Lewis, S.D., Behrmann, J.H., Musgrave, R.J., and Cande, S.C. (Eds.), 1995Proceedings of the Ocean Drilling Program, Scientific Results, Vol. 141

36. DATA REPORT: GEOCHEMICAL LOGGING THROUGH AN ACCRETIONARY PRISM:CHILE TRIPLE JUNCTION1

Elizabeth Lewis Pratson,2 Cristina Broglia,2 Xenia Golovchenko,2 Amane Waseda,3 and Philip Froelich4

ABSTRACT

Geochemical well logs were obtained through sediments at Leg 141 Sites 859 and 863. Corrections have been applied to thelogs to account for variations in borehole size, drilling-fluid composition, and drill-pipe attenuation. Concentrations of Th, U, andGd, as well as oxide weight percentages, have been calculated from the logs and compared with shipboard and shorebasedX-ray-fluorescence (XRF) and carbonate bomb core measurements. The geochemical processing was performed in both open holeand cased sections of Holes 859B and 863B. Comparison of XRF core data to log data is good.

INTRODUCTION

During Leg 141, two sites were drilled into the toe of the accre-tionary prism at the point where the Chile Ridge is about to be (Site859), or has already been (Site 863), subducted. The purposes of drill-ing these sites were to determine the lithologies, depositional envi-ronments of the sediments, and the extent of hydrothermal alteration(Shipboard Scientific Party, 1992a, 1992b). To provide a continuousdetermination of lithology, Holes 859B and 863B were logged withthe geochemical logging tool string (GLT), which measures the majorelements of a formation at intervals of 0.15 m. The GLT measure-ments provide continuous, in-situ chemical measurements of the rockand are not affected by incomplete core recovery or core expansion,a common problem in core-based studies.

This report describes the basic principles of the GLT and outlinesthe post-cruise processing techniques. It then briefly compares geo-chemical log values with lithologic core descriptions and availablecore measurements.

GEOCHEMICAL TOOL STRING

The geochemical logging tool string consists of four tool com-ponents: the natural gamma-ray tool (NGT), the compensated neu-tron tool (CNT), the aluminum activation clay tool (AACT), and thegamma-ray spectrometry tool (GST). (GLT, NGT, CNT, AACT, andGST are trademarks of Schlumberger; Fig. 1.) The combination of thetool components uses three separate modes of gamma-ray spectros-copy for a comprehensive elemental analysis of the formation. TheNGT, located at the top of the tool string, measures the naturallyoccurring radionuclides thorium (Th), uranium (U), and potassium (K)before the formation is irradiated by the nuclear sources contained inthe tools below. The CNT, located below the NGT, carries a low-energycalifornium-252 (252Cf) neutron source to activate the Al atoms in theformation. The AACT, a modified NGT, is located below the 252Cfsource, measuring the activated gamma rays in the formation. By com-bining the AACT measurement with the previous NGT measurement,the background radiation is eliminated, and a reading of formation Al

1 Lewis, S.D., Behrmann, J.H., Musgrave, R.J., and Cande, S.C. (Eds.), 1995. Proc.ODP, Sci. Results, 141: College Station, TX (Ocean Drilling Program).

Borehole Research Group, Lamont-Doherty Earth Observatory, Columbia Univer-sity, Palisades, NY 10964, U.S.A.

3 J APEX Research Center, 1-2-1 Hamada, Mihama-ku, Chiba 261, Japan.4 Lamont-Doherty Earth Observatory and Department of Geological Sciences, Co-

lumbia University, Palisades, NY 10964,U.S.A.

is obtained (Scott and Smith, 1973). The GST, at the base of the string,carries a pulsed neutron generator to induce prompt-capture gamma-ray reactions in the borehole and formation and an Nal(Tl) scintillationdetector to measure the energy spectrum of gamma rays generated bythe neutron-capture reactions. Because each of the elements in theformation is characterized by a unique spectral signature, it is possibleto derive the contribution (or yield) of each of the major elementssilicon (Si), iron (Fe), calcium (Ca), titanium (Ti), sulfur (S), gadolin-ium (Gd), and potassium (K) from the measured spectrum. By com-bining these yields with the elemental concentrations from the NGTand AACT one can estimate the relative abundance in the formation ofeach element above. The GST also measures the hydrogen (H) andchlorine (Cl) in the borehole and formation, but these elements are notused for determining rock geochemistry.

