抄録
The study of Bellani radiation integrator in this country was started by Takeda (1955), who tried several kinds of the instruments. After it, there have been no notable development and no further studies. A radiation integrator suitable in the routine observation is still unavailable.
According to Pereira (1959), Monteith et al. (1960) and McCulloch et al. (1967), characteristics of the Bellani integrator are as follows, the water-filled instrument can be used for measurement of daily radiation in the tropics, but its threshold value is so high that it can not be used in the middle or high latitudes, the alcohol filled model has a better choice for low radiation intensities, but it is not capable of oparation a whole day; in the case of the metal receiver, continued exposure of the instrument after the whole liquid has been distilled into the burette causes excessive heating of the receiver with risk of fracture of the metal and glass seal.
It is of use to produce an integrator, which has low threshold value and high efficiency of distillation and which has the capacity for an amount of alcohol sufficient for a usual daily input of radiation.
In the present trial, two types of radiation integrator were made, one with reddish brown glass receiver containing ethyl alcohol, the other with ultramarine glass receiver containing methyl alcohol. Each radiation receiver, was evacuated to an order of pressure of 10mmHg and was filled with 45cc of alcohol.
Laboratory tests revealed that the relation between integratead radiation and distillation agreed with the formula developed by Worner (1957). When the volume (cm3) of condensed liquid is converted into height (cm) the formula can be written as follows,
E(H)=dS/dH=aZ+C, Z=1/E-H
where E is calibration factors, S circumglobal radiation, H height of liquid in condenser burette, B is a constant showing the final height of liquid in burette, and is determined as the height of liquid column when all the liquid is distilled into the burette, or as the length of burette when the liquid is enough to fill the burette. Therefore, a and C are determined by the least squares analysis.
In the field values of dS/dH at various valus of H were measured and the calibration curve under natural illumination was obtained in the formula mentioned above for each instrument. There was a good agreement between the results obtained by the two instruments calibrated for circumglobal radiation in this way, through they were equipped with different coloured receiver.
Finalliy, comparison between a radiation integrator and a Eppley type horizontal pyranometer gave the curve in Fig. 6 a fairly good correlation was found between the two values of radiation all the year round, though the efficiency of distillation tended to increase in winter. As a matter of course, the integrator was able to be operated a whole day without reset and its threshold value was bellow 75cal/cm2.
These results show that the radiation integrators calibrated are useful for field mesurement of circum-global radiation and for the daily total of the horizontal radiation in this country.
