Michael Batty - The New Science of Cities

The New Science of Cities

The New Science of Cities

Michael Batty

The MIT Press

Cambridge, Massachusetts

London, England

2013 Massachusetts Institute of Technology

All rights reserved. No part of this book may be reproduced in any form by any electronic or mechanical means (including photocopying, recording, or information storage and retrieval) without permission in writing from the publisher.

MIT Press books may be purchased at special quantity discounts for business or sales promotional use. For information, please email .

Library of Congress Cataloging-in-Publication Data

Batty, Michael.

The new science of cities / Michael Batty.

p. cm

Includes bibliographical references and index.

ISBN 978-0-262-01952-1 (hardcover : alk. paper)

1. City planning. 2. Cities and townsGrowth. I. Title.

HT166.B38667 2013

307.1'216dc23

2013004383

10 9 8 7 6 5 4 3 2 1

Science may be described as the art of systematic over-simplificationthe art of discerning what we may with advantage omit. Karl Popper, The Open Universe: An Argument for Indeterminism (1992)

Contents

List of Figures

: System structure and hierarchy: Varieties of population dynamics: Self-similar urban morphologies from population, remotely sensed imagery, and street network representations: Coupled transport networks generating a convoluted dynamics of traffic: Idealized urban patterns generated from the bottom up using modular rules for constructing development among nearest neighbors (from Batty, 2005): Segregation from a random spatial distribution t(time) = 1 to a highly polarized but stable distribution by t = 200: Visualizing aggregate flows by volume, from the full 33 33 matrix to the largest flows, with the upper middle frame set at the flow average: The earliest digital visualization of personal trips on the 1959 Cartographatron display device: Directional flows from (a) origins to all destinations and (b) vice versa: Ravensteins (1885) Currents of Migration: Kohns (1841) flow fields: Flow volumes assigned to explicit and implicit networks: Assignments of flow to various network systems: Distorting flows in various network systems: (a) Directional symmetric flows. (b) Asymmetric flows from origins. (c) Asymmetric flows from destinations.: Flows, connectivity, and networks at different flow thresholds: Directed flows and graphs illustrating a weakly connected system: The transition from complete connection to disconnection: The bipartite flow graph and its binary equivalent: Joining bipartite graphs (a) through destinations and (b) through origins: The correlation flow graph (a) based on origins and (b) its dual based on destinations: Hierarchical clustering of differences between (a) origins based on destination flows and (b) destinations based on origin flows: A planar graph of streets in Central London with the minimal spanning tree imposed as dark lines on the graph: Cumulative distance in the planar graph: Hierarchical transport network graphs in Greater London: Betweenness centrality at the nodes defining stations on the London tube network: Betweenness centrality for the London network presented using the Harel-Koren force-directed algorithm: Network morphologies in (a) centralized and (b) decentralized cities: Exponential change: Proportionate growth: (a) the frequency distribution up to size 100 but noting that the maximum size is 12,000 and (b) the rank-size distribution over its entire range: The additive exchange model generating the Boltzmann-Gibbs distribution: A multiplicative exchange model generating a power-law-like distribution: Visualizing space-time dynamics in terms of rank shift: Possible trajectories defining the morphology of the rank clock: Rank clocks for (a) the World, (b) the United States, (c) Great Britain, and (d) Israel: (a) The Fortune 100 rank clock and (b) the persistence-decline of firms by rank, 19551994: Individual rank trajectories for selected Fortune 500 firms, 19551994: Rank clocks of the top 100 high buildings in (a) New York City and (b) the World from 1909 until 2010: Generating a lognormal distribution using proportionate effect and power-law scaling from proportionate effect with a minimum size threshold: Emergence of the rank-size distribution using proportionate effect with cut-off: Power-law scaling as the population distribution emerges: Lognormal and power-law scaling of the US population based on incorporated places, 1970 to 2000: The first, sixth, twelfth, and eighteenth top-ranked population cells and their progress through the simulation: Consistent scaling behavior for different sizes of lattice: Lognormal distributions generated from proportionate effect with diffusion: Patterns of diffusion: A hierarchy for the pattern at t = 10,000: Top-ranked cells during the 10,000-year simulation: City-size distributions for the agglomeration model at t = 1,000 and t = 10,000: Patterns of network connectivity: Rank-size distribution of the nodal network distribution: The implicit retail hierarchy in Central London: The explicit population hierarchy in Greater London: A strict hierarchy from a treelike graph of the London tube system: Conventional graph-theoretic representation of the street