-brate rapidly with calcite since Leg 15 pH measurementsare consistent with this hypothesis. Squeezed samples fromSites 147 and 148 show slight calcite super saturation andmay have lost CO2 during collection. The undersaturationexhibited by CS samples from Site 149 remains a mystery.

It is interesting to note that 12 is constant to better than20% in each set of measurements and shows no systematic

842


TABLE 4Comparison of ΣCO2 + PQQ and Alk + pH

Sample

147B-1-3147B-1-4147B-1-4147B-2-2147B-2-6147B-6-2148-M148-2-1148-3-3148-4-3148-7-3148-7-3148-8-3148-10-3148-12-4148-18-2148-20-3148-23-4Median

Depth(m)

4.254.755.10

15.02151

61122315959

6786

106159179209

ΣCO

QC

15.0213.4414.63

8.1812.519.503.602.915.214.807.607.497.716.544.712.863.242.88

2 (mmol/kg)

SIO

18.3217.4615.11

8.3915.0510.09

3.963.515.245.047.707.707.706.635.123.683.673.08

(%)

-21.9-29.9- 3.3- 2.5-12.3- 6.1- 9.9-20.9- 0.4- 4.9- 1.3- 1.9

0.0- 1.3- 8.6-28.5-13.2- 7.0- 6.5%

SIO

17.7217.6015.17

8.1813.7510.064.003.515.235.027.617.617.666.725.273.733.633.05

Ealk meq/kg

QC

15.8014.4715.00

8.7413.8010.22

3.842.995.414.997.927.637.866.764.872.863.252.86

Δ(%)

10.817.6

1.1- 6.8- 0.4

1.6

4.014.9

- 3.40.6

- 4.1- 0.2- 2.6- 0.6

7.523.310.5

6.0

+ 0.5%

^ c o 2 (10~3 a t m )

QC

5.973.758.422.662.402.231.392.052.752.483.205.454.723.282.422.782.943.03

/ measured - calculated \\ measured / × 100.

depth dependence. This suggests that the estimated errorsare realistic with pH measurements introducing the majorerror in £2 and that MACE may be reliable to better than10%, even at Site 149 where the largest [Ca++] and[Mg++] changes are observed. It should be pointed out thatan alternative explanation might be advanced to explain thepH and £2 data. It has been suggested (most recently bySuess, 1970) that organic coatings can develop on calcitesurfaces and inhibit rapid solution and precipitation. Sites147 and 148 have significant organic content while Site 149does not. Attributing the difference between PI and WSpHmeasurements to liquid junction potentials, claiming CO2loss occurred at Site 147, and that at Site 149 calcite andwater can equilibrate rapidly but are inhibited by organicshields on calcite crystals at Sites 147 and 148 leads to theconclusion that Sites 147 and 148 water exhibit saturationunder in situ conditions and that little precipitation hasoccurred as cores were retrieved. This hypothesis still failsto explain the CS undersaturation at Site 149 and also theobserved alkalinity increases in the lower portions at Site148. Consequently, the CO2 loss hypothesis is favoredalthough it is difficult to unequivocally ascertain.

SUMMARY

1. The major element data from Leg 15 have beenexamined and appear reliable with the exception of [Ca++]measurements made by atomic absorption.

2. The Garrels and Thompson (1962) model predictsthat the apparent constants for the carbonic acid systemwill be different in pore water than in seawater. Theirmodel is used to develop a computational procedure whichcalculates apparent constants for pore waters by adjusting

experimentally determined seawater constants. This proce-dure predicts that if seawater constants were used, theywould introduce a systematic error of up to 10% incalculating 12, the degree of saturation of calcite, primarilydue to the low [Mg++] in pore waters. It is estimated thatthe procedure is accurate to better than 10%.

3. The several carbonate parameters measured, Σalkalinity, Σ C O 2 , P C O 2 '

ane noted when pore water last contactssediment. This equilibration may be hindered by organicmaterial.

