Map overlay operation in GIS is often followed by a data transfer between incompatible zonal systems. There are various methods of data transfer which are called areal interpolation methods. Any interpolation methods, however, involve error in data transfer because of insufficient information about the distribution of spatial objects in overlaid zones. This paper proposes a probabilistic model of the error in count data transferred through the areal weighting method, and analyze the relationship among the source zonal system, target zone, and the distribution of spatial objects whose counts are recorded in GIS. Both theoretical and empirical findings are shows.

View full abstract

This paper develops a method for analyzing trend curves that is robust against errors and noises. First, we classify the local. topological characteristics of a trend curve in three categories: a "peak", a "bottom" and a "slope". Second, we define "distinctiveness" of these topological categories. Third, we describe the characteristics of a trend curve using a configuration of these peaks and bottoms with the distinctiveness, and introduce the concept of "isomorphic" relations with respect to this configuration. We develop a method of classifying trend curves and define the similarity between two trend curves using the isomorphic relations. Finally, we apply the proposed method to the analysis of land cover change in the Persian Gulf region using trend curves of NDVI obtained from NOAA remotely sensed data over a year from 1982-1993.

View full abstract

Regression models based on cross-sectional data are a traditional but still a very useful tool in regional analysis, where spatial autocorrelation and heteroscedasticity of error terms are important problems. Analogous to the approaches in time series analysis, some regularization methods have been suggested and applied for these problems. In this paper, we focus on these methods and apply them to a simple regression model with the data of officially assessed land price. We suggest that the parameter estimates depend on the regularization methods that we apply and their differences cannot be ignored in measuring the effect of infrastructure projects.

View full abstract

The purposes of this article are (1) to develop the new routing algorithm with genetic algorithm (GAF) and (2) to evaluate the advantages of GAF by comparing with previous algorithm (e.g. flooding type algorithm; FTA) by use of DDM completion. DDM completion is a new index proposed in this article and it indicates whether obtained DDM reflect altitude information of original DEM correctly or not. In this article, the following results are clarified; DDM completions against the original DEM of the GAF are higher than the one of the ETA for the low relief landforms and fans. For the high relief landforms, like mountains, the differences of completions between both algorithms are not clear. The adoptions of weights for mutation which correspond to the slope angles are useful to determine the flow direction on the error points. These results can offer the solution to improve completions of DDM on low relief landforms and fans where the FTA can not calculate correctly.

View full abstract

The data models for GIS are usually static. In those cases, the geographic features cannot move nor mutate. However, the phenomena in the real world commonly have such a characteristic that they are changed among time and be succeeded to the next generations. This paper starts from the definition of the persisten 2D schema. And, the temporal static primitives are introduced. Then, the spatiotemporal schema which can describes the motion (movement and deformation) is suggested. Finally, the simple examples applying these schemas are introduced.

View full abstract

This paper examines each difference between building area measured on a polygon data and building area described on an application for a zoning permit. This paper also calculates the average distance between both of those arcs. Distributions of the distances approximate to normal distribution on many uses of buildings. Type 1 quantification method suggests that the distance consists slightly of systematic error at digitizing polygons and of the length of eaves. Analysis of covariance is used to make a formula that expects the true building area from use, structure, perimeter and area measured on a polygon.

View full abstract

In humid tropics, land degradation driven by increasing human activities such as forest clearance is the most important problem against sustainable biological products. The objective of this study is to make the land classification map as a base map for a land use planning in the Peninsular Thailand. Recognizing what kinds of land nature are found in humid tropics and their distribution patterns, homogenous environmental units were extracted by combining the factors of geology, land form and soils. These factors and cells were ordinated by correspondence analysis and classified by cluster analysis. In the result, the study area was classified into six homogenous units, and these units were represented as the sequential change from mountain environment to coastal one. In addition, land evaluation was carried out by land cover patterns and their changes, and net primary product which were calculated by remote sensing imagery. Compared by each land unit, land cover and its change were significantly related to land classification types, and net primary product was significantly different among land classification types.

View full abstract