那些觉得我胆子大的朋友,她们不知道其实我有多怕. 为了能去见识四年一回世界级的古生物盛会,我必须和来自世界各国的学者交流我自己的研究成果.想象自己站在台上用英语娓娓道来,我就觉得申请口头报告一定是自己熬夜熬多了……

Reconstructing the diet of extinct species is a fundamental goal in vertebrate paleobiology. Dental calculus is the calcium phosphate deposits on teeth, which captures a large number of food particles and contains the food information of ancient animal. It is possible to explore ancient animal dietary and investigate the environmental information from dental calculus. A large number of starch granules were found in dental calculus of Dicros gansuensis excavated from the earliest Late Miocene Guonigou fauna of the Linxia Basin, Gansu Province. It is the first time that starch granules from Late Miocene were found. We classified the ancient starch granules by its morphological characteristics and also analyzed the starch granules of leaves of modern plants around the living environment of D. gansuensis to find modern starch granules and compare them with ancient ones. The results indicate that D. gansuensis may eat not only shrubs leaves, such as the plant from Caprifoliaceae, but also some tree leaves from Juglandaceae (walnut), and maybe some herbs from Ranunculaceae and Polygonaceae. It is consistent with the results of previous morphological studies, and it also provides more information. Our study extends the range of time and object for dental calculus research, and offers more possibilities for research on feeding habits of ancient mammals.

Snow cover, as the most widely distributed element in the cryosphere, plays a critical role in the climate change and hydrological cycle.Microwave remote sensing is an important technique to monitor snow cover, because of its all-weather, all-time capability and ability to penetrate.In this study, FY-3Csatellite's passive microwave brightness temperature data acquired by FY-3C MWRI, snow cover products obtained by MERSI and VIRR, MOD10C1 and MOD11C1, are used to develop a new Snow identification algorithm in western China.In this algorithm, the passive microwave brightness temperature of different land types are firstly extracted, and then they are analyzed using cluster analysis.The analysis results exhibit that TB19V-TB19 H, TB19V-TB37 V, TB22V, TB22V-TB89 V, (TB22V-TB89V) - (TB19V-TB37V) can be used as the criterion for identifying snow cover from other scatters.Finally, MODIS snow cover products are used to validate the identification accuracy as a reference, and the results show that the overall accuracy of this algorithm in western China is 87.1％, the omission rate is 4.6％, the commission rate is 23.3％.The overall accuracy of Grody algorithm is 78.6％, the omission rate is 9.8％, and the commission rate is 30.7％.The accuracy of this algorithm is higher than the Grody algorithm.The Kappa coefficient of this algorithm is 47.3％, which is higher than the Grody algorithm's Kappa coefficient of 39.9％, indicates that the algorithm's snow identification results are more consistent with the MODIS snow product identification results.