TdL-‘Sls _‘ lllllllllllllllllllllllllllllllll 293 00994 7874 Un. ~......,.. This is to certify that the thesis entitled COMPOSITION OF ISLAND ARC DETRITUS FROM SOUTHWEST JAPAN: PROVENANCE AND TECTONIC HISTORY presented by Julie Marie Taylor has been accepted towards fulfillment of the requirements for MaSter' 5 degree in GEOIOQX Date 6-17-82 0-7639 MS U is an Affirmative Action/Equal Opportunity Institution MSU LIBRARIES .-,—.. RETURNING MATERIALS: Place in book drop to remove this checkout from your record. FINES will be charged if book is returned after the date stamped below. THE COMPOSITION OF ISLAND ARC DETRITUS FROM SOUTHWEST JAPAN: PROVENANCE AND TECTONIC HISTORY By Julie Marie Taylor A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Geology 1982 ABSTRACT THE COMPOSITION OF ISLAND ARC DETRITUS FROM SOUTHWEST JAPAN: PROVENANCE AND TECTONIC HISTORY By Julie Marie Taylor The average composition of sediments from site 298 located in the fore-arc region of S.W. Japan is MQ56 PQ” HF9 LP 21. The most abundant feldspar is albite and the second most abundant is orthoclase (91-100 Wt 96 OR). The most abundant type of plagioclase (other than albite) is in the range of 50-70 Wt 96 AN. This interval reflects the andesitic volcanics associated with island arcs. Mineralogic comparison of these fore-arc sediments to the back-arc sediments of Japan reflects the different provenance of these sediments. Comparison of several fore-arc regions illustrates the importance of the tectonic history of an island arc. Arcs underlain by a "continental" crust may be associated with fore-arc sediments rich in quartz and sedimentary rock fragments. Island arcs underlain by oceanic crust may be associated with quartz-poor fore-arc sediments rich in volcanic Iithic fragments. ACKNOWLEDGEMENTS I would like to thank my husband, Tom, for being so wonderful, encouraging and supportive. I would also like to thank my parents for their life-long encouragement to do what I want to do. Many thanks to my advisor, Duncan Sibley, and committee members, Kazuya Fujita, John Wilband and Tom Vogel for their valued advice. Special thanks to Kaz for the gracious use of his library and to Jay Silber for the use of his programmable calculator. I would also like to thank the AAPG Grant-in—Aid Committee for awarding me financial support for this research. TABLE OF CONTENTS LIST OF FIGURES . . . .......... . . . . . ..... iv LISTOFTABLES.... ......... .......... v INTRODUCTION........................l GEOLOGICSETTING...................... 3 METHODS...........................ll PETROGRAPHIC ANALYSIS OF THE FORE-ARC SEDIMENTS . . . . l4 MICROPROBE ANALYSIS OF THE FORE-ARC SEDIMENTS ..... l9 PROVENANCE OF THE FORE-ARC SEDIMENTS . . . . . . . . . . 21} COMPARISON OF THE FORE-ARC AND BACK-ARC BASINS . . . . 29 THE IMPORTANCE OF ISLAND ARC CRUSTAL TYPES ....... 31} SUMMARY 0 I ....... O O O O O O O O O O O O ...... 38 BIBLIOGRAPHY ........... . ............ #0 iii Figure 1. Figure 2. Figure 3. Figure ‘5. Figure 5. Figure 6. Figure 7. LIST OF FIGURES Locations of D.S.D.P. sites 298, 299 and 301 . Bathymetric surface of the Shikoku slope . . Geotectonic division of Japan (Pre-Neogene). Geologic sketch map of Japan . . . . . . . Grain mount from site 298 . . . . . . . . Histogram of feldspar compositions in the fore-arcsediments. . . . . . . . . . . . Location of Quaternary volcanoes in Japan iv IO 17 23 Table 1. Table 2. Table 3. Table 4. Table 5. Table 6. Table 7. Table 8. Table 9. LIST OF TABLES Mineral loss during fusion. ..... . ......... Petrographic analyses . . . . . . . . . . . . . . . Analysis of the > .0625 mm size fraction in silty claysandclayeysilts . . . . . . . . . . . . . . . Microprobeanalyses. . . . . . . . . . . . . . . . Distribution of feldspar compositions in the fore-arc . . Fore-arc mineral proportions . . . . . . . . . . . . Occurrence of feldspar compositions in the fore-arc and baCk-ar C baSInS o o o o o o o o o o o o o o o 0 Mineral proportions in the fore-arc and back-arc basins Mineral proportions of various fore-arc sediments . . . 12 16 18 21 25 28 32 33 37 INTRODUCTION The variation of sediment compositions with tectonic setting has received increased attention in recent years. Particularly important to this paper is the composition of detritus associated with island arcs. Crook (1974) suggested island arc detritus will consist of quartz-poor graywackes reflecting a basic volcanic and ultramafic provenance. Dickinson and Valloni (1980) and Valloni and Maynard (1981) have presented data on island .arc sediment compositions from various island arcs showing the sediments to be low in quartz with plagioclase as the predominant feldspar. This paper provides a petrographic analysis of hemipelagic muds from DSDP site 298 located in the fore-arc region of Japan. The composition of these sediments does not fit the models cited above. These sediments are not primarily volcanic in composition yet are derived from an active volcanic arc. Most studies of clastic sediments have utilized the sand size fraction of a sediment for analysis. However, mudrocks are the most abundant rock type in the sedimentary record (Picard, 1971). The advantages of analyzing sands are the ease of mineral identification and the ability to study rock fragments. The analysis of muds requires special sample preparation and the study of rock fragments in these sediments is very limited, if possible at all. However, analysis of muds may often provide a more complete analysis of