Department of Geography and Planning, The University of New England, Armidale, Australia.
PLoS One. 2012;7(5):e37754. doi: 10.1371/journal.pone.0037754. Epub 2012 May 30.
Understanding the influence of the built environment on human movement requires quantifying spatial structure in a general sense. Because of the difficulty of this task, studies of movement dynamics often ignore spatial heterogeneity and treat movement through journey lengths or distances alone. This study analyses public bicycle data from central London to reveal that, although journey distances, directions, and frequencies of occurrence are spatially variable, their relative spatial patterns remain largely constant, suggesting the influence of a fixed spatial template. A method is presented to describe this underlying space in terms of the relative orientation of movements toward, away from, and around locations of geographical or cultural significance. This produces two fields: one of convergence and one of divergence, which are able to accurately reconstruct the observed spatial variations in movement. These two fields also reveal categorical distinctions between shorter journeys merely serving diffusion away from significant locations, and longer journeys intentionally serving transport between spatially distinct centres of collective importance. Collective patterns of human movement are thus revealed to arise from a combination of both diffusive and directed movement, with aggregate statistics such as mean travel distances primarily determined by relative numbers of these two kinds of journeys.
理解建筑环境对人类活动的影响需要从广义上量化空间结构。由于这项任务的难度，运动动力学的研究往往忽略了空间异质性，仅通过行程长度或距离来研究运动。本研究通过分析来自伦敦市中心的公共自行车数据，揭示了尽管行程距离、方向和出现频率在空间上是可变的，但它们的相对空间模式在很大程度上保持不变，这表明存在固定的空间模板的影响。提出了一种方法来描述相对运动方向、远离和围绕地理或文化意义上的位置的基本空间，从而产生两个场：一个是收敛场，一个是发散场，这两个场能够准确地重建观察到的运动空间变化。这两个场还揭示了短程旅行仅仅是从重要地点扩散和长程旅行是有意在空间上不同的集体重要中心之间运输之间的分类区别。因此，揭示了人类集体运动模式是由扩散和有向运动的组合产生的，平均旅行距离等综合统计数据主要由这两种旅行的相对数量决定。
