Education

46 GEOCHRONOLOGY OF BASALTIC ROCKS RECOVERED BY DSDP LEG 41, EASTERN ATLANTIC OCEAN

Description
46 GEOCHRONOLOGY OF BASALTIC ROCKS RECOVERED BY DSDP LEG 41, EASTERN ATLANTIC OCEAN Robert A. Duncan, School of Oceanography, Oregon State Unversty, Corvalls, Oregon, and Everett D. Jackson, U.S. Geologcal
Categories
Published
of 6
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
46 GEOCHRONOLOGY OF BASALTIC ROCKS RECOVERED BY DSDP LEG 41, EASTERN ATLANTIC OCEAN Robert A. Duncan, School of Oceanography, Oregon State Unversty, Corvalls, Oregon, and Everett D. Jackson, U.S. Geologcal Survey, Menlo Park, Calforna INTRODUCTION Attempts at obtanng relable K-Ar age determnatons on gneous materal collected from the ocean floor have been complcated by nondeal crystallzaton and subsequent alteraton n the submarne envronment. Magma erupted onto the ocean floor under a klometer or more of seawater may be rapdly quenched so that some amount of radogenc argon nherted from the magma's source regon s retaned, leadng to an erroneously old age of crystallzaton. Dalrymple and Moore (1968) and Dymond (1970), among others, have shown that ths effect s mportant n the outermost few centmeters of pllow lavas, but for more slowly cooled, coarser-graned basalts, suffcent degassng has occurred durng crystallzaton that excess argon s probably not sgnfcant. Radogenc argon formed after crystallzaton from the natural decay of 40 K to 40 Ar may be lost ether by low-temperature alteraton (converson of prmary mneral phases and ntersttal matrx to secondary clay mnerals and zeoltes that, owng to ther looser structure, leak argon) or from glass whch has an unpredctable argon retentvty because of ts amorphous structure. A thrd problem results from addton of potassum to the sample from seawater, leadng to an underestmate of the sample age (Sedemann, 1976). These three effects must be consdered n attemptng K-Ar age determnatons on DSDP basalts. For the present study, basaltc materal was provded from near the bottom of the holes drlled at Stes 367 and 368 of Leg 41. We selected samples for geochronology on the bass of mcroscopc examnaton. s were elmnated because of extensve lowtemperature alteraton or presence of devtrfed glass. None of the selected samples met the usual crtera, but four samples appeared suffcently fresh to attempt age determnatons. Conventonal K-Ar age determnatons were performed for the four whole-rock samples. The results of these analyses appear n Table 1 and follow the technques descrbed by Dalrymple and Lanphere (1969). Ages on addtonal alquants of these samples were then determned by the 40 Ar/ 39 Ar total fuson technque (Merrhue and Turner, 1966). In ths method, samples are rradated n a fast neutron flux (USGS TRIGA reactor n Denver, Colorado). A proporton of 39 K n the samples s converted to 39 Ar. Because a montor sample of known age s also rradated wth each batch of unknowns, the converson effcency, expressed as the parameter J, can be measured. Argon s then extracted n the conventonal manner and the ratos of four sotopes ( 40 Ar, 39 Ar, 37 Ar, and 36 Ar) are measured by gas-source mass spectrometer. The sample age s then calculated from the sotopc ratos, the J parameter determned from the montor sample and known decay constants. Partculars of ths method appear n Dalrymple and Lanphere (1971). Results are presented n Table 2. To offset the nconvenence of sample rradaton, ths method offers several advantages over the conventonal method. All measurements are determned on a sngle alquant of the sample ( 39 Ar drectly measures the potassum content), whch elmnates the problem of obtanng chemcally dentcal alquants from nhomogeneous samples. It also follows that a much smaller sample weght can be used an advantage when sample qualty and quantty are low. Perhaps the most attractve aspect of ths method les n the ablty to nvestgate the dstrbuton of radogenc argon wthn a sample by heatng n ever-ncreasng temperature ncrements the Stepwse, or ncremental heatng technque. Ideally, ages measured from the sotopc compostons of successve fractons of gas from a heated sample should be dentcal, producng a plateau on a plot of age versus temperature or fracton of gas released. Commonly, radogenc