provenance than sands. Analysis of only coarse grained material in a sedimentary sequence composed mostly of fine grained sediments may provide biased results. Deep sea samples from site 298 consist predominantly of clayey silt (stone) and silty clay (stone). The coarse silt fraction of sediments from the fore-arc basin of S.W. Japan are studied here to further the investigation of the provenance of muds and mudrocks and the relation between sediment composition and tectonic setting. GEOLOGIC SETTING Samples from the fore-arc region of S.W. Japan were taken on Leg 31 of the Deep Sea Drilling Project from site 298 (Karig and Ingle, et al., 1975). The samples are Quaternary in age and consist of interbedded hemipelagic muds and turbidites deposited on the lower, inner slope of the Nankai Trough off Shikoku Island (Figuresl and 2). The term fore-arc region is used in this paper to describe the oceanic side of the arc; between the arc proper and the trench. Sediment deposition at site 298 is affected by several geomorphological features on the Shikoku slope. Sediment accumulates on benches and in old submarine channels along the slope (Hilde and others, 1969). Active submarine channels off Shikoku transport some sediment to the trough, but much of the sediment is trapped behind the outer edge of the Tosa Terrace (Figure 2). A major input of sediment into the Nankai trough is through the Suruga Submarine Canyon on the northeastern end of the trough. Sediment flows through the Suruga Submarine Canyon down the axis of the trough to the southwestern end of the trough (Hilde and others, 1969). The Median Tectonic Line provides the major drainage divide on the southwestern islands of Japan (Figure 3). The divide provides the northern limit on the source area supplying detritus to the fore-arc region. Rivers flowing from the divide to the fore-arc region travel through a variety of rock types. The terrain consists of sedimentary, metamorphic and intermediate and basic intrusive and extrusive rock types (Figure 11). However, the metasediments of the Shimanto belt cover the largest area of the source region. It is predominant on the trough side of the divide down the length of the trough (Figures 3 and 4). The Shimanto belt is a eugeosynclinal belt and is described by Matsuda and Uyeda (1971) as: a thick sequence of turbidite graywacke and shale, interbedded irregularly with conglomerate, chert, submarine basaltic lavas and tuffs. Ultramafic intrusions are also found. These rocks...are regionally metamorphosed up to the greenschist facies through the pumpellyite-prehnite metagraywacke facies. North of the Shimanto belt are the Sambosan, Chichibu, and Sambagawa belts (Figures 3 and it). These belts also consist of eugeosynclinal sediments. The Sambosan belt is characterized by Middle Permian to early Jurassic limestones, cherts and mafic volcanics. The Chichibu belt consists largely of weakly metamorphosed upper Paleozoic elastic rocks, cherts, limestone: and mafic volcanic rocks. The Sambagawa belt is now a high pressure, low temperature metamorphic belt characterized by crystalline schist. Quaternary volcanoes and Miocene volcanics are located on Honshu near the Suruga Submarine Canyon. Quaternary volcanoes are also located on Kyushu (T anaka and Nozawa, 1977). HOKKAIDO PACIFIC OCEAN ' O 100 200 Ian w Figure 1. Locations of D.S.D.P. sites 298, 299 and 301. Figure 2. Bathymetric surface of the Shikoku slope. Locations of the Tosa Terrace, Suruga Submarine Canyon and Nankai Trough included (after Hilde and others, 1969). .59 .om. Mm. a: . . Em: wismNU f U0<¢¢Wh <90» [an 9 . .u 1.3 \ s x ‘ v - n o N. 2923 We. misiaaw «oazam .6? .mm. .m. cm. a? .v? a? a? .zr Figure 2. Figure 3. Geotectonic division of Japan (Pre-Neogene) (after Tanaka and Nozawa, 1977). SOUTHWEST JAPAN INNER SIDE 1. Hida Belt 2. Sangun Belt 25. Maizuru Structural Belt 2'. Hida Marginal Belt 2". J oetsu Belt 3. Tamba Belt 3'. Ashio Belt 4. Ryoke Belt OUTER SIDE 5. Sambagawa Belt 6. Chichibu Belt 7. Sambosan Belt 8. Shimanto Belt NORTHWEST KYUSHU 9. Ainoshima Belt 10. Nishisonogi Belt 11. Tsushima Belt NORTHEAST JAPAN 12. Abukuma Belt 13. South Kitakami Belt 13h. Hayachine Structural Belt 14. North Kitakami Belt 15. Iwaizumi Belt 16. Taro Belt CENTRAL HOKKAIDO 17. Ishikari Belt 18. Kamuikotan Belt 19. Hidaka Belt 20. Tokoro Belt EAST HOKKAIDO 21. Nemuro Belt SEA OF JAPAN 5 Tanakura Tectonic Lino I ,ta: SHIIIDIIU Median Tectonic Llno Butsuzo ‘lbctonlc Line PACIFIC OCEA N [-0 100 . _390 km Figure 3. lO KYUSHU 0 100 200k!!! Figure it. Geologic sketch map of Japan (after Tanaka and Nozawa, 1977). ' Early Neogene volcanic terranes ("Green Tuff" region). —" Main Paleozoic and Mesozoic sedimentary terranes. 