Abstract Quantifying and effectively communicating the spatio-racial distribution of urban residencies is important for taking the measure of how the multiracial society organizes itself in an urban environment. Most currently used approaches to this problem center around the calculation of segregation metrics; as such, they pertain to only a single pattern's feature and they lack a compelling visualization component. In this paper, we propose a reimagined approach to spatio-racial analysis based on the concept of landscape and landscape analysis. This approach unites quantification and visualization components of the analysis. It also quantifies the entire racial topography, not just segregation. Key novel concepts are the racial landscape (RL) and the exposure matrix. RL is a high-resolution grid in which each cell contains only inhabitants of a single race. The exposure matrix tabulates adjacencies between neighboring cells weighted by the local density of adjacent subpopulations; it provides a concise quantification of the RL pattern. Two information-theoretical metrics, derived from the exposure matrix, quantify diversity, and segregation of the RL. Segregation is quantified from cell adjacencies without the need for subdivision of the region of interest. Thus, the entire region, as well as its arbitrary subregions, are RLs quantified by their diversities and segregations. Coloring cells in RL according to combinations of their race and local densities provides a natural visualization of racial topography which serves as an “observation” that provides check on numerical metrics. The RL method is described and its application is demonstrated on Cook County, IL. An implementation of the RL method in R package accompanies this paper.

中文翻译：

---

种族景观——一种基于模式的、无区域的种族地形分析和可视化方法

摘要 量化和有效地传达城市居民的空间-种族分布对于衡量多种族社会如何在城市环境中自我组织非常重要。大多数当前使用的解决此问题的方法都围绕隔离指标的计算；因此，它们仅与单个模式的特征有关，并且缺乏引人注目的可视化组件。在本文中，我们提出了一种基于景观和景观分析概念的重新构想的空间种族分析方法。这种方法结合了分析的量化和可视化组件。它还量化了整个种族地形，而不仅仅是种族隔离。关键的新概念是种族景观 (RL) 和暴露矩阵。RL 是一个高分辨率网格，其中每个单元格仅包含一个种族的居民。暴露矩阵将相邻单元格之间的邻接关系制成表格，这些单元格由相邻亚群的局部密度加权；它提供了 RL 模式的简明量化。源自暴露矩阵的两个信息理论指标量化了 RL 的多样性和隔离。隔离是从细胞邻接中量化的，无需细分感兴趣的区域。因此，整个区域及其任意子区域都是通过其多样性和隔离来量化的 RL。根据种族和当地密度的组合对 RL 中的细胞进行着色，提供了种族地形的自然可视化，作为“观察”，可以检查数值指标。描述了 RL 方法，并在伊利诺伊州库克县演示了其应用。本文附带了 R 包中 RL 方法的实现。