超级计算机上矩阵乘的并行计算与实现

数值预报系统中经常要用到矩阵乘运算。在分布式超级计算机 (如IBM-SP) 上, 矩阵乘的并行计算需要较多的数据移动, 有效的数据传输对矩阵乘的实现至关重要。该文讨论了两种矩阵乘的并行算法, 一种是基于矩阵的列-行划分方式, 一种是基于矩阵的网格划分方式。在IBM-SP计算机上的实验结果表明, 网格划分的矩阵乘并行算法通讯开销更小, 并行效率更高, 其并行加速比较列-行并行算法改善约10 %。     

起伏地形下我国太阳直接辐射空间制图

建立了起伏地形下太阳直接辐射分布式计算模型 ;成功地解决了起伏地形中地形相互遮蔽对太阳直接辐射影响的难题 ;采用数据集群技术 ,探讨了不同数据集情况下太阳直接辐射计算模式的时空有效性 ;以 1km× 1km分辨率的DEM数据作为地形的综合反映 ,完成了我国 1km× 1km分辨率各月气候平均太阳直接辐射空间分布制图     

Application of Distributed Temperature Sensing for coupled mapping of sedimentation processes and spatio‐temporal variability of groundwater discharge in soft‐bedded streams

The delineation of groundwater discharge areas based on Distributed Temperature Sensing (DTS) data of the streambed can be difficult in soft‐bedded streams where sedimentation and scouring processes constantly change the position of the fibre optic cable relative to the streambed. Deposition‐induced temperature anomalies resemble the signal of groundwater discharge while scouring will cause the cable to float in the water column and measure stream water temperatures. DTS applied in a looped layout with nine fibre optic cable rows in a 70 × 5 m section of a soft‐bedded stream made it possible to detect variability in streambed temperatures between October 2011 and January 2012. Detailed monthly streambed elevation surveys were carried out to monitor the position of the fibre optic cable relative to the streambed and to quantify the effect of sedimentation processes on streambed temperatures. Based on the simultaneous interpretation of streambed temperature and elevation data, a method is proposed to delineate potential high‐groundwater discharge areas and identify deposition‐induced temperature anomalies in soft‐bedded streams. Potential high‐discharge sites were detected using as metrics the daily minimum, maximum and mean streambed temperatures as well as the daily amplitude and standard deviation of temperatures. The identified potential high‐discharge areas were mostly located near the channel banks, also showing temporal variability because of the scouring and redistribution of streambed sediments, leading to the relocation of pool‐riffle sequences. This study also shows that sediment deposits of 0.1 m thickness already resulted in an increase in daily minimum streambed temperatures and decrease in daily amplitude and standard deviation. Scouring sites showed lower daily minimum streambed temperatures and higher daily amplitude and standard deviation compared with areas without sedimentation and scouring. As a limitation of the approach, groundwater discharge occurring at depositional and scouring areas cannot be identified by the metrics applied. Copyright © 2015 John Wiley & Sons, Ltd.     

Heterogeneous surface displacement pattern at the Hatchobaru geothermal field inferred from SAR interferometry time-series

We estimated surface displacements using persistent scatterer SAR interferometry (PS-InSAR) around the Hatchobaru geothermal field, Japan, from 18 ALOS/PALSAR images acquired from July 2007 to December 2010. Generally, geothermal fields, covered with natural targets such as rocky terrain and vegetation, have been one of the difficult targets for PS-InSAR analysis. However, we applied space adaptive filtering to increase the number of pixels for measuring surface displacement. The results of our analysis demonstrate ground subsidence with decaying velocity over the observation period around the geothermal field. The spatial pattern of ground subsidence includes sharp boundaries of subsidence that can be interpreted as fault traces. We demonstrated the usefulness of PS-InSAR analysis with the space adaptive filtering to estimate surface displacements with high spatial resolution and high spatial density around a geothermal field.     

ArcTOP:TOPKAPI与GIS紧密连接的分布式水文模型系统

基于遥感（RS）和地理信息系统（GIS）的分布式水文模型已成为当前水文科学发展的前沿，地理信息建模系统是目前地理信息系统研究的热点问题之一。目前实现通用GIS空间分析功能与水文模型的集成主要有四种途径：水文模型嵌入CIS平台．GIS功能嵌入水文模型．水文模型和GIS的松散连接以及水文模型和CIS的紧密连接。本文提出的ArcTOP分布式水文模型系统，运用ArcViewGIS提供的宏语言，将以物理概念为基础的分布式水文模型TOPKAPI和CIS紧密连接，实现分布式水文模型和GIS的完全集成。     