18 Main metamorphic terranes. METHODS Thirty-two samples were examined from site 298. These samples were taken from all levels of the core and range from clayey silt (stone) to silty clay (stone) with < 5% sand. The samples were processed in order to allow identification of quartz and feldspar grains. The sample preparation techniques are applicable to well-indurated mudrocks to allow comparative studies with ancient rocks. The samples were first broken up by hand and ultrasonically cleaned. Each sample was then heated in alternating solutions of .5 m NaOH and 10% HCl until warm. This served to remove organics, iron oxides and carbonates. After the sample was dried, it was boiled in potassium hydrogen sulfate for 20 minutes to remove phyllosilicates (Blatt and Schultz, 1976). Upon cooling, each sample was boiled for 5-10 minutes in 10% HCI to remove remaining potassium hydrogen sulfate and iron oxides liberated by the pyrosulfate fusion. Finally, the sample was brought to a boil in .5 m NaOH to remove any silica wreckage resulting from the breakdown of the clays. The residual grains were quartz and feldspar plus a few surviving rock fragments. The samples were wet sieved through 270 mesh and 325 mesh screens. Measurement of the long axis of grains with the microsc0pe revealed the size range to vary from .031 mm to .125 mm (coarse silt to very fine sand) although most grains were in the .044 mm to .053 mm (coarse silt) range. A test of this process was made on an artificial sample consisting of coarse silt to very fine sand sized grains of quartz, plagioclase and alkali feldspar. The composition of this sample before processing was determined by point counting. Three hundred grains were counted resulting in 26% monocrystalline ll 12 quartz (MQ), 16% high feldspar (HF, n > 1.54) and 58% low feldspar (LP, n < 1.54). The composition of the sample after processing was found to be MQ30 HF 14 LF56. A chi-square test was performed on the data resulting in a x2 value of 1.58. The critical value of x2 is 5.99, a = .05. Therefore, a significant difference between the processed and unprocessed sample cannot be shown at the 95% level of confidence. Jones (1979) determined the effects of pyrosulfate fusion on quartz and feldspar in the coarse silt fraction. The results of his work are listed in Table 1. The largest loss is found in the bytownite samples, which in this study amounts to a maximum loss of 1% of the total feldSpar. Table 1. Mineral Loss During Fusion. Mineral Weight% Lost During Fusion Quartz .8 Microcline 2.5 Albite 3. 5 Oligoclase 4.7 Andesine 1 1 . 7 B ytownite 23 . 6 Grain mounts of the 32 samples were examined by the ribbon method (Galehouse, 1969) on the petrographic microscope. Approximately 300 grains were identified on each slide in order to determine the percentage of MQ, PQ (polycrystalline quartz, including chert), HF and LF in each sample. An average of 25 feldspar grains were analyzed from each grain mount on the electron microprobe. Feldspars were chosen by the ribbon method to insure randomization (Galehouse, 1969). Counts for K, Ca, and Na were simultaneously 13 recorded at 20 nanoamps for 10 seconds. Readings were taken at least three times on each feldspar grain. Weight percent anorthite, albite and orthoclase were determined in the following manner (after Bence and Albee, 1968): ut% AN = Ca CPSU Ca Ca CPSAN Std ° BAN Std Ca Na K x 100 cps cps CPS u + u + u Ca Ca Na Na K K CPSAN Std ' BAN Std cPSAB Std ' BAB Std CPSOR Std ° B0R Std utz OR = K CPSU K K CPSOR Std ° B0R Std x 100 cpsK CPSca cpsNa ___11 + u + u K K Ca Ca Na Na CPSOR Std ° B0R Std cPSAN Std ° BAN Std cPSAB Std ' BAB Std utz AB = Na cpsU Na Na CPSAB Std ' BA3 Std N Ca K X 100 CPS a CPS cps u + u + u Na Na Ca Ca K K CPSAN Std ° BA3 Std CPSAN Std ' 8AN Std CPSOR Std ' Boa Std CPS = Counts per second U = Unknown 8 = Correction factors (Sweatmen and Long, 1969) PETROGRAPHIC ANALYSIS OF THE FORE-ARC SEDIMENTS The average composition of the samples examined from site 298 is MQ 56 (s = 9.33) PQ14 (s = 7.64) HF9 (s = 3.06) LF21 (s = 5.59). A complete list of this data is included in Table 2. Most of the polycrystalline quartz is chert (microcrystalline and cryptocrystalline quartz, shown in Figure 5). Trace amounts of felsite were noted in some samples and it is possible that some felsite may have been recorded as chert (Pettijohn, et al., 1973; Wolf, 1971; Dickinson, 1970; Blatt, 1967). Felsite and chert were distinguished by the presence of feldspars or glass shards in felsite when visible. The feldspars were rarely twinned or zoned in this size fraction (Figure 5). Therefore, these characteristics were not often utilized in determining composition or provenance. Feldspar grains of every composition often contain numerous inclusions. Thus, the presence of cloudy albite grains does not unequivocally indicate albitization has occurred. Rock fragments were not analyzed because they are too difficult to identify in the silt size fraction and many rock fragments may be destroyed during pyrosulfate fusion. Harrold and Moore (1975) examined seven samples from sand units at site 298. They found a Q (total quartz)/F (total feldspar) ratio of .94 for the > .0625 mm size fraction. These sands range in size from very fine sands to coarse sands. The Q/F ratio determined in this study for the coarse silt fraction of the silty clay and the clayey silt units is 2.30. The > .0625 mm size fraction was examined in three samples from silty clay and clayey silt units (Table 3). These sands range in size from very fine to fine sand. The Q/F ratio of these sands is 2.05. The difference in the quartz to feldspar ratios of the sand and silt l4 15 units may be due to the abundance of sedimentary rock fragments in the sand size fraction. Sedimentary rock fragments composed of quartz and feldspar in an argillaceous matrix comprise 41% of the sand units. Natural disaggregation of these sedimentary rock fragments will provide an increase of quartz and feldspar grains in a finer grain size. It appears that the additional quartz found in the silty units may be bound in sedimentary rock fragments in the sand units. Disaggregation of sand sized sedimentary rock fragments in the silty units by pyrosulfate fusion may provide some additional silt sized quartz and feldspar. However, the sand fraction of the silty units is < 5%. It does not appear that the disaggregation of sand sized sedimentary rock fragments by pyrosulfate fusion techniques would provide the volume of silt sized quartz needed to account for the difference in the Q/F ratios of the sand and silt units. 