network: Space syntax representation: Space syntax as bipartite graphs: Variations in similarity between direct distance and indirect step-distance: The basic data for Gassin: Points and lines reflected in the matrix A : Scatter plots of access measures from the data and against the direct and indirect distance measures: Comparison of distance measures for the primal and dual problems: Surface interpolation from the line distances: Surface interpolation from the point distances: A new mapping for space syntax: Adapting surface interpolations to the building and street patterns: A traditional planar graph-street network and a space syntax representation: Dual and primal networks for the two map problems: Axial lines and junctions in the Regent Street area of Central London: (a), (b) Primal networks embedded in and across Euclidean space. (c), (d) Dual networks embedded in and across Euclidean space: Space syntax on the Manhattan grid: Uniform accessibilities: Space syntax on a fractal tree: The primal is the dual: Accessibility surfaces for the primal and dual problems from the simple planar and axial maps shown in figures 7.1(a) and 7.1(b): Key accessibility measures for the primal and dual pure syntax analysis of Gassin: The street grid for Central Melbourne with the Underground Rail Loop: Line accessibility surfaces based on (a) out-degrees, (b) step-distances, (c) weighted distances, and (d) Euclidean distances: Point accessibility surfaces based on (a) out-degrees, (b) step-distances, (c) weighted distances, and (d) Euclidean distances: Relative proximities for the primal and dual pure syntax analysis of Gassin: Relative proximities for the primal and dual analysis of Central Melbourne: Constructing a space-filling curve: The Koch snowflake curve: The hierarchy of composition in constructing a fractal: Literal hierarchies: Transport from a central source: Space-filling hierarchies: Generating clustered city growth using diffusion-limited aggregation: The organically evolving network of surface streets in Greater London classified by traffic volume: Two hundred years of urban growth in Baltimore: Simulating growth using diffusion-limited aggregation, or DLA, in the spatial landscape centered on the city of Cardiff: Cellular automata: How cells are developed: Classic CA models. (a) Nearest neighbor physical diffusion on a grid. (b) Koch-like fractal diffusion. (c) Oriented diffusion-limited aggregation: Regular diffusion using CA: Patterns reminiscent of idealized Renaissance city plans: The colonial plan for Savannah, Georgia: Windows comprising the basic interactive model template: Small multiples of graphic output from exploration of the model data: Visualizing thematic map layers, flows, and histograms using Google Earth as an external viewer linked to the desktop interface: Predictions of residential population from the model: Building scenarios on the fly: Inputting a new heavy rail line from the western airport at Heathrow to the West End of the central business district: The impact of a doubling of road travel costs for private car on location: A noncontiguous partition of the system leading to population relocation: (a) Before and (b) after accessibility by road, and (c) their ratio: Factors represented as partial solutions to the location problem: Factors represented as desirability surfaces for best locations: A comparison of different linear syntheses: The graph of the problem and its associated adjacency matrix: A maximal spanning tree and related hierarchy: Hierarchical structure based on the application of Wards algorithm with and without contiguity constraints: The resultant lattice-like structure: A comparison of the syntheses derived from the hierarchical methods: The power structure as a directed graph: The Markov process as sequential averaging: The spatial averaging in constraint map form: An illustrative example of a communications network and the convergence of opinions: A classification of design problems as Markov digraphs: Factor maps for the Alexander-Manheim highway location problem: The Alexander-Manheim hierarchical design solution: Solutions generated by the design: Weighting structures generated by the design machine: Selection of the optimal design machine from two alternative systems: The solution generated by the optimal design machine: Structure of the decision-making system: Formal relationships comprising the decision-making system: Interaction between actors through problems: Interaction between actors through policies: Interaction between actors through problems and policies: Interaction between problems: Interaction between policies: Prior and posterior distributions of power over problems: Predicted distributions from the decision-making model: Weighting of factors in the composition of policies: Observed and predicted allocations of resources: Built form and land ownership in Tolmers Square: Dimensions and sparsity of the four models: Equilibrium distribution of resources: Equilibrium distribution of investment values: Equilibrium exchange between actors: Equilibrium predictions of land ownership patterns: Range of desired and observed optimal budget allocations: Predicted optimal budget allocationsNext page