Mapping uncertain geographical attributes: incorporating robustness into choropleth classification design

ABSTRACT

Choropleth mapping provides a simple but effective visual presentation of geographical data. Traditional choropleth mapping methods assume that data to be displayed are certain. This may not be true for many real-world problems. For example, attributes generated based on surveys may contain sampling and non-sampling error, and results generated using statistical inferences often come with a certain level of uncertainty. In recent years, several studies have incorporated uncertain geographical attributes into choropleth mapping with a primary focus on identifying the most homogeneous classes. However, no studies have yet accounted for the possibility that an areal unit might be placed in a wrong class due to data uncertainty. This paper addresses this issue by proposing a robustness measure and incorporating it into the optimal design of choropleth maps. In particular, this study proposes a discretization method to solve the new optimization problem along with a novel theoretical bound to evaluate solution quality. The new approach is applied to map the American Community Survey data. Test results suggest a tradeoff between within-class homogeneity and robustness. The study provides an important perspective on addressing data uncertainty in choropleth map design and offers a new approach for spatial analysts and decision-makers to incorporate robustness into the mapmaking process.     

Color map design for visualization in flood risk assessment

Visualizations of flood maps from simulation models are widely used for assessing the likelihood of flood hazards in spatial planning. The choice of a suitable type of visualization as well as efficient color maps is critical to avoid errors or bias when interpreting the data. Based on a review of previous flood uncertainty visualization techniques, this paper identifies areas of improvements and suggests criteria for the design of a task-specific color scale in flood map visualization. We contribute a novel color map design for visualizing probabilities and uncertainties from flood simulation ensembles. A user study encompassing 83 participants was carried out to evaluate the effects of this new color map on user’s decisions in a spatial planning task. We found that the type of visualization makes a difference when it comes to identification of non-hazardous sites in the flood risk map and when accepting risks in more uncertain areas. In comparison with two other existing visualization techniques, we observed that the new design was superior both in terms of task compliance and efficiency. In regions with uncertain flood statuses, users were biased toward accepting less risky locations with our new color map design.     

DEM误差对滑坡危险性评价模型的影响

基于数字高程模型(DEM)计算得到的坡度、坡向等地形属性是滑坡危险性评价模型的重要输入数据, DEM误差会导致地形属性计算结果不确定性, 进而影响滑坡危险性评价模型的结果。本文选择基于专家知识的滑坡危险性评价模型和逻辑斯第回归模型, 采用蒙特卡洛模拟方法, 研究DEM误差所导致的滑坡危险性评价模型结果不确定性。研究区位于长江中上游的重庆开县, 采用5 m分辨率的DEM, 以序贯高斯模拟方法模拟了不同大小(误差标准差为1 m、7.5 m、15 m)和空间自相关性(变程为0 m、30 m、60 m、120 m)的12 类DEM误差场参与滑坡危险性评价。每次模拟包括100 个实现, 通过对每次模拟分别计算滑坡危险性评价结果的标准差图层和分类一致性百分比图层, 用以评价结果不确定性。评价结果表明, 在不同的DEM精度下, 两个滑坡危险性评价模型所得结果的总体不确定性随空间自相关程度的变化趋势并不相同。当DEM空间自相关性程度不同时, 基于专家知识的滑坡危险性评价模型的评价结果总体不确定随着DEM误差增加而呈现不同的变化趋势, 而逻辑斯第回归模型的评价结果总体不确定性随着DEM误差大小增加而单调增加。从评价结果总体不确定性角度而言, 总体上逻辑斯第回归模型比基于专家知识的滑坡危险性评价模型更加依赖于DEM数据质量。     

