About 93 percent of the Kiso drainage basin is occupied by mountainous areas including the Central Japan Alps where are developed wide fossil periglacial slopes covered with coarse debris. Occurrence of such slopes that a large amount of the debris and gravels have been produced under cold climatic conditions the Last Glacial Age. The sediments underlying the river terraces formed during the late Pleistocene and those underlying alluvial fans in the Kiso drainage basin mainly consist of these debris and gravels. Alluvial fans have, however, become small in size during the Holocene or the Post Glacial Age because the supply of coarse debris and gravels has relatively decreased. By contrast, a large amount of finer floating materials have been deposied in the western subsiding part of the Nobi Plain, being carried by the rivers. As the lower reaches of the Kiso “Three” Rivers, the Kiso, the Nagara and the lbi, have flowed and gathered to the western part of the lowland, their river beds have been successively raised in such depositional environment. The river bed of the Kiso is the highest and that of the lbi is the lowest among three. The inundation have, therefore, concentrated to the lowest lbi and their river banks were broken by the attacks of river flow at inundation. The flood control to the Kiso “Three” Rivers in historical time has been carried out in such fluvial conditions.
The Kiso River basin is located in a central part of Japan and covers one of the greatest and most active area in both social and economical sense. The Kiso River basin has three main streams, that is the Kiso, the Nagara and the Ibi River. From old days. these three main streams has mutual connection to each other through many tributories, and had an intimate relation by causing submersion damage from one stream to the others. Division of these three river channels had been considered as a master counterplan for mitigating flood damage. The Japanese Government during the Meiji era proceeded to dividing three river streams, known as Meiji River Improvement, with the assistance of Hollander Civil Engineer, Johannis De Rijke, from about 1870. This paper aims at describing the historical and hydrological circumstances of the counterplan in the Kiso River basin until Meiji River Improvement. It is known that human activity taken for river improvement has an unexpected influence on geographical, hydrological and social activities in a river basin. This interaction between rivers and human society under various governmental regime is also explained from historical viewpoint to a certain extent.
(1) In the lowland of the Kiso, Nagara and Ibi rivers, the people constructed embankments around the villages and arable land for the flood control to protect the villages from the flood. The flood control community is called Waju in Japanese. (2) People have long dwelled in the Waju region and a part of the dike was already built in the ancient time. But it is only in the early 17th century when the main parts of the Waju embankment were established for the first time. The construction of Waju increased since then. (3) When a Waju was established, the earth and sand were accumulated in the river bed. As a result, the river bed became so higher that the embankments were frequently broken in the feudal time. To cope with this situation, they raised the ground level of the building land and there built the mizuya in Japanese for an emergency evacuation. They used mizuya as a temporary house and a store-place of flood. Moreover, they raised the ground level with the earth produced by the digging of a moat in a part of the paddy. Through this procedure, the reduction of the rice production was prevented (horita in Japanese). (4) Since the Meiji era, the extensive river improvements by the government decreased the flood damage. And Waju, which was an unofficial organization in the feudal period, was controlled by a lower branch of the government's river administration. (5) In the Waju region, wet rice fields changed into dry ones in the Showa era. With increasing urbanization and industrialization, it became difficult to maintain the organization of the flood control. On the other hand, a new type of flood has also appeared.
The fact that the cross-profile of glacial valley can be well approximated by parabolas (Y=aX2) is explained by the variation principle, assuming that the glacier erosion works towards minimizing the bottom friction under the constant length of contact surface. The variation principle proves that the ideal or fully-developed profile of the glacial valley be a catenary, if the friction is proportional to the ice thickness. Maclaurin's expansion of the catenary shows that a parabola is a very good approximation of the catenary. Hence, the good approximation of the cross-profile by parabolas is explained together. Different catenaries are generated by changing the form ratio FR (depth to rim width) and are then approximated by Y=aXb by the method of least squares. The obtained b values for catenaries bocome fractionally larger than 2.0 with increasing form ratios up to 1.0, indicating that b value would practically range between 1.0 and about 2.0. The relationship is plotted on the 'b-FR diagram' with FR in the abscissa and b in the ordinate, respectively. Two types of developmental succession of glacial valleys are distinguished in the diagram. The one means that glacier valleys started from shallow V-valleys get larger b value in pace with increase of FR. The other shows the opposite trend starting from rather deep V-valleys. The first type is called the Rocky Mountain model here after its source of data, and represents over-deepening of the glacial valley as it develops. The second type is called the Patagonia-Antarctica model likewise, which represents widening instead of deepening process of the profile. The difference is attributed to the nature of glaciers which produced these valleys, i.e., alpine glaciers and continental ices. The variation principle also shows that the circular profile gives the maximum area for a constant perimeter length. This means that the glacier ice be most easy to flow down as a psuedo-plastic material under a constant boundary shear. However, it seems unreasonable to consider that the boundary shear proportional simply to shear rate within the ice and not to the weight of it would control the erosional process of the glacier valley. The morphology of cross-valley profiles also suggests that a catenary gives better figuration for U-shaped glacier valley than an arc.