4. Clay minerals, where present, release H+ ions as theyare warmed. The reaction CaCO3 + H+ Ca

+ + + HCO3~occurs, increasing alkalinity and decreasing pH. In regionswhere clay minerals are rare, CaCO3 precipitation occurs assediment is warmed, causing a decrease in alkalinity. Theexpected accompanying decrease in pH is not clearlyapparent.

ACKNOWLEDGMENTS

I am indebted to the members of the scientific andtechnical staff of the Glomar Challenger for their assistancein data collection and especially to Dr. W. S. Broecker whomade the punch-in pH measurements, provided stimulating

843

D. E. HAMMOND

TABLE 5Comparison of Calculated Measured pH

Sample

Site 147147B-1-2147B-1-3147B-1-4147B-1-4147B-1-4147B-2-3/4147B-2-2147B-2-6147B-4-3/4147-6-4147B-6-2147-8-2/3147B-7-4147-10-3147-10-3/4

147B-9-4147B-11-3147-14-3147C-2-1147C-2-2

147C-4-4147C-7-3147C-7-4

Site 148

148-1-2148-1-4148-2-1148-2-3148-3-3148-4-3148-5-2148-6-4148-7-3

148-8-3148-9-4148-10-3148-12-4148-14-3148-16-3148-18-2148-20-3148-23-4148-26-2

Site 149149-2-2149-2-5149-3-5149-4-3149-5-3149-6-4149-7-2149-8-4149-9-5149-10-2

Depth

24445

815212847

5163638282

83100119126128

148175176

36

111322

31395159

67

7986

106122141159179209232

48

172332

4349627278

.5

.2

.7

.8

.1

.5

.0

.0

ColdSqueeze

pH

7.367.287.587.627.56

7.497.307.377.25

7.557.217.34

6.97

7.057.17

7.12

7.11

7.43

7.647.537.467.617.48

7.457.467.487.41

7.46

7.657.617.727.507.43

7.617.4217.407.49

7.517.497.287.357.46

7.387.327.327.24

T

27.627.828.028.028.0

27.328.226.427.9

27.928.428.4

26.729.429.0

27.9

27.2

27.4

29.529.028.028.626.8

28.128.529.329.4

29.0

28.928.027.828.628.227.428.030.030.1

30.030.029.028.426.9

27.025.826.026.8

WarmSqueeze

pH

6.776.207.627.577.52

7.297.29

7.59

7.35

6.957.05

6.94

6.98

7.42

7.547.437.927.517.44

7.427.427.447.49

7.41

7.607.717.537.427.307.517.427.737.42

7.477.467.377.357.44

7.307.277.327.24

T

27.627.827.728.028.0

28.226.4

27.9

28.4

29.429.0

27.9

27.2

27.4

29.528.528.028.626.8

28.128.529.329.4

29.028.928.027.828.628.227.428.030.030.1

30.030.029.028.426.9

27.025.826.026.8

Punch-In

pH

6.937.017.367.167.19

7.056.94

7.03

6.76

6.64

6.716.72

7.347.377.897.337.42

7.437.427.377.43

7.41

7.477.527.367.517.337.38

7.447.407.307.327.357.337.367.297.337.38

T

22.023.524.022.024.8

25.025.7

27.0

28.0

27.0

26.027.5

19.520.018.022.019.0

19.018.019.015.5

17.517.019.020.020.021.522.0

20.020.020.020.020.020.020.020.020.020.0

cs

7.387.307.617.657.59

7.517.327.387.27

7.577.237.36

6.98

7.077.19

7.13

7.12

7.45

7.687.567.487.647.49

7.477.497.517.44

7.497.697.647.757.537.457.637.447.447.53

7.557.537.317.377.477.397.337.337.25

WS

6.786.217.647.607.54

7.317.30

7.61

7.37

6.977.07

6.95

6.99

7.44

7.587.467.957.547.457.447.457.477.53

7.447.637.747.567.457.327.537.447.787.46

7.517.517.407.377.457.317.287.337.25

pH (25°

PI

6.927.007.357.147.19

7.056.94

7.04

6.77

6.65

6.716.73

7.307.347.827.317.38

7.397.377.337.36

7.367.417.477.327.477.317.36

7.407.367.277.197.30

7.307.337.267.307.34

C)

