Laboratory of Populations
Joel E. Cohen
Abby Rockefeller Mauzé Professor
Populations exhibit phenomena that are difficult to deduce from the characteristics of an isolated member. For example, counterintuitively, the proportion of elderly in a human population is more strongly affected by the population's birthrate than by its life expectancy. Dr. Cohen and his colleagues use mathematical tools to analyze observations of concrete problems in demography, epidemiology and ecology. They have a dual goal: first, to understand better the populations they analyze; and second, to develop new and improved conceptual tools for analyzing future population problems. Their empirical research spans studies of human population growth, infectious diseases and food webs.
A major challenge in the coming decades is to understand how demographic, economic and cultural changes interact with Earth's physical, chemical and biological environments, now and in the future. Dr. Cohen's laboratory has analyzed the spatial distribution of Earth's human population in relation to geophysical factors such as elevation, distance from coasts or navigable rivers, temperature and precipitation. Who cares about such abstract studies? Soon after the publication of initial results, a major soap manufacturer asked for details of the resulting information because the formulation of soaps depends on the altitude at which people use them. A major semiconductor manufacturer asked for the details of the resulting information because the cooling of personal computers depends on the density of air, which varies strongly with altitude. The manufacturer wanted to know the size of markets at different altitudes. These examples illustrate a repeated finding of the Laboratory of Populations: basic quantitative research on populations frequently has unexpected practical applications.
In another example of unexpected connections of basic population research with practical concerns, Dr. Cohen developed new methods to assess the uncertainty of population projections. As a result, Dr. Cohen and his colleagues were asked to serve as neutral scientific experts for federal courts to predict future claimants of asbestos-related diseases and injuries, and to assess the uncertainty of such predictions. The methods they developed and used may provide models for future mass-tort litigation as well as for long-term assessments of the impact of other environmental contaminants.
Dr. Cohen has also studied Chagas disease, an insect-borne infectious disease that is a New World relative of African sleeping sickness. Chronic Chagas disease afflicts millions of people in Latin America and has no vaccine or cure. Dr. Cohen has collaborated with Argentine colleagues in a field study of the control of Chagas disease in rural northwest Argentina. They have developed a mathematical model of the risk of household transmission to humans, which enables householders to reduce risks of infection by better household management.
Because ecological communities strongly affect human well-being, Dr. Cohen's research extends to nonhuman species as well. One approach focuses on a food web, a flowchart of who eats whom that describes the major pathways of food energy and of chemical and biological toxins. Dr. Cohen and his colleagues developed a new food web graph that plots species and feeding links in the plane spanned by species' average body mass and numerical abundance. They analyzed unique data on soil food webs to understand how environmental variables, human land uses and below-ground food webs interact.
Dr. Cohen's laboratory also studies international migration in collaboration with colleagues at the United Nations Population Division. Newly developed mathematical and statistical models make it possible to account for more than half of the variability in the annual numbers of migrants among 229 countries or regions from 1960 to 2004. These models can be incorporated in deterministic or probabilistic population projections in combination with standard demographic techniques for projecting births and deaths.
The laboratory seeks to understand how stochasticity - random influences such as changes in weather or resources, or chance events of birth and death - creates novel patterns in nonlinear dynamics of population change. Dr. Cohen and his colleagues studied contained populations of flour beetles, cannibalistic insects that have long been used to study population dynamics. From time series of counts of the numbers of larvae, pupae and adults in jars, the researchers derived "power spectra" to compare the accuracy of predictions about the beetle population generated by their model with that of predictions from the linearization methods of physics, a more traditional means of understanding randomly perturbed dynamical systems. In most cases, predictions from their model were far more successful in describing the experimental data.
Updated 2010-04-25