The only major rock-forming elements not measured by the GSTare magnesium (Mg) and sodium (Na); the neutron-capture crosssections of these elements are too small relative to their typical abun-dance to be detected by the GST. A rough estimate of Mg + Na can bemade by using the photoelectric factor (PEF), measured by the litho-density tool. This measured PEF is compared with a calculated PEF (asummation of the PEF from all of the measured elements). The sepa-ration between the measured and calculated PEF is, in theory, attribut-able to any element left over in the formation (i.e., Mg and Na). Furtherexplanation of this technique is found in Hertzog et al. (1989). The Mgcalculation was not attempted for this leg, because including it in thenormalization with the other elements induced noise into all otherelements, this problem is pervasive for ODP geochemical logs (Pratsonet al., 1993). MgO + Na2O + MnO values from core data are includedin the normalization step of the processing. This is explained further inStep 5 of the data-reduction section below.

DATA REDUCTION

The well-log data from the Schlumberger tools are transmitteddigitally up a wireline and are recorded and processed on the JOIDESResolution in the Schlumberger Cyber Service Unit (CSU). The re-sults from the CSU are made available as "field logs" for initial,shipboard interpretation. Subsequent reprocessing is necessary tocorrect the data for the effects of fluids added to the well, loggingspeed, and drill-pipe interference. Processing of the spectrometrydata is required to transform the relative elemental yields into oxideweight fractions.

The processing is performed with a set of log-interpretation pro-grams written by Schlumberger, which were modified to account forthe lithologies and hole conditions encountered in ODP holes. Thesteps are summarized in Figure 2 and described in more detail below:

427

DATA REPORT

πTCCB

NGT

CNT-G

AACT

D

D

GST

D

i

Telemetry cartridge

Natural gamma-ray tool: measures naturally radioactiveelements thorium, uranium, and potassium

Dual-porosity compensated neutron tool: Measuresneutron porosity in the thermal and epithermal energy ranges

Thermal detectors

Californium 252 source

Epithermal detectors

Aluminum activation clay tool: Measures aluminumactivation and natural count rates as it passes the formationactivated by californium 252.

Gamma-ray spectrometry tool: Measures concentrationof calcium, silicon, iron, sulfur, gadolinium, titanium,hydrogen, chlorine, and formation capture cross section

Boron sleeve - 3.75 in.

Neutron accelerator

Figure 1. Schematic drawing of the Schlumberger geochemical logging toolstring used in the Ocean Drilling Program.

1. Reconstruction of Relative Elemental Yieldsfrom Recorded Spectral Data

This first processing step compares the measured spectra from thegamma-ray spectrometry tool with a series of "standard" spectra todetermine the relative contribution (or yield) of each element. Each"standard" approximates the spectrum of each element and is com-bined at each depth with the recorded spectrum in a weighted, least-squares inversion to determine the relative elemental yields.

Six elemental standards (Si, Fe, Ca, S, Cl, and H) are used to pro-duce the shipboard yields, but three additional standards (Ti, Gd, andK) can be included in the post-cruise processing to improve the fit ofthe spectral standards to the measured spectra (Grau and Schweitzer,1989). Although Ti, Gd, and K often appear in the formation in verylow concentrations, they can make a large contribution to the mea-sured spectra because they have large neutron-capture cross sections.For example, the capture cross section of Gd is 49,000 barns, whereasthat of Si is 0.16 barns (Hertzog et al., 1989). Therefore, including Gdis necessary when calculating the best fit of the standard spectra to themeasured spectra.

