空间数据统计分析的思想起源与应用演化

系统总结了空间数据统计分析的发展历程,并分为五个时期：① 早期孕育（计量革命之前）,其重要思想是19世纪初德国的区位论;② 计量革命（1950-1960年代）,主要是经典统计学的应用和理论探索;③ 空间统计学（1970-1980年代）,重点是空间点数据、面数据和空间连续性数据的分析;④ 成熟与扩散（1990-2000年代）,空间数据统计分析发展成熟并快速向其他领域扩散;⑤ 时空大数据（2010年以后）。换句话说,计量革命开始后的空间数据统计分析大约每20年有重要的新技术或方法出现,到现在已经具有成熟、系统化的方法和显著的社会效益。而在当前的时空大数据时期,其发展需要计算机科学家、统计学家和地理学家等不同学科领域人员的共同努力。     

Using geovisual analytics to compare the performance of geographically weighted discriminant analysis versus its global counterpart,linear discriminant analysis

Geographically weighted spatial statistical methods are a family of spatial statistical methods developed to address the presence of non-stationarity in geographical processes, the so-called spatial heterogeneity. While these methods have recently become popular for analysis of spatial data, one of their characteristics is that they produce outputs that in themselves form complex multi-dimensional spatial data sets. Interpretation of these outputs is therefore not easy, but is of high importance, since spatial and non-spatial patterns in the results of these methods contain clues to causes of underlying non-stationarity. In this article, we focus on one of the geographically weighted methods, the geographically weighted discriminant analysis (GWDA), which is a method for prediction and analysis of categorical spatial data. It is an extension of linear discriminant analysis (LDA) that allows the relationship between the predictor variables and the categories to vary spatially. This produces a very complex data set of GWDA results, which include on top of the already complex discriminant analysis outputs (e.g. classifications and posterior probabilities) also spatially varying outputs (e.g. classification function parameters). In this article, we suggest using geovisual analytics to visualise results from LDA and GWDA to facilitate comparison between the global and local method results. For this, we develop a bespoke visual methodology that allows us to examine the performance of global and local classification method in terms of quality of classification. Furthermore, we are also interested in identifying the presence (or absence) of non-stationarity through comparison of the outputs of both methods. We do this in two ways. First, we visually explore spatial autocorrelation in both LDA and GWDA misclassifications. Second, we focus on relationships between the classification result and the independent variables and how they vary over space. We describe our visual analytic system for exploration of LDA and GWDA outputs and demonstrate our approach on a case study using a data set linking election results with a selection of socio-economic variables.     

An ex-ante analysis of housing location choices due to housing displacement: The case of Bristol Place

Urban renewal has been criticized for its unintended effects on the displacement of residents to housing situations where they are worse off. The literature concerned with housing displacement focuses on after the fact (ex post) analysis of mobility, hence providing only partial insights to prospective outcomes for residents. In this article, we develop a methodology for predicting before the fact (ex ante) spatial location outcomes for residents facing forced displacement. We ask where in the city will displaced residents locate and how do potential new locations compare to the current one? We apply this methodology to an ongoing urban renewal project in Illinois in order to predict spatial relocation patterns. Our findings show that residents facing forced displacement are likely to relocate to low-priced neighborhoods that are on average as bad or worse off in terms of housing adequacy, accessibility to labor, and shopping. These findings not only bear important implications for urban redevelopment policy, but also help to reveal spatially determined costs of moving which are not factored into current redevelopment impact evaluations.     

