CYCLOSTRATIGRAPHY A METHODOLOGICAL APPROACH MARIO SPROVIERI 1, ANGELO BONANNO 2., SALVATOREMAZZOLA 2b E. BERNARDO PATTI Z' Recebecl July 1 5, 2AA1; accepted JudnuatT 9, 2AA2 Keyusords : Cyclostratigraph,v, insolation curve, spectral anal,v- sis, astronomical calibration. Rittssunto- In questo lavoro vengono descritte e commentate sinteticamente le metodologie pìù usate di analisi ciclostratigrafica per la calibrazione astronomica di sequenze sedimentarie. Dapprrma sono descritte dir.erse soluzioni numeriche della curva di insolazione e, in particolare, della più nota curva astronomica La9O1r,ry. Vengono quin- di, in ordine, descritte le metodologie di calibrazione astronomica i) su base lito-ciclostratigrafica e ii) sulla base di metodoiogie di analisi spet- trale applicate a segnali faunistici e geochimici che rìspondono ìn fase alle dìverse forzanti astronomiche. Abstract. Aim of this study is r synrhetic description of the most used methodologies of cyclostratigraphic analysis for the astro- nomical calibration of sedimentary sequences. The different numerical solutions of the insolation cune and, in particular, the most used astronomìc cur-ve La901r,ry are analtzed. Then. a detailed description of the methodologies for the astronomic calibration of different sedi- menrar)- sequences i1 on the basis of a iitho-cyclostratigraphic approach and ii) on the basis of spectral analysis applied to faunal and geochemical climate-sensitive records, is proposed. Introduction a- ;-^^*,.-. ^1rr of this volume is declicated to the reconstruction of an astronomically calibrated time scale from about 1 1 Ma to about 1,3.7 Ma, on the basis of cyclostratigraphic analyses of different land-based sec- tions. In the last years, a restricted group of researchers provided to the international scientific community a reliable astronomical calibration of the late Neogene geological record te.g. Hilgen l99la, 199 lbl Hilgen et al. 1995; Shackleton et al. 1995; Lourens et aI. 1996; Hilgen et al. 2000). They also developed the procedures to tune sedimentary sequences, characterised either by homogeneous or cyclically-repeated lithologies, to dif- ferent numerical solutions of the insolation curve. The most used numerical and cyclostratigraphic techniques applied for calibration of sedimentary records are here synthesized. fwo different kinds of sedimentary records are generally used for astronomical calibration: i) sequences characterised by lithologic alternations organised in well-recognisable cluster patterns and ii) sequences characterised by homogeneous lithology and studied by means of spectral methodologies applied to selected cli- mate-sensitive records. When possible, the combination of these two cyclostratigraphic approaches represents the most reli- able system to calibrate geological time series with the astronomical curves. The insolation curve The insolation on the Earth depends on its orbital parameters. Until 1988, the solution adopted for paleo- climate computation consisted of the orbital elements of the Earth, complemented by the computation of its pre- cession and obliquity (Berger 1928,1988). In 1988, Laskar proposed a solution for orbital elements of the Earth, which was obtained in a new manner, making use of vast analytical computation and numerical integration (Laskar 1988). In this solution the precession and obliq- uity parameters, necessary for paleoclimate computa- tions, were integrated at the same time, insuring a good consistency of the solutions. lJnfortunately, for various reasons, this latter solution for the precession and obliquity was considered reliable only for the last 3 Ma (Berger & Loutre, 1,991).Later on, Laskar et al. (1993) produced a new solution (hereafter named La 90) that is a slight improvement of the previous one. This solution allows evaluation of orbital parameters, precession and obliquity, up to the last 20 Ma. The orbital solution LagO is obtained by numerical ' Dipartimento (CFTA) - Via Archirafi 3(),9a1.23 Palermo, Itall', e-mail: marios@unipa.ir ' Istituto IRMA CNR - Via Luigi Vaccara, 61,, 91,A26 Mazara del Vallo, Italv, e-mail: 2" bonanno@irma.pa.cnr.tt;2b mazzola@irma.pa.cnr.it; 2 b p atti @)irma.pa. cnr. it 180 M. Sprooieri, A. Bonanno, S. Mazzola & B. Patti integration of an extended averaged sysrem that repre- sents the mean evolution of the orbits of the planets. All the 8 main planets of the solar system are taken inro account, as well as the main lunar and relativistic perrur- bations. The use of numerical integration for computing the solution of the secular system is one of rhe reasons for the good quality of this solution, which was checked