The spectral analysis was performed using the spectral standardsfor H, Si, Ca, Cl, Fe, Ti, S, and Gd. The spectral standards for K wasnot used, because this element exist in concentrations below theresolution of the tool, and the inclusion of K was found to signifi-cantly increase the noise level of all the other yields. A straight,five-point (2.5 ft, 0.762 m) smoothing filter was applied to all theyields to reduce the noise in the data during this reconstruction step.An additional 10 point (7 ft, 2.13 m) smoothing filter was applied to

the yields to further reduce the noise level in the normalization factor(explained in step 5), which affects the overall character of the finalelemental yields.

2. Depth Shifting

Geochemical processing involves the integration of data from thedifferent tool strings; consequently, it is important that all the data aredepth correlated to one reference logging run. The NGT, run on eachof the logging tool strings, provides a spectral gamma-ray curve withwhich to correlate each of the logging runs. A reference run is chosenon the basis of low cable tension (the logging run with the leastamount of tool sticking) and high cable speed (tools run at fasterspeeds are less likely to stick and are less susceptible to data degrada-tion caused by ship heave). The depth-shifting procedure involvesselecting several reference points where log characters are similar andthen utilizing a program that stretches or squeezes sections of thematching logging run to fit the reference logging run. The geochemi-cal tool string was the reference logging run for Hole 859B; the thirdpass of the Formation MicroScanner tool string was the reference runfor Hole 863B.

3. Calculation of Total Radioactivity and Th, U,and K Concentrations

The third processing routine calculates the total natural gammaradiation in the formation as well as concentrations of Th, U, and K,using the counts in five spectral windows from the natural gamma-raytool (Lock and Hoyer, 1971). This routine resembles shipboard pro-cessing; however, the results are improved during post-cruise pro-cessing by including corrections for hole-size changes andtemperature variations. A Kalman filtering (Ruckebusch, 1983) isused in the CSU processing at sea to minimize the statistical uncer-tainties in the logs, which can otherwise create erroneous negativevalues and anti- correlation (especially between Th and U). An alphafilter has been introduced more recently and is now recommended bySchlumberger for shore-based processing. This filter stronglysmooths the raw spectral counts but keeps the total gamma-ray curveunsmoothed before calculating the Th, U, and K (Charles Flaum, pers.comm., 1988). The outputs of this program are K (wet wt%), U (ppm),and Th (ppm), as well as total gamma-ray and computed gamma-ray(total gamma ray minus U contribution) curves.

4. Calculation of Al Concentration

The fourth processing routine calculates the concentration of Al inthe formation using four energy windows recorded with the AACTDuring this step, corrections are made for natural radioactivity, bore-hole-fluid neutron-capture cross section, formation neutron-capturecross section, formation slowing-down length, and borehole size.Porosity and density logs are needed to convert the wet-weight per-centages of K and Al curves to dry-weight percentages. The neutronporosity curve was used in both Holes 859B and 863B. In Hole 859B,this curve had to be rescaled to match cores using the formula:

0, = (p,!-35)×O.54, (1)

where:0t = percentage porosity, and pπ = neutron density.

The reason for this large rescaling of porosity is the fact thatthe neutron source in this tool is a 252Cf, instead of an Americium-Beryllium (AmBe) source which the tool was originally calibrated for.

A correction is also made for Si interference with Al; the 252Cfsource activates the Si, producing the aluminum isotope 28A1 (Hertzoget al., 1989). The program uses the Si yield from the gamma-ray

DATA REPORT

FMS CNT LDT DITE SDT NGT ACT

DEPTHSHIFT alf data to one reference run.