16 Table 2. Petrographic Analyses. Sample Number of Grains Counted Age Core Section Interval MQ PQ HF LF Pleistocene- Holocene 2 2 48-52 196 10 40 54 2 2 92-94 191 20 33 56 2 100-101 158 55 31 56 Late Pleistocene 4 1 68-70 169 35 27 63 4 2 99-101 167 19 44 71 4 2 100-101 156 37 23 84 5 1 75-76 209 28 19 44 5 1 103-105 189 43 22 45 7 1 52-54 125 103 30 42 7 1 100-101 193 16 45 46 7 2 100-102 144 39 40 77 Early Pleistocene 8 2 50-52 123 49 33 88 8 2 96-98 125 73 38 64 8 2 100-101 132 69 26 73 10 1 48-50 144 64 20 69 10 2 30-31 172 45 25 58 10 2 52-54 231 6 20 43 10 3 100-102 184 10 43 64 11 2 100-101 131 83 14 72 11 3 100-101 140 56 19 85 ll 4 100-101 171 27 41 61 12 2 100-101 135 74 20 71 12 3 100-101 173 38 14 75 12 4 100-101 137 33 26 104 13 2 100-101 167 24 17 92 13 3 100-101 199 20 16 65 13 5 100-101 168 54 36 42 14 2 100-101 207 35 24 34 14 4 100-101 159 59 27 55 15 3 100-101 169 37 26 68 16 2 40-41 166 58 23 53 16 5 100-101 201 33 22 44 MQ = monocrystalline quartz PQ = polycrystalline quartz and chert HF = high feldspar, n > 1.54 LF = low feldspar, n < 1.54 Figure 5. Grain mount from site 298 (width of field is .17 mm). 18 Table 3. Analysis of the > .0625 mm size fraction in silty clays and clayey silts. Sample MQ PQ HF LF 2-2-100-101 131* 63 46 60 5-1-103-105 144 53 32 71 14-2-100-101 151 61 32 56 *N umber of grains counted out of 300. MICROPROBE ANALYSIS OF THE FORE-ARC SEDIMENTS Microprobe analysis of 860 feldspars from site 298 revealed the following pr0portions of feldspar types: 35% plagioclase (minus albite), 27% albite, 35% alkali feldspar and 3% ternary feldspar. Ternary feldspars are defined as feldSpars with > 50 Wt% AB > 10 Wt% AN and > 10 Wt% OR. Albite is defined as feldspars with < 10 Wt% AN and < 10 Wt% OR. A complete list of this data is provided in Table 4. A histogram of the frequency of feldspar types occurring in the sediments from site 298 is shown in Figure 6. Compositional groups in the histogram were chosen according to the plagioclase series and this was extended to the alkali feldspars for convenience and clarity. The histogram shows highest frequencies for albite and feldspars with 91-100 Wt% OR and lowest frequencies for feldSpars with 91-100 Wt% AN and 31-50 Wt% OR. The most abundant type of plagioclase (other than albite) is in the range of AN 50_70. This interval reflects the calc-alkaline volcanic series associated with island arcs. In order to compare microprobe and microscope feldspar determinations, all alkali feldspars and plagioclase with < 22 Wt% AN were grouped as low feldspar grains. All plagioclase grains with > 22 Wt% AN were grouped as high feldspar grains. A chi-square test was performed on the data for each sample. Yate's correction was applied for the 2 x 2 probabilities. For samples which contained cells with expected numbers < 5 an approximation of the chi-square test was used (Mainland, et al., 1965). Significant difference between micrOprobe and microsc0pe feldspar analyses were found in only two samples; 11-2 (100-101) and 14-4 (100-101). The amount of HF identified with the micr0probe is higher than expected in both 19 20 samples. The data obtained by use of the microsc0pe is consistent with data for other samples from the site. Therefore, the microprobe data for these samples have been left out of all analyses in this study. 21 OO-d—t—t—iON—lo—lo—to nNa-Nmnmmonam—«xo ~mo~~n~~oo~~o~ N-«No—INNNOONHOO o—«oo—tmooo—t—«oo—t a—tN—IMMW—t—‘Jficd'fil #N—‘O—INNON—‘H—‘Né o—r—to—toooo—cmooo \DNN—iONOO—IM—dom—s nmneamwmmmnmno IAN—l—q—lNNo—iNM—Id'd' “WNMd'G—‘Nonffiéfi—d NB—‘d'NklfiMlfima'MMO‘ fiodlooa mauve anon Nofiloofi uofiloofi anan nofilnou mulnn fiofilood fiodlmm oulww mofiloofi emlmm Nut»: NNNfiNNH—Ifi—‘NNNN NNNd-d'd'hlfikhkoooow ooT—m emu: out; out; emu: oTo can; can: out: our: emu: m0 8“? 95560 meflmeo mo conEaz oTo Z< R33 328.:— anmm 0.60 vac—mm 609292 30.59.22 .e o—nmh 22 .z< ea; co mo eat» me dead 5 8282 $553.3 or; z< ea; n 55 a8_ Ea mo *5 n 55 m8_ 53 mascot o 0 ~ g o a n o g a m a w _o~loo~ n m— N m d o o a N o o Q ~ — n fierce N w~ — n N u o N ~ g N ~ ~ n N fiofiloofi m n— o a o ~ o m c 0 ~ a d n - fio~loo— N ¢~ o e N ~ o d a o o N o N N —c~looq w m~ N ~ o o o — o ~ a N a a m~ qofilood m m~ g m N o o m N g d o n N N ~o_loo~ N n~ o m n m d a o 0 ~ w N m w ~o~rco~ e N~ o m ~ 9 d e N o A e n N N financed m N~ o e d N o n n o o N d _ m ~o~loo~ N Nd a w _ g o m N o m m m d w fiofiloo_ e ~— ~ w n o — N N m o N N N n do—lood m flu ~ N N N o o N fl Q fl 0 ~ ~ No~loo~ m o~ — e o o o _ o o d N ~ ~ m enan N o~ ~ e N o d m N o m a g A m ~Mlom N o— — N H o o N fl 0 o g n a N owl»: H ed H can; can: Othn onlan emu: Sic coal; can: ONlR can: emu: 07o 3385 035mm 0.80 m0 893 Z< $93 emu—Sou 38.5 no .5955 oEEmm .Aeoscflcouv momNEc< mack—00:2 .e 2an 23 20-5 14.11.. 22-; 20.