QC

7.858.02

7.67

7.968.20

8.15

7.957.65

7.76

7.78

7.857.567.65

7.787.80

7.477.507.42

LDGO

7.146.84

6.816.96

6.756.616.89

6.85

6.776.92

844


TABLE 5 - Continued

Sample

149-11-4149-12-5149-14-3149-16-4149-18-3149-20-4149-23-4149-26-2149-29-3149-31-1149-33-1149-35-4149-37-3149-40-1149-41-5149-42-2

Depth

90100116135153

173201226254271

289313329354369374

ColdSqueeze

PH

7.237.127.217.067.137.036.886.986.986.976.986.966.987.017.007.00

T

29.029.029.029.029.029.029.228.528.428.028.028.029.029.029.228.0

WarmSqueeze

pH

7.237.217.146.957.127.036.916.926.947.006.946.987.017.017.057.07

T

29.029.029.029.029.029.029.228.528.428.028.028.029.029.029.228.0

Punch-In

pH

7.387.407.407.01

7.05

T

20.020.020.020.0

20.0

CS

7.267.147.237.087.157.056.907.007.006.986.996.977.007.037.027.01

WS

7.257.237.166.977.147.056.936.946.967.016.956.997.037.037.077.09

pH

7776

7

(25 C)

PI QC LDGO

.34

.37

.37

.98

.02

discussion throughout the study, and with Mr. S. Emerson,reviewed the final manuscript. Drs. Gieskes, Presley, andSayles are thanked for fruitful discussion and the use oftheir data prior to publication. Drs. Takahashi and Simpsonand Mr. Jaw-Long Tsou made helpful comments. MarylouZickl and Tom Zimmerman did the typing and drafting.

Financial support from the National Science Foundationby contrast NSF C-482 to collect the data presented hereand by the Atomic Energy Commission grant AT(ll-l)2185 to interpret the results is gratefully acknowledged.

REFERENCES

Ben-Yaakov, S. and Kaplan, I. R., 1971. Deep sea in situcarbonate saturometry. J. Geophys. Res. 76, 722.

Berner, R. A., 1965. Activity coefficients of bicarbonate,carbonate, and calcium ions in sea water. Geochim.Cosmochim. Acta. 29, 947.

Bohr, C, 1899. Definition and Methode zur Bestimmungder Innvasions-und Evasionskoefficienten beiderAuflösung von Gasen in Flüssigkeit. Ann. Phys. Chem.68,500.

Buch, K., 1951. Das Kohlensaüre-Gleichgewichtssystem imMeerwasser. Hansforshkn Inst. Skr. No. 151. Helsingfors.

Disteche, A. and Disteche, S., 1967. The effect of pressureon the dissociation of carbonic acids from measurementswith buffered glass electrode cells: The effects of NaCl,KC1, Mg++, CΣL++, SC>4=, and of boric acid with specialreference to sea water. J. Electrochem. Soc. 114, 330.

Garrels, R. M. and Thompson, M. E., 1962. A chemicalmodel for sea water at 25°C and one atmosphere totalpressure. Am. J. Sci. 360, 57.

Harned, H. S. and Davis, R., 1943. The ionization constantof carbonic acid in water and the solubility of carbondioxide in water and aqueous salt solutions from 0° to50°C. J. Am. Chem. Soc. 65, 1030.

Kester, D. R. and Pytkowicz, R. M., 1967. Determinationof the apparent dissociation constants of phosphoricacid in sea water. Limnol. Oceanog. 12, 243.

Krough, A., 1904. On the tension of carbonic acid innatural waters and especially sea water. Medd. Groen-land. 26,331.

Li, Y.-H., Takahashi, T.} and Broecker, W. S., 1969. Degreeof saturation of CaCO3 in the oceans. J. Geophys. Res.74,5507.

Li, Y.-H. and Tsui, T. F., 1971. The solubility of CO2 inwater and sea water. J. Geophys. Res. 76, 4203.

Lyman, J., 1957. Buffer mechanism of sea water. Ph.D.Thesis, Univ. California, Los Angeles. 196 p.