argon and artfcally produced 39 Ar have been lost or redstrbuted among the varous meltng phases so that gas fractons yeld nonunform ages. The ncremental heatng mode has been used on two samples, one each from Stes 367 and 368. Results appear n Table 3. RESULTS AND DISCUSSION Ste 367 Ste 367 s located n the Cape Verde Basn, southeast of the Cape Verde Islands and wthn the magnetc quet zone along the eastern margn of the Atlantc Ocean. Basalt was recovered at ths ste after penetraton of a 1153-meter sedmentary secton. Reddshbrown argllaceous lmestone, marl, and claystone that overle the basalt are Upper Jurassc but not older than Oxfordan age (see Ste Chapter 367, ths volume). s from Cores and are nearly aphyrc fne-graned vescular basalts composed prncpally of plagoclase and pyroxene, whch have been partally replaced by smectte, calcte, and pyrte, 1113 R. A. DUNCAN, E. D. JACKSON TABLE 1 Leg 41 K-Ar Whole-Rock Ages (Interval n cm) Rad 40 Ar %K 2 O ( lo~ 11 mol/g) Rad 40 Ar 100 X Total 40 Ar Age (m.y.) ±1 S.D , ( 39, 40 ( 45, 38 ( 0.400, 88 ( 99, 92 ( 0.638, «0.635, j 0.413, , ± ± ± 18.7 ± 16.3 ± 18.1 ± 18.6 ±0.6 TABLE 2 Leg Ar/ 39 Ar Total Fuson Ages (Interval n cm) 40Ar/39 Ar 37 Ar 39 Ar a 36 Ar/ 39 Ar jb 40 Ar rad 36 Ar Ca 39 Ar,Ca Age (m.y.) ±1 S.D , ± ± ± ±0.6 Corrected for 37 Ar. b Irradaton parameter determned by montor of known age. λ ç = ~ 10 yr~ 1 ; λ j3 = 4.72 X 10~ 10 yr~ 1 ; 40 K/K = 4 mol/mol. TAJ LE3 Leg 41 Ages from 40 Ar/ : 'Ar Incremental Heatngs (Interval n cm) 39 and Temp. ( C) 40Ar/ Δr /39 Ar 37 Ar/ 39 Ar a 36 Ar /39 b 40 Ar Ar Age (m.y.) ±1 S.D Fuson Total gas age 105 /= ± ± ± ± ± ± Fuson Total gas age 19.8 /= ± ± ± ± ± 20.4 ±1.5 Corrected for 3' Ar decay. b Percent released durng ncrement, λ, ~ 10 yr~ 1 ; λ ß = 4172 X 10~ 10 yr~ 1 ; 40 K/K = 1.19 X 10~ 4 mol/mol. and glass that has been converted to brown smectte. Vescles are flled by calcte and clays. The sngle sample selected for geochronology (Table 4) was the freshest avalable to us. 1114 GEOCHRONOLOGY OF BASALTIC ROCKS TABLE 4 Petrographc Data on Dated Leg 41 Basaltc Rocks (Interval n cm) Phenocrysts (or Mctophenocrysts) Average Sze Mneral Percent (mm) Percent Vescles Average Sze (mm) Percent Groundmass Average Sze (mm) Domnant Texture Alteraton Remarks , Plagoclase Aphyrc - - Aphyrc - - Aphyrc 4.1 None None None Plagoclase Clnopyroxene Glass Opaques Botte Plagoclase Clnopyroxene Glass Opaques Botte Plagoclase Pyroxene Glass Opaques Botte X 0.8 X X Intersertal or ntergranular Dabasc Dabasc Dabasc 75% of vescles flled wth calcte; 25% partally flled wth brown smectte; glass altered to olve-brown smectte; some plagoclase altered to calcte some pyroxene altered to smectte; rare chlorte Glass mostly altered to smectte; very mnor smectte alteraton n plagoclase and pyroxene; very few thn calcte vens Glass less altered than cm nterval Small amounts of smectte alter plagoclase pyroxene and glass about equally Contaned less orgnal glass than other samples avalable; opaque oxdes more evenly dstrbuted Glass n rregular patches that contan opaques and apatte Opaques scattered evenly through secton Conventonal whole-rock K-Ar ages (Table 1) for ths sample ndcate a Cretaceous age (~ 90 m.y.). In vew of the pervasve alteraton, ths must be consdered a mnmum age. Total fuson 40 Ar/ 39 Ar (Table 2) determnes an age whch s about 10% older ±2.0 m.y. Loss of 39 Ar from smectte between tmes of rradaton and fuson may partally explan the dfference n ages determned by the two methods. If loss of radogenc argon through alteraton has led to an erroneously young age by the conventonal method, then the total fuson age may be regarded as a truer estmate of the sample's age, but s stll a mnmum age. More nformaton about the dstrbuton of 40 Ar and 39 Ar (= K) s provded by the ncremental heatng data n Table 3, whch are llustrated n Fgure 1. Sx successve temperature fractons have been analyzed for ther argon sotopc composton. Determned ages range from 124 to 95 m.y., generally decreasng from the low-temperature fractons to complete fuson. Fgure la s an sochron dagram, plottng 40 Ar aganst 39 Ar, each normalzed to 36 Ar. For deal behavor, all components of a chemcal system (.e., lava) at tme zero (crystallzaton) wll