The seasonal migration of rainbelt in the low latitudes has generally been considered to be caused by that of the ITCZ. However, air streams such as trade winds and equatorial westerlies, as well as the ITCZ, have been pointed out as important factors which account for it. The purpose of this paper is to clarify the seasonal migration of rainbelt and its main causes along the southeast coastal region of Africa. Daily rainfall in 1963 was classified into the four types in terms of the causes of rain based on weather chart; rain in the NE Trade Winds (Type N), rain in the ITCZ (Type I), rain in the SE Trade Winds (Type S) and rain in the polar frontal zone in the Southern Hemisphere (Type F).
Type S was classified in terms of pressure gradient into two types; rain in the regime of the weak SE Trade Winds (Type S₁) and rain in the regime of the strong SE Trade Winds (Type S₂). The former is accompanied by the air mass with high dew point temperature, the latter by the air mass with low one.
Then the seasonal migration of rainbelt caused by each factor was demonstrated by using the three isoplethes (E_N, Es and W in Fig. 1) of daily rainfall. The feature of rainbelt caused by each factor was summerized as follows:
1. Type I extends to the north and south side of the ITCZ on weather chart, which in detail dominates to the south of it along the east coast (E_N and Es) and to the north of it along the west coast (W).
2. Type S concentrates along the east coast (E_N and Es) which faces windward and rarely appears along the west coast (W) because of the shadow effect of the high mountain ranges of Madagascar.
3. Type S₁ is mainly caused by the passage of cold front without cold surge along E_N and by easterly wave and the other factors along Es.
4. Type S₂ is mainly caused by the cold surge without the passage of cold front along E_N and Es, Then easterly wave rarely appears on weather chart.
5. Type N appears along E_N, not along Es and W.
6. Type F appears along Es, not along E_N and W.
Seasonal change of air streams was summerized as follows (Fig. 9):
1. From January to February, the ITCZ migrates most southward and crosses Madagascar.
2. In March, the ITCZ migrates northward to East Africa and the regime of the weak SE Trade Winds covers Madagascar.
3. From April to May, since the ITCZ migrates northward to the Northern Hemisphere, the regime of the weakest SE Trade Winds covers all the southeast coastal region of Africa and polar frontal zone temporalily shifts northward to the south of Madagascar.
4. A few days before the onset of the Indian monsoon, the regime of the weakest SE Trade Winds is suddenly taken place of by that of the strong SE Trade Winds and outburst of cold air begins.
5. From June to July, the regime of the strongest SE Trade Winds is established all over the region.
The seasonal change of air streams causes the seasonal migration of rainbelts as follows (Fig. 9):
1. From January to February, Type I appears over Madagascar (Es and W) and Type N appears along E_N.
2. In March, Type I appears along E_N and Type S₁ appears along Es.
3. From April to May, Type S₁ appears along E_N and Type F appears in the south o f Es.
4. From June to July, Type S₂ appears along E_N and Es.
The factors which cause the maximum monthly rainfall in 1963 were extracted and the the results were generalized in comparison with the situation of average year as follows: 1. The months of maximum rainfall along E_N and Es (March to May) correspond to the establishment of the regime of the weak SE Trade Winds as soon as the ITCZ migrates northward and are early in the south, late in the north.

View full abstract

The balanced condition and the formation of “fluvial terrace” were studied using a tilting laboratory flume (Fig. 1). The experiment was performed on three tilting rates: 1. 78, 0. 444 and 0. 118‰/h for RUN 1, 2 and 3 respectively. Sand was supplied with a constant rate of 0.50g/s during all the experimental RUN, and the discharge was fixed at 100cm³/s. The median diameter of sand was 1.5φ, and the specific gravity was 2. 70±0. 04. Each RUN consisted of two procedures: 1) appearance of the stationary state in the flume which was fixed at a moderate gradient, and then 2) reaction in a gradually tilting flume. The result was as follows:
In the procedure 1), the channel reached the stationary state after an adequate running time. Nogami et al. (1975) who conducted a similar experiment using the same flume, reported that in a stationary state the slope is a linear function of sand supply:
I=19.6s+26.1 (1)
where I is the slope in ‰ and s is the sand supply rate (sediment load) in g/s. This relation was verified, though the two experiments differed in the diameter of sand.
In the early stage of tilting, a “fluvial terrace” was formed from the upper reach to the down (for example, RUN 1 in Figs. 3 and 4). The slope of the “flood plain” increased as the flume tilted before the emergence of the “terrace” surface (Fig. 2-II and III, middle column). The output sand also increased as long as the “flood plain” mentioned above existed in the flume (Fig. 2, lower column). The relation between the tilting rate and the increase rate of the output sand (Fig. 5) shows that Eq. (1) can be established dependently on time. These “terrace” surfaces were not formed under the stationary state.
In RUN 3 (Fig. 2-III, after about 90h), the model reached approximately the stationary state under which erosion was balanced with the tilting movement. Under this condition the value of the output sand (0.73g/s) and the slope of water surface (41‰) were also agreeable to Eq. (1). At the same time the channel was waving, and “terraces” numbered T3-2, T3-3 and T3-4a groups (Figs. 2-III, 3 and 4) were formed.
According to these results it is concluded that Eq. (1) which states the hydraulic balance holds under wider conditions of time-independent and dependent than expected, and that the mechanism of fluvial terrace formation should be examined from such a point of view.

