基于机器学习模型的区域土壤重金属空间预测精度比较研究

为识别区域土壤重金属的空间变异特征并厘清其影响因素,本研究构建了多元线性回归（MLR）、弹性网络回归（ENR）、随机森林（RF）、随机梯度提升（SGB）、堆叠（stacking）集成模型、反向传播神经网络（BP-ANN）、基于模型平均的神经网络集成（avNNet）、线性核支持向量机（SVM-L）和高斯核支持向量机（SVM-R）共九种机器学习模型,利用山东省中部土壤重金属（Cd、Cu、Hg、Pb和Zn）和环境辅助变量数据,开展区域土壤重金属空间预测精度比较研究。结果表明：RF对五种重金属空间预测的决定系数（R²）介于0.263~0.448之间,平均绝对误差（MAE）和均方根误差（RMSE）分别小于8.408和10.636,预测值/实际值（P/O）均接近于1,对五种重金属的预测效果均较为理想,是研究区土壤重金属空间预测的最优模型;SVM-R整体预测性能仅次于RF,各项精度评价指标均相对稳健,可作为备选模型;其余七种模型的预测性能均明显低于RF和SVM-R。RF的空间预测结果显示,研究区五种重金属呈现出相似的空间分布格局,含量均由研究区东北部向西南部递减,包括东北部、北部和南部3个高值区,且高值区与当地工业–交通密集区的分布格局一致,反映出人类活动是研究区土壤重金属空间分异的主要影响因素。本研究可为区域土壤污染调查、评价和管控提供科学参考。     

Predicting the visualization intensity for interactive spatio-temporal visual analytics: a data-driven view-dependent approach

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.     

һ�ֻ���Ƶ�׷�����AR�����Ķ������ӳ�Ԥ��ģ��

?????????????????????ZTD????????????????????￡???????????????????????????????????????????????ZTD???????????????????????????ZTD???н?????????????????????????????????????????ZTD?????????????б????????????????????????в???AR?????и???????????????????????????????AR??????ZTD??????????????????????????μ??????????cm????     

AVO与流体替换相结合识别深层Ⅰ类气藏

传统的“亮点”技术适合检测Ⅲ类AVO含气异常,对于中、深层含气砂岩表现出来的Ⅰ类AVO现象无法识别。利用AVO技术,结合流体替换,寻找差异,运用G属性体,可以有效识别深层Ⅰ类AVO含气砂岩,利用这种方法预测了某油气田a主力气藏的分布。预测结果得到B井证实。     

Synthetic information prediction system for crisis mine based on GIS

Reserves of some kinds of the crisis mines will be lack now or from now on, because of lacking seriously reserves of mineral resources and the crisis of exploring bases in support. So that it is urgent to predict, appraise, development and utilize the replaceable resources of the crisis mines. The mineral resources prediction software system of synthetic information is intelligent GIS which is used to quantitative prediction of large-scale synthetic information mineral target. It takes the geological body and the mineral resource body as a unit. And it analyzes the ore deposit genesis and metallotect, knows the spatial distribution laws of the ore deposit and ore body, and establish the prospecting model based on the concept of establishing the three-dimensional space of a mine. This paper will primarily discuss some important problems as follows: the secondary development of various kinds of data(including geology, geophysical prospecting, geochemical prospecting and remote sensing, etc); process synthetically and establish the synthetic information interpretative map base; correspond prospecting model with synthetic information of ore deposit; divided into statistical units of metallogenic information synthetic anomalies based on the synthetic information anomalies of ore control, then research the metallogenic information variable of unit synthetically and make quantitative prediction according to choose the quantitative prediction math model which is suitable to the demands of large-scale precision; at last, finish the target area optimization of ore deposit (body).     

面向未来的海面变化研究

长期以来以地质时期海面变化过程等理论性研究为主要方向的海面变化研究经过７０年代在观念、方法和资料积累方面的突破，自８０年代初期以来进入了以“面向未来”为基调的成熟发展的新阶段，出现了一大批关于未来海面变化幅度及其影响和对策的研究成果，形成了较为完整的研究体系。９０年代初以来的几年间，世界海面研究的活跃领域是地质历史时期（例如ＬＧＭ，５ｅ等阶段）的海面实况、极地冰原对气候变化的反应、海面控制原理、海面模型以及绝对海面的航天测量技术与应用等方面。这一情况在一定程度上体现了“预测热”之后的“冷思考”。中国海面变化研究近年来取得了令世人瞩目的迅速发展，今后应注意引进先进技术方法，加强学科间的交叉合作，发掘自身的优势，向纵深发展     

