A machine learning approach for predicting computational intensity and domain decomposition in parallel geoprocessing

ABSTRACT

High performance computing is required for fast geoprocessing of geospatial big data. Using spatial domains to represent computational intensity (CIT) and domain decomposition for parallelism are prominent strategies when designing parallel geoprocessing applications. Traditional domain decomposition is limited in evaluating the computational intensity, which often results in load imbalance and poor parallel performance. From the data science perspective, machine learning from Artificial Intelligence (AI) shows promise for better CIT evaluation. This paper proposes a machine learning approach for predicting computational intensity, followed by an optimized domain decomposition, which divides the spatial domain into balanced subdivisions based on the predicted CIT to achieve better parallel performance. The approach provides a reference framework on how various machine learning methods including feature selection and model training can be used in predicting computational intensity and optimizing parallel geoprocessing against different cases. Some comparative experiments between the approach and traditional methods were performed using the two cases, DEM generation from point clouds and spatial intersection on vector data. The results not only demonstrate the advantage of the approach, but also provide hints on how traditional GIS computation can be improved by the AI machine learning.     

An approximate method for evaluating the shear band thickness in saturated soils

A formula for the thickness of a shear band formed in saturated soils under a simple shear or a combined stress state has been proposed. It is shown that the shear band thickness is dependent on the pore pressure properties of the material and the dilatancy rate, but is independent of the details of the combined stress state. This is in accordance with some separate experimental observations. Copyright © 2004 John Wiley & Sons, Ltd.     

GPS/LEO掩星观测的变分同化技术

在简单介绍GPS/LEO掩星探测大气的发展历史和科学意义之后,详细阐述了反演的基本原理;分析了标准反演中存在的问题,并说明一维变分同化(1DVAR)在反演方法中的重要性;给出了一维变分同化中价值函数的求解,以及各种同化因子;简单介绍了对当前气象学中普遍使用的四维变分同化(4DVAR);重点讨论了各种同化方法,以及使用各种同化因子的优缺点。最后,通过CHAMP卫星的观测实例分析,验证了GPS数据在数值天气预报(NWP)中的作用,以及相对于标准反演法一维变分对气象要素的改进。     

Approach to Mountain Hazards in Tibet, China

Tibet is located at the southwest boundary of China. It is the main body of the Qinghai-Tibet Plateau, the highest and the youngest plateau in the world. Owing to complicated geology, Neo-tectonic movements, geomorphology, climate and plateau environment, various mountain hazards, such as debris flow, flash flood, landslide, collapse, snow avalanche and snow drifts, are widely distributed along the Jinsha River (the upper reaches of the Yangtze River), the Nu River and the Lancang River in the east, and the Yarlungzangbo River, the Pumqu River and the Poiqu River in the south and southeast of Tibet. The distribution area of mountain hazards in Tibet is about 589,000 km^2, 49.3% of its total territory. In comparison to other mountain regions in China, mountain hazards in Tibet break out unexpectedly with tremendously large scale and endanger the traffic lines, cities and towns, farmland, grassland, mountain environment, and make more dangers to the neighboring countries, such as Nepal, India, Myanmar and Bhutan. To mitigate mountain hazards, some suggestions are proposed in this paper, such as strengthening scientific research, enhancing joint studies, hazards mitigation planning, hazards warning and forecasting, controlling the most disastrous hazards and forbidding unreasonable human exploring activities in mountain areas.     

Optical monitoring of the quasar 4C 38.41

The results of a photometric monitoring of the quasar 4C 38.41, performed at the optical R and B bands in 2002 February–March, are presented. With a 60/90 cm Schmidt telescope at the Xinglong station of the National Astronomical Observatories of China, we observed the source exhibiting amplitude variations of up to 0.78 mag in both bands during the whole campaign. Intraday and even intranight variations are detected as well. A typical variability time-scale of about 36 d is derived from our 2-month observations at the optical bands, which is identical to that found at a radio wavelength of 92 cm, suggesting a common origin for the variations in 4C 38.41 from optical to radio bands.     

SIMULATION AND PREDICTION OF DEBRIS FLOW USING ARTIFICIAL NEURAL NETWORK

Debris flow is one of the most destructive phenomena of natural hazards. Recently, major natural haz-ard, claiming human lives and assets, is due to debris flow in the world. Several practical methods for forecasting de-bris flow have been proposed, however, the accuracy of these methods is not high enough for practical use because of the stochastic and non-linear characteristics of debris flow. Artificial neural network has proven to be feasible and use-fill in developing models for nonlinear systems. On the other hand, predicting the future behavior based on a time se-ries of collected historical data is also an important tool in many scientific applications. In this study we present a three-layer feed-forward neural network model to forecast surge of debris flow according to the time series data collect-ed in the Jiangjia Ravine, situated in north part of Yunnan Province of China. The simulation and prediction of debris flow using the proposed approach shows this model is feasible, however, further studies are needed.     

The FAST/SKA site selection in Guizhou province

Many karst depressions with diameters of 300 m to 500 m, suitable for constructing Arecibo-style radio telescopes, were identified in the south of Guizhou Province by Remote Sensing (RS) and Geographic Information System (GIS) technologies together with field investigations. Fundamental topography and landform databases were established for 391candidate depressions, and using GIS the3-dimensional images of depressions, at a scale of 1:10000, were then simulated to fit a spherical antenna. This revised version was published online in July 2006 with corrections to the Cover Date.     

平阳矾山明矾石矿床岩石学特征及其成因浅析

在野外地质工作的基础上，通过对浙江平阳研山明矾石矿床的岩石学及地球化学特征进行认真的研究，进一步证实了矾山明矾山矿床是酸性火山气热液与中酸性火山碎屑岩内的碱性长石的交代中形成的。为了解释矿床的均一性，成矿的周期性和矿床单向依变分带等重要特征，我们重点研究了矿床中明矾石化和叶蜡石化的热力学性质，得出了K 活度和pH值是控制交代产物的重要因素的新结论，提出了该矿床交代成矿作用的新模型，并据此成功地解释了矿床的一些重要特征。     

滇西北和邻区均衡重力异常特征及其壳内构造

本文以爱黎-海斯卡宁均衡补偿假设模式为基础,利用圆模板编制了滇西北及邻区均衡重力异常图。分析了异常的基本特征与壳内构造的关系,并初步探讨了均衡重力异常和强震活动的相关性。     

重力插值方法研究

为了根据离散观测数据构制连续空间重力变化图像，分析和讨论了３种数值插值方法，计算结果表明多面函数方法插值精度最高。由于逐步回归分析筛选核函数中心点的计算繁琐，文中提出根据分形理论和Ｓｈａｎｎｏｎ取样定理来确定核函数中心点。对滇西试验场进行模拟试算，插值精度可达到４～５（１０－８ｍｓ－２）。