-Caliper— i 1Borehole Environmental

Corrections

neutron porosity

GST

Waveform Spectra

f YeildReconstruction

9 relative yieldsdetermined byweighted least

squares inversion^ of spectra >

Relative yieldsof Si, Ca, Fe, S,Cl, H, Ti, Gd, K

tManual peak to peak matching or auto-correlative techniques used. |

density resistivity

Core comparison to et best porosity log

Caliper,Porosity

Cable Speed

( NGT Calculation λ

Calculation of wet ILolume percent Th, Uj\^ and K j

IK, Th, U

Caliper,Density,Porosity

Al CalculationCalculation of dry weight percent Al and

converts WPOT to dry weight %

.1.

OxidesMultiply weight percent]

elementsby oxide factors

S1, CA, FE.TI, S,

Al, K (dry wt. %)

Normal izatin

Normalization of WAL &WPOT with the GST

elemental yields to calculateelemental weight fractions,

• estimation of Mg from PEF .

Figure 2. Flow chart of the processing steps involved in arriving at final oxides from raw geochemical logs.

spectrometry tool to determine the Si background correction. Theprogram calculates dry-weight percentages of Al and K, which areused in the calculation and normalization of the remaining elements.

5. Normalization of Elemental Yields from the GSTto Calculate Elemental Weight Fractions

Relative concentrations of the GST-derived elemental yields canbe determined by dividing each elemental yield by a relative spectralsensitivity factor 5,. These factors are constants, which can be mea-sured in the laboratory and are principally related to the thermalneutron-capture cross sections and the gamma-ray production anddetection probabilities of the element (Hertzog et al., 1989). Therelative elemental concentrations are related to the desired absoluteconcentrations by a depth-dependent normalization factor F, as de-fined by the relationship:

where:Wtj = absolute elemental concentration, F, = relative elemental yield,Sj = relative spectral sensitivity factor, and F = depth-dependentnormalization factor.

The normalization factor is calculated on the basis that the sum of theelemental weight fractions is 100%. The closure model accounts forcarbon and oxygen, which are not measured by this tool string, byapproximating their amounts in combination with each of the measur-able elements as a single carbonate or oxide factor. The dry-weight per-centages of Al and K are normalized with the reconstructed elementalyields to determine F at each depth interval with the following equation:

+ XKWtK + XAlWtAl= 100, (3)

Wt( = (2)

where:Xj = oxide factor: atomic wt of oxide or carbonate ÷ atomic wt ofelement i; XK - oxide factor of K: atomic wt of oxide of element K ÷

4?.9

DATA REPORT

Table 1. Oxide factors used in normalizing elements to 100% and con-verting elements to oxides.

Element

SiCa < 67r6%>Ca< 12<7rCa> \2%FeKTiAl

Oxide/carbonate

SiO,CaOCaO and CaCO,CuCO,FeX>,K,OTiO,Al,O_,

Conversion factor

2.1391.3991.399-2.490*2.4901.4301.2051.6681.889

Note: (*) = linearly interpolated

atomic wt of K; WtK - dry-weight % of K measured from NGT; XAl

= oxide factor of Al: atomic wt of oxide of element Al ÷ atomic wt ofAl, and WtAI = dry-weight % of Al measured from the AACT.

The value X; accounts for the C and O associated with eachelement. Table 1 lists the oxide factors used in this calculation. All themeasured elements associate with C and O in a constant ratio in theselithologies, except for Ca, which associates with C and O in one oftwo ways: CaCO3 or CaO (Table 1). To convert the measured yieldsto elements, a dominant oxide factor must be assumed at each depthlevel. A routine that combines both these oxide factors is implementedhere, as suggested by Jim Grau at Schlumberger-Doll Research (pers.comm., 1992). When the elemental form of Ca is less than 6%, CaOis assumed, and an oxide factor of 1.39 is used. When the elementalform of Ca is greater than 12%, CaCO3 is assumed, and an oxidefactor of 2.49 is used. When the elemental form of Ca is between 6%and 12%, both forms are assumed to be present, and the oxide factoris linearly interpolated between 1.39 and 2.49.