by comparing with the ephemeris over a short time scale (Laskar 1988). The La 90 solution is in quasi-periodic form over 10 Mr, but these representations are s1owly convergent, which prevents good accuracy of the solu- tion. The reason for this slow convergence is due to the presence of multiple resonance in secular periodicity of the inner solar system (Laskar 1990). Because of these resonance the motion of the solar sysrem is chaotic in the iong term, and not quasi-periodic. The Lyapunov exponents of the solar system, that quantify the average grow of small initial error in the calcuiation, was esri- mated to l/5yr-r. This implies that it is not possible to give any precise solution for the motion of the Earth over more than about 10 Mr, and most probably, ephemerids can only be given with good precision for no more than 20 Mr. The solution of Laskar et al. (1993\ can be "recali- brated" at different ages because it gives the possibility to vary the values of the dynamical ellipticity and totai dissipation by the Sun and Moon. Lourens et aI. (1996) carried out a careful correla- tion between Plio-Pleistocene Mediterranean land-based sedimentary sequences characterised by well-recognis- able lithologic rhythms and well calibrated from a cyclostratigraphic point of view, and the different solu- tions of the insolation curve La 90. The best fit between the geological record and the insolation curve has been obtained using the numerical solution of Laskar et al. (1993) with unitary values of both the dynamical ellip- ticity and the total dissipation by the Sun and moon La9O.11.11. Successive works (Hilgen et al. 1995; Sprovieri et aI. 1999: Hilgen et al. 2000) confirmed these results for the Neogene stratigraphic records. In the papers of Iaccarino (2002), the LagO 1r,r; solution has been used ro calibrate the studied sedimen- tary records. Astronomical calibration of cyclic land-based sedimen- tary sequences The studies of Hilgen and co-workers (e.g., Hilgen 1991,a, 1991b; Hilgen et aL. 1995, 2OO0) on Mediterranean sections demonstrated that the high fre- quency alternations of different lithologies, recognised throughout several land-based sedimentary sequences, can be directly related to the precessional forcing and that the alternation of lonqer sedimentary intervals char- acterised by absence/presence of high frequency litho- logic cycles can be related to the eccentricity periods of t i ^^ | /^^ I aDout tuu ancl +uu Kvr. These papers demonstrated that the same kind of cyclic pattern - with clusters on different scales super- imposed on the basic sedimentary cycle - can be recog- nised in the Mediterranean from the Tortonian up to rhe Plio-Pleistocene. Alternation of homogeneous grey marls and sapropelitic layers represents the classic high frequency cyciicity recognised in most of the Mediterranean records. Alternativeiy, these two lithologies can be sub- stituted by similar variants (e.g., grey marls tone /car- bonate layers, light coloured/ grey coloured marls rone, etc.). Individual sapropelitic layers or substitutive sedi- mentary expressions correspond to precession minima or the equivalent summer insolation maxima, while small- and large-scale lithologic clusrers can be related to I '^^ ^^^ | r^^ ^^^the 100.000 and 400.000 years eccentricity maxima cycles. This simple sedimentary response to insolation forcing remained in practical identical over the last 1O Myr, with the same phase relations between sedimenta- ry and orbital cycles described for the Plio-Pleistocene. The assumption of a constant phase relationship between lithological cycles and astronomic forcing for all the Neogene, ailow us to identify a rigorous stepwise methodological approach for calibrating sedimentary sequences, characterised by lithological alternated records. Firstly, a detailed lithologic description of the studied sections allows the identification and the charac- terization of the sedimenr^ry rhythms along the records. Secondly, the studied successions musr be accu- rately dated with classic bio-magnetostratigraphy. In particular, polarity chrons musr be identified on rhe basis of Cande tr Kent (1995;CK95). Then, the ages of CK95 and Shackleton er aL (1995a; SCHPS95) for the poiarity reversals, and especially for the youngest ones, have to be adopted as starting point for the actual tun- irg. Thirdly, a first-order tuning between the sedimen- tary records and the insolation curve is obtained by a first-order correlation of large-scale sapropel clusters ro the 400,000-year eccentricity maxima and a direct corre- lation of the smal1-scale sapropel ciusters to the 10O.OOO- year insolation cycles