基于氢氧同位素和水化学的祁连山老虎沟冰川区径流过程分析

基于2012-2013年两个消融期在祁连山老虎沟冰川区连续2 a采集的冰川融水径流、雪冰以及降水样品,分析探讨了冰川区水体介质中氢氧同位素和水化学要素(主要化学离子、pH值、TDS和电导率等)在消融期的变化过程及特征。结果表明:祁连山老虎沟雪冰融水中的氢氧同位素值(δD和δ¹⁸O)表现出明显的消融期随月份波动,先升高再降低的趋势,在7月份表现出高值,反映了冰川消融强弱程度的变化过程。冰川径流中同位素含量与冰雪融水接近,且处于当地降水线上,其主要来自冰雪融水和降雨补给。老虎沟冰川融水径流水化学主要表现为Ca-Na-HCO₃-SO₄和Ca-Mg-HCO₃-SO₄型,其组成特征也表现出随消融过程而变化。对氢氧同位素和化学要素组成在消融期(6~9月)随时间的变化过程进行了分析,表明结合冰川区氢氧同位素和化学要素(包括化学离子、TDS、pH值和电导率等)的组成可以区分雪坑和新雪、河水的组分变化,可以反映冰川融水径流在消融期的变化过程。     

Scale-dependent analysis of satellite imagery for characterization of glacier surfaces in the Karakoram Himalaya

Information regarding process-structure relationships and change in the Karakoram Himalaya is of great importance in studying glacier hydrollogy, mass balance, and dynamic environmental change. Such information is not readily available. Detailed spatio-temporal assessment requires field investigation coupled with quantitative remote sensing studies. We conducted an investigation of the large Batura Glacier in Pakistan to determine if spectral variability can be quantified and used to characterize glacier surfaces. Specifically, SPOT Panchromatic satellite data were evaluated for differentiating features of glacier structure resulting from ice movement, ablation, and supraglacial fluvial action. Image semivariogram analysis was conducted. for assessing spectral variability patterns and fractal analysis was used to examine scale-dependent variation in the data. Results indicate that spectral variability from fields of ice seracs can exhibit fractal characteristics, although most surface features on the glacier exhibit a change in the fractal dimension over different ranges in scale. The fractal dimension was found to be useful for differentiating between glacier surfaces such as white ice and debris-covered ice. Characteristics of the debris-load and the scale-dependent nature of calculating the fractal dimension ultimately determined the potential of class separability.     

Exploratory analysis of the interrelations between co-located boolean spatial features using network graphs

Visual data mining of spatial data is a challenging task. As exploratory analysis is fundamental, it is beneficial to explore the data using different potential visualisations. In this article, we propose and analyse network graphs as a useful visualisation tool to mine spatial data. Due to their ability to represent complex systems of relationships in a visually insightful and intuitive way, network graphs offer a rich structure that has been recognised in many fields as a powerful visual representation. However, they have not been sufficiently exploited in spatial data mining, where they have principally been used on data that come with an explicit pre-specified network graph structure. This research presents a methodology with which to infer relationship network graphs for large collections of boolean spatial features. The methodology consists of four principal stages: (1) define a co-location model, (2) select the type of co-association of interest, (3) compute statistical diagnostics for these co-associations and (4) construct and visualise a network graph of the statistic from step (3). We illustrate the potential usefulness of the methodology using an example taken from an ecological setting. Specifically, we use network graphs to understand and analyse the potential interactions between potential vector and reservoir species that enable the propagation of leishmaniasis, a disease transmitted by the bite of sandflies.     

Metadata and provenance for spatial analysis: the case of spatial weights

Within a CyberGIS environment, the development of effective mechanisms to encode metadata for spatial analytical methods and to track the provenance of operations is a key requirement. Spatial weights are a fundamental element in a wide range of spatial analysis methods that deal with testing for and estimating models with spatial autocorrelation. They form the link between the data structure in a GIS and the spatial analysis methods. Over time, the number of formats for spatial weights implemented in software has proliferated, without any standard or easy interoperability. In this paper, we propose a flexible format that provides a way to ensure interoperability within a cyberinfrastructure environment. We illustrate the format with an application of a spatial weights web service, which is part of an evolving spatial analytical workbench. We describe an approach to embed provenance in spatial weights structures and illustrate the performance of the web service by means of a number of small experiments.     

