The Late Cretaceous environment of the Arctic: A quantitative reassessment based on plant fossils

Late Cretaceous megafossil floras from the palaeo-Arctic of northeastern Russia and northern Alaska are reviewed in respect of their age, composition, structure and floral dynamics. Palaeofloral correlations and comparisons are made between the two regions. Nine angiosperm-rich, predominantly Cenomanian to Coniacian, floras from the palaeo-Arctic are re-evaluated using Climate Leaf Analysis Multivariate Program (CLAMP) calibrated using a global gridded (0.5° × 0.5°) climate data set derived from that used in climate modelling. Additional floras from lower palaeolatitudes were used to derive latitudinal temperature gradients: seven from N. America, five from around 30 °N palaeolatitude in Europe and one from Kazakhstan. The Arctic climatic determinations, similar to previous estimates, support the existence of a northern Pacific Ocean cold gyre and a warm Arctic Ocean. At palaeolatitudes greater than 80 °N floras are insufficiently diverse in woody dicot taxa to use CLAMP, but using CLAMP-derived latitudinal temperature gradients Arctic Ocean coastal environments at 70 Ma and 82 °N, and which supported a diverse dinosaur magafauna, are predicted to have experienced a mean annual temperature of 6.3 ± 2.2 °C, a warm month mean of 14.5 ± 3.1 °C and a cold month mean no colder than − 2.0 ± 3.9 °C. All uncertainties are 2σ. The new estimates are in good agreement with a wide range of non-palaeobotanical climate proxies and render as an outlier warmer temperature estimates for the Arctic Ocean derived from the TEX86 proxy. Modelling, however, shows that land to ocean temperature gradients could have been steep. The CLAMP estimates also suggest high values for humidity and precipitation consistent with sedimentological indicators and, coupled with warm temperatures, support the existence of a persistent polar cloud cap that helped maintain high terrestrial air temperatures throughout prolonged periods (up to 5 months) of winter darkness.