maxima. Fourthiy, the astronomical calibration is complet- ed by tuning individual sapropels to precession minima and the corresponding summer insolation maxima. According to Hilgen (1991a, 1991b) and Lourens et al. (1996), a phase lag of -3 ky between mid-summer Northern Hemisphere insolation and lithologic cycies has to be considered for the final tuning. During the tuning exercise, larger-scale eccentric- ity cycle patterns must be taken into account, because Cyclostratigraphy, a methodological approach 181 they provide a direct check on the validity of the cali- bration and prevent accumulations of errors due to tun- ing of the individual cycles (i.e., several 20 kyr cycles could be missing or not recognisable). The resulting tuning is probably accurate to the 1eve1 of the individual precession/insolation cycle in certain intervals but may be off by one cycle in others. The final calibration pro- vides astronomical ages for the sedimentary cycles, as well as for the bio-events recognised throughout the studied record. It should also provide refinement of duration of individual polarity chrons. Astronomical calibration of homogeneous sedimentary sequences When lithological rhythms are not present rn a sedimentary sequence, the acquisition of detailed time series of climate sensitive records (oxygen isotopes, geo- chemical elements, planktonic and benthic foraminifera species, nannofossils species, etc.), known to be forced by astronomic periodicities, can be used to calibrate the studied sequence. In this case application of spectral analysis allows the recognition of a set of periodicity directly correlatable to astronomic parameters. The first step for comparing proxy records to the insolation curve is an appropriate sampling rate of the section (selected for a proper analytical resolution of the highest frequen- cy band) and the selection of the tuning straregy thar enables the establishment of the phase reiationships among the different chosen signals and the astronomic record for the different frequency bands (generally the three Milankovitch frequencies of precession, obliquity and short and long-eccentricity). The different records have to be converted in the time domain using the classic procedures suggested by Martinson et al. (1982) and the numerical algorithms of the inverse conjugate. Identification of a set of tie points (already inde- pendent dated biohorizons and/ or magnetostratigraphic intervals recognized along the section) allows the defini- tion of a first approximation of the sedimentarion rate of the studied succession. It is necessary to transform the available signals from the space- to the time-domain and to identify in the power spectra, estimated for the selected signals, the classic Milankovitch periodicities. The filtering of the original signals in the long- and short-eccentricity frequency bands and the direct comparison with the same cyclicity of the insolation curve allows a first-order astronomical calibration of the sequence. Then, filtering of the signals in the precession fre- quency bands and correlating them with the 19-23 kyr cycles present in the astronomic curve, produces a detailed short-time calibration of the studied record. Cross-spectral analysis between the paleoclimate signals and insolation should confirm with high coherency val- ues for the Milankovitch frequency bands the correct- ness of the obtained calibration. At this point all the stratigraphic events recognised throughout the succes- sion can be dated. Shackleton et al. (1997) suggested that application of numerical techniques of complex demodulation could be used as useful tool for calibrating sedimentary sequences, characterised by high variance concentrated in the precession frequency band. The complex demod- ulation method is a tool for examining the instantaneous amplitude and phase of that portion of the variability that is in a particular frequency band of the signal spec- trum. For the mathematical procedure, reader is referred to Shackleton et al. (1995b) and Pisias & Moore (1981). There are several applications of such a method for investigating the amplitude modulation in the pre- cession band, or in the obliquity band (Pisias & Moore 1e8 1) . The complex demodulation can be used as a method to assess the correctness of a timescale. Shack- leton et al. (1995b) pointed out thar ir is extremely dif- ficult to use cross-spectral analysis to investigate whether signal and forcing share the same amplitude modulation (e.g., the eccentricity modulation of the pre- cession signal), so that this aspect of the validity of a timescale must be evaluated by complex demodulation