Simulation and construction of the glacier mass balance in the Manas River Basin,Tianshan, China from 2000 to 2016

The glacier mass balance(GMB) is an important link between climate and water resources and has remarkable regulatory functions in river runoff. To simulate changes of the GMB and to analyze the recharge rates of glacier meltwater to runoff in the Manas River Basin(MRB) during 2000–2016, MOD11 C3, TRMM 3 B43 and other multi-source remote sensing data were used to drive the degree-day model. The results showed that:(1) the accuracy of the remote sensing meteorological data can be corrected effectively by constructing the temperature and precipitation inversion models, and the characteristics of glacial climate can be finely described through downscaling. The average annual temperature was –7.57 °C and the annual precipitation was 410.71 mm in the glacier area of the MRB. The zone at an altitude of about 4200 m was a severe climate change zone, and above and below that zone, the temperature drop rates were –0.03°C/100 m and –0.57°C/100 m, respectively, while precipitation gradients were –2.66 mm/100 m and 4.89 mm/100 m, respectively.(2) The overall GMB was negative with a cumulative GMB of up to –9811.19 mm w.e. and the average annual GMB fluctuated between –464.85 and –632.19 mm w.e. Besides, the glacier melted slowly during 2000–2002 and 2008–2010, but rapidly for 2002–2008 and 2010–2016, while the most serious loss of the glacier occurred in 2005–2009. Moreover, the vertical changes of the GMB increased at 244.83 mm w.e./100 m in the ablation zone but only at 18.77 mm w.e./100 m in the accumulation zone.(3) The intraannual runoff strongly responded to the change of the GMB especially in July and August when the loss of the GMB accounted for 75.4% of the annual loss, and when runoff accounted for 55.1% of the annual total. Due to differences in the annual precipitation and snow meltwater outside the glacier, the interannual glacier meltwater recharge rates fluctuated between 19% and 31%. The recharge rate of glacier meltwater to runoff in the MRB was close to that for other basins in the Tianshan Mountains, which may be used as a basis to confirm the reliability of the estimated GMB results. Furthermore, based on the present findings, it is recommended that the research community pursue studies on the GMB in other alpine river basins.     

我国医学生死亡教育实施及研究述评

The degree-day model is one important method to estimate glacier melt, which is based on the specific relationship between glacial melting and the sum of daily mean temperatures above the melting point. According to the observation data on the Koxkar Glacier(KG) from 2005 to 2010, we analyzed the temporal and spatial variation of degree-day factors(DDF) and its influential factors. The results indicate that the average value of DDF was 7.2~10.4 mm/(°C·d) on the KG from 2005 to 2010. It showed a decreasing trend between 3,700 m and 4,200 m, and the deceasing trend was more obvious in the upper part of the KG. On a spatial scale, the DDF increased evidently with increasing altitude. The DDF ranged from3.6 to 9.3 mm/(°C·d) at 3,700 m a.s.l., with the average value of 9.3 mm/(°C·d). It varied from 6.9 to 13.0 mm/(°C·d) at4,000 m a.s.l., with the average value of 10.2 mm/(°C·d). During the period of ablation, the fluctuation of DDF was not significant at the lower altitude(3,700 m a.s.l.), but it decreased at the higher altitudes(4,000 m a.s.l. and 4,200 m a.s.l.).The debris changes the transmission of heat, which accelerates the melting of a glacier; and the DDF showed high value.This paper will provide the reference for temporal–spatial parameterization schemes of DDF on Tuomuer glaciers of the Tianshan Mountains.     

Contextual neural gas for spatial clustering and analysis

This study aims to introduce contextual Neural Gas (CNG), a variant of the Neural Gas algorithm, which explicitly accounts for spatial dependencies within spatial data. The main idea of the CNG is to map spatially close observations to neurons, which are close with respect to their rank distance. Thus, spatial dependency is incorporated independently from the attribute values of the data. To discuss and compare the performance of the CNG and GeoSOM, this study draws from a series of experiments, which are based on two artificial and one real-world dataset. The experimental results of the artificial datasets show that the CNG produces more homogenous clusters, a better ratio of positional accuracy, and a lower quantization error than the GeoSOM. The results of the real-world dataset illustrate that the resulting patterns of the CNG are theoretically more sound and coherent than that of the GeoSOM, which emphasizes its applicability for geographic analysis tasks.     