- 10-. lo-_ 14__, PERCENT PILDSPAI 104.. l-u- ..l- ..._. Cfin- _l 0.1 , ossooosozosaoaososoesr AN OR Figure 6. Histogram of feldspar compositions in the fore-arc sediments. The median values of the class intervals are listed on the X-axis. Class interval T represents Ternary feldspars. PROVENANCE OF THE FORE-ARC SEDIMENTS The provenance of site 298 sediments can be inferred from the feldSpar composition, Q/F, HF/F and PQ/Q ratios of the sediment. The range of feldspar compositions indicates a diverse source region of acid and basic rocks (Table 5). Approximately 15% of the feldspars are ternary feldspars or have a composition 1 l-OR5O which indicates a volcanic source (T revena and Nash, 1981). Nineteen percent of the feldspars have a composition between AN between OR 31-70 which reflects the calc-alkaline nature of the volcanic arc. The fourteen percent of the sample having a composition of OR indicates the presence 91-100 of plutonic or metamorphic rocks in the source region. Examination of Figure 4 shows a variety of probable sources for the fore- arc sediments. The Shimanto belt and Quaternary volcanoes within it (Figure 7) may provide volcanic input, the Shimanto belt and Sambagawa belts may provide a metamorphic source, and ultramafic intrusions and granitic plutons occurring in the Shimanto and Chichibu belts may provide a plutonic source. The predominance of the metasediments of the Shimanto belt in the source areas for the detritus in the fore-arc region may explain the relative abundance of quartz to feldspar and low feldspar to high feldspar in the sediments at site 298 (Table 6). Detritus enriched in monocrystalline quartz relative to feldspar and low feldspar relative to high feldspar would be expected from a sedimentary source because quartz is more stable than feldspar and alkali feldspars are more stable than Ca- plagioclase. 24 25 Table 5. Distribution of feldSpar compositions in the fore-arc. Composition Percentage of Feldspar Grains AN 91-100 AN 71-90 AN 51-70 11 AN 31-50 8 AN 11-30 9 AN 0-10 and 27 OR 0-10 OR 11-30 10.5 OR 31-50 1.5 OR 51-70 3.5 OR 71-90 5.5 OR 91-100 14 TERNARY 3 26 Figure 7. Location of Quaternary volcanoes in Japan. c Tholeiitic basalt ° Pyroxene andesite it Pyroxene andesite and associated hornblende andesite u Hornblende andesite with or without biotite phenocryst O Dacite or rhyolite + Alkali basalt and/ or other alkali rocks c Pyroclastic flow deposit 0 Caldera whose diameter is larger than 10 km 27 Figure 7. 28 Table 6. Fore-arc mineral proportions. Q/F MQ/F .-. 1.84 PQ/Q: .20 2.30 HF/ F = .30 The Q/F and PQ/Q ratios found for the fore-arc sediments fit the "tectonic" provenance model for subquartzose sandstones suggested by Dickinson (1970). Sands indicative of a "tectonic", mixed provenance, as described by Dickinson (1970) will have a Q/F ratio of up to 2, including a significant amount of chert. Dickinson (1970) writes: The source rock are mainly chert and sedimentary or metasedimentary strata of argillite, slate, phyllite, etc., with variable amounts of intercalated volcanic and metavolcanic strata...derived from the erosional destruction of volcano-plutonic orogens or magmatic arcs and deposited in adjacent oceanic trenches and associated troughs (Ojakangas, 1968; Dickinson, 1970). Thus, the tectonic provenance model suggested by Dickinson (1970) seems to well fit the fore-arc sediment composition and its most likely source, the Shimanto belt. Sugimura and Uyeda (1973) provide a summary of papers on the source area of the eugeosynclinal sediments of the Shimanto belt. It is suggested on the basis of sedimentary structures and conglomerates within the sediment that there was once a land source with a continental crust to the south of the Shimanto belt. The fore-arc sediments at site 298 appear to be multicycle with a source consisting, in part, of eroded sialic continental crust. COMPARISON OF THE FORE-ARC AND BACK-ARC BASINS Sibley and Pentony (1978) examined back-arc sediments from DSDP sites 299 and 301 in the Sea of Japan (Figure l). The difference between the composition of Quaternary sediments from sites 299 and 301 are very small compared to the differences between these sites and site 298 from the fore-arc area. Therefore, the data from sites 299 and 301 have been combined to make comparisons between the fore-arc and back-arc basins. The average composition of Quaternary back-arc basin samples of the very fine sand to coarse silt size fraction was determined as MQ41 HF LF35 (Sibley and Pentony, 1978). Only 24 trace amounts of polycrystalline quartz or chert were found in these samples. Comparison of the back-arc mineralogy to the fore-arc sediment composition of MQ56 PQ14 HF9 LF shows striking contrast. Statistical 21 comparison of the occurrence of MQ, HF, and LP in the two basins by x2 showed a significant difference between the mineralogy of the two basins 2 = 1114.33, (1 = .05). The percentages of MQ and Q are significantly higher in (x the fore-arc sediments than the back-arc sediments by t tests (a = .05). The percentages of HF and LF are significantly lower in the fore-arc sediments than in the back-arc sediments by t tests ( = .05). Table 7 lists the percentage of feldspar types found in the samples from the fore-arc and back-arc basins. All of the data is restricted to Quaternary samples only, as sites 299 and 301 penetrated Tertiary sediments and site 298 did not. Sibley and Pentony (1978) did not analyze "dirty" feldspars with the microprobe. These feldspars have an albite composition. Thus the number of albite grains analyzed with the micr0probe does not indicate the total 29 30 percentage of albite grains in the back-arc basin. Therefore, samples with the composition AND-10 or ORG-10 are not included in the data listed in Table 6. Significant differences were found by t tests in every category between the two basins except in the OR91-100 and AN7 1-100 