Abstract In the past few years the visualization and modeling of land use change over time has been carried out by summarizing the total amount, types, and locations of change but without a means to query the spatial pattern of the individual changes or investigate the probability of change in or between land cover or land use classifications. We show how the stochastic process known as Markov chains afford a powerful descriptive and predictive model for land use changes and for future land use distributions. We also present an automated interface for change analysis that employs a Geographic Information System and an application that facilitates the construction and spatial query of change mechanisms. This interactive spatial query tool, conceived for the investigation of the impact of urban growth management polices, is applied to remote sensing data for 1984 and 1988–1989 of the Chesapeake Bay region (U.S.A.). Matrix menus written in ARC/INFO Macro Language (AML) permit the spatial display of the different land cover classes and are also an excellent tool for the visualization of dynamic change and for hypothesis formulation.
[1]
John Pastor,et al.
Using Simulation Models and Geographic Information Systems to Integrate Ecosystem and Landscape Ecology
,
1992
.
[2]
Earl J. Bell,et al.
Markov analysis of land use change--an application of stochastic processes to remotely sensed data
,
1974
.
[3]
J. E. Dobson.
Commentary: a conceptual framework for integrating remote sensing, GIS, and geography
,
1993
.
[4]
David C. Streatfield,et al.
THE SANTA CRUZ MOUNTAINS REGIONAL PILOT STUDY; EARLY WARNING SYSTEM..
,
1970
.
[5]
Robert J. Naiman,et al.
Watershed Management: Balancing Sustainability And Environmental Change
,
1994
.
[6]
Jerome E. Dobson,et al.
A Coastal Land Cover Classification System for the NOAA Coastwatch Change Analysis Project
,
1993
.