圣保罗大都市区社会空间分异研究——多元统计方法在城市连绵区的应用

空间数据和地理信息系统在城市规划和决策中应用的重要性日见凸显。主要原因在于：重要的人口数据和社会变动经常表现出一定的空间特性,这种特性可以通过空间分析和统计方法被认识和解释。应用多元分析的空间分类方法编制圣保罗大都市区社会分异地图并进行相关分析。研究的主要数据来自2000年巴西全国人口普查,其中包括了圣保罗大都市的所有行政区和39个自治市的21774个人口普查区。为了把都市连绵区从数据全集中分离出来,我们采用混合技术进行互补分析,即在2000年4月30日的陆地卫星7号图像中绘制一个个多边形,这些被识别出来的多边形就是人口普查区。然后,通过目视解译出假彩色多边形集合。应用空间分类评分程序将这些多边形分成五类,并建立人口普查区的数目、覆盖的面积和都市连绵区之间的关系。这种多元分析方法是基于变量的均衡化来生成易于用分级统计图描述平均值,以促进可视化和后续的空间分布分析。基于多元分析的空间分类方法研究,清楚地展现了圣保罗大都市最重要的社会特征,也说明城市社会地图方法和多元分析的空间分类方法在大都市区的管理、公共政策规划和复杂决策中具有重要的应用价值。     

Increasing the spatial resolution of agricultural land cover maps using a Hopfield neural network

Land cover class composition of remotely sensed image pixels can be estimated using soft classification techniques increasingly available in many GIS packages. However, their output provides no indication of how such classes are distributed spatially within the instantaneous field of view represented by the pixel. Techniques that attempt to provide an improved spatial representation of land cover have been developed, but not tested on the difficult task of mapping from real satellite imagery. The authors investigated the use of a Hopfield neural network technique to map the spatial distributions of classes reliably using information of pixel composition determined from soft classification previously. The approach involved designing the energy function to produce a ‘best guess’ prediction of the spatial distribution of class components in each pixel. In previous studies, the authors described the application of the technique to target identification, pattern prediction and land cover mapping at the sub-pixel scale, but only for simulated imagery. We now show how the approach can be applied to Landsat Thematic Mapper (TM) agriculture imagery to derive accurate estimates of land cover and reduce the uncertainty inherent in such imagery. The technique was applied to Landsat TM imagery of small-scale agriculture in Greece and largescale agriculture near Leicester, UK. The resultant maps provided an accurate and improved representation of the land covers studied, with RMS errors for the Landsat imagery of the order of 0.1 in the new fine resolution map recorded. The results showed that the neural network represents a simple efficient tool for mapping land cover from operational satellite sensor imagery and can deliver requisite results and improvements over traditional techniques for the GIS analysis of practical remotely sensed imagery at the sub pixel scale.     

Classification of kangaroo habitat distribution using three GIS models

Abstract

In this paper, we compare three techniques for mapping wildlife habitat, termed BIOCLIM, CART and a new classification method based on nonparametric techniques. These techniques model dependent map layers of species distributions, where the areas to be mapped are large and the plot data is sparse. The techniques recognise pattern in the (independent) plot data, available to natural resource managers. In this case, the independent data set comprised 12 climate surfaces, that attempt to represent the range of temperature and precipitation important in determining the habitat of kangaroos across Australia. With this particular data set, the CART (decision tree) model was most accurate, but more time consuming to initialise. The relative performance of these models depends on the quality of the data set, and skill of the GIS analyst. Where possible, GIS analysts should implement all available methods, and compare output.     

中国西部干旱区生态景观制图的若干问题

制图方法是认识和揭示客观现象与过程的重要途径。生态景观制图作为专题制图学的一个重要方向,通过建立分类体系,制定分类方法,编制专题图件,可以反映自然要素和人为作用的特点,揭示生态景观要素的耦合关系。生态景观制图中专题要素的表现方式,制图过程的数据采集,制图单元的划分,制图的模式选择,以及遥感、GIS与数字制图手段的应用,是十旱Ⅸ景观生态制图的关键问题。生态景观制图巾的地图信息理论、地图模式论、地图分析理论、地图要素综合理论、制冈工艺学等的发展,对于开拓资源环境制图学的新领域具有理论价值,并对生态规划与生态建设具有重要促进作用。     