possess atmospherc 40 Ar/ 36 Ar (= 295.5). Because potassum s heterogeneously dstrbuted among the varous mneral phases, radogenc argon accumulates at dfferent rates. By separatng temperature fractons, a range of 40 Ar/ 36 Ar and 39 Ar/ 36 Ar composton results. These compostons should be related lnearly, wth ntercept 40 Ar/ 39 Ar = and slope proportonal to the age of the sample. Fgure la shows that ths sample has not behaved deally. Isotopc compostons from sx temperature fractons scatter about the best fttng straght lne (determned by York's, 1969, technque for correlated errors). The sochron age s ±7.7 m.y. wth ntercept 298 ±25. Even though the atmospherc composton s ndstngushable from the estmated ntercept, the departure of compostons from the bestft straght lne suggests that radogenc argon ( 40 Ar) and potassum (= 39 Ar) have been lost or redstrbuted snce crystallzaton. Hence ths sochron age cannot be consdered sgnfcant. The low-temperature fractons of the ncremental heatng whch yeld ages close to 122 m.y. may be the closest to the true age of the basalt. It s unlkely, however, that these ages are sgnfcant and our best estmate of the sample age remans that derved from overlyng sedments (Upper Jurassc). In plottng age aganst cumulatve percent 39 Ar released, the so-called age spectrum dagram (Fgure lb), no plateau age s dscernble. The dstnctve decrease n apparent age from low-temperature to hghtemperature gas fractons may ndcate 39 Ar recol effects (Turner and Cadogan, 1974). In fne-graned rocks composed of K-rch and K-poor phases, recol of 39 Ar from neutron bombardment may cause a transfer of 39 Ar from the K-rch phases to the K-poor phases. Because K-poor phases (e.g., pyroxene) generally retan argon to hgher temperatures than K-rch phases (e.g., plagoclase, mesostass), the rradaton-nduced redstrbuton of 39 Ar ncreases apparent ages for lowtemperature gas fractons by ncreasng 40 Ar/ 39 Ar and decreases apparent ages for hgh-temperature gas fractons by decreasng 40 Ar/ 39 Ar. Such redstrbuton of 39 Ar may have occurred durng rradaton of ths sample. If no 40 Ar has been lost pror to rradaton, a meanngful age may be obtaned by combnng all gas fractons (equvalent to a total fuson age). As 40 Ar loss 1115 R. A. DUNCAN, E. D. JACKSON 1100,400 (b) IΛI / FUSION 80- ó j AR/ 36.AR AR RELEASED Fgure 1. Incremental heatng for cm. (a) ^Ar-^Ar sochron, ^Ar^Ar, and ^Arj^Ar for sx temperature fractons are corrected for calcum and potassum nterferences-uncertantes n age and ntercept are 1 S. D. (York, 1969); (b) apparent ^Arj^Ar age versus fracton of^ar released-dashed lnes are 1 S. D. for the apparent ages. has probably occurred, the recombned total gas age of ±1.8 m.y. s a mnmum age. Ste 368 Ste 368 s located northeast of the Cape Verde Islands on the Cape Verde Rse. Drllng at ths ste penetrated 985 meters of predomnantly terrgenous sedments before ntersectng three dabase slls nterstratfed wth black shale. The shales are Upper Cretaceous n age (Ste Chapter 368, ths volume). The lthostratgraphc data suggest that ths ntrusve epsode s related to the Mocene volcanc actvty on the Cape Verde Islands and around Dakar, Senegal. Volcanc ash s also found n upper Mocene sedments of Cape Verde Rse (also Ste 368), probably manfestng the same volcanc epsode. Three samples of dabase from the lower, thcker unt at Ste 368 were chosen for geochronology. These rocks are much fresher than the sample collected from Ste 367. They are medum- to coarse-graned aphyrc basalts wth dabasc texture (Table 4). Glass s abundant, but both the amount of glass and the extent of ts smectte alteraton decrease downward n the sll. There s some suggeston from our samples that gran sze also ncreases downward (see Natland, ths volume). Conventonal K-Ar whole-rock ages ndcate that the dabase recovered at Ste 368 s of early Mocene age and thus postdates the Cretaceous sedments nto whch t must have been ntruded. Two of the three samples ( cm and , cm) yeld reproducble ages and agree wth one another at m.y. (Table 1). The thrd yelded a slghtly younger age, 16.3 m.y. Total-fuson 40 Ar/ 39 Ar (Table 2) age determnatons on the same three samples show a smlar consstency, wth slghtly older mean age, 19.0 m.y. From the