The parameters 6% and 12% were chosen according to observa-tions of how Ca occurs in nature. CaO is not likely to occur inquantities greater than 12%, therefor CaCO3 is a logical assumptionin these instances. When Ca is less than 6%, even if CaCO3 wasincorrectly assumed, the error would be very small when 1.39 is usedas the oxide factor instead of 2.49. The linear interpolation is done inorder to provide a smooth transition and avoid invoking any errone-ous character changes on the final processed logs. This procedure forCa gives the best model in most cases and minimizes the error whenthe model is not exactly correct.

The Mg- and Na-content curves cannot be calculated from thelogs, because the neutron-capture cross sections of these elements are

too small relative to their typical abundance for detection by the toolstring; therefore, available core information is included. A constantvalue of 6.38% MgO + Na2O + MnO was used in the normalizationof Hole 863B. This value was derived from the average measuredcore values. No renormalization was performed in Hole 859B toaccount for these oxides, due to lack of representative XRF data at thetime of processing.

6. Calculation of Oxide Percentages

The final routine converts the elemental weight percentages intooxide/carbonate percentages by multiplying each by its associatedoxide factor, as shown in Table 1.

7. Calculation of Error Logs

The calculated statistical uncertainty of each element is calculatedfrom each of the elements measured with the GST and NGT (Grau etal, 1990; Schweitzer et al., 1988). This error is strongly related to thenormalization factor, which is calculated at each depth (Equation 2).The normalization factor is displayed to the right of the oxide logs. Alower normalization factor represents better counting statistics andhigher quality data.

COMPARISON OF GEOCHEMICAL LOGSWITH CORES FROM HOLES 859B AND 863B

Site 859

The processed natural gamma-ray curves for Hole 859B, shownin Figure 3, are displayed adjacent to core-recovery and litho-stratigraphic columns. The processed NGT curves are from the geo-chemical logging tool string. Figure 4 displays the oxide-weightfractions estimated from the logs from Hole 859B, along with calcu-lated statistical uncertainties of each element (Grau et al., 1990;Schweitzer et al., 1988). Core measurements of XRF major elementalanalyses are displayed as solid circles for comparison with the oxide-weight fractions derived from the log data. XRF data are listed inTable 2, representing both shipboard and shore-based data.

The sediments at Site 859 are composed of fine-grained terrigenousclastic material (dominated by glacial rock flour) derived from theadjacent continental (Andean) volcanic arc and crystalline basement.Both the elemental logging data and the XRF data reflect this compo-sitionally homogeneous detrital source.

Table 2. Shipboard and shore-based XRF analyses, Site 859.

Source

ShoreShoreShoreShoreShipShoreShoreShipShoreShipShipShinShoreShoreShoreShipShoreShoreShoreShipShipShoreShoreShore

Hole

859A859A859A859A859 B859B859B859A859A859A859A859B859B859B859B859B859B859B859B859B85915859B859B859B

Core

IH2H4H6XIRIR4R1IX

I4X20XIOR1 IRI4RI9R19R21R22R

36R30R33R38R

Section

1142i

i

1311

cc513Λ

431421411

Depth(mbsf)

0.052.68

21.6136.4553.9454.0158.5071.9078.1087.91

136.16203.22206.94238.70287.12289.22306.16313.26346.80381.57381.57394.50418.92467.58

SiO,

57.7556.4059.8761.6060.6357.9858.5058.5458.9460.6960.3058.7563.0555.8657.8758.4957.9857.3260.9160.7164.7760.3866.8162.34

CaO

5.094.514.714.794.744.435.573.433.413.453.814.352.574.894.814.355.1 14.662.6X5.872.102.951.712.13

Fe2O^

5.897.346.795.897.357 357 327.567.317.398.637.746.297.167.017.897.157.467.087.585.476.725.466.35