categories (Table 7). The most outstanding contrast between the basins is the relative enrichment of plagioclase in the back-arc basin. The most abundant type of feldspar in the back-arc basin listed in Table 7 is in the compositional range of AN31_70. This composition reflects the caIc-alkaline volcanic series associated with the island arc. Sibley and Pentony (1978) showed the islands of Japan to be a common source area for the back-arc sites within the Asian continent supplying additional detritus to site 301. The most likely source area for the bulk of detritus in the region of sites 299 and 301 is N.E. Honshu and S.W. Hokkiado. This area is known as the "Green T uff' region (Figure 4) and is characterized by volcanic effusives of Miocene age (Tanaka and Nozawa, 1977). Quaternary volcanics have continued to be distributed in this region and also in S.W. Hokkaido (Figure 7). Plutonic, metamorphic, and volcanic rocks are present in the source areas for both basins. However, the relative abundance of metasediments in the fore- arc source region and the extensive volcanics in the back-arc source region appear to explain the major differences in the mineralogy of the fore-arc and back-arc basin sediments. As shown in Table 8, the fore-arc sediments have larger Q/F and MQ/F ratios than the back-arc basin sediments. The fore-arc detritus also has a much lower HF/LF ratio than the back-arc basin sediments. This corresponds to electron microprobe analyses indicating a relative enrichment of plagioclase feldspars in the back-arc basin and of K-spars in the fore-arc. The abundance of plagioclase feldSpars in the range of AN30-70 in the back-arc basin reflects the predominantly volcanic nature of the source for these 31 sediments. The abundance of feldspars of varying compositions in the fore-arc sediments reflects the mixed provenance of these sediments. The mineralogical differences between the fore-arc and back-arc basin sediments may be largely attributed to the metasediments occurring in the Shimanto belt in the fore-arc source region and the Green Tuff and Quaternary volcanics occurring in the back-arc source region. Both source regions have, from time to time, been partially overlain by marine sediments, but it appears that the lower strata account for the major mineralogical differences between the fore-arc and back- arc detritus. 32 Table 7. Occurrence of feldspar compositions in the fore-arc and back-arc basins. Feldspar Composition Percent of Total Feldspar T Test Results (Minus Albite) Fore-arc Back-arcl 0R9l-100 20 21 N.S. 0R51_90 12 4 S.D. OR 1 1_ 50 17 1 S.D. AN 1 1-30 12 18 S.D. AN 3 140 26 49 S.D. ANN-100 9 6 N.S. TERNARY 4 l S.D. 1Data from Sibley and Pentony, 1978 33 Table 8. Mineral pr0portions in the fore-arc and back-arc basins. 1 Mineral PrOportions Fore-arc Back-arc T Test Results Q/F 2.30 .70 S.D. MQ/F 1.84 .70 S.D. PQ/Q .20 - S.D. HF/LF .44 .70 S.D. HF/F .30 .42 S.D. 1Data from Sibley and Pentony, 1978 THE IMPORTANCE OF ISLAND ARC CRUSTAL TYPES Comparison of fore-arc detritus from island arcs underlain by "continental" and oceanic crust types are made with the limited amount of data presently available. The following comparisons are based on samples of varying grain Sizes. Some petrographic data is available from the silt and sand size fractions from all sites, except the fore-arc region of Sumatra. Due to the varying amounts of petrographic data available on the silt to coarse sand fractions direct comparison of petrographic data values may not provide exacting compositional limits on fore-arc detritus. However, based on the limited amount of data available, the provenance of sediments at each site does not appear to vary with grain size. Petrographic data from the fore-arc regions of Puerto Rico (Breyer and Ehlmann, 1981) and the central Aleutians (Stewart, 1978; Scholl, et al., 1971) reflects the volcanic provenance for the sediments in both of these regions. Petrographic data from the fore-arc regions of S.W. Japan and Sumatra (Moore, 1979) do not reflect a primarily volcanic source. Comparison of the provenance of these fore-arc sediments provides insight as to the nature of the mineralogic differences seen in the fore-arc detritus derived from magmatic arcs. The volcanics of S.W. Japan, Sumatra, Puerto Rico, and the central Aleutians (Andreanof Islands) are predominantly calc-alkaline (Windley, 1977; Katili, 1975; Mattson, 1977; Coats, 1952). Windley (1977) utilizes the predominant volcanic series on an island arc as a measure of the relative maturity of an arc. The maturity of an arc increases from the tholeitic series to 34 35 the calc-alkaline to the alkaline series. By this definition, all four arcs are at a similar stage of evolution. Petrographic data from the fore-arc sediments of S.W. Japan, Sumatra, Puerto Rico and the central Aleutians is shown in Table 9. Considerable variation occurs between all four sites. Variability in P/F ratios may be due to the alteration of feldspars. Moore (1979) suggests original K-spar in melange sediments off of Sumatra may have been altered by reaction with seawater. Altered feldspars were noted at most sities. In general, the fore-arc sediments of S.W. Japan and Sumatra are enriched in quartz and sedimentary- metasedirnentary rock fragments compared to the fore-arc sediments of Puerto Rico and the central Aleutians. A correlation between fore-arc detritus mineralogy and the nature of the crust underlying the adjacent island arc is found for these island