Development of a Global Population Database

A principal motivation for the NCGIA Global Demography Project is that global change research largely neglects the human dimensions of environmental degradation processes. One reason for this is the lack of suitable data sets describing socioeconomic variables consistently and with suitable detail for large areas. This point has been made convincingly in Clarke and Rhind's assessment of "Population Data and Global Environmental Change." As a first step towards increased availability of such data, we intend to develop a GIS based global population database. This work will be conducted under the umbrella of the upcoming NCGIA research initiative on Multiple Roles for GIS in Global Change Research (I-15).

Waldo Tobler is the principal investigator on this project that is supported by the Consortium for International Earth Science Information Network (CIESIN) and the Environmental Systems Research Institute, Inc. (ESRI). In addition, we are cooperating with several international institutions involved in global research, including the Global Resource Information Database of the United Nations Environment Programme (UNEP/GRID) and the World Resources Institute (WRI).

The first step in the project is to compile consistent GIS data sets of subnational administrative boundaries and associated population figures for the whole world. As much as possible we are relying on existing data sets that have been produced by various agencies and universities. From these data, continuous surfaces of population density will be developed. The conversion of the data into regular grid format has the advantage that the resulting data will be compatible with many global physical data sets. For instance, one of the statements in the chapter on Demographic Dynamics and Sustainability of Agenda 21, the Report of the U.N. Conference on Environment and Development, calls for the disaggregation of demographic data by ecological regions; a task that clearly requires data of the kind just described.

From a methodological point of view some interesting issues are involved. The problem of converting socioeconomic data collected for discrete spatial units into continuous representations has already been addressed in work related to NCGIA initiatives I-1 on The Accuracy of Spatial Databases and I-14 on GIS and Spatial Analysis. However, these approaches have so far focused on data sets of fairly small geographic extent. Assembling and reconciling many heterogeneous databases consisting of data sets collected for different time periods at different scales poses several interesting questions related to database accuracy and methods of data integration. It is clearly important to find ways of communicating the uncertainty inherent in data products to end users in the global change community. Once the data sets have been completed, they can be used in many different modeling exercises, such as predicting the population distribution for future time periods under various scenarios, or combining demographic with environmental data bases to test hypotheses concerning human induced degradation processes. Further topics for investigation include the development of alternative methods of representing global data, for example by means of spherical harmonics.

For more information contact Waldo Tobler at

This is also available as NCGIA Technical Report 95-6.

Return to population related activities at NCGIA