基于线弹性位错模型反演1997年西藏玛尼Mw7.5级地震的干涉雷达同震形变场——Ⅱ滑动分布反演

1997年11月8日西藏M_w7.5级玛尼地震是干涉雷达技术应用于地震观测以来的一次重要事件.在第一部分中，我们应用广泛使用的Okada线弹性位错模型，假设断层的各个分段滑动量均匀，反演得到断层各个分段的几何参数和均匀滑动量.本部分的反演进一步去除滑动均匀假设，并利用更能反映断层真实状态的角形元位错模型（线弹性），在第一部分反演得到断层几何的基础上，反演断层面的静态位错分布.反演结果表明，线弹性滑动分布模型能够更好地解释观测数据，进一步提高反演的数据拟合程度.最终得到了断层面上的走滑和倾滑位错分布.首次得到的断层面滑动分布显示断层面滑动在浅部（0～12 km）比较集中，地震破裂长度约170 km，最大左旋走滑位移达4.8 m；反演结果还表明局部段落存在较大倾滑位移，量值达到1.9 m，这在断层模型中是不能忽略的，它可能是断层两侧形变不对称的原因之一；反演得到的标量地震矩为2.18×10²⁰ N·m，相当于矩震级M_w7.5，与Velasco等利用地震波形反演得到的结果一致.     

基于Powerbuilder+SQL Server的天津市机井管理信息系统的开发及其分布式应用

以SQL Server为后台数据库管理机井数据，并把机井数据库作为主要研究对象，以Powerbuilder为前台开发工具，利用其提供的优良性能与丰富界面。开发了机井管理信息系统，可对机井信息进行科学管理与分布式应用。     

流域水文模型计算域离散方法

常用的概念性水文模型 ,能够很好地模拟水文时间变化过程 ,但没有考虑水文变量和水文参数的空间变化与空间不均匀性。随着空间数据的获取手段的增多以及空间离散技术的发展 ,考虑水文参数和水文变量空间变化的分布式水文模型得到了极大的发展。本文详细介绍了分布式流域水文模型中用到的几种不同计算域离散方法 ,并讨论了河道汇流模型中常用到的有结构网格和无结构离散网格。地理信息系统技术对计算域离散有辅助作用 ,其有利于无结构离散网格的自动生成和交互修改 ,并可结合遥感技术 ,使水文模型能获取精确的空间分布的水文参数和水文变量。     

内陆河流域分布式日出山径流模型——以黑河干流山区流域为例

应用常规的气象水文数据并结合GIS,建立了一个适合西北干旱区内陆河山区流域的以日为步长的分布式径流模型,并对黑河干流山区出山径流进行了模拟计算和讨论。模型以子流域作为最小的产流、汇流单元,根据植被覆盖将各子流域分为裸地区、乔木区、牧草区和冰川区,并根据实际调查将土壤分为 3层,各分区单独进行水量平衡计算。产流过程以土壤储水能力和储水量表征,而储水能力和储水量等则由土壤的孔隙度、干密度和厚度等表征。入渗原理基于土壤储水率平衡原理,并考虑重力势的作用。实际蒸散发与蒸发力和土壤体积含水量的乘积成正比,不同的土壤和植被具有不同的调节参数。模拟结果表明,模拟效果较差的原因是区域日降水过程具有较大的随机性,难以用有限的站点合理计算区域日降水量。寻找一个合适的区域日降水量计算方法是目前少资料大型流域分布式水文模型模拟成功的关键。     

Capturing the essential spatial variability in distributed hydrological modelling: Infiltration parameters

Selecting the correct resolution in distributed hydrological modelling at the watershed scale is essential in reducing scale-related errors. The work presented herein uses information content (entropy) to identify the resolution which captures the essential variability, at the watershed scale, of the infiltration parameters in the Green and Ampt infiltration equation. A soil map of the Little Washita watershed in south-west Oklahoma, USA was used to investigate the effects of grid cell resolution on the distributed modelling of infiltration. Soil-derived parameters and infiltration exhibit decreased entropy as resolutions become coarser. This is reflected in a decrease in the maximum entropy value for the reclassified/derived parameters vis a vis the original data. Moreover, the entropy curve, when plotted against resolution, shows two distinct segments: a constant section where no entropy was lost with decreasing resolution and another part which is characterized by a sharp decrease in entropy after a critical resolution of 1209 m is reached. This methodology offers a technique for assessing the largest cell size that captures the spatial variability of infiltration parameters for a particular basin. A geographical information system (GIS) based rainfall-runoff model is used to simulate storm hydrographs using infiltration parameter maps at different resolutions as inputs. Model results up to the critical resolution are reproducible and errors are small. However, at resolutions beyond the critical resolution the results are erratic with large errors. A major finding of this study is that a large resolution (1209 m for this basin) yields reproducible model results. When modelling a river basin using a distributed model, the resolution (grid cell size) can drastically affect the model results and calibration. The error structure attributable to grid cell resolution using entropy as a spatial variability measure is shown.