Using the isotope enabled ECHAM4, GISS E and HadCM3 GCMs, the spatial distribution of mean 6180 in precipitation, mean seasonality and the correlations of 6180 in precipitation with temperature and precipitation amoun...Using the isotope enabled ECHAM4, GISS E and HadCM3 GCMs, the spatial distribution of mean 6180 in precipitation, mean seasonality and the correlations of 6180 in precipitation with temperature and precipitation amount are analyzed. The simulated results are in agreement with stable isotopic features by GNIP observations. Over East Asia. the distribution of ~180 in precipita- tion is of marked latitude effect and altitude effect. The latitude effect is covered by the continent effect in some regions. The larg- est seasonality of^lSo in precipitation appears in eastern Siberia controlled by cold high pressure, and the smallest seasonality is in the western Pacific controlled by the subtropical high. Relatively weak seasonality appears in middle latitudes where oceanic and continental air masses frequently interact. However, three GCMs show significant systematic lower ~180 for inland mid-high lati- tudes than GNIP data, which is related to the used isotopic scheme in GCMs. Temperature effect occurs mainly in inland mid-high latitudes. The higher the latitude and the closer the distance to inland is, then the stronger the temperature effect. Amount effect occurs mainly in low-mid latitudes and monsoon areas, with the strongest effect in low-latitude coasts or islands. However, three GCMs provide virtually non-existent amount effect in arid regions over Central Asia. The enrichment action of stable isotopes in falling raindrops under a cloud base, which is enlarged by these modes, is responsible for such a result. A significant difference between spatial distributions of δ^18O statistics by GCMs simulations and by GNIP observations is that the standard deviation of GCMs statistics is greater than that of GNIP statistics. In contrast, by comparing parallel time series at a single station, the standard deviations of GCMs simulations are smaller than that of GNIP observations.展开更多
基金supported by the National Natural Science Foundation of China (Grant Nos. 41171035,40871094)the Construct Program of the Key Discipline in Hunan Province (No. 2012001)+1 种基金Open Fund of Key Laboratory of Tibetan Environment Changes and Land Surface Processes,CAS (No. 2011004)Scientific Research Fund of Hunan Provincial Education Department (No. 09A056)
文摘Using the isotope enabled ECHAM4, GISS E and HadCM3 GCMs, the spatial distribution of mean 6180 in precipitation, mean seasonality and the correlations of 6180 in precipitation with temperature and precipitation amount are analyzed. The simulated results are in agreement with stable isotopic features by GNIP observations. Over East Asia. the distribution of ~180 in precipita- tion is of marked latitude effect and altitude effect. The latitude effect is covered by the continent effect in some regions. The larg- est seasonality of^lSo in precipitation appears in eastern Siberia controlled by cold high pressure, and the smallest seasonality is in the western Pacific controlled by the subtropical high. Relatively weak seasonality appears in middle latitudes where oceanic and continental air masses frequently interact. However, three GCMs show significant systematic lower ~180 for inland mid-high lati- tudes than GNIP data, which is related to the used isotopic scheme in GCMs. Temperature effect occurs mainly in inland mid-high latitudes. The higher the latitude and the closer the distance to inland is, then the stronger the temperature effect. Amount effect occurs mainly in low-mid latitudes and monsoon areas, with the strongest effect in low-latitude coasts or islands. However, three GCMs provide virtually non-existent amount effect in arid regions over Central Asia. The enrichment action of stable isotopes in falling raindrops under a cloud base, which is enlarged by these modes, is responsible for such a result. A significant difference between spatial distributions of δ^18O statistics by GCMs simulations and by GNIP observations is that the standard deviation of GCMs statistics is greater than that of GNIP statistics. In contrast, by comparing parallel time series at a single station, the standard deviations of GCMs simulations are smaller than that of GNIP observations.
