UC Berkeley

Abstract

Heterogeneity within Indian cities:

Methods for empirical analysis

by

Krishnachandran Balakrishnan

Doctor of Philosophy in Landscape Architecture and Environmental Planning

University of California, Berkeley

Professor Louise Mozingo, Chair

This dissertation proposes new methods for analyzing the heterogeneity within Indian cities. The research focuses on developing methods which rely mainly on secondary datasets and demonstrates the application of the methods using data for Bangalore city. The dissertation is structured as a set of three papers, each of which focuses on a specific dimension of heterogeneity within Indian cities. The first paper uses data from the Census of India to generate sub-city typologies based on socio-economic attributes, housing quality, and access to water and sanitation infrastructure. The second paper proposes a predictive framework for high-resolution population density estimation in Indian cities. The method uses data on land-use, land-cover, street network, building height and asset ownership to predict population at a resolution of 30m. This paper also demonstrates the application of a new method for generating building height estimates at a city-wide scale using satellite stereo imagery. The third paper in this dissertation focuses on understanding heterogeneity in the volume of domestic piped water availability across parts of Bangalore city. It combines the high-resolution population mapping method of the second paper with data from the local water utility, to analyze domestic piped water availability. Using normative demand scenarios, it also estimates the deficit in domestic piped water availability and the extent of direct or indirect dependence on groundwater in a spatially disaggregated manner. The dissertation concludes with a discussion on the patterns of spatial inequality within Bangalore, as revealed by the three papers.