従来、いわゆる「慶長日本図」と称されてきた国立国会図書館所蔵日本総図の慶長期成立は誤認であって、正しくは島原の乱直後の寛永15年（1638）に作成された日本総図の写であるという筆者の見解は現在では大方の認知を得たようである。幕府大目付の井上筑後守（政重）は寛永15年に日本総図の作成を理由に中国筋諸国に国絵図の提出を要請したことが明らかになっていた。しかし著者は日本総図を編集するのにどうして中国地方の国絵図のみでこと足りたのかが理解できなかった。ところがこの度新しく発見された南葵文庫所蔵の『日本全国図』によってその疑問が解消された。この『日本全国図』は寛永15年日本図の下図とみなされるが、この図では中国筋諸国に限って渡河方法の注記が著しく欠如している。幕府が島原の乱後に軍事的理由で作成した寛永15年日本図（国会図書館蔵日本総図）は街道筋の渡渉地点で舟渡りか歩渡りかといった渡河方法を注記するのが内容上の目立った特徴である。幕府は寛永15年日本図の編集に際して中国地方での渡河方法の情報を補う必要から中国筋諸国に限って国絵図の調進を要請したのである。

抄録全体を表示

撤回

抄録全体を表示

In this study, construction of a super fine scale multi-layer vegetation map was attempted by overlaying three different egetation map layers using GIS.
First, a classical plant-sociological vegetation survey method, the “Uppsala School Method,” was introduced and its ignificance for multilayer vegetation mapping was evaluated.
Second, a case study was implemented in Kadonokami bog, which is almost entirely covered by the largest patch of Alnus japonica in southwest Japan.
Three layers of vegetation were independently surveyed: a tall tree layer, a shrub layer, and an herb layer. The vegetation from each layer was classified and a vegetation map based on each layer was drawn using GIS. The resolution of the vegetation maps was set at 2×2 m². The three vegetation maps were then overlaid, to generate78 combinations of the three vegetation layers. Among these, 17 combinations that occupied more than 1% of the total bog area were selected as legends for the multi-layer vegetation map.
In our view, the super fine scale multi-layer vegetation map has a number of advantages. Multilayer methods can help to identify and display smallsized patches of different combinations of vegetation layers. Identification of these small vegetation patches is not possible using Braun-Blanquet methods. This combined method can help in practical vegetation management; for example, in this case study, we identified an evergreen shrub layer that shades important wetland herb species.

抄録全体を表示

The authors have produced various types of landscape ecological maps. Basic legend of landscape ecological maps consists of the combination of vegetation classification and landform classification.
The airborne laser survey (LIDAR Survey) is expected for the tool of producing landscape ecological map of forest area. Because, last pulse LIDAR data in winter season is useful for detection of micro landform under forest area, and vegetation classification has been down using three dimensional vegetation structure detected by the difference between LIDAR data in summer and winter seasons. In this paper, the authors introduce the trial produce of three dimensional vegetation structure maps by LIDAR (LIDAR vegetation map) in two different type areas. One is Shinjuku Gyoen Park, where is typical city park in Tokyo Metropolitan area. Another is the south east foot of Mt. Rausu of Shiretoko Peninsula, where is Natural Heritage Area of Japan, Hokkaido Island.
On Shinjuku Gyoen Park, the authors got 1m grid DSM (Digital Surface Model) and DEM (Digital Elevation Model) in summer season and 1m grid DEM in winter season by LIDAR survey. On south east foot of Mt. Rausu, the authors got 0.5m grid DSM and DEM in summer season and 2m grid DSM and DEM in spring or autumn season.
Algorithm of producing LIDAR vegetation maps is as follows. For vegetation classification, the authors used three dimensional vegetation structures such as vegetation height, thickness of crown, and the difference of single layer deciduous forests, double layers deciduous forest and evergreen forest. Using these vegetation structures, it is possible to classify vegetation map with eleven categories legend. In this paper, the authors show 1m grid LIDAR vegetation maps of Shinjuku Gyoen Park and south east foots of Mt. Rausu.
The results of comparison between LIDAR vegetation map and Actual Vegetation Map with 1/25,000 scale by the Ministry of Environment shows that LIDAR vegetation map is corresponding to Actual Vegetation Map. The results of comparison between LIDAR vegetation map and ground troth data shows that LIDAR vegetation map is corresponding to ground truth data in city park, but LIDAR vegetation map is not corresponding to ground truth in natural forest, because of the size difference between crown size of tree and grid side of LIDAR vegetation map.

抄録全体を表示

We aimed to assess the spatial distribution of biodiversity on a regional scale and proposed a method for extracting important areas for conservation of diversity of bird species by using previously recorded information about distribution. The areas targeted in this study were the Ishikari lowlands, Hokkaido, and the surrounding areas (6,800km²); the data of birds (approximately 75,000 items) obtained in these areas were used. The unit of analysis was set to 5-km mesh. The following 3 indices were calculated to facilitate extraction of important areas from the perspective of biodiversity conservation; 10% of the top mesh was regarded as important areas. The first index is the total number of recorded species, as normalized for each environment type. The second index is the abundance of valuable species. The third index is the number of potential species inferred from the relationship between the area of each landscape factor (5 habitats: forest, grassland, urban, river and lake, and coast) and the number of species recorded. Analysis of component ratios of landscape factors in these important areas revealed that diversity of birds tended to be higher in areas with plural landscape factors than in those with a dominant landscape factor. Subsequently, gap analysis for important areas and conservation areas revealed that approximately 70% of the important areas are designated as protected areas for the total number of recorded species and for the number of valuable species, whereas 33% are designated as protected areas for the number of potential species. The higher percentage of the former type of protected areas indicates that the conservation area was established on the basis of previous information. The lower percentage of the latter type of protected areas indicates that investigation of conservation in these areas is needed in the future. We found that important areas for the total number of recorded species and those for the valuable species nearly overlapped, and this finding suggests that many valuable species inhabited the regions wherein the diversity of birds was high. Three important regions in the Nopporo Forest Park and Lake Utonai were found to overlap; these 3 areas are the most important areas in this region. The findings show that the method proposed in this study for extracting important regions based on the relation between the previously recorded information and landscape factors is of practical value. Moreover, the results suggest that this method can contribute to assessment of wider areas for biodiversity conservation.

抄録全体を表示