Jie LinFollow

Date of Completion


Embargo Period



isopleth mapping; areal interpolation; remote sensing; GIS

Major Advisor

Dr. Robert G. Cromley

Associate Advisor

Dr. Daniel L. Civco

Associate Advisor

Dr. Dean M. Hanink

Associate Advisor

Dr. Chuanrong Zhang

Field of Study



Doctor of Philosophy

Open Access

Campus Access


Interpolation procedures used in isopleth mapping are closely tied to methods of areal interpolation, which is a technique used to transfer attribute information from source zones with known values to the target zones with unknown values. This research first presents and describes new local polycategorical methods for solving areal interpolation problems. Two different types of local neighborhoods for selecting observations used to estimate density coefficients are presented: one spatially based and one statistically based. The new local polycategorical methods are evaluated against a number of existing methods. Results suggest that polycategorical methods provide comparable results to the local statistical areal interpolation models, especially with enhanced land cover data sets, but with much less computational requirements.

Remotely sensed land cover data are widely used as control data in intelligent areal interpolation methods. This research then examines the usefulness of different publicly available remotely sensed land cover data sets as control data used in conjunction with different areal interpolation methods. Results show that greater detail in the land use classification of developed areas was important for areal interpolation. Even more important is the further enhancement of remotely sensed land cover categories by incorporating local road or parcel data layers. Overall weaker methods using enhanced remote sensing-derived land cover data produce more accurate results than overall stronger methods using only the original land cover data. The results also show that parcels produce better enhancements than road buffers

Finally, this research reunites areal interpolation with the isopleth mapping process to construct intelligent isopleth maps. Two types of interpolated population density surfaces are used as inputs for pycnophylactic smoothing procedures, one is target zone population density surface and the other is control zone population density surface. Results suggest that control zone population density surface-derived isopleth maps has more peaks and variations in population distributions than target zone population density surface-derived isopleth maps. And intelligent isopleth maps with control data are better than simple isopleth map derived from areal weighting results. Furthermore, intelligent isopleth maps based on parcel enhanced land cover data sets have less overall error values than those derived from land cover data sets without any enhancement.