The purpose of this study is to develop a quantitative site grouping method as a supportive measure for prehistoric settlement researches. One of the basic procedures in settlement archaeological studies is to group individual sites. Grouping scale must be determined in relation to hierarchical level of the social units (e.g. micro level, semi-micro level, macro level). Conventional methods (i.e. Voronoi, Buffer) for site grouping had a difficulty in taking hierarchical structure into account. In this study, hierarchical grouping is attempted by utilizing the scale flexibility of cluster analysis. Spatial extents of the units were also measured through 'time-distance', which simulates the Jomon people's sense of distance. Sites were clustered into groups by suitable threshold values, which were estimated by spatial statistical scores calculated by K-function. Effectiveness of the method was verified by observing its correspondence to the site distribution with specific hierarchical structures. The verification confirmed the developed method to be relevant for quantitative and hierarchical grouping of Jomon-sites.

View full abstract

Research on urban land-use-change phenomenon evolves multi-disciplinary theories and methodologies, with factors of population, economy, land and natural conditions intertwining internally. Cellular Automaton (CA) provide simple discrete deterministic mathematical modeling method for such multi-discipline field. With Geographic Information System (GIS) providing spatialdata processing and analyzing environment, efforts were made to integrate Geographic Location Theory with CA to address land-use-change mechanism in urban region. "Land Cellular Automaton (LCA)" prototype was proposed in this paper and developed into GIS model through case study performed in Beijing City of China. Various issues related with LCA were explored and discussed in details.

View full abstract

From the lessons learned in Great Hanshin-Awaji Earthquake, the importance of collection and usage of the disaster information is widely recognized. Most of Japanese local governments have disaster information systems include helicopters with live camera, surveillance camera on top of the building, and so forth. The role of decision support function in disaster information systems will be more important, as the amount of information collected in these systems becomes huge and hard to understand. The authors propose an information visualization system for disaster information, which helps humans to analyze and understand situations from collected data. A prototype system is built and the usability of the system is confirmed with experiments.

View full abstract

This paper first defines a function (called the cross M-function) that measures similarity between two distributions of line objects in terms of the intersection area of the buffer zones generated by the two distributions with respect to the width of the buffer zones. Second, the cross M-function for the distribution generated by random lines is obtained. The comparison between an observed cross M-function and its expected cross M-function indicates the degree of the similarity. Third, computational methods for the cross M-function method are developed, and they are implemented into a user-friendly tool.

View full abstract

The objective of this paper is to examine quantitative and spatial changes in woodland vegetation for the last 40 years in the Kanto region, where has been exposed to the strongest pressure of urban development in Japan. According to the national census of agriculture and forestry from 1960 to 1980, the woodland cover decreased by 4.3 percent (618km²). This decline was significantly correlated with the increase in the human population density mostly on the peri-urban areas. During the same period, many broadleaved woodlands were converted into conifer plantations in the rural areas. The results of the third (1983-86) and fifth (1994-98) national vegetation surveys showed further decrease in the woodland cover, especially in oak and pine woodlands located on topographically flatter lands. In contrast with the 1960s and 1970s, the shift (succession) from coniferous woodlands to broadleaved woodlands seems to have been underway in many areas since the 1980s. This shift is considered to be partly because many conifer plantations have been abandoned, and partly because Pinus densiflora has widely died back.

View full abstract

One of the most fundamental spatial analytical methods on a two-dimensional plane is the cell count method. This paper extends the ordinary cell count method defined on a plane, called the planar cell count method, to a method defined on a network, called the network cell count method. These two methods are applied to the same distribution of points in a small study area in Shibuya, Tokyo. The statistical tests by the goodness-of-fit test and a spatial autocorrelation analysis reveal that the planar cell count method tends to overestimate the extent of dispersion in comparison with the network cell count method.

View full abstract

In this paper, we discuss the tendencies of telephone call frequencies in Japan. In our analysis, we developed a new data-visualization method using backward-calculated distances from gravity model. The tendencies of the phone traffic data seem to follow the gravity model as a whole, but if we analyze details, the tendencies are very complicated. The reason might be influences of big cities or regional geographical features. We do not develop a model that explains the general phone traffic structure, but we develop a new method for visualizing the regional structre of phone-traffic. We use the gravity model as a basis, and using the concept of backward-calculated distance. Finally, the multi-dimensional scaling method is applied for visualizing the data. The concept of backward calculated distance was developed by Plane (1984), in analyzing the migration space. Our method is unique because we apply the backward calculated distance into multi dimensional scaling method. We calculated distances from non-constrained gravity model, and it enables us to analyze regional structure in many aspects simply. We analyze Japanese phone traffic data which is between 567 MAs'.

View full abstract

Many kinds of phenomena, including traffic accidents, road kills of wild animals and distribution of urban facilities, occur on a network. To analyze their spatial patterns, this paper proposes a new spatial analytical method, called NT-VCM (Network Variable Clumping Method), as an extension of Variable Clumping Method on a plane, which we developed earlier and named PLN-VCM (Planar Variable Clumping Method). The major function of NT-VCM is to statistically detect significant multi-level network clumps in the distribution of points on a network using Monte Carlo simulations. Applying NT-VCM and PLN-VCM to the same data set, this paper shows that NT-VCM preserves connectivity of points on a network better than PLN-VCM.

View full abstract

When a real urban world is projected into 3-D virtual space, buildings are major objects in this space. To realize 3-D CG urban model, it is important to generate building models efficiently. 3-D urban model is the important information infrastructure that can be utilized in several fields, such as landscape evaluation, urban planning, disaster prevention simulation, etc. In this paper, we propose the GIS and CG integrated system to generate 3-D building models automatically from building polygons on the digital map stored by GIS. Some building polygons contain noise edges to be filtered out. We present the algorithm for filtering out noise edge and breaking down complicated polygon into primitive ones. The system automatically generates 3-D urban model so quickly that it meets the urgent demand to realize another alternative urban planning. In the last chapter, we show proposed 3-D urban models automatically generated for urban planning.

View full abstract

This research tested the applicability of landscape indices for monitoring landscapes in a heavily disturbed environment. Landscape indices were calculated from land cover maps based on Landsat TM data using FRAGSTATS. According to the results, considerable land cover change occurred from 1985 to 1999 due to economic development. In the assessment of landscape structure, forest and agriculture patch types were progressively subdivided into smaller, morphologically simpler as a result of human land use activities. Diversity and configuration of landscape patterns increased, especially in the patches adjacent to forest edge and agriculture. Future research will be directed towards the detection of correlations between landscape patterns and specific environmental and socioeconomic landscape functions.

View full abstract

We evaluated the success rate and accuracy of GPS collar telemetry on a free ranging Japanese monkey (Macaca fuscata). Total success rate was 39.7%, and higher in daytime than nighttime. At nighttime, the monkey was resting in the forest. The weakness of GPS signals and deterioration of ephemeris data in the forest environment may have led to the low success rate. The 50th-95th percentile errors of the GPS at 3D fix status was 11.8m to 39.9m and at 2D fix status (DOP<6) was 15.7 to 53.1m, respectivly. The accuracy of the GPS on a free-ranging animal was almost the same as at a fixed position.

View full abstract