D. Salon's Current Research

Although many developing cities experience traffic congestion and severe vehicular air pollution problems, most residents of these cities do not own cars. Most developing city residents are already making the transportation choices we hope to encourage in developed cities. They live in densely populated areas, and extensively use non-motorized modes of transportation such as walking and bicycling. Where it exists, transit is popular as well.

The problem is that in developing cities today, these choices are being made largely for economic reasons and not because dense, car-free living is actually more convenient and comfortable. As the residents of these cities grow richer, the challenge for local planners is to provide for housing and transportation options that support an increasing standard of car-free living. Developing cities present an opportunity for local planners to guide their transportation-land use systems towards an environmentally and socially sustainable future of low car ownership and use.

Since starting my current position at the Earth Institute, one of the main foci of my research has been travel behavior and land use in Nairobi, Kenya - a city where all of the above is true. My work in Nairobi is divided into two distinct projects. One of them is an analysis of data from a medium-scale household travel survey that was collected in 2004 by our Kenyan partners at the Kenya Institute for Public Policy Research and Analysis (KIPPRA). Two of my main collaborators on this project are Elliott Sclar, Director of the Center for Sustainable Urban Development and Eric Aligula of KIPPRA. The second project focuses on the travel behavior of Nairobi's slumdwellers using a more general household survey data set collected by the World Bank, also in 2004. My collaborator on this project is Sumila Gulyani, assistant professor in the urban planning program at Columbia. The first analysis will be a mode and location choice model, methodologically similar to my work on New York City. The slumdweller analysis will be a simpler mode and destination choice model, reflecting the available data. What is exciting about the slumdweller analysis is that this may be the best data ever collected specifically on slumdwellers. We are hoping to use it to better understand the needs of this underserved population. To find out more about this project, look in the papers section of this website!

My dissertation research used data from New York City to create a discrete choice model of the joint choices of residential location, car ownership status, and transport mode for both work and non-work trips. The model is spatially-explicit, allowing me to link the statistical choice model to a GIS framework. In the GIS environment, I can identify the spatial distribution of choice behavior across the city, as well as simulate spatially-explicit policy scenarios. In my view, this spatially-explicit simulation is the most exciting aspect of this research. This combined statistical-GIS model is able to produce not only information about the likely behavioral response to transportation and land use policy changes, but also a map of where the people whose behavior is most sensitive to policy are located within New York City. My dissertation advisers were Dan Sperling, Susan Handy, and Jim Wilen. To find out more about this project, check out my papers!

In 2000, the city of Bogotá embarked on a grand land use and transportation system experiment. The city government, under then-mayor Enrique Peñalosa, implemented a series of new policies regarding the use of public space and began to make substantial investments in public infrastructure - all of which aimed to increase the standard of car-free living for Bogotanos. These investments have continued under subsequent city governments, and Bogotá has been transformed into what some would call a model city for the 21st century.

The specifics of the transformation of Bogotá include taking back the sidewalks for people, building approximately 350 kilometers of bicycle paths - many of them through poor neighborhoods - and building the TransMilenio bus rapid transit (BRT) system. The TransMilenio is a city-wide system that offers speed and convenience similar to that of an underground metro. Buses run in dedicated lanes, and riders purchase tickets as they enter covered bus-stops. The planned system is huge - with 400 kilometers of dedicated trunk routes plus feeder buses - and is still under construction. The first lines opened in December of 2000, and additional lines have opened approximately at 9 month intervals since then. Currently, there are over 80 kilometers of dedicated busways and approximately 500 kilometers of feeder bus routes. After only 6 years of operation, the system moves more than 1 million passengers each day (Cain et. al., 2006).

Working with our Colombian partners at CERAC, Darby Jack (a fellow post-doc here at the Earth Institute) and I are using data from a labor market survey to test the following hypothesis: by reducing commuting times and thus expanding the radius within which the poor can search for jobs, BRT systems help improve the job prospects for the poor. Check back here to see what we find!

The policymaker who would like to encourage walking as a transportation mode to promote public health has a problem. Walking behavior, car ownership, and residential location choice are integrally linked. All else equal, the more money people have, the more cars they will buy. The more cars they buy, the less they walk, bike, and use transit. The less they walk, bike, and use transit, the less people will consider local access to services and public transit in their residential location choice. The less urban residents consider local access to services and public transit in their residential location choice, the more spread out our cities become.

The close linkage between the choices of non-motorized transport modes, car ownership, and residential location makes it difficult to disentangle the effects of policy on any one of these choices. This problem has been described in the literature as the "self-selection" problem. Self-selection is the idea that perhaps people who choose to live in more traditional neighborhoods (higher density, pedestrian friendly places) and/or not to own cars make these choices because they like to walk.

How much of the relationship between the built environment and the choice to walk can be explained by self-selection? This research project endeavors to answer this question using a discrete choice model of the joint choice of walking, car ownership, and residential location. The explicit inclusion in the model of the choice of residential location makes it possible to separate the effect of the built environment on location choice and the effect of the built environment on the choice to walk. Details of the initial analysis are available in Chapter 5 of my dissertation. I'm continuing this work together with Pat Mokhtarian and Susan Handy - additional papers will be available soon!

	CURRENT AND RECENT RESEARCH Transport in Nairobi, Kenya NYC Mode-Location Choice Transit and Job Access in Bogotá, Colombia NYC Focus on Walking		Photo by Nathan Parker 2006.
			Over the course of my career in transport-related research, my personal research program has evolved from a focus on the energy used in the transport sector to a focus on greenhouse gas emissions from the sector to a focus on urban travel behavior and land use. This evolution is the outcome of my attempt to both contribute to the academic literature in a meaningful way and to study issues that I feel are important for us as a society to understand better. I have not switched my research focus to travel behavior and land use because I no longer am concerned about our global environment. On the contrary, I am deeply concerned about our environment. And the way we use our land and choose to travel have an enormous impact on the environment. In addition, urban land use and travel behavior can have a large impact on urban human communities and the quality of life that urban residents enjoy. The rest of this web page describes a number of my past and current research projects so you can read a bit about them without having to read a whole paper. To go directly to a particular project, click on the link!