Using the recent compilation of the isotopic composition data of surface snow of Antarctic ice sheet, we proposed an improved interpolation method of δD, which utilizes geographical factors (i.e., latitude and altit...Using the recent compilation of the isotopic composition data of surface snow of Antarctic ice sheet, we proposed an improved interpolation method of δD, which utilizes geographical factors (i.e., latitude and altitude) as the primary predictors and incorporates inverse distance weighting (IDW) technique. The method was applied to a high-resolution digital elevation model (DEM) to produce a grid map of multi-year mean δD values with lkm spatial resolution for Antarctica. The mean absolute deviation between observed and estimated data in the map is about 5.4‰, and the standard deviation is 9‰. The resulting δD pattern resembles well known characteristics such as the depletion of the heavy isotopes with increasing latitude and distance from coast line, but also reveals the complex topographic effects.展开更多
基金Under the auspices of National Natural Science Foundation of China (No. 40825017, 40576001)100 Talents Project of Chinese Academy of SciencesNational Key Technologies R&D Program of China (No. 2006BAB18B01)
文摘Using the recent compilation of the isotopic composition data of surface snow of Antarctic ice sheet, we proposed an improved interpolation method of δD, which utilizes geographical factors (i.e., latitude and altitude) as the primary predictors and incorporates inverse distance weighting (IDW) technique. The method was applied to a high-resolution digital elevation model (DEM) to produce a grid map of multi-year mean δD values with lkm spatial resolution for Antarctica. The mean absolute deviation between observed and estimated data in the map is about 5.4‰, and the standard deviation is 9‰. The resulting δD pattern resembles well known characteristics such as the depletion of the heavy isotopes with increasing latitude and distance from coast line, but also reveals the complex topographic effects.