Maclntyre, W. G., 1965. The temperature variation of thesolubility product of calcium carbonate in sea water.Fisheries Res. Board Canada Manuscript Rept. Ser. No.200. 153 p.

Simpson, H. J., 1970. Closed basin lakes as a tool ingeochemistry. Ph.D. Thesis, Columbia University. 325 p.

Suess, E., 1970. Interaction of organic compounds withcalcium carbonate—I. Association phenomena and geo-chemical implications. Geochim. Cosmochim. Acta. 34,157.

Takahashi, T., Weiss, R. F., Culberson, C. H., Edmond,J.M., Hammond, D.E., Wong, C. S., Li, Y.-H. andBainbridge, A. E., 1970. A carbonate chemistry profileat the 1969 GEOSECS intercalibration station in theeastern Pacific Ocean. J. Geophys. Res. 75, 7648.

Weyl, P. K., 1967. The solution behavior of carbonatematerials in sea water. Studies Trop. Oceanogr. Miami. 5,178.

OTHER REFERENCES NOT CITED

Edmond, J. M., 1970. The carbonic acid system in seawater. PhD Thesis, Univ. California, San Diego, 172 p.

Edmond, J. M. and Gieskes, J. M. T. M., 1970. On thecalculation of the degree of saturation of sea water withrespect to calcium carbonate under in situ conditions.Geochim. Cosmochim. Acta. 34, 1261.

Sverdrup, H. W., Johnson, M. W. and Fleming, R. H., 1942.The Oceans. Englewood Cliffs, N. J. (Prentice Hall).1060 p.

Wattenberg, H. and Timmermann, E., 1936. Uber dieSattigung des Seewassers an CaCθ3, und die anorgano-gene Bildung von Kalksedimenten. Ann. Hydrogr. Mar.Meteoral. 23.

845

D. E. HAMMOND

50h

IOO µ

I5θh

200

ZAIk. meg / kg

20

OX

s

-

-

× SIO

o SIO

r I i

1

X

(cs)(ws)

1

' X

×

XD

×

×

, , 1

X X

X

' ' 1

HOLE

x< o

, I 1

147 "

-

o —

×

ox

-

Figure 13. Site 147alkalinity profile.

ioo

X 200

Q

300

T

HOLE 149

ZAIk. meg /kg

1.0 2.0

× SIO (cs)

o SIO (ws)

× ox C× c

× Ox O

XO

× O

400 ' ' '

Figure 15. Site 149 alkalinity profile.

Σ Alk. meg / kg

2 4 6

1001-

2001-

300 h

1 1

" HOLE 148

X

× SIO (cs)

o SIO (ws)

1 1

×l

a

o

|

~π—o×

C

×

×

×o

×o

o

1

1X O

×oo ×

-•» × o

X 0

1

1

0 ×

XO

1

1

0

×o

—

-

-

-

-

-

-

1

Figure 14. Site 148 alkalinity profile. C = contaminationwith fresh water during drilling.

TABLE 6The Effect of Warming on Alkalinity Changes

Region

Site 147

Site 1483-26m27-63m64-23 2m

Site 1494-239m240-3 74m

WS-CS

mmol/kgCaCθ3

Dissolved

+0.7

+0.2-0.2+0.3

+0.2-0.1

meq/kg

Acid

Released

1.2

0.3-0.2+0.3

+0.2-0.1

Sediment Type

Calcareous and dolomiticclay

Marls and calcareous clays

Marls, clays, and chalkRadiolarian chalk(abundant radiolaria)

846


6.6 7.0

pH (25°(

7.4 7.8

100o

150 -

200

,_o 1 ' o1 'Δ

ΔΔ

T Δ

ΔT

Δ

T

X

Δ ΔT V ×

- T Δ

T

T °

T

T

I . I . I

3 X

X

OX

1 1

× × l '×o * » θ ' ' ü

ex0 X

× o

xo

HOLE

× Cold

o Warm

Δ Punct

' a •

a

D

147

Squeeze

Squeeze

in

1

G "

-

-

-

-

CK a QC (calculatedfrom cs Σ C O J + ' 3 C 0 2 )

* LDGO (calculated fromEAlk + gas pocket PC(J-

• 1 i i i 1 i 1 i 1 *

Figure 16. Site 147 pHprofile (calculated at 25°C).

