• Journal of the Korean association of regional geographers
• /
• v.10 no.2
• /
• pp.389-406
• /
• 2004

with the increasing popularity of regional studies, the importance of regional data has been recognized dramatically in recent years. However, due to potential problems from the intrinsic characteristics of aggregate regional data for the research, and incompatible regional units between source and target regional data units, the method for reorganizing the regional data units for a given research analysis should be required. In this regard, the purpose of this research is to review the significant interpolation methods for reorganizing the data units and, based on it, to propose the population weighted interpolation method. For the first purpose, areal weighted interpolation method, pycnophylactic method, dasymetric method, area-to-point method were reviewed. The proposed population-weighted interpolation method was applied to the case study of population census regional data in Erie County, NY, compared with areal weighted interpolation method, pycnophylactic method in terms of several statistical characteristics.

• Journal of the Korean Geographical Society
• /
• v.46 no.1
• /
• pp.67-80
• /
• 2011

Choropleth mapping of population distribution is based on the assumption that people are uniformly distributed throughout each enumeration unit. Dasymetric mapping technique improves choropleth mapping by refining spatially aggregated data with residential information. Further, pycnophylactic interpolation can upgrade dasymetric mapping by considering population distribution of neighboring areas, while preserving the volumes of original units. This study proposed a combined solution of dasymetric mapping and pycnophylactic interpolation to improve the accuracy of population density distribution. Specifically, the dasymetric method accounts for the spatial distribution of population within each census unit, while pycnophylactic interpolation considers population distribution of neighboring area. This technique is demonstrated with 1990 census data of the Athens, GA. with land use land cover information derived from remotely-sensed imagery for the areal extent of populated areas. The results are evaluated by comparison between original population counts of smaller census units (census block groups) and population counts of the grid map built from larger units (census tracts) aggregated to the same areal units. The estimated populations indicate a satisfactory level of accuracy. Population distribution acquired by the suggested method can be re-aggregated to any type of geographic boundaries such as electoral boundaries, school districts, and even watershed for a variety of applications.

• Journal of Korean Society for Geospatial Information Science
• /
• v.11 no.2 s.25
• /
• pp.17-28
• /
• 2003

Since 1970s, Seoul Metropolitan Area has been growing in physical and demographic aspect. A number of new urban fringes, New City, had been particularly developed from the early of 1990s. To examine the urban growth, the population density maps are generally used to the efficient urban management tool. The density maps from political boundaries, however, were traditionally used to estimate an urban concentration, there is problems to apply directly to urban management decision making due to (i) the abrupt changes between boundaries and (ii) the inclusion of green areas and forests in these areas. The mass-preserving interpolation method, the Pycnophylactic interpolation, could provide more realistic density maps. In addition, the classified urban areas from satellite images corresponding years would turn out to be more reliable results since populations were only applied to urbanized areas. Even though the Pyconophylactic method makes the density larger, it would be useful to produce a general urban growth trend at large scale. Consequently, four different density maps are compared and reviewed for this study, and the cross-sectional analysis provided to glimpse of population density around the city center.

• Proceedings of the Korean Association of Geographic Inforamtion Studies Conference
• /
• 2003.04a
• /
• pp.191-198
• /
• 2003

서울은 인구 1000만이 넘는 세계적인 대도시로 발전하였다. 1970년대 이후로 급속적인 경제발전과 더불어 서울의 도시화는 급속히 이루어졌으며 이에 따른 인구집중 역시 빠른 속도로 진행되었다. 본 연구에서는 서울과 그 주변 도시를 대상으로 인구자료와 행정구역도 그리고 Landsat 위성영상을 사용하여 인구밀도의 변화에 따른 도시의 성장형태를 분석하여 도시의 관리와 발전계획을 위한 기초 자료로 활용하고자 한다. 도시성장 분석을 위한 기존의 연구방법은 행정구역에 따른 인구밀도를 통해 수행되었으나 이는 행정구역 내에 일률적인 인구데이터의 분배로 실제 경우와 다른 해석이 가능할 수 있는 오류를 포함하고 있다. 이에 본 연구에서는 실질적인 대부분의 인구가 도시지역에 거주한다는 가정 하에 1985년부터 2000년까지의 5년 간격의 Landsat 위성영상을 사용하여 도시지역을 추출하고 이를 기초로 행정구역 내에 포함되어 있는 도시지역에만 인구를 배분하는 새로운 방식으로 인구밀도의 변화 추이를 나타내었다. 연구결과 기존에 방법에서는 발견되지 않았던 서울의 확장형태를 알 수 있었으며 또한 인구데이터의 경계현상을 Pycnophylactic 보간 알고리즘을 통해 제거함으로써 보다 실질적인 도시지역 인구밀도의 변화를 알 수 있었다. 이러한 인구밀도의 변화는 도시의 성장과 밀접한 상관관계를 갖기 때문에 이를 통해 서울 및 주변도시의 성장 형태를 확인할 수 있었다. 마지막으로 4장의 서울-경기 도시지역의 인구밀도 변화도를 제작하여 GIS를 이용한 관련 분야의 활용에 도움이 되고자 한다.

• Journal of the Korean association of regional geographers
• /
• v.22 no.1
• /
• pp.251-266
• /
• 2016

Intelligent interpolation methods such as dasymetric mapping are considered to be the best way to disaggregate zone-based population data by observing and utilizing the internal variation within each source zone. This research reviews the advantages and problems of the dasymetric mapping method, and presents a geographically weighted regression (GWR) based method to take into consideration the spatial heterogeneity of population density - land cover relationship. The locally adaptive intelligent interpolation method is able to make use of readily available ancillary information in the public domain without the need for additional data processing. In the case study, we use the preclassified National Land Cover Dataset 2011 to test the performance of the proposed method (i.e. the GWR-based multi-class dasymetric method) compared to four other popular population estimation methods (i.e. areal weighting interpolation, pycnophylactic interpolation, binary dasymetric method, and globally fitted ordinary least squares (OLS) based multi-class dasymetric method). The GWR-based multi-class dasymetric method outperforms all other methods. It is attributed to the fact that spatial heterogeneity is accounted for in the process of determining density parameters for land cover classes.