arcs. Sumatra (Hamilton, 1977; Katili, 1975) and S.W. Japan (Sugimura, 1968) are underlain by sialic crust of continental thicknesses > 20 km. Puerto Rico (Donnelly, 1964) and the central Aleutians (Coats, 1962) are underlain by nonsialic oceanic type crust of < 20 km. Thus, there is a correlation between the occurrence of relatively quartz rich detritus with abundant sedimentary rock fragments in fore-arc regions and the presence of sialic crust underlying the adjacent island arc. The source areas for sediments from the fore-arc regions of Sumatra (Moore, 1979) and S.W. Japan are characterized by the presence of metasediments and the exposure of metamorphic, plutonic, and volcanic rocks on the magmatic and frontal arcs of these arc systems. The source areas of the fore-arc sediments off Puerto Rico and the central Aleutians are primarily volcanic and from the adjacent arc (Stewart, 1978; Breyer and Ehlmann, 1981). Puerto Rico and the central Aleutians, although having the same volcanic 36 maturity as S.W. Japan and Sumatra, do not have such exposures of metasediments and metamorphic rocks (Coats, 1962; Breyer and Ehlmann, 1981). It is possible that the uplift, exposure, and erosion of a sialic "continental" type crust accounts for the mineralogy of the fore-arc detritus in S.W. Japan and Sumatra. Sugimura and Uyeda (1973) provide a summary of papers on the source area of the eugeosynclinal sediments of the Shimanto belt. It is suggested on the basis of sedimentary structures and conglomerates within the sediment that there was once a land source with a continental crust to the south of the Shimanto belt. The sedimentary/metasedimentary source rocks on Sumatra are part of a Tertiary arc complex. Moore (personal communication, 1982) suggests that the metasediments exposed on Sumatra were derived from the exposed plutonic roots of an earlier arc system and from the erosion of the Malaysian continent. The fore-arc basin sediments of Japan and Sumatra appear to be multicycle with an original source consisting, in part, of an uplifted, exposed and eroded "continental" crust. The lack of such a crust under Puerto Rico and the central Aleutians may account for the differences seen in fore-arc sediments. 37 .8322: A33 .ccmEEm 0cm 0.59.8 0.0 0:03 05 co 00.59. 33:05:08 0 0300205 no: on can mama—0 973.. .02: 0.3 35633 0.09. Neon—05:03 805 no ~82 A8 .om. fl 3:06:03. 085 e3 030.. LEE 05. .mm. «o 62.55:. m «c3050.. 53 06.0.. "tn 05 .Z< we“? 3 A :23 flame—3 3 00:30.0 0.3 05mb 0.0202me E .Nw. «o 033 83:58:. 0 3:00 :3 030.. "—E 05 :05 .mO *3? 2 v 53 30333 am 00500 0.3 v.50.» 03—8mwfla 3 .338 LE EEEEE “Em 82.5me 035.5 3 00m: :03 0>0c 395903. c0300? 05 :33 005.5300 meoEmon—Eoo .8333 .320.» 5.3.0 53 Eafimmuoe no BenoEm 20:5 53 maficw 0:21.. a .8205. ed 2 6033503 9::me .3 00550300 0.03 033 05 E .003: 330.. «in .053 =< .3059.qu 5300—0 05 :33 005,533 mm? enema .B.m mo 3:08:03 0.8.33 05 no 35.. Em 05. 60:02:03 =00?»ch 5 305.3320 no 03003 033 £5 S 003: 330.. "(a .058 05 3 0333500 3300.50 0n no: .35 :mamn .B.m mo cowwe 0.3.9.3 05 603 3:» «o own. "in 05. A5 .MNE .40 “0 .0330 ea... :2de £2 .53me .32 55520 cam been a .32 .28: a 53 .28: Ba Beam 2 so: can .<.z .<.z .<.z .<.z .<.z 8. 3 Ba... .<.z .<.z .<.z .<.z .<.z a. s accuser? .<.z .<.z .<.z .<.z .<.z nm. N n50 Eu 8. ea. R. 3. we. 2. R 23 ”5082 .8200 Eva. R. 3. ea. an. S. 2 mesa :0. 82 ocean 09:30.5 R. a. ma. 2; so. he E mesa 83:85 27.3 803 ea. R. 3. ”we 8. Rs 2 «came 5285 235% we. R. N». .<.z mm. a. e secs 358 t 0:3 50> .<.z .<.z 3 £4 8. 84.. NM Ed 838 5am.” $3, .5 d> "E "to: 90a ":0 835% Sudan 05 8:83 mo .38: Z .mpc0EMv0m 03:83 got? Ho meoflcoaoce 305.2 .m 033. SUMMARY Comparison of the fore-arc basin detritus of S.W. Japan to back-arc basin detritus from the Sea of Japan and other fore-arc sediments exemplifies the importance of provenance and tectonic history in the analysis of island arc detritus. Provenance proved to be particularly important in the comparison of the fore-arc and back-arc detritus of Japan. Here, the source area for back-arc detritus is a volcanically active area overlying an unexposed sialic crust. The source area for the fore-arc detritus is an area of relatively little volcanic activity with large exposures of eugeosynclinal sediments. These different source areas are directly reflected in the fore-arc and back-arc basin petrology. Comparison of several fore-arc basins illustrates the importance of the tectonic history of an island arc. Arcs underlain by a "continental" crust may be associated with forebarc sediments rich in quartz and sedimentary rock fragments. Likewise, island arcs underlain by "oceanic" crust may be associated with quartz-poor detritus rich in volcanic lithic fragments. However, it should be noted that island arcs underlain by oceanic crust with large exposures of plutonic bodies may shed local pockets of detritus rich in quartz. It is also possible that extreme tr0pical weathering may provide detritus enriched in quartz (Breyer and Ehlmann, 1981). It is questionable whether back-arc basin detritus, like that from Japan (Q/F = .70, predominant feldSpar is plagioclase) may be distinguished from fore- arc regions such as that of Puerto Rico (Q/F = .68, predominant feldspar is plagioclase) on the basis of sediment compositions. Clearly, more petrographic data is needed to define the compositional limits of tectonic basins. 