compatblty of conventonal K-Ar ages wth total-fuson ages, we conclude that 39 Ar loss after rradaton has not been a sgnfcant problem. Results of 40 Ar/ 39 Ar ncremental heatng for one sample from Ste 368 appear n Table 3 and Fgure 2. Dsregardng the 400 C ncrement, n whch fractonaton of 36 Ar relatve to 40 Ar has led to a negatve age, and the 750 C ncrement, whch requred an addtonal clean-up so that ts extreme age may also be due to fractonaton, the remanng temperature steps ndcate a Mocene to Olgocene age. Fgure 2a plots the sotopc compostons of these fractons and the sochron resultng from the best-ft straght lne to three of them (525 C ncrement s not used because of lkely 36 Ar or 39 Ar loss). The sochron yelds an age of 19.1 ± m.y. wth an ntercept of 301 ±4, slghtly above the atmospherc 40 Ar/ 36 Ar. The plateau age developed by the three temperature ncrements (Fgure 2b) s 2 ± m.y., a weghted mean. Hence ages determned by the three methods are n good agreement and ndcate an age of about 19 m.y. (early Mocene, Berggren, 1972). 1116 GEOCHRONOLOGY OF BASALTIC ROCKS (a) O FUSION 300 ^301.5 ± AR/ 36.AR AR RELEASED Fgure 2. Incremental heatng for cm. (a) ^Ar-^Ar sochron, ^Ar^Ar, and ^Ar/^Ar for four temperature fractons are corrected for calcum and potassum nterferences-uncertantes n age and ntercept are 1 S. D. (York, 1969); (b) apparent ^Arj^Ar age versus fracton ^Ar released-dashed lnes are 1 S. D. for the apparent ages. Alkalc volcanc rocks erupted n the Cape Verde Islands durng Mocene and Quaternary tmes, last eruptng n 1951 (see Dllon and Sougy, 1974, for a summary). Volcanc actvty of a smlar age occurred near Dakar, Senegal, and t s natural to nclude the submarne ntrusons wth ths larger regon of tectonc and volcanc actvty. Dllon and Sougy (1974) suggest that the collson of the Afrcan plate wth the Eurasan plate n Mocene tme precptated ths volcansm. Alternatvely, the volcansm may be a manfestaton of plume actvty (Morgan, 1972) whch, because of the very slow movement of the Afrcan plate snce Mocene tme, has produced a restrcted regon of prolonged actvty rather than a well-defned lneament. In the Ste 368 Summary (ths volume) t s proposed that the uplft of the Cape Verde Rse, whch occurred n the mddle Mocene, was related to ths volcanc actvty. The age of the dabase sll n Ste 368 ndcates that the ntrusve event somewhat predated the uplft and suggests that longer term thermal conducton may account for the uplft. ACKNOWLEDGMENTS We are grateful to G.B. Dalrymple who suggested ths project and gave helpful advce. We thank J. Von Essen and S.E. Sms for techncal assstance and G.P. Kraker for sample rradatons. J.V. Gardner kndly provded the samples. We also thank R.J. Fleck and J.V. Gardner for helpful comments n revewng the manuscrpt. REFERENCES Berggren, W.A., A Cenozoc tme scale some mplcatons for regonal geology and paleobogeography: Lethaa, v. 5, p Dalrymple, G.B. and Lanphere, M.A., Potassumargon datng: San Francsco (Freeman)., Ar/ 39 Ar technque of K-Ar datng: A comparson wth the conventonal technque: Earth Planet. Sc. Lett., v. 12, p Dalrymple, G.B. and Moore, J.G., Argon-40: Excess n submarne pllow basalts from Klauea Volcano, Hawa: Scence, v. 161, p R. A. DUNCAN, E. D. JACKSON Dllon, W.P. and Sougy, J.M.A., Geology of West Afrca and Canary and Cape Verde Islands. In Narn, A.E.M. and Stehl, F.G. (Eds.), The ocean basns and margns: New York (Plenum Press), p Dymond, J.R., Excess argon n submarne basalt pllows: Geol. Soc. Am. Bull., v. 81, p Merrhue, C. and Turner, G., Potassum-argon datng by actvaton wth fast neutrons: J. Geophys. Res., v. 71, p Morgan, W.J., Deep mantle convecton plumes and plate motons: Am. Assoc. Petrol. Geol. Bull., v. 56, p Sedmann, D.E., K-Ar dates for basaltc rocks from Stes 319 and 321, Leg 34. In Yeats, R.S., Hart, S.R., et al., Intal Reports of the Deep Sea Drllng Project, Volume 34: Washngton (U.S. Government Prntng Offce), p Turner, G. and Cadogan, P.H., Possble effects of 39 Ar recol n 40 Ar- 39 Ar datng. In Ffth Lunar Sc. Conf. Proc. (Suppl. 5? Geochm. Cosmochm. Acta): v. 2, p York, D., Least squares fttng of a straght lne wth correlated errors: Earth Planet. Sc. Lett., v. 5, p
Search
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks