MAGNETIC POLARITY STRATIGRAPHY AND CORRELATION OF THE LEISEY SHELL PITS, TAMPA BAY, HILLSBOROUGH COUNTY, FLORIDA Bruce J. MacFaddeni ABSTRACT Oriented paleomagnetic samples were collected from plaster jackets containing fossils, in situ bivalves, and surrounding matrix from the Leisey Shell Pit. Analysis in the paleomagnetic laboratory indicates that the magnetization is probably carried in fine-grained magnetite. After both alternating field and thennal demagnetization, all of the five sites sampled are interpreted to be of reversed polarity f6r the local section at Leisey. In conjunction with the stage of evolution oflhe fossil mammals and interbedded Sr isotopic data, the Bermont Formation in the area of Leisey correlates to the late Matuyama Reversed Chron between about 1.66 and 1.4 Ma. RESUMEN Se colectaron muestras de orientaci6n paleomeagn6ticas desde moldes de yeso que contenian fasiles, bivalvos in situ y matriz circundante, provenientes de la Excavaci6n de Conchuelas de Leisey. Andlisis realizados en el laboratorio paleomeagn6tico sefialan que probablemente la magnetizaci6n es transportada en magnetila de grano fino. Desputs de la desmagnetizacitn con campos alternos y ttnnica, se considera que los cinco sitios muestreados en la secci6n local de Leisey tienen polaridad revertida. En conjunto con el estado evolutivo de los mamiferos f6siles y la formaci6n de los is6topos de Sr en capas entrelaminadas, la Formaci6n de Bermont en el drea de Leisey se correlaciona con el Cron Revertido del Matuyana tardio, de cerca de 1.66 y 1.4 Ma. 1 Tk author is Curator b Vertebrate Palcontology, Florida Mu,eum of Natural Histog, and Professor of Geology ani Zoology, Univemity of Florida, P. O. Boot I ll800, Gainesville FL 32611-7800, U.S.A. MACFADDEN, B. J. 1995. Magmetic Polarity stratigraphy and correlation of the Leisey Shell Pits, Tampa Bay, Hillsborough County, Florida. Bull. Florida Mus. Nat. Hist. 37 Pt. I(3):107-116. 108 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL 37, Fr. 1, No. 3 INTRODUCTION The fossil record of Cenozoic mammals in Florida is excellent and unparalleled elsewhere in eastern North America. Despite this abundance, however, this sequence suffers from the lack of long continuous exposures, which are needed for paleomagnetic stratigraphy, and furthermore, there are no outcrops of potentially datable volcanic units. Thus, until recently, the age determinations of the rich Florida sequence were obtained from stage of evolution comparisons with dated faunas from the western United States. The results from the paleomagnetic and isotopic dating presented in 1his volume therefore represent a new direction in understanding the calibration of mammal evolution in this important region. In this paper I present the results from the paleomagnetic sampling and laboratory analyses of samples collected from the Leisey Shell Pit. As will be explained below, this represents a novel approach because in contrast to conventional practice in which oriented sediments samples are collected, paleomagnetic samples were taken from sediment infillings (steinkerns) of oriented, in situ bivalves. The resulting polarity data are consistent with the other chronological results presented elsewhere in this volume and synthesized in the article by Morgan and Hulbert. ACKNOWLEDGMENTS I thank Michael J. Whitelaw for devising a method for hardening the poorly indurated samples and for his excellent sample analysis in the University of Florida Paleomagnetics Laboratory. This is University of Florida Contribution to Paleobiology number 398. PALEOMAGNETIC SAMPLING AND LABORATORY PROCEDURES The paleomagnetic study of the Bermont Formation at Leisey Shell Pit was initiated in 1987, and as such there were no in situ exposures from the highly fossiliferous lA locality, which had previously been excavated and was then destroyed during mining operations. Accordingly, paleomagnetic samples ("sites") were taken from oriented (top/bottom and some with azimuth) plaster jackets that were still available in the Vertebrate Paleontology collection at the FlaMNH. In the worst case scenario where only top/bottom data were available, the resulting polarity determinations could therefore be made from the vertical component (inclination) of the pateomagnetic vector (this procedure yields data similar to that of azimuthially unoriented deep-sea cores). In contrast exposures from Leisey 3A were still available, and in addition to those taken from oriented plaster jackets, MACFADDEN: LEISEY SHELL PIT MAGNETIC POLARITY 109 ' 41 f 4 Figure 1. In mu bivalves Vercenaria) showing paleomagnetic sampling. After the flat surface is formed by scraping away part ofthe shell, the fossil is oriented with an azimuth. In the laboratory, these fossils are sawed open and the internal sediment cast (steinkern) is impregnated with a non-magnetic hardener. Thereafter, the sample is cut into a 2.5 cm oriented cube to be analyzed in the paleomagnetic laboratory. (Photograph from author.) 110 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL- 37, Fr. I, No. 3 PALEOMAGNETIC SAMPLING COMPOSITE SECTION COMPOSITE SECTION LEISEY SHELL PIT I LEISEY SHELL PIT 3 depth below ground surface (In) -0 0.4 m, modern soil zone 0.4 m, modem. soil zone (removed)- -8 F 0 - 0 1.3 m, unconsolidated 11!pun 2.4 m, E quartz sand 1 0 H ld unconsolidated .. - quartz sand = lill~ ~ 1U~:rz=r=~:~= S *sandy, massive shell .2 @- bed -@- 645'61 •an< BONC BED - - UPPER SHELL BED i - 8g 3.0 m, buff, gandy, E LOWER SHELL BED 8 5 massive shell bed O 5- & 1 3.3 4 cream to blu194 1 E3 2 Ul_ _ _ - 2 - aandy, massive shell bed 0.4 m. shelly,dolom, -hard layer- 8 ~ ~ - 0.1 -~. iholjy, mu«*, BO~E BE<- E I.- 1.0 m, dark, sandy mud; C freshwater shells; bones LOWER SHELL BED ~ E 4.3 m, bluish, 1 -sandy, massive shell bed ~ PALEOMAGNETIC SITE -10 darl4sandy - - X -mill , modern mean sea level u, u Z Etan, hard, massive, W 2 phosphatic dolomile 0 0 E 1 1 Figure 2. Stratigraphic sections showing LEisey lA (dere plaster jackets were collected prior to this study) and Lcisey 34 and position ofpaleomagnetic samples. The site numbers correspond to those listed in Table 1. standard paleomagnetic samples of homogeneous fine-grained sediments were supplemented with oriented in situ bivalves (Mercenaria; from which hardened steinkerns were cut into paleomagnetic samples, see Fig. 1). In total, three to five separately oriented paleomagnetic samples were taken from each of five horizons at Leisey, including the bone-bearing units (Fig 2.). In the laboratory these samples were prepared into standard 2.5 cm cubes. The poorly MACFADDEN: LEISEY SHELL PIT MAGNETIC POLARITY 111 indurated nature of these sediments required that the samples be hardened. White glue or sodium silicate were used, both of which are non-magnetic (based on our measurements of test samples). After preparation, samples were analyzed in the Paleomagnetics Laboratory at the University of Florida. In order to determine the dominant magnetic mineral, isothermal remanent acquisition experiments were carried out on selected samples (Fig. 3.). Magnetization was measured in a magnetically shielded room and measurements were made on a SCT cryogenic magnetometer. Demagnetization was done on Schonstedt alternating field (AF) and thermal demagnetizers. As is standard procedure, at least one sample from each locality was demagnetized in a stepwise regime in alternating fields of between 5 and 50 mT (milliTesla), some . r LEISEY 12.3 60 40 IN DU CE D M AG N ET IZ AT IO N (m A/ rn ) 20 0 41 1 0 1 2 3 4 APPLIED FIELD (Tesla) Figure 3. Isothermal remanence (IRM) experiment using representative example from Leisey (sample 12.3). In lower applied fields there is a progressive increase in magnetization until about 1 T. Thereafter, regardless of the increase in applied field, the induced magnetization remains the same, i.e. it is saturated. Behavior such as this is characteristic of low-coercivity minerals and probably represents fine-grained magnetite. 112 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL 37, PT. I, No. 3 were demagnetized further up to 100 mT. A second sample from each site was also subject to a stepwise regime of thermal demagnetization, usually in 12 or more steps between 100° and 620° C (Fig. 4). The subsequent treatment of the other samples from each site was based on the results of the first and second samples. PALEOMAGNETIC RESULTS As exemplified in Figure 3, the results of the IRM acquisition experiments indicate that the dominant magnetic mineral in the Leisey samples saturates in relatively low fields of about 1 T or less. This kind of behavior is characteristic of low coercivity magnetic minerals and is probably carried in fine-grained magnetite. This observation is important to note, because, as such, the results from the AF demagnetization should be as valid as those of the thermal demagnetization. As it turned out, the polarity data are consistent from both of these laboratory techniques and suggest the lack of a high-coercivity component attributable to minerals such as hematite or geothite, which are oftentimes encountered in terrestrial sediments. Intensities of magnetization lie within the general range that would be expected from fine-grained sedimentary magnetite. Prior to demagnetization, the mean NRM magnetization was 0.5 mA/m (N=17). After demagnetization, mean intensities ranged from 0.1 mA/m at 40 mT (N=9), to 0.14 mA/m at 500° (N=6), to 0.05 mA/m at 600' C (N=6). The large drop in intensity between 500° and 600° further suggests that magnetite is the dominant carrier of the paleomagnetic remanence. The paleomagnetic data from the Leisey sites are summarized in Table 1. Values were taken from single steps at higher demagnetizations, thereby minimizing low-coercivity of low-blocking temperature components of the NRM. With the possible exception of 11.2, all samples are of negative inclination, and those with declinations lie within the southern hemisphere. As such, all of the paleomagnetic data indicate a reversed polarity for the Bermont Formation, and they yield a unit mean of dec. 157.24°, inc. -34.97° using the statistical method of Fisher (1953). Comments on the Quality of the Paleomagnetic Data.- It is clear from the data presented in Table 1 that the quality of the paleomagnetic data is less than what would be hoped for. For samples 1 and 3, this could not be avoided because, as mentioned above, they were taken from sediment samples encased in plaster jackets. The data from the other samples (10-12) are also less than desirable, with only two samples per site giving results from which the polarity could be interpreted. In particular, the original data from these sites show considerable dispersion in declinations, and this possibly resulted from post-mortem rotation of the Mercenaria bivalves after acquisition of remanence. Despite these problems, MACFADDEN: LEISEY SHELL PIT MAGNETIC POLARITY 113 A SAMPLE 1.1 - AF DEMAGNETIZATION N(UP) V H NAM 10 65mT# WI , , i ,-E 8(DOWN) B SAMPLE 1.3 - THERMAL DEMAGNETIZATION N(LP) H NRM V 1000 400°C w j .1~ EV EACH DIVISION 0.25 mA/M 8(DOWN) Figure 4. Plots ofthe behavior oflhe horizontal and vertical components of magnetization after alternating field (A) and thennal (B) dernagnet,7-,tion from Leisey. Although the declination is unoriented (see text) the vertical component indicates reversed polarity. The trend to the origin in higher demagnetization steps is interpreted to indicate removal of secondary components and isolation of the stable, "characteristic" component of magnetization that is used to interpret the polarity of the sediments at, or soon after, time of deposition- 114 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL 37, PT. I, No. 3 Table 1. Paleomagnetic data from the Lizisey Shell Pit used to interpret the site polarities. Also see Figure 2 for