SIMULATION OF CLIMATE ACROSS THE PERMIAN-TRIASSIC BOUNDARY WITH AN EMPHASIS ON THE PHYTO-GEOGRAPHICAL DISTRIBUTIONS

An example of rapid climate change into a hothouse world are environmental fluctuations near the Permian-Triassic boundary (PTB, 251.9 ± 0.024 Ma). During this period the largest mass extinction of the Phanerozoic era occurred, causing an extinction of more than 90% of marine species and 70% of terrestrial species. The present paper focuses on deriving the paleo-climate using modeling studies as well as determining the floristic patterns using published fossil data and the paleo-biology database across the end Permian (Wuchiapingian and Changhsingian) and the early Triassic stages (Induan and Olenekian). In this study, output form PTB climate simulations from a fully coupled comprehensive model, the Community Climate System Model (CCSM4) are used to predict environmental changes affecting distribution of floral species. The modeling results for the surface air temperature show extreme seasonality over Gondwana that can be attributed to the increased continentality during the Permian, thus giving rise to Winter hemisphere in North and Summer hemisphere in South. The maximal temperature in the interior of Gondwana reached up to 50°C, whereas the lowest temperature simulated was about -20°C. The simulation of precipitation across PTB shows widespread aridity with ≤ 2mm of rainfall over most of the interior of Gondwana, except over the ITCZ and near-coastal region of the paleo-Tethys ocean, which had about 8 mm to 11 mm of rainfall per day. The evaporation exceeds the precipitation for most continental interior in the summer hemisphere and equator-ward to about 25° latitude in the winter hemisphere. The modeled biomes, classified based on Walter’s classification scheme, are compared to the paleo-reconstructions of terrestrial climate represented by the fossil data. To determine correlation of fossil genera with locality a Correspondence Analysis is applied. The fossil data used is classified into 14 coarser morphological categories with a total of 89 genera. The localities and genera that are closely related plot close together and vice versa, during the correspondence analysis that allows the estimation of variance in the data. The comparison between these data-derived biomes and modeled results will also be discussed.