a challenge to science as well as to practice. Based on the foundation laid in Chapters 1 through 7, the committee concludes that responses by the discipline and by its external constituencies are needed to:
Geographers must also improve the practice of relating the ''front end" of geographic analysis—conceptualization and data selection/sampling design—with "back-end" modeling and analysis. Without thoughtful and intellectually robust linkages between these two elements of the research enterprise, geographic analysis will be inherently incomplete. At least as important, the capacity of geographic analysis to address issues of complex systems and nonlinear dynamics needs to be improved in order to fulfill geography's potential to contribute to the body of science. The improvement of capabilities for data collection and analysis should therefore be high on the discipline's research agenda.
In addition, it is important to recognize the value of utilizing a variety of methodologies in seeking better understandings of the world, combining geography's characteristic appreciation of diversity with its recognition that there is no single "foolproof method" for producing knowledge.
A particular challenge is that of analyzing and modeling relationships among natural science and human science phenomena and processes, which are so often separated by boundaries of epistemology, professional specialization, data categories, and units of measurement. Besides the technical challenges, such as relating economic and ecological indicators, this is also a challenge to individual scientists to transcend conventional boundaries for understandings of other kinds of processes and linkages.