A12O,

13.7215.5015.6614.6316.8916.0316.0818.31

18.0718.3917.8415.7415.3315.0717.1915.4615.7116.4716.6715.0716.0215.2916.07

MgO* K2O

6.40 1.757.09 2.117.23 1.949.16 1.713 72 2 056.79 2.14

11.10 1.763.83 2.799.57 2.586.29 2.797.16 2.663.90 2.345.69 2.407.47 2.18

10.70 1.664.41 2.357.98 1.637.76 1.939.81 2.674.09 1.232.73 2.566.33 2.688.04 2.605.47 2.96

TiO2

0.840.950.900.860.880.941.050.860.960.820.950.880.S61.010.920.840.980.930.940.920.650.870.770.87

Notes: The shore-based analyses were done by Amane Waseda. MgO* represents the combination of MgO + NaiMnO.

430

DATA REPORT

Table 3. Shipboard and shore-based XRF analyses, Site 863.

Source

ShoreShoreShipShoreShipShipShipShoreShipShoreShipShipShoreShoreShipShoreShipShoreShipShoreShoreShore

Hole

863A863 A863A863 A863 A863A863 A863 A863A863 A863A863 B863 B863 B863 B863 B863 B863 B863 B863 B863 B863 B

Core

IH5H5H7.X9X17X18X21X24X25 X

4X4X7N9X10R12R17R23 R

31 R49 R

Section

41213

CCcc

1

cc113

CC1151224

Depth(mbsf)

5.3337.8239.7056.4675.52

153.34162.66191.59221.01230.18259.37329.55329.72356.51361.64371.00386.47441.64493.36522.93570.94738.70

SiO2

59.6358.6961.6958.9466.0062.4463.9562.0862.3057.9964.5558.3662.7558.5360.0858.2260.6958.7761.7862.5856.8659.05

CaO

4.483.733.213.543.894.944.973.584.206.834.655.245.135.545.935.835.234.634.764.918.025.54

Fe 2O,

6.747.196.926.655.99

6.266.146.645.746.076.126.116.507.256.147.166.047.035 33

5.68

Al2Oλ

15.3015.0817.5215.3415.0615.7515.2715.0716.4813.8115.4614.6514.2914.5116.4013.9716.3713.9716.2113.4413.0213.14

MgO* 1

7.27 :6.24 :6.33 :6.43 ;6.126.786.506.25 :6.42 ;5.856.26

10.266.246.86

11.136.61

11.026.42

11.145.855.585.75

C2O

..04

..42

..93

..89.89.70.65

..09

..48.72.90.58.51.62.93.59.92.69.61.57.54.48

TiO 2

0.940.900.800.920.750.860.810.880.790.800.730.750.840.900.820̂ 870.840.850.850.760.780.79

Note: The shore-based analyses were done by Amane Waseda. MgO* represents the combination of MgO + Na:O +MnO.

Site 863

The processed natural gamma-ray curves for Hole 863B are dis-played in Figure 5. The through-pipe logs were not corrected, and thespikes seen in the upper 110 m of data are due to pipe joints, notlithologic changes. Open-hole logs begin at 231 mbsf. Processedgeochemical logs for Hole 869B are displayed in Figure 6. Attemptshave been made to correct for through-pipe conditions in the upper200 m of the hole.

The sediments at Site 863 are composed of clayey silt and sand-stones, reflecting a coarser (than Site 859) detrital origin from thenearby continental source. The sandstones below 240 mbsf are verti-cally tilted and cemented sandstone (distal turbidite) unit. Both the coredata and the logging data are thus effectively along-bedding profilesbelow 240 mbsf. The logging data and XRF data (listed in Table 3) arecompositionally homogeneous, reflecting both the detrital source ofthe material and the along-bedding attitude of the profiles.

The logging and XRF major oxide data from Hole 863B show noevidence of the hydrothermal (metasomatic) mineralization (carbon-ate, zeolite, and pyrite cements and veins) observed in the cores belowabout 400 mbsf (Shipboard Scientific Party, 1992b). Mineralizationin the basal portion of Site 863 was likely initiated upon collision ofthe toe of the accretionary prism with the subducting ridge axis, whichoccurred only about 50,000 years ago. Thus insufficient time haselapsed to render a measurable effect on the bulk composition of therocks, although the mineralized veins and crystals are clearly visiblein the cores themselves.