基于MODIS 数据的长江三角洲地区土地覆盖分类

长江三角洲地区是我国经济最发达的地区之一, 人类活动对自然环境产生了很大影响。为了研究该地区人类活动与生态环境的相互作用, 利用250 m 分辨率MODIS 数据进行土地 覆盖制图研究, 采用的主要数据为增强型植被指数EVI 数据、反射率数据和DEM 数据。通过基于时间序列的滤波方法消除EVI 的噪声, 通过PCA 变换压缩数据量, 并计算均质度来表征空间维的纹理信息, 构造了一个综合性的分类数据矩阵, 依据高分辨率影像选取了训练区, 采用最大似然法进行分类, 并采用缓冲区分析技术进行分类修正, 得到长江三角洲地区的土地覆盖分类结果。利用高分辨率影像解译信息对分类结果进行了精度评价, 并将分类结果与 MODIS 土地覆盖产品进行了对比, 精度分析表明分类结果很好的反映了研究区的土地覆盖信息, 显示了本研究分类方法与技术处理在实践中的可行性及250 m 分辨率EVI 时间序列数据在区域尺度土地覆盖分类方面的优势与潜力。     

小城镇信息图谱初探

小城镇建设将成为我国农村经济发展的一种重要模式。为了引导小城镇的可持续发展,有必要对小城镇演变的内在规律作深入研究。运用“图谱”的表达方式和GIS分析工具,从小城镇图谱的基本构成要素入手,对小城镇的群体空间、外部形态和内部结构的演变规律进行了探讨。并以泰州市城镇群、兰溪市永昌镇为例,作了实证研究。研究表明,采用图谱的表达方式有助于我们对城镇发展进行了历史总结、规律挖掘,客观地把握城镇的发展阶段,从而了解不同时期城镇工作的重点,以有的放矢地采取措施,部署长远规划。     

地理信息系统空间分析方法及其若干应用

地理信息系统中的核心是空间属性数据库，而管理，分析和应用数据库中数据的空间分析功能和应用模型则是发挥数据库作用的关键部分。衡量一个地理信息系统的好坏，不仅决定于其数据库的数据质量，数据量和更新速度，关键要看其空间功能是否强大，实用具有灵活性。一般来讲，地理信息系统的空间分析有很强的综合性和专业性，也具有很大的灵活性。     

Sensitivity analysis of a decision tree classification to input data errors using a general Monte Carlo error sensitivity model

We analysed the sensitivity of a decision tree derived forest type mapping to simulated data errors in input digital elevation model (DEM), geology and remotely sensed (Landsat Thematic Mapper) variables. We used a stochastic Monte Carlo simulation model coupled with a one‐at‐a‐time approach. The DEM error was assumed to be spatially autocorrelated with its magnitude being a percentage of the elevation value. The error of categorical geology data was assumed to be positional and limited to boundary areas. The Landsat data error was assumed to be spatially random following a Gaussian distribution. Each layer was perturbed using its error model with increasing levels of error, and the effect on the forest type mapping was assessed. The results of the three sensitivity analyses were markedly different, with the classification being most sensitive to the DEM error, than to the Landsat data errors, but with only a limited sensitivity to the geology data error used. A linear increase in error resulted in non‐linear increases in effect for the DEM and Landsat errors, while it was linear for geology. As an example, a DEM error of as small as ±2% reduced the overall test accuracy by more than 2%. More importantly, the same uncertainty level has caused nearly 10% of the study area to change its initial class assignment at each perturbation, on average. A spatial assessment of the sensitivities indicates that most of the pixel changes occurred within those forest classes expected to be more sensitive to data error. In addition to characterising the effect of errors on forest type mapping using decision trees, this study has demonstrated the generality of employing Monte Carlo analysis for the sensitivity and uncertainty analysis of categorical outputs that have distinctive characteristics from that of numerical outputs.     

