We report experimental observation of a sizable elastic anisotropy in a polycrystalline sample of ferromagnesian silicate in post-perovskite (ppv) structure. Using a novel composite X-ray transparent gasket to contain and synthesize ppv in a panoramic diamond-anvil cell along with oblique X-ray diffraction geometry, we observed the anisotropic lattice strain and {1 0 0} or {1 1 0} slip-plane texture in the sample at 140 GPa. We deduced the elasticity tensor (cij), orientation-dependent compressional wave velocities, polarization-dependent shear-wave velocities, and the velocity anisotropy of the silicate ppv. Our results are consistent with calculations and indicate that with sufficient preferred orientation, the elastic anisotropy of this phase can produce large shear-wave splitting. 相似文献
Large area forest inventory is important for understanding and managing forest resources and ecosystems. Remote sensing, the Global Positioning System (GPS), and geographic information systems (GIS) provide new opportunities for forest inventory. This paper develops a new systematic geostatistical approach for predicting forest parameters, using integrated Landsat 7 Enhanced Thematic Mapper Plus (ETM+) images, GPS, and GIS. Forest parameters, such as basal area, height, health conditions, biomass, or carbon, can be incorporated as a response variable, and the geostatistical approach can be used to predict parameter values for uninventoried points. Using basal area as the response and Landsat ETM+ images of pine stands in Georgia as auxiliary data, this approach includes univariate kriging (ordinary kriging and universal kriging) and multivariable kriging (co-kriging and regression kriging). The combination of bands 4, 3, and 2, as well as the combination of bands 5, 4, and 3, normalized difference vegetation index (NDVI), and principal components (PCs) were used in this study with co-kriging and regression kriging. Validation based on 200 randomly sampling points withheld field inventory was computed to evaluate the kriging performance and demonstrated that band combination 543 performed better than band combination 432, NDVI, and PCs. Regression kriging resulted in the smallest errors and the highest R-squared indicating the best geostatistical method for spatial predictions of pine basal area. 相似文献