Grau, J.A., Schweitzer, J.S., and Hertzog, R.C., 1990. Statistical uncertaintiesof elemental concentrations extracted from neutron induced gamma-raymeasurements. IEEE Trans. Nucl. Sci., 37:2175-2178.

Hertzog, R., Colson, L., Seeman, B., O'Brien, M., Scott, H., McKeon, D.,Wraight, J., Grau, J., Ellis, D., Schweitzer, J., and Herron, M., 1989.Geochemical logging with spectrometry tools. SPE Form. Eval., 4:153—162.

Lock, G.A., and Hoyer, W.A., 1971. Natural gamma-ray spectral logging. LogAnalyst, 12:3-9.

Pratson, E.L., Broglia, C, and Jarrard, R., 1993. Data report: geochemical welllogs through Cenozoic and Quaternary sediments from Sites 815,817,820,822, and 823. In McKenzie, J.A., Davies, PJ., Palmer-Julson, A., et al.,Proc. ODP, Sci. Results, 133: College Station, TX (Ocean Drilling Pro-gram), 795-817.

Ruckebusch, G., 1983. A Kalman filtering approach to natural gamma rayspectroscopy in well logging. IEEE Trans. Autom. Control, AC-28:372-380.

Schweitzer, J.S., Grau, J.A., and Hertzog, R.C., 1988. Precision and accuracyof short-lived activation measurements for in situ geological analyses. J.Trace Microprobe Techn., 6:437-451.

Scott, H.D., and Smith, M.P., 1973. The aluminum activation log. Log Analyst,14:3-12.

Shipboard Scientific Party, 1992a. Site 859. In Behrmann, J.H., Lewis, S.D.,Musgrave, R.J., et al., Proc. ODP, Init. Repts., 141: College Station, TX(Ocean Drilling Program), 75-157.

, 1992b. Site 863. In Behrmann, J.H., Lewis, S.D., Musgrave, R.J.,et al., Proc. ODP, Init. Repts., 141: College Station, TX (Ocean DrillingProgram), 343-446.

ACKNOWLEDGMENT

We thank Mary Ann Cusimano for providing shipboard XRF data.

REFERENCES*

Grau, J.A., and Schweitzer, J.S., 1989. Elemental concentrations from thermalneutron capture gamma-ray spectra in geological formations. Nucl. Geo-phys., 3:1-9.

Abbreviations for names of organizations and publications in ODP reference lists followthe style given in Chemical Abstracts Service Source Index (published by AmericanChemical Society).

Date of initial receipt: 26 July 1993Date of acceptance: 5 November 1993Ms 141SR-032

431

DATA REPORT

50-

100-

150-

2 0 0 -

2 5 0 -

300-

SPECTRAL GAMMA RAY

I .COMPUTED

|θ API units 1 O o

TOTAL POTASSIUM THORIUM URANIUM

API units 1 O o I 0 wt.% 5 I 0 ppm 10 I 0

<

50

- 100

- 150

-200

- 2 5 0

300

Figure 3. Processed natural gamma-ray data, Hole 859B.

DATA REPORT

SPECTRAL GAMMA RAYCOMPUTED

0 API units 100

TOTAL

300-

3 5 0 -

4 0 0 -

POTASSIUM THORIUM URANIUM

API units 10010

f

5 I 0 ppm 10 I 0 ppm

300

- 3 5 0

- 4 0 0

Figure 3 (continued).

