The expansion of the buckwheat area is planned in Ogimi village, but some farmer's fields showed lower buckwheat yield. Therefore, choosing fields suitable for buckwheat is important. From site investigation, we found that fields with discontinuous and discontinuous buckwheat cultivation formed several clusters. Thus, this study is intended to develop a land classification method for finding suitable fields for buckwheat in Ogimi village. For developing the method to find suitable fields, we focused on spatial hierarchy of agricultural land, namely, regions, fields block, and fields in order from wide to narrow range. The survey and data analyses were performed separately for these three spatial hierarchy units. In the regions survey, topography and soil data were collected. Also, data of agricultural land improvement were collected in fields blocks survey, while physical and chemical soil feature were collected in the field survey. Analyzing of the data obtained thorough these hierarchal surveys suggested that the region should be classified at first according to their topography groups. The regions classified to topography group I and II, where fields conditions are humid, are not suitable for buckwheat cultivation, whereas those classified topology group III and IV, where fields condition are dry, are suitable for its cultivation. Here, we could propose the fundamental land classification method to find suitable fields for buckwheat cultivation in Ogimi village. To extend this method to other districts or other crops, it would be effective to collect and analyze data considering spatial hierarchy as performed in this study.

In order to conserve terraced paddy landscapes in mountainous areas in Japan, strategic land-use zoning for terraced paddies, surrounding forests and villages is important. For this purpose, it is essential to understand spatial distribution of original terraced paddy parcels before intensive afforestation and mechanism of land-use changes. In this study, we interpreted large-scale 1953 aerial photographs and constructed vector parcel dataset, and conducted spatial analysis in consideration of landforms and the current land covers.
As the results, we identified that many terraced paddy parcels in 1953 were very tiny lots with its sizes were less than 100m². It was suggested that terraced paddy parcels in nearby areas along the Arida River tended to be still cultivated until now, whereas those on the steep slope over 20 degree in upper river area are mostly abandoned. And our results also showed 40% of the parcels were planted with conifer trees and 40% could be changed into broadleaf forests as natural forest succession in this area.
Our parcel dataset could contribute to develop a deeper understanding of the land use change process of terraced paddy field and its backgrounds, and to make a new conservation strategy with not only conceptual design but also numerical goals based on scientific evidence.