上海及周边主要城镇城市用地扩展空间特征及其比较

基于多时段TM遥感影像资料,运用缓冲分析法对上海市区及其周边主要城镇城市用地扩展的时空特征进行了分析和比较。研究结果表明:1)区域城市土地利用扩展过程主要受距中心市区(CBD)的距离的影响;2)城市用地扩展在距CBD10km范围内主要表现为上海市区扩展的空间异向性,距CBD10km范围之外城市扩展的异向性则源于不同时期、不同方向上周边城镇的异速扩展过程,并受河流、渠系等小尺度地貌格局差异的影响;3)上海主要郊区城镇扩展可分为标准型、被动扩展型、平缓扩展型和不规则扩展型。     

A representativeness-directed approach to mitigate spatial bias in VGI for the predictive mapping of geographic phenomena

Volunteered geographic information (VGI) contains valuable field observations that represent the spatial distribution of geographic phenomena. As such, it has the potential to provide regularly updated low-cost field samples for predictively mapping the spatial variations of geographic phenomena. The predictive mapping of geographic phenomena often requires representative samples for high mapping accuracy, but samples consisting of VGI observations are often not representative as they concentrate on specific geographic areas (i.e. spatial bias) due to the opportunistic nature of voluntary observation efforts. In this article, we propose a representativeness-directed approach to mitigate spatial bias in VGI for predictive mapping. The proposed approach defines and quantifies sample representativeness by comparing the probability distributions of sample locations and the mapping area in the environmental covariate space. Spatial bias is mitigated by weighting the sample locations to maximize their representativeness. The approach is evaluated using species habit suitability mapping as a case study. The results show that the accuracy of predictive mapping using weighted sample locations is higher than using unweighted sample locations. A positive relationship between sample representativeness and mapping accuracy is also observed, suggesting that sample representativeness is a valid indicator of predictive mapping accuracy. This approach mitigates spatial bias in VGI to improve predictive mapping accuracy.     

空间分析方法在房地产市场研究中的应用--以北京市为例

基于北京市2003年普通住宅数据,利用空间分析中点模式分析、空间自相关分析和空间插值方法等,对北京市房地产,尤其是普通住宅的空间格局进行分析。研究表明,北京市房地产发展在空间上具有强烈的集聚特点,而房价的空间自相关特性也非常明显。空间分析方法提供了准确认识、评价和综合理解空间位置和空间相互作用的手段,为定量研究空间格局提供了支持。在房地产发展等社会经济现象研究中,空间分析方法强调了“位置”因素的重要性,是刻画房地产空间格局的理想工具。     

巢湖西湖岸新石器-商周遗址空间分布规律及其成因

将地理信息系统空间分析方法引入巢湖西湖岸新石器-商周遗址考古工作,通过点密度分析、空间距离分析、三维分析、缓冲区分析等方法研究遗址空间分布规律及其影响因素。研究表明自新石器至商周时期研究区内遗址时空分布呈现:随时间推移由湖岸边逐渐向西北部扩展,后迁移至南部,再均匀扩散的遗址迁移轨迹;先民多选择靠近水源、地势平坦、土壤肥沃的自然岗地、河谷阶地、山麓面居住,导致古遗址空间上大致呈线状、团聚状、分散状等分布特征,具有明显的河谷谷地指向性、阶地岗地指向性、土壤指向性等规律。提出遗址分布在早期可能主要受到气候水文、地貌、植被土壤等自然因素影响,后期生产力发展水平、经济生活方式等人文因素影响加重。本研究为GIS支持下区域考古研究提供了合适的研究实例,同时GIS方法得到一些推论假设仍需要田野考古调查与发掘等进一步佐证。     

Using spatial analysis and Bayesian network to model the vulnerability and make insurance pricing of catastrophic risk

