EARTH SCIENCES CENTRE G?TEBORG UNIVERSITY C21 1999 EARLY PALEOGENE WARM CLIMATES AND BIOSPHERE DYNAMICS INTERNATIONAL MEETING JUNE 9-13,1999 G?TEBORG, SWEDEN ABSTRACT VOLUME Edited by Fredrik P. Andreasson, Birger Schmitz, and Elisabet I. Thompson :rk Department of Marine Geology G?TEBORG 1999 PLANT-INSECT ASSOCIATIONS RESPOND TO PALEOCENE-EOCENE WARMING Peter Wilf and Conrad C. Labanddra Department of Paleobiology, National Museum of Natural Histoiy, Smithsonian Institution, Washington, DC 20560-0121, USA Terrestrial plants and insects today comprise the majority of the Earth's biomass and biodiversity, and almost half of insect species are herbivores. Consequently, imderstandii^ how plant-insect associations respond to warming events is a l?tal componet? of global change studies. Nevertheless, there have been no investigations of plant-insect interactions in the context of past climate change. The fossil record offers a unique opportunity to examine plant-insect response over long time intervals tiirough analysis of insect dam^e occurring on fossil plants. Deposits bearir^ fossil plants simultaneously provide data on host taxonomy, paleoclimate. and stratigraphy, to which insect damage data can be tied directly. In modem insect faunas, decreasing latitude is associated with several broad trends, including increased diversity per host plant and increased herbivore pressure, the latter expressed in higher attack frequency. We used insect damage on fossil plants to test for these trends at constant latitude, in the context of the Paleocene-Eocene global warming inter\*al. We also examined whether herbivoty and its increase more affeaed abundant hosts, as predicted by observations of extant associations, and addressed herbivore turnover. The paleobotanicai record from the Great Divide, Green River, and Washakie basins of southwestern Wyoming, USA, includes diverse and abundant floral assemblages from the Paleocene-Eocene interval that contain well- preserved insect damage. Speci?cally, we compared two floral samples, from the latest Paleocene (Clarkforkian) and the mid-early Eocene (mid-Wasatchian) (Wilf, submitted). Both samples were originally deposited on humid, swampy floodpl??ns, allowing an "isotaphonomic" approach that helps to factor out biases such as depositional re^me, paleotopography, and past moisture levels {Behrensmeyer & Hook 1992). Previous analysis of these samples showed that from the latest Paleocene to the mid-early Eocene: (a) mean annual temperatures rose from 14.4*0 ? 2.5^ to 21.rC ? 2J2?C; (b) plant species turnover exceeded 80%; (c) all dominant plant taxa were replaced; (d) alpha, beta, and ^rama diversity increased significantiy; and (e) an immigraion of thermophiiic plants occurred (Wilf, submitted). Presence-absence data were tabulated for 39 Paleocene and 49 Eocene species of terrestrial flowering plants at 49 Paleocene and 31 Eocene localities. Forty-two types of insect damage were identified. A database was constructed in which the presence or absence of each damage type was scored for each species in each of the two samples. In addition, we field- censused the four plant localities with highest diversity and best preservation in the Paleocene and Eocene for insect damage on dicot leaves (two each. 749 Paleocene and 791 Eocene specimens). This analysis permiued the observed damage frequencies to be related as directly as possible to the relative abundance of host plants in the source forests. The census data VMU.Y PAUOCE?IE WARM CLfMATES AND BtOSPt' (24), higher than the more abundant Paleocene confemilial, Corylites (18). As the plant wnth the highest herbivore diversit>" and attack frequency, we hypothesize Almts to have been apparent and rapidly grovving. with limited investment in chemical defenses. Alnus palatability in the Eocene was probably enhanced by elevated leaf nitrogen content resulting from an actinorhizal association with nitrogen-fixing symbionts, as in all modem Alnus. The Paleocene-Eocene floral turnover appears to have accompanied a turnover in the herbivore fauna. Twenty-four percent of damage types made their last appearances in the Paleocene, whereas 28% made their first appearances in the Eocene. Many of these damage types are generalized and relatively more likely to have been produced by unrelated insects. However, if only the 27 most specialized damage types are counted, which are each more likely to correspond to restricted taxonomic groups of herbivore culprits, the results are similar Paleocene last appearances are 26% and Eocene first appearances 33%. The obsen.aiions of rare, specialized damage types are derived from intensively sampled hosts. t.-\.UV F.M.l U?KSi; W.-WIM Cl.I-NtMES ASD BIOSPHbRI: DVNAMiCS. E.ARTH SCIENCES CKNTRE. G?TE?OR? t^SlVARSITY. C21 !?* which supports our view t?iat iheir turnover is not a saniplipg arti&ct. The percentages given above should be regarded as minima g^ven the difficulty of evalualmg imnover in moie generalized feeding groups. We conclude that a Paleocene-Eooene turnover of at least 25% occurred in the herbivore &una, in parallel with the cUmatic and ?oristic charige already documented in the Rocky Mountain region. References Behrcnsmeyer, A.K. & Hook, R.W., 1992: Paleoenv?ioninental contexts and laphoaom?c modes. In A.K B?ncnsmcyer, J.D. Dcunuifa, ? al. (eds.): Terrestrial Ecosystems Throigk Time. 15-136. University of Chicago Press. Feeny, PJ*., 1976: Plant apparency and choni?il defense. In J. Wallace &. RX. Mansell (eds.): Biochemical Interaction Between Plants and Insects, 1-40. Plenum Press. Wilt P.. siUxnitted: Late Paleocene-eariy Eocene climate changes in sotObwesteni Wyoming: Paleoboanical analysis. Geological Society of America Bulletin. EARLY PALEOOENE WARM CLftlATES AND BIOSPHERE DYNAMICS. EARTH SCIENCES CESTRE. COTEBCffiG UNIVERSITY, ai 1999