DATA REPORT

CaO-SiO2 CaCO3 Fβ2O3 AL,O3 K2O TiO2

Normalization Q 9Gd Factor | u o l

5θlθ wt% 2010 wt% 7T0 wt% 7J0 ppm 50lθ 10001

50

100-

150-

2 0 0 -

2 5 0 -

300-

50

- 100

- 1 5 0

- 2 0 0

- 2 5 0

300

Figure 4. Estimates of major oxide-weight fractions from geochemical logs, Hole 859B. Solid circles represent XRF measurements (Shipboard Scientific

Party, 1992a).

434

DATA REPORT

SiO2

CaO-CaCO3

F β 2 ° 3 I A I 2 O 3 I K 2 ° T i O 2Normalization 9 <£

Gd Factor ü ü

50I0 wt% 2010 wt% 40I0 wt% 7I0 wt% 7I0 ppm 50I0 1OOθl

300- 300

3 5 0 -

4 0 0 -

450-

-350

-400

450

Figure 4 (continued).

DATA REPORT

SPECTRAL GAMMA RAYCOMPUTED

0 API units 100

TOTAL POTASSIUM THORIUM URANIUM

0 API units 1Oo 10 wt.% 3 I0 ppm 10 I0 ppm

50-

100-

150 -

2 0 0 -

2 5 0 -

300-

X

DRILL PIPEOPEN HOLE

50

- 100

-150

- 2 0 0

- 2 5 0

300

Figure 5. Processed natural gamma-ray data, Hole 863B.

436

DATA REPORT

SPECTRAL GAMMA RAY

COMPUTED

0 API units 100

TOTAL POTASSIUM THORIUM URANIUM

API units 1 O n | 0 wt.% 3 1 0 ppm 10 I 0 Ppm 5

300- 300

3 5 0 -

4 0 0 -

450 H

5 0 0 -

550-

• Λ -•

f

- 3 5 0

- 4 0 0

- 4 5 0

- 5 0 0

550

Figure 5 (continued).

437

DATA REPORT

SPECTRAL GAMMA RAY

.COMPUTED

0 API units 1 O o

TOTAL

550-

6 0 0 -

650-

7 0 0 -

POTASSIUM THORIUM URANIUM

API units 1 O o 10 wt.% 3 I0 ppm 10 iθ PPm 5

550

- 6 0 0

- 6 5 0

- 7 0 0

Figure 5 (continued).

DATA REPORT

oNormalization Q ^

SiO2 |CaO-CaCO3| Fe2O3 AI2O3 K2O TiO2 Gd Factor | ü θ |

**% 5010 wt% 15I0 wl% 501o wt% 2I0 wt% 1 o |o PPm 50lo 10001

50

100-

1 5 0 -

2 0 0 -

2 5 0 -

300-

50

- 100

-150

- 2 0 0

- 2 5 0

300

Figure 6. Estimates of major oxide-weight fractions from geochemical logs, Hole 863B. Solid circles represent XRF measurements, triangles represent carbonate

bomb measurements from core (Shipboard Scientific Party, 1992b).

439

DATA REPORT

SiO2 ICaO-CaCCU Fe2O3 Aip3 K2O TiO,Normalization Q Q

Factor O ü

|θ wt%ioolθ wt% 50I0 wt% 15I0 wt% 50I0 wl% 2 I 0 wt0/" 1θlθ PPm 5010 1OOθl

3 5 0 -

4 0 0 -

4 5 0 -

5 0 0 -

550- I

, J1

>

- 3 5 0

- 4 0 0

- 4 5 0

- 5 0 0

550Figure 6 (continued).

440

DATA REPORT

SIO2 |CaQ-CaCO3| Fe2O3 | AI2O3 | K2O Gd"TIC}

I 0 w t % i o θ l θ wt% 50I0 wt% 15I0 wt% 5010 wt% 2T0 wt% 1 0 | o ppm 50I0

ONormalization ^ ^g

Factor | ü θ |

10001

550-

6 0 0 -

6 5 0 -

7 0 0 -

- 6 0 0

550

- 6 5 0

- 7 0 0

Figure 6 (continued).

44 i