Vulnerability refers to the degree of an individual subject to the damage arising from a catastrophic disaster. It is affected by multiple indicators that include hazard intensity, environment, and individual characteristics. The traditional area aggregate approach does not differentiate the individuals exposed to the disaster. In this article, we propose a new solution of modeling vulnerability. Our strategy is to use spatial analysis and Bayesian network (BN) to model vulnerability and make insurance pricing in a spatially explicit manner. Spatial analysis is employed to preprocess the data, for example kernel density analysis (KDA) is employed to quantify the influence of geo-features on catastrophic risk and relate such influence to spatial distance. BN provides a consistent platform to integrate a variety of indicators including those extracted by spatial analysis techniques to model uncertainty of vulnerability. Our approach can differentiate attributes of different individuals at a finer scale, integrate quantitative indicators from multiple-sources, and evaluate the vulnerability even with missing data. In the pilot study case of seismic risk, our approach obtains a spatially located result of vulnerability and makes an insurance price at a finer scale for the insured buildings. The result obtained with our method is informative for decision-makers to make a spatially located planning of buildings and allocation of resources before, during, and after the disasters.     

A spatial analysis of the geographic distribution of musculoskeletal and general practice healthcare clinics in Auckland,New Zealand

Despite the high incidence of musculoskeletal health problems in New Zealand, the use of specialised musculoskeletal healthcare providers appears to be limited to a small subsection of society. Analysis of the spatial distribution of musculoskeletal and general practice clinics in Auckland, New Zealand was undertaken to understand the geographical component of the issue of musculoskeletal healthcare accessibility. The locations of all physiotherapy, chiropractic, osteopathic, podiatry, acupuncture and general practitioner (GP) clinics in urban Auckland were mapped and analysed using a combination of spatial statistical tools. Results showed a contrast between the distribution of the predominantly privately-funded musculoskeletal clinics, which appeared clustered in central areas of the city, and predominantly publicly-funded GP clinics, which appeared more evenly spread across the study area. Several physical and social environmental variables appeared to be correlated with clinic location, including clinic proximity to major roads and urban centres, and residential socioeconomic status and ethnicity. This study represents a first step in understanding the underlying causes of the demographic divide between users and non-users of musculoskeletal healthcare. These results may help to inform the development of strategies to improve the accessibility of musculoskeletal healthcare services for people living or working in areas with low provision of musculoskeletal clinics.     

北京市长安街沿线的扒窃案件高发区分析及防控对策

本文利用空间分析方法,对北京市长安街沿线的扒窃类案件的空间分布特征进行了研究。首先分析了长安街沿线的犯罪分布状况,发现该长安街沿线共存在3个异常明显的犯罪高发区域,分别对应于西单商业圈(A)、建国门路段(B)和大望路—四惠路段(C),通过将长安街沿线的常住人口与兴趣点(Point of Interest, POI)分布进行比较,发现案件与POI密度较高的区域存在较高的相关性。随后,利用核密度估计方法和时空热点矩阵方法分析了3个异常明显的案件高发区域内的犯罪活动时空分布模式,结果表明：高发区域A的犯罪高发时段为上午10时至下午18时,其中峰值出现在正午12时,犯罪活动主要集中在区域内的一些大型购物场所内;高发区域B和C的犯罪高发时间均为上午6时和下午18时前后,与早晚交通高峰时间比较一致,犯罪活动主要集中在区域内比较重要的公共交通站点附近。最后,根据长安街沿线案件高发区域内犯罪活动的时空规律性特征,提出了犯罪防控措施与建议。     

鲁甸地震灾后重建的综合地理分析与对策研讨

地震灾区是地表各圈层相互作用最剧烈、人地关系最紧张的区域。深入开展综合地理研究对于协调这一区域的人地关系,促进区域可持续发展有着重大的意义。本文首先对鲁甸灾前的资源环境特点与区域发展特征进行了评价和分析,发现该地区生态环境脆弱、人口密度大、贫困面广、生产生活对资源的依赖程度高、资源环境长期处于超载状态。进而,针对性地提出了尽早启动资源环境承载能力评价,科学制定重建规划,体制机制创新,以及开展整个青藏高原边缘地带及近邻地区的防灾减灾系统研究和整体规划等对策建议。