GEOPHYSICAL RESEARCH LETTERS, VOL. 11, NO. 6, PAGES 611-613, JUNE 1984 TRANSITION STRATIGRAPHY AND THE PROBLEM OF REMANENCE LOCK-IN TIMES IN THE SIWALIK RED BEDS L. Tauxe 1and C.Badgley 2 1Scripps Institution of Oceanography, University of California, San Diego, La Jolla, CA 92093 2Museum of Paleontology, University of Michigan, Ann Arbor, MI48109 Abstract. The record of paleomagnetic polarity transitions can provide relative time information in the stratigraphic record on scales of hundreds to thousands of years. Transitionstratigraphy is here used to determine the relative lock-in times of several components of magnetization in the Siwalik red beds. Introduction Many geologic processes, such as the migration of fluvial channel belts, the acquisition of early post-depositional remanence, soil formation, etc., occur over time spans of hundreds to thousands of years. For all but the most recent history, these events are considered instantaneous because of insufficient time resolution. However, since paleomagnetic polarity transitions have a duration considered to be less than 10,000 years and often displaya continuum of directions,it is possible to use the transition record to provide relative chronologic information on geologically short time scales. In this paper, we present an illustration of "transition stratigraphy", using a record of a magneticpolarity transition in fluvial sediments to determine the relative timing of acquisition of various components of remanence. The results are preliminary and are intended primarily as an example of the method rather than as a finished product. Transition Stratigraphy in the Siwalik Red Beds Magnetostratigraphy is the principle means of linking the important Siwalik vertebrate fossil record to an absolute time scale(e.g. Johnson et al., 1982;Tauxe and Opdyke,1982). The interpretationof the paleomagnetic data as a magnetochronology relies on the assumptionthat the characteristic magnetizations were acquired penecontemporaneously with deposition. Recent contributions to the understanding of red-bed aliagenesis, however, suggest the possibilityof long-term, chemical alteration of the magnetic mineralogy of red beds(e.g. Walker et al., 1981; Larson et al., 1982). Such aliagenetic changes may affect the natural remanent magnetization of the rock in a complex manner. Since the magnetic remanence of Siwalik rocks is carried by hematite, questionsarise as to the age and fidelity of remanence in the Siwaliks. We have identified two distinct grain-size fractions of hematite; both contribute to the magneticremanence(Tauxe et al., 1980). The coarse-grained fraction (specular hematite) has a maximum blocking temperature of 685øC, but a low coercivity and was shown by means of an intraformational conglomerate test to carry a remanence which was early acquired. The fine- grained fraction (pigmentary hematite)is characterized by lower blocking temperatures(< 625øC), high coercivitiesand was shown to carry several components, including present field directions and ancient but secondary components of magnetization. The discovery of a record of a paleomagnetic polarity transition in the sediments exposed in Ganda Kas (Tauxe and Opdyke, 1982), suggested a means of establishing the timing of remanence acquisition (see Channellet al., 1982). Copyright 1984 by the American Geophysical Union. Paper number 4L6005. 0094-82 76 / 84/004L-6 005 $03.00 In this paper, we present our analysesof samples magnetized during a late Miocene paleomagnetic transition. Several components of remanence were acquired during the reversal, but were lockedin at slightlydifferent times. Hence, they follow one another along the path of the transition. The order of acquisition of the various components is readily apparent,and inferencescan be made as to the origin and fidelity of the Middle Siwalik paleomagneticrecord. Methodology The GandaKas transition (GK of Tauxe and Opdyke, 1982), first sampled in 1977, was resampled in 1982. The new series (GKA) was sampled with one site located approximately every meter. A parallel section (TR) is located 100 m to the west and was sampled in 1980. Representative examples of step-wise thermal demagnetization are shown in Figure 1. Several components of magnetizationare evident from the vector diagrams. Although the blocking temperature spectrum of each component varies slightly from specimen to specimen, there is a broad tendency for linear segmentsto be isolated between the limits of NRM- 250 ø , 250-450 ø , 550-650 ø and 650-670 ø . These components are designated A through D respectively. Few specimens have all four components.For example,GKA12C (Figure ld) has only the B component from 250-450ø. The directions of the various components were calculated using principle component analysis as discussedby Kirschvink (1980). Components were all calculated using at least three data points and were not considered linear if their Maximum Angle of Deviation, or MAD sensu Kirschvink exceeded 10 ø. Directions are considered significantly different if their MAD angles do not overlap. In the following section, we discussthe implications of the relationship of the components to the transition path for the timing of remanenceacquisition in the GKA specimens. Results The magnetostratigraphy based on the B component directions of the TR and GKA sectionsis shown in Figure 2. The B componentwas used primarily for the sake of continuity. We make no assumption about its age relative to that of the rock. We do assume, however, that the B components were acquired in superpositional order. The lithostratigraphic sections shown on Figure 2 were measured by us and by A.K. Behrensmeyer. The M.G., U.G. and "U" units correspond to the Middle Grey, the Upper Grey and the "U" strata respectively (Behrensmeyer and Tauxe, 1982). The magnetozones N1 through R3 refer only to these two sections. The zone N1 and R3 correspond to DN4 and DR4, respectively, of Tauxe and Opdyke (1982). The zones N2 through R3 are part of the transitional record and indicate that the transitional path is fairly complex. Since the same polarity zones occurred in sections with different patterns of sedimentation, we consider the paleomagnetic stratigraphy a reliable indicator of the reversal history. The original transition was discovered between 26.0 and 19.0 meters above the Middle Grey unit in the GK section (identical to the GKA). Transitional specimens were not found in the TR series (100 m away) which led us to sample the GKA section. The reason for 611