or other means like band-pass filtering, etc. However, demodulation works only if the paleoclimatic record is modulated (as the insolation curve). Then, application of this numerical technique must be proceeded by a ver- ification of the temporal relationship between the select- ed climate sensitive-record and the primary forcing. In the following papers we preferred to use the similar method of the band-pass filtering technique. In the papers presented in Iaccarino (2002), the SPAGE,OS software package (Bonanno et aI. 1996) has been used for application of all the above-described methodologies of spectral analvsis. Acknozoledgmenrs. \We thank E. helpful crìtìcal comments. This research Cofin 98. Erba and L. Lanci for theìr has been 5upporred b1 Mur.t 182 M. Sprovieri, A. Bonanno, S. Mazzola & B. Patti REFERENCES Berger A.L. (1978) - A simple algorithm to compute long-term variations of daily or monthly insolation. Contrib., Inst. Astr. Geophys. Geogr. Lemaitre, Unfu. Cathol. De Lou- vain,Y.18, 17 pp., Louvaine. Berger A. (1988) - Milankovitch theory and climate. Rev. Geo- pbys ., 26: 621-657 . Berger A. Ec Loutre M.F. (1991) - Insolation values for the cli- mate of the last l0 m.y. Quat. Sci. Rev., L0:297-317, Amsterdam. Bonanno A.,MazzoIa S., Sprovieri M. & Patti B. (1996) - SPA- GEOS, a system for the anal1'sis in time-and in fre- quency-domain of paleoceanographic and sedimento- logical data. Paleopelagos,6: 371-384, Roma. Cande S.C. & Kent D.V (1995) - Revised calibration of the geomagnetic polarity timescale for the late Cretaceous and Cenozoic. J. Geopbys. Res., 1OO (84): 6093-6095, \X/ashington D.C. Hilgen F.J. (1991a) - Astronomical calibration of Gauss to Matuyama sapropels in the Mediterranean and implica- tion for the Geomagnetic Polaritv Time Scale. Earth Planet. Sc. Letters, 1a4: 226-244, Amsterdam.. Hilgen F.J. (1991b) - Extension of the astronomically calibrat- ed (polarity) time scale to the Miocene/Pliocene boundary. Eartb Planet. Sc. Letters,107: 349-368, Ams- terdam. Hilgen F.J., Krijgsman 'il{, Langereis C.G., Lourens L.J., Santarelli A. & Zachariasse \MJ. (1995) - Extending the astronomical (polarity) time scale into the Miocene. Eartb Planet. Sc. Letters,136: 495-51Q, Amsterdam. Hilgen F.J., Krijgsman \{1, Raffi I., Turco E. &'Zacharjasse WJ. (2000) - Integrated stratigraphy and astronomical cali- bration of the Serravallian/Tortonian boundary section at Monte Gibliscemi (Sicily, Italy). Mar. Micropal., 38: 1,81-211, Amsterdam. Iaccarino S.M. (ed.) QAA4 - Integrated Stratigraphy and Pale- oceanosraohv of the Mediterranean Middle Miocene..--.-*_'"b'_r_'/ "' Rip. It. Paleont. Strat., L08: 175-353, Milano. LaskarJ. (1988) - Secular evolution of the solar system over 1O million years. Astron. Astrophys., 1.98: 341.-362, Vashing- ton D.C. Laskar J. (1990) - The chaotic motion of the solar system: a numerical estimate of the size of the chaotic zone. Icarus, 88: 266-291, Orlando. LaskarJ.,Joutel F. & Boudin F. (1993) - Orbital, precessional, and insolation quantities for the Earth from -20 Myr to *10 Myl Astron. Asrroplrys.,27A: 522-533, \ù/ashington D.C. Lourens L.J., Antonarakou A., Hilgen F.J., Van Hoof A.A.M., Vergnaud-GrazziniC. & Zachariasse \(J. (1996) - Eval- uation of the Plio-Pleistocene astronomical timescale. Pale o c e an o grap hy, L L : 39 1. - 4 13, \lashin gton D. C. Martinson D.G., Menke \( & Stoffa P (1982) - An inverse approach to signal correlation. J. Geophys. Res., 87: 4807-48 1 8, Washington D.C. Pisias N.G. & Moore T.C. (1981) - The evolution of the Pleis- tocene climate: a time series approach. Earth Planet. Sc. Let ters., 52: 450-458, AmsLerdam. Shackleton N.J., Crowhrust S., Hagelberg T., Pisias N.G. & Schneider D.A. (1995a) - A new late Neogene timescale. Application to leg 138 sites. Proc. ODP Resuhs, 138 73-101, College Station (TX) Shackleton, N.J., Hagelberg, T.D. & Crowhurst, S.J. (1995b) - Evaluating the success of astronomical tuning: pitfalls of using coherence as a criterion for assessing pre-Pleis- tocene times c ales. Pal e o c ean o grdp hy, Ia : 69 3 - 69 7, Vash- ington D.C. Shackleton N.J. Ec Crowhrust S. (1q97) - Sediment fluxes based on an orbitally tuned time scale 5 Ma to 14 Ma, site926: AppÌication to ODP leg 154 sites. In: Shackleton N.J., Curry \í8., Richter C. and Bralower T.J. (Eds.) Proc. ODP Resuhs., L54: 69-82, College Station (TX). Sprovieri M., Bellanca A., Neri R., Mazzola S., Bonanno 4., Patti B., & Sorgente R. (1999) - Astronomical calibra- tion of Late Miocene stratigraphic events and analysis of precessionally driver paleoceanographic changes in the Mediterranean basin. Mem. Soc. Geol. It., 54 7-24, Roma.