Evaluation of a Spatial Relationship by the Concept of Intrinsic Spatial Distance

Yee Leung, Deyu Meng, Zongben XuGeographical 2013 · DOI

TL;DR

This paper introduces "Intrinsic Spatial Distance" (ISD), a novel metric that quantifies spatial relationships by considering both physical proximity and attribute similarity between locations. Unlike traditional methods, ISD accounts for complex relationships and reveals transitional paths, offering a more nuanced understanding of spatial associations.

What Problem Does This Paper Address?

Traditional spatial analysis often relies on measures like Moran's I or semivariograms, which primarily focus on spatial autocorrelation and can be influenced by Tobler's "first law of geography" (things are more related to nearby things). These methods can struggle to capture nuanced relationships where attribute similarity might outweigh physical distance, or vice versa, especially in the presence of spatial heterogeneity. The authors aim to develop a more comprehensive measure that integrates both physical and attribute closeness to provide a richer understanding of spatial relationships between any two points, regardless of their geographical proximity.

Methods

The core of the paper is the introduction of the Intrinsic Spatial Distance (ISD). This is a distance measure that considers both the physical separation ($d^g$) and attribute difference ($d^a$) between two points. The ISD is calculated by finding the "shortest spatial path" (SSP) between two points, which is defined as the path that minimizes a global measure of spatial relationship. This global measure is derived from a local measure, the $\epsilon$-spatial-distance ($d_\epsilon$), which penalizes abrupt changes in attribute values along a path. The $\epsilon$-spatial-distance itself is a function of both physical and attribute distances, with a parameter $c$ that controls the penalty for attribute variation. The algorithm for calculating ISD involves constructing a neighborhood graph and then applying shortest-path algorithms.

Datasets

The authors evaluate their ISD method using:

  • A synthetic dataset of 1,000 spatial points.
  • Two climatic datasets from China: daily cloud coverage and daily humidity (from 641 stations).
  • A gross domestic product (GDP) dataset for 287 Chinese cities.

These datasets are used for comparison with existing methods like Moran's I, local Moran's I, and semivariograms.

Key Insights and Ideas

  • Beyond Physical Proximity: The ISD highlights that spatial relationships are not solely determined by how close things are geographically. Attribute similarity can be a powerful driver of relatedness, even over larger distances.
  • Transitional Paths: The concept of a "shortest spatial path" (SSP) is crucial. This path isn't just the shortest geographical route but the one that best balances physical and attribute closeness, revealing the intrinsic connection between two points.
  • Not Biased by Tobler's Law: Unlike many traditional methods, ISD is not inherently predisposed to the idea that only nearby things are related. It can identify strong relationships between distant points if their attributes are similar.
  • Rich Information: ISD provides a more detailed account of spatial relationships than global measures like Moran's I, offering insights into both local and global contexts.

Key Results

  • The ISD measure was shown to be more elaborate than Moran's I, local Moran's I, and semivariograms in analyzing spatial relationships across synthetic and real-world datasets.
  • In climate data analysis, ISD revealed that stations at similar elevations (and thus similar microclimates) were more related in terms of cloud coverage, even if geographically distant, a pattern not fully captured by other methods.
  • For GDP data, ISD showed stronger relationships among coastal cities due to economic linkages, and identified specific transitional paths that bypassed less-developed areas.
  • The paper proves that the ISD function satisfies the conditions of a distance metric.

Limitations and Future Work

  • Single Attribute Focus: The current analysis primarily focuses on single attribute cases, although the authors note the method is generalizable to multiple attributes.
  • Extension to 3D and Temporal Domains: The authors suggest future work could extend the framework to 3-dimensional space (e.g., topographical profiles) and to incorporate the temporal domain for spatio-temporal relationships.
  • Comparison with Other Methods: Further research is suggested for thorough comparisons with other advanced spatial weighting techniques.
  • Application to Image Processing: The ISD method could be applied to image processing, such as evaluating intensity values in image filters.

Summary updated

Abstract

We propose the concept of intrinsic spatial distance ( ISD ) for the study of a spatial relationship between any two points in space. ISD is a distance measure that takes into account the separation of two points with respect to their physical and attribute closeness. We construct an algorithm to implement this concept as an ISD measurement. Based on the ISD concept, two points in space are related through a transitional path linking one to the other. As an ISD measurement decreases, the spatial relationship between two points becomes increasingly stronger. We argue theoretically and demonstrate empirically that the ISD concept is not predisposed in favor of the first law of geography, but directly considers variance of nearness in physical distance and attribute distance to derive the extent to which two points are spatially associated. Specifically, in single attribute cases, the information uncovered by an ISD measurement is more elaborate than that revealed by M oran's I , local M oran's I , and a semivariogram, giving a meticulous account of relatedness in both local and global contexts. The ISD concept is also sufficiently general to be used to study multiple attributes of relationships. Proponemos el concepto de distancia espacial intrínseca ( intrinsic spatial distance ‐ISD) para el estudio de la relación espacial entre dos puntos en el espacio. La ISD es una medición de distancia que tiene en cuenta la separación de dos puntos con respecto a su cercanía física y de atributo. Construimos un algoritmo para aplicar este concepto y crear una medición ISD. De acuerdo con en el concepto ISD, dos puntos en el espacio están relacionadas a través de un camino de transición que los vincula uno al otro. A medida que la medición ISD disminuye, la relación espacial entre dos puntos se refuerza. Argumentamos teóricamente y demostramos empíricamente que el concepto ISD no está predispuesto a favor de la primera ley de la geografía (de Tobler), pero considera directamente la varianza de la cercanía de la distancia física y la distancia del valor del atributo para derivar una medición que cuantifica el grado en el que dos puntos están asociados espacialmente. En concreto, en los casos de atributos individuales, la información revelada por una medición ISD es más sofisticada que la que proporcionada por el índice de Moran, el índice de Moran local, y el semivariograma, pues da cuanta minuciosa de la relación espacial y de atributo en contextos locales y globales. El concepto de ISD también es lo suficientemente general para ser usado en el estudio de relaciones entre múltiples atributos. 本文提出了内蕴空间距离(ISD)来研究空间中任意两点之间的空间关系。ISD是考虑空间上分离的两点各自物理和属性接近度的一种距离度量。通过构造一个算法将该概念转变为ISD可度量方式。基于ISD概念,空间上的两点通过连接两者迁移路径相关联。当ISD度量减小时,两点之间空间关系变得越来越强。理论论证和实例演示表明ISD概念并不倾向于符合地理学第一定律,但直接考虑两点间物理距离和属性距离临近性的变化可以推导两点在空间上的联系程度。特别是在单一属性案例中,相比于Moran’s I,局部Moran’s I和半方差图,ISD度量揭示了全局与局部环境中详尽的空间相关信息。同时,ISD概念在研究多属性的空间关系中也是足够通用的。
BibTeX
@inproceedings{leung2013evaluation,
  author = {Yee Leung and Deyu Meng and Zongben Xu},
  title = {{Evaluation of a Spatial Relationship by the Concept of Intrinsic Spatial Distance}},
  booktitle = {Geographical},
  year = {2013},
  doi = {10.1111/gean.12021},
}
Figure