38 39 The main compositional control of island arc detritus appears to be provenance. The provenance of the sediments appears to be largely affected by the tectonic history of the arc, i.e., development of an arc on continental or oceanic crust; the occurrence and location of magmatic activity. Thus, the composition of arc detritus may provide insight as to the composition and tectonic history of ancient arcs. BIBLIOGRAPHY BIBLIOGRAPHY Bence, A. E. and Albee, A. L., 1968, Emperical correlation factors for the electron microanalysis of silicates and oxides: Jour. Sed. Petrology, v. 76, p. 382-1103. Blatt, H., 1967, Provenance determinations and recycling of sediments: Jour. Sed. Petrology, v. 37, p. 1031-1044. Blatt, H. and Schultz, D. 3., 1976, Size distribution of quartz in mudrocks: Sedimentology, v. 23, p. 857-866. Breyer, J. A. and Ehlmann, A. 3., 1981, Mineralogy of arc derived sediment: siliclastic sediment on the insular shelf of Puerto Rico: Sedimentology, V. 28, p. 61.7“. Coats, R. R., 1952, Magmatic differentiation in Tertiary and Quaternary volcanic rocks from Adak and Kanaga Islands, Aleutian Islands, Alaska: Geol. Soc. America Bull., v. 63, p. 485-511}. Coats, R. R., 1962, Magma type and crustal structure in the Aleutian Arc, in MacDonald, G. A. and Kuno,I. (eds.), The Crust of the Pacific Basin: American GeOphys. Union, Mon. 6, p. 92- 109. Crook, K. A., 1974, Lithogenesis and geotectonics: the significance of compositional variation in flysch arenites (graywackes), in Dott, R. H., J r. and Shaver, R. H. (eds.), Modern and Ancient Geosynclinal Sedimentation: Soc. Econ. Paleo. Min. Spec. Publ., no. 19, p. 304- 308. Dickinson, W. R., 1970, Interpreting detrital modes of graywacke and arkose: 3 our. Sed. Petrology, v. 40, p. 695-707. Dickinson, W. R. and Valloni, R., 1980, Place settings and provenance of sands in modern ocean basins: Geology, v. 8, p. 82-86. Donnelly, T. W., 1964, Evolution of the eastern Greater Antillean island arc: Amer. Assoc. Petrol. Geol. Bull., v. #8, p. 680-696. Galehouse, J. S., 1969, Counting grain mounts: number percentage vs. number frequency: 3 our. Sed. Petrology, v. 39, p. 812-815. Hamilton, W., 1977, Subduction in the Indonesian region, in Talwani, M. and Pitman, W. C., III (eds.), Island Arcs Deep Sea Trenches and Backarc Basins: Amer. Geophy. Union, Maurice Ewing Series no. 1, p. 15-32. #0 41 Harrold, P. J. and Moore, J. C., 1975, Composition of deep-sea sands from marginal basins of the northwestern Pacific, in Karig, D. E., Ingle, J. C., Jr., et al., Initial Reports of the Deep Sea Drilling Project, v. 31: Washington, U.S. Government Printing Office, p. 507-514. Hilde, T. W. C., Wageman, J. M. and Hammond, W. T., 1969, The structure of the Tosa Terace and Nankai Trough of S.E. Japan: Deep Sea Res., v.16, p. 67-750 Jones, R. L., 1979, Mineral diSpersal patterns in the Pierre Shale, Unpublished Ph.D. Thesis, Oklahoma University. Karig, D. E., Ingle, J. C., Jr., et al., 1975, Initial Reports of the Deep Sea Drilling Project, v. 31: Washington, U.S. Government Printing Office, p. 317-3500 Katili, J. A., 1975, Volcanism and plate tectonics in the Indonesian island arcs: TectonoPhysics, v. 26, p. 165-188. Mainland, D., Herrera, 1.. and Sutcliffe, M. I., 1956, Statistical Tables for use with Binomial Samples: New York, Dept. of Medical Statistics, New York Univ. College of Medicine, 83p. Matsuda, T. and Uyeda, 5., 1970, On the Pacific - type orogeny and its model extension of the paired belts concept and possible origin of marginal seas: Tectonophysics, v. 11, p. 5-27. Mattson, P. H., 1977, Geological character of Puerto Rico, in Mattson, P. H. (ed.), West Indes Island Arc (Bench mark papers in geology), v. 33: Stroudsburg, Dowden, Hutchinson and Ross, Inc., p. 179-192. Moore, G. F., 1979, Petrography of subduction zone sandstones from Nias Island, Indonesia: Jour. Sed. Petrology, v. 49, p. 71-84. Pettijohn, F. J., Potter, P. E. and Siever, R., 1973, Sand and Sandstone: New York, Springer-Verlag, 618p. Picard, M. D., 1971, Classification of fine-grained sedimentary rock: Jour. Sed. Petr0108y, V. #1, p. 179-195. Scholl, D. W., Creager, J. S., et al., 1973, Initial reports of the Deep Sea Drilling Project, v. 31: Washington, U.S. Printing Office, p. 217-277. Sibley, D. F. and Pentony, K. J ., 1978, Provenance variation in turbidite sediments, Sea of Japan: J our. Sed. Petrology, v. 48, p. 1241-1208. Stewart, R. J., 1978, Neogene volcaniclastic sediments from Atka Basin, Aleutian Ridge: Amer. Assoc. Petrol. Geol. Bull., v. 62, p. 87-97. Sugimura, A., 1968, Spatial relations of basaltic magmas in island arcs, in Hess, H. H. and Poldervaart, A. (eds.), Basalts: New York, Intersciences, p. 537-572. €12 Sugimura, A. and Uyeda, 5., 1973, Island arcs: Japan and its environs: Amsterdam, Elsevier Scientific Publishing Co., 2117p. Sweatman, T. R. and Long, J. V. P., 1969, Quantitative electron-probe microanalysis of rock-forming minerals: Jour. Petrology, v. 10, p. 332-379. Tanaka, K. and N ozawa, ‘1'. (eds.), 1977, Geology and Mineral Resources of Japan, v. 1: Hisamoto, Kawasaki-shi, Geol. Surv. of Japan, l#30p. Trevena, A. S. and Nash, W. P., 1981, An electron micrOprobe study of detrital feldspar: Jour. Sed. Petrology, v. 51, p. 137-150. Valloni, R. and Maynard, J. B., 1981, Detrital modes of recent deep-sea sands and their relation to tectonic setting: a first approximation: Sedimentology, v. 28, p. 75-83. Windley, B. F., 1977, The evolving continents: New York, John Wiley and Sons, 385p. Wolf, K. H., 1971, Textural and compositional transitional stages between various lithic grain types: J our. Sed. Petrology, v. 41, p. 328-332.