While the inversion of electromagnetic data to recover electrical conductivity has received much attention, the inversion of those data to recover magnetic susceptibility has not been fully studied. In this paper we invert frequency-domain electromagnetic (EM) data from a horizontal coplanar system to recover a 1-D distribution of magnetic susceptibility under the assumption that the electrical conductivity is known. The inversion is carried out by dividing the earth into layers of constant susceptibility and minimizing an objective function of the susceptibility subject to fitting the data. An adjoint Green's function solution is used in the calculation of sensitivities, and it is apparent that the sensitivity problem is driven by three sources. One of the sources is the scaled electric field in the layer of interest, and the other two, related to effective magnetic charges, are located at the upper and lower boundaries of the layer. These charges give rise to a frequency-independent term in the sensitivities. Because different frequencies penetrate to different depths in the earth, the EM data contain inherent information about the depth distribution of susceptibility. This contrasts with static field measurements, which can be reproduced by a surface layer of magnetization. We illustrate the effectiveness of the inversion algorithm on synthetic and field data and show also the importance of knowing the background conductivity. In practical circumstances, where there is no a priori information about conductivity distribution, a simultaneous inversion of EM data to recover both electrical conductivity and susceptibility will be required. 相似文献
Accurate rainfall distribution is difficult to acquire based on limited meteorological stations, especially in remote areas like high mountains and deserts. The Hexi Corridor and its adjacent regions (including the Qilian Mountains and the Alxa Plateau) are typical districts where there are only 30 available rain gauges. Tropical Rainfall Measuring Mission (TRMM) data provide a possible solution. After precision analysis of monthly 0.25 degree resolution TRMM 3B43 data from 1998 to 2012, we find that the correlations between TRMM 3B43 estimates and rain gauge precipitation are significant overall and in each station around the Hexi Corridor; however, the biases of annual precipitation differ in different stations and are seriously overestimated in most of the sites. Thus, Inverse Distance Weighting (IDW) interpolation method was used to rectify TRMM data based on the difference between TRMM 3B43 estimates and rain gauge observations. The results show that rectified TRMM data present more details than rain gauges in remote areas where there are few stations, alt- hough they show high coherence of distribution. Precipitation decreases from southeast to northwest on an annual and seasonal scale. There are three rainfall centers (〉500 mm) including Menyuan, Qilian and Toson Lake, and two low rain- fall centers (〈50 mm) including Dunhuang and Ejin Banner. Meanwhile, precipitation in most of the study area presents an increasing trend; especially in northern Qilian Mountains (〉5 mm/a), Badain Jaran Desert (〉2 mm/a), Toson Lake (〉20 mm/a) and Qingtu Lake (〉20 ram/a) which shows a significant increasing trend, while precipitation in Hala Lake (〈-2 mm/a) and Tengger Desert (〈-3 mm/a) demonstrates a decreasing trend. 相似文献
The continually increasing size of geospatial data sets poses a computational challenge when conducting interactive visual analytics using conventional desktop-based visualization tools. In recent decades, improvements in parallel visualization using state-of-the-art computing techniques have significantly enhanced our capacity to analyse massive geospatial data sets. However, only a few strategies have been developed to maximize the utilization of parallel computing resources to support interactive visualization. In particular, an efficient visualization intensity prediction component is lacking from most existing parallel visualization frameworks. In this study, we propose a data-driven view-dependent visualization intensity prediction method, which can dynamically predict the visualization intensity based on the distribution patterns of spatio-temporal data. The predicted results are used to schedule the allocation of visualization tasks. We integrated this strategy with a parallel visualization system deployed in a compute unified device architecture (CUDA)-enabled graphical processing units (GPUs) cloud. To evaluate the flexibility of this strategy, we performed experiments using dust storm data sets produced from a regional climate model. The results of the experiments showed that the proposed method yields stable and accurate prediction results with acceptable computational overheads under different types of interactive visualization operations. The results also showed that our strategy improves the overall visualization efficiency by incorporating intensity-based scheduling. 相似文献