长江口滨岸湿地环境信息系统的建立与应用

长江口滨岸湿地环境信息系统 (Yangtze Estuary Tidal Wetland Environmental Information System,简称YETWEIS)是一个适用于长江口滨岸湿地环境管理与决策的软件系统。该系统基于MapObjects组件技术、数据挖掘技术和数学建模方法,以Visual Basic为开发平台,功能包括对长江口滨岸湿地环境信息的显示、编辑、查询检索、信息统计、空间分析、专题地图编制和环境质量评价等。本文首先介绍了YETWEIS的体系结构、实现技术和主要功能,然后重点对专题地图编制模块和环境质量评价模块的实现思想进行了阐述,同时对水体、沉积物、大气环境质量评价方法、污染因子权重确定方法、基于熵值法的环境质量综合评价方法进行了深入探讨,并利用YETWEIS分析了长江口滨岸湿地2002年重金属和持久性有机污染物的空间分布特征,并对长江口滨岸湿地2003年环境质量进行了系统评价。     

利用NOAA-AVHRR数据进行积雪监测与制图的方法研究

本文在总结分析国内外现有方法的基础上,并根据大量的实验研究,介绍了利用NOAA－AVHRR遥感数据进行积雪监测和流域雪盖制图的方法。文章侧重介绍了AVHRR数据在积雪制图应用中的数据处理方法。也较为详细地介绍了利用遥感数据和地形数据的复合进行积雪信息提取的方法。研究表明：在地形相对复杂的地区使用AVHRR数据进行积雪监测和制图,采用监督分类的方法可获得较为理想的分类结果。利用GIS的空间分析方法,实现遥感和地形数据的复合分是积雪遥感制图高效和实用的方法之一。     

A citizen data-based approach to predictive mapping of spatial variation of natural phenomena

The vast accumulation of environmental data and the rapid development of geospatial visualization and analytical techniques make it possible for scientists to solicit information from local citizens to map spatial variation of geographic phenomena. However, data provided by citizens (referred to as citizen data in this article) suffer two limitations for mapping: bias in spatial coverage and imprecision in spatial location. This article presents an approach to minimizing the impacts of these two limitations of citizen data using geospatial analysis techniques. The approach reduces location imprecision by adopting a frequency-sampling strategy to identify representative presence locations from areas over which citizens observed the geographic phenomenon. The approach compensates for the spatial bias by weighting presence locations with cumulative visibility (the frequency at which a given location can be seen by local citizens). As a case study to demonstrate the principle, this approach was applied to map the habitat suitability of the black-and-white snub-nosed monkey (Rhinopithecus bieti) in Yunnan, China. Sightings of R. bieti were elicited from local citizens using a geovisualization platform and then processed with the proposed approach to predict a habitat suitability map. Presence locations of R. bieti recorded by biologists through intensive field tracking were used to validate the predicted habitat suitability map. Validation showed that the continuous Boyce index (B_cont(0.1)) calculated on the suitability map was 0.873 (95% CI: [0.810, 0.917]), indicating that the map was highly consistent with the field-observed distribution of R. bieti. B_cont(0.1) was much lower (0.173) for the suitability map predicted based on citizen data when location imprecision was not reduced and even lower (?0.048) when there was no compensation for spatial bias. This indicates that the proposed approach effectively minimized the impacts of location imprecision and spatial bias in citizen data and therefore effectively improved the quality of mapped spatial variation using citizen data. It further implies that, with the application of geospatial analysis techniques to properly account for limitations in citizen data, valuable information embedded in such data can be extracted and used for scientific mapping.     

