In this paper, the author intended to make clear the dependences of wind speed (U) and net radiation (Rn) above the bare ground on stability parameter measured by sonic anemometer thermometer. As the stability parameter, we take ζ evaluated from z/L, withz=1.1 m. In the calculation of ζ, L is Obukhov's scale height defined by L=_??_. Here, κ(_??_0.40) is von Kármán's constant, θ is the absolute temperature, g is the acceleration of gravity, U* is the friction velocity and T, * is the scaling temperature. The following results were obtained. Values of show systematical diurnal change taking positive values at night and negative values at daytime. They correspond to the stable and unstable conditions respectively. Whenζ reverses the sign, the outgoing net radiation is balanced with the incoming. The decrease of log|ζ| with increasingU is occurred by increasing turbulent mixing as shown in Fig. 5-a and Fig. 5-b. The relation between log |ζ|and Rn is approximately linear in the case ofζ< 0, under the unstable condition, as shown in Fig. 6-a, and it is fitted by the line log|ζ|=0.048Rn-1.900, In the case ofζ>0, namely under the stable condition, as shown in Fig. 6-b, it is expressed by the equation log|ζ|=-0.050Rn-1.096. These relationships mean that the closer Rngets to zero, the smallerlog|ζ|becomes. Un-der the unstable conditions, the ratio of Rn/log|ζ|is remarkably large. The correlation between log|ζ|andRn is larger than that between log |ζ|andU. Under the stable conditions, on the other hand, large correlation is observed between log |ζ| andU as given in Table 2. From the facts described above, it can be concluded that the stability para-meter is estimated from the relationships between U and Rn as shown in Fig. 7. A tentative classification of the relations between U and Rn is given in Fig. 8.
A series of eruptions of Mt. Usu since August 7, 1977, had brought ejecta of 8.3×107 cubic meters over the vicinity (see Fig. 1). Due to rainfalls after the eruptions, mud-flows occurred on the piedmont of Mt.Usu. In September, the largest mud-flow occurred in Ichino-sawa, Izumi District of Abuta-cho, causing a gully erosion of 1, 100 meters long with transported boulders up to 1.7 meters in diameter. Mud-flows are generally caused by the heavy rainfall through changes of runoff characteristics effected by the deposit of volcanic ejecta. The rate of runoff and the amount of peak discharge were probably increased by the deposition of the volcanic ejecta in Mt. Usu area due to low permeability of the ash. Since we did not know the real amount of the discharge at the time of the mud-flow occurrence, a maximum possible discharge was computed utilizing the rational formula: _??_ where Q is peak discharge (m3/s), f is runoff coefficient, R is rainfall intensity(mm/h) and A is drainage area (km2) The values, A=0.66km2, R=15mm/h estimated from the rainfall data and f=1 were used to obtain a value of 2.8m3/s for a maximum possible discharge. Because this amount of discharge is generally considered to be insufficient to cause mud-flows, other factors than the changes of runoff characteristics should have been responsible. In order to account for the enormous energy required to induce the mud-flow, the hypothesis that blocking and damming-up of the stream by slope failures on the valley sides and subsequent burst of the dam caused the mud-flow was postulated. This hypothesis was investigated by geomorphological analyses comparing basin characteristics of Ichino-sawa (with mud-flow occurrence) and Kousu-gawa (without mud-flow occurrence) as shown in Figure 6. The slope maps of Figure 8 show that steep portion (_??_35°) of the valley sides, where slope failures are likely to occur, is much more extensive in the middle part of Ichino-sawa than in Kousu-gawa. Detailed analyses of aerial photo-graphs indicate the occurrence of many slope failures in Ichino-sawa and appear to support this hypothesis (see Fig. 9). Geological weakness of the Bunsei nuées ardente deposits and a small amount of pumice relative to the total amount of deposits are also thought to have contributed to the occurrence of the mud-flow (see Fig. 10).