东海区近海强风预报模式

本文探讨东海区近海强风预报模式，首先介绍建立预报模式的思路和步骤，及使用说明，为了验证强风预报模式和判别条件，用１９８６、１９８７上日本传真天气图进行试报，准确和基本准确度为９１．１％，并给出了１９８９、１９９５年随船试验预报结果，准确和基本准确率为９０％以上。     

长江口年代际冲淤演变趋势预测与应对策略

受流域来水来沙条件变化及河口大型工程建设的综合影响,长江口呈现新的冲淤格局,为预测未来演变趋势,本研究基于前期研究中建立的长江口年代际冲淤演变预测模型(Delft3D),未来情景考虑不同来沙量条件和相对海平面上升速率。预测结果表明,到2035年长江口整体以冲刷为主,口内河段主槽和浅滩边缘冲刷较明显,仅高滩局部淤积;到2050年口内河段保持净冲刷状态,拦门沙地区在现状来沙量条件下略有淤积,但在极端低来沙量条件下转变为净冲刷状态,海平面上升对拦门沙地区冲刷具有一定抵消作用,但不会使冲淤状态产生本质改变。本研究分析认为,长江口局部区域未来冲淤趋势可能对河口综合治理与保护产生不利影响,针对新格局条件下的滩槽河势稳定、重要洲滩保护、重大工程安全评估、冲刷致灾研判以及海堤防护标准再评估等方面提出了对策建议,可为新时期长江口综合治理与可持续发展提供参考。     

顾及地理-语义动态的城市热点预测框架

城市热点时空预测是城市管理和智慧城市建设的一项长期而富有挑战性的任务。准确地进行城市热点时空预测可以提高城市规划、调度和安全保障能力并降低资源消耗。现有的区域级深度时空预测方法主要利用基于地理网格的图像、给定的网络结构或额外的数据来获取时空动态。通过从原始数据中挖掘出潜在的自语义信息,并将其与基于地理空间的网格图像融合,也可以提高时空预测的性能,基于此,本文提出了一种新的深度学习方法地理语义集成神经网络(GSEN),将地理预测神经网络和语义预测神经网络相叠加。GSEN模型综合了预测递归神经网络(PredRNN)、图卷积预测递归神经网络(GCPredRNN)和集成层的结构,从不同的角度捕捉时空动态。并且该模型还可以与现实世界中一些潜在的高层动态进行关联,而不需要任何额外的数据。最终在3个不同领域的实际数据集上对本文提出的模型进行了评估,均取得了很好的预测效果,实验结果表明GSEN模型在不同城市热点时空预测任务中的推广性和有效性,利用该模型可以更好地进行城市热点时空预测,解决一系列如犯罪、火灾、网约车预订等等现代城市发展中亟需解决的相关问题。     

The simulation of LUCC based on Logistic-CA-Markov model in Qilian Mountain area,China

The Qilian mountain area was examined for using the Logistic-CA-Markov coupling model combined with GIS spatial analyst technology to research the transformation of LUCC, driving force system and simulate future tendency of variation. Results show that:(1) Woodland area decreased by 12.55%, while grassland, cultivated land, and settlement areas increased by 0.22%, 7.92%, and 0.03%, respectively, from 1986 to 2014. During the period of 1986 to 2000, forest degradation in the middle section of the mountain area decreased by 1,501.69 km~2. Vegetation cover area improved, with a net increase of grassland area of 38.12 km~2 from 2000 to 2014.(2) For constructing the system driving force, the best simulation scale was 210m×210m. Based on logistic regression analysis, the contribution(weight) of composite driving forces to land use and cover change was obtained, and the weight value was more objectively compared with AHP and MCE method.(3) In the natural scenarios, it is predicted that land use and cover distribution maps of Qilian mountain area in 2028 and 2042, and the Lee-Sallee index test was adopted. Over the next 27 years(2015–2042), farmland, woodland, grassland, settlement areas show an increasing trend, especially settlements with an obvious change of 0.56%. The area of bare land will decrease by 0.89%. Without environmental degradation, tremendous structural change of LUCC will not occur, and typical characteristic of the vertical zone of the mountain would remain. Farmland and settlement areas will increase, but only in the vicinity of Qilian and Sunan counties.