100

200

300

7.21 •

-

9 ΔΔ

Δ

Δ

UP

1 HOLE

X

o

Δ

a

1 .

Cold

PH

7.41 'o 1 '×

× oΔ O XΔ o x

Δ o ×O X

Δ × o aΔ O X

ΔΔ

0

O Δ X

X

Δ G O

« a

D x

o ×

148

Squeeze

Warm Squeeze

Punch in

QC

1

( calculated from

I I I ,

(25°C)7.6

cs

1

> P * 1

a0 X

X

X

Σ C O 2+

, 1

a

o

X

7.8

a

G

a

o

o

1 1 1

8.0

-

-

-

-

-

-

-

-

-

-

-

i 1

Figure 17. Site 148 pH profile (calculated at 25°C). Ccontamination with fresh water during drilling.

7.0

pH (25°C)

7.2 7.4 7.6

100 -

x 200

IdQ

300 -

400

" HOLE 149

o ×

Δ •

— x o

0 X

O X

× o

0 X

× o

X O

•

xx °o

X

ax

Δ

c

Δ

o×0

Δ ×

Δ

)

Δ

X

0

Δ

Δo

•

X

X

Δ

ΔΔ

Cold

Warm

Punch

i

'8×

o ×

Squeeze

Squeeze

in

×

c -

-

-

-

Figure 18. Site 149 pH profile (calculated at 25°C). Ccontamination with dye during drilling.

847

D. E. HAMMOND

0

TEMPERATURE (°C)

10 20

-7.0

Figure 19. Effect of temperature on pH of pore water withconstant ΣCC>2 and Σ alkalinity.

TABLE 7Comparison of pH (Recalculated to 25° Q Measurements

(Median Differences in pH Units)

TABLE 8Errors in Calculation of S2

Source

[Ca+ +]

Σ alkalinity

a H +

MACE

Total

Magnitude

±5%

±0.5%

±0.05 pH units ±12%

±10%

±28%

CALCITE SATURATION, [ C α ^

0.0 1.0 2.0

s] / KS P

3.0

5 0 -

Region

Site 147

Site 148Site 149

0-66 m

67-262 m

263-374 m

CS-WS

+0.06

+0.06

+0.02

-0.03

WS-PI

+•0.30

+0.14

+0.10

-0.07

MedianTemperature

of PIMeasurement

25.4°C

19.0°

20.0°C

20.0°C

20.0° C

Sediment Type

Calcareous and dolo-mitic clayMarls and clays

Marls and chalk, firstradiolaria appear at194 meters

Radiolarian ooze andchalk

Q.LJQ

Fi

~ 100I

150-

200

o

_

-

-

-

-

-

×oΔT

^ ' & ×g • •××x

Δ ×Δ X O

X

T Δ ×

T ×x

T

COLD SQUEEZEWARM SQUEEZE _PUNCH IN

GAS POCKET

, , 1 . , , .

o * o

o

T

Δ

T

1

— ' i 1 1—

HOLE

o

X

o

X

ex

× o

1 '

147 .-

-o -

-

o -

-

-

-

_

1 .

Figure 20. Site 147. Degree of saturation for calcite (atsample collection temperature and pressure).

848


CALCITE SATURATION, [Cα+ +][Cθà] / KSP

0.0 1.0 2.0

5 0 -

~ 100

xQ.

Q

150 -

200-

1 ' ' ΔX

-

-

-

-

-

-

-

-

×

×

' '

D. E. HAMMOND

TEMPERATURE (°C)

10 20 30

100

Lüccz>COCOUJcr

200

300

400

VENEZUELABASIN

Figure 23. Pressure and temperature history of samples.Dashed lines represent environmental changes whichshould cause caldte precipitation and solid lines thosewhich result in caldte solution.

850

11. interstitial water studies, leg 15 - a comparison of … · 2007. 5. 17. · 11. interstitial...

Documents