Mapping transit‐based access: integrating GIS,routes and schedules

Seabed sediment textural parameters such as mud, sand and gravel content can be useful surrogates for predicting patterns of benthic biodiversity. Multibeam sonar mapping can provide near-complete spatial coverage of high-resolution bathymetry and backscatter data that are useful in predicting sediment parameters. Multibeam acoustic data collected across a ～1000 km² area of the Carnarvon Shelf, Western Australia, were used in a predictive modelling approach to map eight seabed sediment parameters. Four machine learning models were used for the predictive modelling: boosted decision tree, random forest decision tree, support vector machine and generalised regression neural network. The results indicate overall satisfactory statistical performance, especially for %Mud, %Sand, Sorting, Skewness and Mean Grain Size. The study also demonstrates that predictive modelling using the combination of machine learning models has provided the ability to generate prediction uncertainty maps. However, the single models were shown to have overall better prediction performance than the combined models. Another important finding was that choosing an appropriate set of explanatory variables, through a manual feature selection process, was a critical step for optimising model performance. In addition, machine learning models were able to identify important explanatory variables, which are useful in identifying underlying environmental processes and checking predictions against the existing knowledge of the study area. The sediment prediction maps obtained in this study provide reliable coverage of key physical variables that will be incorporated into the analysis of covariance of physical and biological data for this area.     

Remote sensing mapping of the regeneration of coastal natural habitats in Singapore: Implications for marine conservation in tropical cities

Rapid urbanization has resulted in the loss of coastal and marine habitats in cities worldwide. The effective conservation of urban coastal ecosystems requires detailed knowledge of their spatial distribution, necessitating high-resolution mapping. Our study produces a high-resolution coastal and marine habitat map and shoreline map for the tropical city-state of Singapore created through pixel-based supervised classification of satellite imagery, bathymetry data and expert ground knowledge. These maps can be used as a base reference for multiple applications including ecological research, conservation and urban planning. They also help identifiy trends in the extent of key coastal habitats, providing insight into their differing levels of vulnerability to loss and potential for restoration to ensure long-term resilience. The method used for mapping shoreline typologies and resulting insights gained, can guide other rapidly urbanizing coastal cities on strategies to assemble useful spatial knowledge for effective conservation of their urban coastal ecosystems.     

A case study of model selection and parameter inference by maximum likelihood with application to uncertainty analysis

One of the uses of geostatistical conditional simulation is as a tool in assessing the spatial uncertainty of inputs to the Monte Carlo method of system uncertainty analysis. Because the number of experimental data in practical applications is limited, the geostatistical parameters used in the simulation are themselves uncertain. The inference of these parameters by maximum likelihood allows for an easy assessment of this estimation uncertainty which, in turn, may be included in the conditional simulation procedure. A case study based on transmissivity data is presented to show the methodology whereby both model selection and parameter inference are solved by maximum likelihood.     

基于GIS技术的泥石流风险评价研究

为了满足对自然灾害预测不断增长的紧迫要求，泥石流风险评价成为帮助决策过程重要的基础工具之一。即使泥石流风险性各组分的评价很困难，但地理信息系统可辅助提出这种风险性制图的有关方法。我们以云南省为研究区，选取6个成因因子参与泥石流危险度敏感性分析，通过将研究区易损性评价图与危险性评价图叠加分析，编制出云南省泥石流风险评价图。该图描述了在现有自然条件和人类活动下的泥石流风险敏感区。研究成果为全面反映灾情，确定减灾目标，优化防御措施，进行减灾决策提供了重要依据。     

三峡库区土地资源的多源信息综合制图法

本方法用来编制1:10万长江三峡库区土地资源评价图,以卫星数据计算机监督分类为基础,配合以野外调查和遥感、非遥感的资料。把这些资料作为多源信息,加以综合后,编制出的土地资源评价图,经实地校核,与SPOT卫星像片、卫星数据计算机监督分类图相套合,准确率分别为88,91,93(％)。     

基于地理信息元组的分类预测理论及应用——以中亚矿集区预测为例

地理信息科学的迅速发展为矿集区预测提供了新的理论依据,使地理信息理论、相应的技术方法组合及应用的整体研究成为可能。在大量理论和实验研究的基础上提出了基于地理信息元组的分类预测理论体系,该理论体系由六个概念、一个空间信息本质、两个空间信息特性以及六个空间信息相关方法和约定的构成。并在该理论体系基础上结合空间数据分析方法和探索式空间数据分析方法,建立了基于地理信息元组的分类预测模型。应用该模型在中亚区域实现了Au矿集区预测目标。为中亚成矿域中其它区域(如新疆)的矿集区预测研究奠定了理论与方法论基础。