location ofsamples Site Sample Treatment Declination Inclination Polarity 1 .1 55 mT NAl -33.8 Reversed 55 mT NA -32.2 Reversed .3 400° C NA -10.5 Reversed .4 30 mT NA -58.3 Reversed .5 70 mT NA -70.3 Reversed .6 250« C NA -3.3 Reversed Leisev 3A 3 .2 350® C NA -31.2 Reversed .4 350' C NA -32.2 Reversed .6 350« C NA -31.2 Reversed (Other samples .1,.3, and .5 gave inconsistent results) 10 .1 40 mT 115.6 -14.0 Reversed .2 590' C 102.2 -10.2 Reversed 11 .1 35 mT 185.4 -4.0 Reversed .2 550' C 270.5 32.3 ?Normal 12. .1 60 mT 224.9 -58.6 Reversed -2 500' C 200.4 -43.2 Reversed 10-12 Combined (except 11.2) 157.2 -34.97 Reversed (N=5, R=3.54, a95=57.400, k=2.73) 1 1 however, the data indicate that reversed polarity predominates in the Leisey section. In its most useful form, magnetic polarity stratigraphy is the technique of establishing the pattern of reversals in a local section and correlating it to the time scale. One might argue that, in situations such as those encountered at the Leisey Shell Pit and other recently analyzed late Cenozoic deposits that contain less than optimal sampling regimes and only single polarity patterns (e. g., Australia, Whitelaw 1991; Jones et al. 1991), paleomagnetism is of dubious geochronological value. On the contrary, magnetic polarity data, particularly if reversed (i.e, suggesting the lack of a present-day overprint) and if taken in conjunction with other, associated geochronological data increase our ability to make valid correlations to the time scale. MACFADDEN: LEISEY SHELL PIT MAGNETIC POLARITY 115 MAGNETIC POLARITY TIME SCALE 0 (Ma) 0 i -0.5 0.73 m THICKNESS LEISEY (Meters) POLARITY 5- - - 1.0 Z O - 0 LU 1.4CO acr UJ - 1.5>ul ,1.66 0 ~ H .-> 0 0 OLDUVAl lilli W 1.88 i < - 30- ' %-2,0 2.47 - r GAUSS Figure 5. Correlation to the time scale (using Berggren et al. 1985). Given the reversed polarity represented at Leisey, as well as the other age constraints provided by the Irvingtonian age and Sr isotopic determinations (see text and other articles in this volume), the Bermont Formation and Leisey Shell Pit correlate to the late Matuyama Reversed Chron, and are therefore younger than 1.66 Ma (post Matuyama) and have an upper age limit (based on the Sr confidence limits) of about 1.4 Ma. DISCUSSION: CORRELATION TO THE TIME SCALE As also discussed in the summary article by Morgan and Hulbert (this volume), several lines of evidence are relevant to establishing the precise age of the Leisey Shell'Pit and its contained fauna: (1) the mammalian fauna is of late early Irvingtonian North American Land Mammal age, which constrains the age to between 1.6 Ma to about 1.0 Ma (Lundelius 1987); (2) the Sr isotopic age determinations for the interbedded molluscs are about 1.75 Z 0.35 Ma (Jones, this 116 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 37, Fr. L No. 3 volume); and (3) the entire section at Lcisey is interpreted to be of reversed polarity. Given these constraints, the Bermont Formation and fossil mammals from the Leisey Shell Pit correlate best to the middle part of the Matuyama Reversed Chron above the Olduvai Normal Subchron (Fig. 5). L]TERATURE CITED Berggren, W. A., D. V. Kent, J. J. Flyim, and J. A Van Couvering. 1985. Cenozoic geochronology. Geol. Soc. Amen Bull. 96:1407-1418. Fisher, R. A 1953. Dispersion on a sphere. Proc. Royal Soc., London. 217:295-305. Jones, D. S., B. J. MacFadden, S. D. Webb, P. A Mueller, D. A Hodell, and T. M. Cronin. 1991. Integrated geochronology of a classic Pliocene site in Florida: Linking marine and terrestrial biochronologies. J. Geol. 99:637-648. L.undelius, E. L (Chairman). 1987. The North American Quaternary sequence. Pp. 211-235 in M. 0. Woodburne (ed.). Cenozoic Mammals ofNorth America: Geochronology and Biostratigraphy. Univ. California Press. Berkeley. Whitelaw, M. J. 1991. Magnetic polarity stratigraphy of Pliocene and Pleistocene fossil vertebrate localities in southeastern Australia. Geol. Soc. Amer. Bull. 103:1493-1503.