View full abstract

The methods for revealing the groundwater flow system include observation of hydraulic heads, computer simulation and tracer method using environmental isotopes as tracers. Among these, the tracer method is the most advantageous one in regional scale studies. Tritium, an isotope of hydrogen, forms a part of water molecule and is distributed in the hydrologic cycle by natural processes. Therefore, it can be used as an ideal tracer.
Since 1952, when the first fusion weapon tests started, tritium levels in precipitation have risen as much as two orders in magnitude. The release of tritium was in pulses with its peak concentration in 1963 and thereafter tritium levels have slowly dropped. When the resultant artificial tritium reaches a groundwater reservoir, it can be used to trace the flow. The investigation area of this study is a downstream area of the Yoro River in Ichihara City, Chiba Prefecture. Topography of the area is roughly classified into diluvial uplands and valley bottom plain of the Yoro River. The aquifer is composed of the Quaternary system called the Shimosa Group. The facies of the Shimosa Group are alternation of sand beds and mud beds, though sand beds dominate. The Shimosa Group shows monoclinic structure gently dipping northwestward, so it have been assumed that groundwater flowed northwestward along the bedding plane in accordance with the geologic structure.
Well waters were sampled during 1981 to 1982 and analyzed for tritium. The results are shown in Figs. 3 to 5.
The results and conclusions of this study are summarized as follows;
1. The groundwater which shows relatively high tritium concentration with its residence time less than 30 years is distributed mainly in the upland regions and the groundwater of low tritium concentration is distributed in the valley plain of the Yoro River. The residence time of the groundwater near the Yoro River can be estimated to be older than 30 years.
2. The distribution pattern of tritium concentration in groundwater suggests that upland regions are recharge area and the valley bottom plain of the Yoro River is the discharge area. Groundwater moves from the upland regions to the Yoro River.
3. At the upland regions or nearby area of the Yoro River, the vertical component of groundwater flow is effective in the flow system. At the upland regions, downward component of groundwater flow exists, and the groundwater recharged after 1953 is entering the groundwater flow system. At the valley bottom plain of the Yoro River, tritium concentration of groundwater is very low owing to the upward component of groundwater flow.
4. The estimated groundwater flow system is formed in accordance with the topography. The altitude of the water table in the upland regions is higher than that of the valley bottom plain, and groundwater flows to the direction that the potential difference will disappear. The mud beds in the aquifer have little effect on the general flow pattern formed by topography.
5. The groundwater with relatively high tritium concentration recharged at the upland regions after 1953 has not reached the Yoro River. This suggests that the velocity of groundwater flow is less than several tens of meters per year.

View full abstract

Following three types are generally found on rocky coasts (Fig. 1): Type A is a platform that slopes seaward without significant topographic break extending to the nearshore bottom; Type B is a platform that has a nearly horizontal surface with marked topographic break, a scarp, at the seaward margin; and Type C is a plunging-cliff coast with no platforms. Differences among these types should be attributed to the relative intensity of assailing force of waves and resisting force of rocks. This study attempts to investigate a quantitative relationship between these coastal types and the lithological condition (resisting force of rocks) selecting an area where regional differences of both input wave energy and tidal conditions are small. The study area is the eastern coast of Chiba Prefecture, facing the Pacific Ocean (Fig. 2).
To obtain the physical and mechanical properties of coastal rocks, compressive strength, tensile strength, shear strength, abrasion hardness, and the velocity of longitudinal waves were tested in a laboratory. The longitudinal wave velocity was also measured in situ. Results (Tables 2 and 3) show large regional variations depending upon locations. Since these properties are closely related and not independent of one another (Fig. 6), compressive strength was used as representative of other rock properties. To quantitatively express the effect of cracks in a rock body in the fields a nondimensional parameter, V_pf/V_p, was employed, where V_pf=longitudinal wave velocity in situ and V_p=longitudinal wave velocity measured in a laboratory for specimens without visible cracks. The quantity (V_pf/V_p) S_c was employed as an index for the resisting force of rocks against waves:
f_r_??_(V_pf/V_p)S_c (1)
where f_r=rock resisting force and S_c=compressive strength of rocks. The three coastal types are clearly distinguished by this index (Fig. 7): Type A landform is formed on a coast with smaller value of (V_pf/V_p) S_c, i. e., 8_??_20kg/cm²; Type C is formed in a place where (V_pf/V_p) S_c is of larger value, i, e., 260_??_510kg/cm²; and Type B is produced in an area where this parameter shows intermediate value, i. e., 50_??_100kg/cm².

View full abstract