The uses of radiation instruments for measuring artificial light sources are common recently. But the instruments are designed for use in the field so that the performance of the instruments in the laboratory may be different from in the field. Therefore, the purpose of the present investigation is to find the performance of errors of the radiation instruments, Eppley type pyrheliometer and Moll-Gorczynski type solarimeter, under very low light intensity in the growth cabinet. The tests were made: (1) Linearity (2) Temperature coefficient (3) Farred response (4) Transient response.
(1) Linearity
The linearities of the outputs of the instruments were checked by comparison with responses of a corrected photometer at various distances under a tungsten lamp (500 watt). In figs 1 and 2, the outputs of Eppley type pyrheliometer and Moll-Gorczynski type solarimeter are plotted against the readings of the photometer. The nonlinearity was not so large over 0.2-10.0mv (0.02-1.0cal/cm².min) in both instruments. But, according to the calculation using eq. (2), Eppley type had more linear than Moll-Gorczynski type. For spot readings in the laboratory, the linearity is unimportant if the output can be uniquely related to energy through a calibration curve.
(2) Temperature coefficient
MacDonald and Courvoisier reported that the temperature coefficient for Eppley was -0.041--0.127% per °C, and Anderson for Moll-Gorczynski type was -0.17% per °C. However, these values had been measured under relatively high level of incident radiation, so that the temperature coefficient under low light intensity was checked in the laboratory. Figs. 3 and 4 show data obtained for Eppley type and Moll-Gorczynski type. The responses in figs 3 and 4 are in terms of response at 27.0°C which is assigned the value of 100 percent. It is noted from the figures that: (1) The temperature coefficient is affected by light intensity, (2) Under high intensity (0.15cal/cm².min) linear relation exists between output and temperature, but under low intensity (0.014cal/cm².min) nonlinear relation, (3) In Eppley type, the temperature coefficient increases the diversion from 100% with higher temperature, but in Moll-Gorczynski type increases with lower temperature. And it is obviously that the observed values for small I will be seriously affected by temperature.
(3) Farred response
The spectral responses of the radiation instruments are characterized by using a glass dome which transmit about 0.3-4.0μ range. Fluorescent lamps emit the radiation of about 400-700nm range. Therefore, when the energy emitted by fluorescent lamp is measured by a radiation instrument, the energy of about 400-700nm range is able to obtain without the calibration of spectral response of the detector. However, when the output of the instrument and the total radiation (included long-wave radiation) were measured under Day-light fluorescent lamps in combination with farred lamps, the output of the instrument increased with increasing radiation of farred lamps (fig. 5). It is recognized that the output of the instrument is included not only to 400-700nm but to farred energy. It seems the reason that the thermal radiations of glass dome and through glass dome affect the output, as the energy of 400-700nm range is extremely small.
(4) Transient response
The time constant for Eppley type is of the order of 20 seconds and for Moll-Gorczynski type, 10 seconds. But, it is assumed that the response may be delay at low level incident radiation, so that the responses of transition were tested in the case of steady- and unsteady-state between the instrument temperature and ambient temperature. Figs. 6 and 7 show the results for Eppley type and Moll-Gorczynski type, respectively. Transient time under high intensity was more rapid than under low intensity, and in steady-state than in unsteady-state. Especially, in unsteady-state in addition to under

抄録全体を表示

著者が先に発表した単色光を混合して作物に照射した実験では, 白色光に近い単色光の混合光より, 赤色部が多い混合光の方が草丈・乾物重などの生育がよかった。しかし, 今まで発表された文献には赤色光より白色光の方が生育がよいという結果が数多くある。この矛盾を解決するため, つぎのような実験を行った。
1) ゴルチンスキー日射計の光量を一致させて, 栽培用螢光灯とこれに近似の波長分布をもつ単色光の混合光 (R12-Y2-G4-B12) をトマト・キュウリに照射したところ, 栽培用螢光灯の方がトマトの草丈を除いて著しく生育がよかった。そのため両光源の約400～700nmの光量と全放射量を測定したところ, 図6のように日射計光量を一致させても, 有効光量 (400～700nm), 全放射量が異ることが判明した。
2) 有効光量と全放射量を一致させた, 赤色光・栽培用螢光灯・R12-Y2-G4-B12と, 有効光量のみ一致させた栽培用螢光灯をトマト・キュウリに照射したところ, 栽培用螢光灯とR12-Y2-G4-B12ではほぼ同じような生育を示したが, これらの区は赤色光より生育が劣った (図7, 8)。
3) 赤色光に橙色光・黄色光を混合して, 赤色部の波長巾を拡げた混合光 (R20-010, R20-Y10) でトマト・キュウリを生育させたところ, トマトは赤色光のみの光源より, これら混合光源で照射した方が乾物重が非常に小さかったのに対して, キュウリはトマトほど大きな差はなかった (図9)。
以上の結果から光量の測定は, 正確に有効光量を測定しなければならないことが判明した。

抄録全体を表示

The subject of this paper is to make clear the characteristics of long-term change and variability of agroclimatic resources influencing crop production. Meteorological data of about 50 stations over the northern hemisphere of the earth were used to study the long-term change, latitude distribution and variability of the sum of daily mean temperature during the period with daily mean above 10°C (sum of effective temperature, ΣT_10°C) which is a most important element of agroclimatic resources. The results obtained can be summarized as follows.
1. The time traces presented in Figure 1 reveal a systematic fluctuation of ΣT_10°C, although there is some difference in the amplitude and the phase among stations. The Arkhangelsk curve shows that the during the last quarter of the 19th century and the first one tenth of the 20th century the value of ΣT_10°C was lower than the overall average by about 200°C day, followed by a net increase of about 400°C day between the 1910's and the 1940's. Since then it has decreased to date. A quite similar trend can be also seen at other stations with the exception of Lagos (Nigeria) near the equator. It is very interesting to note that the difference between lower and higher extremes of ΣT_10°C is approximately independent of the latitude and reaches about 400°C day.
A more detailed analysis of the time fluctuation of ΣT_10°C was made by using correlogram and spectrum analysis methods. As shown in Figure 2, the two types of the correlogram become clearly distinguishable, one type is characterized by the drastic decrease of correlation coefficient with increasing the lag time and mainly predominant in latitudes higher than 30°N. The other type of the correlogram has a relatively high and long tail and is mainly observed in the tropical or subtropical zone. Figure 3 illustrates the results of power spectrum analysis. At Arkhangelsk, period lengths of about 60 and 2 years are prominent. The former seems to correspond roughly to a large fluctuation with the period of about 70 years as recognized on Figure 1. It is characteristic in the tropical and subtropical zones that fluctuations in a higher frequency range above 0.1 cycles year^-1 and fluctuations in a lower frequency range below 0.1 cycles year^-1 both contribute approximately evenly to the variance of ΣT_10°C.
2. The normals of ΣT_10°C (‹ΣT_10°C›) and of the duration of effective period (‹D_10°C›) both show a slight change near the equator. Poleward of latitude 25°N these quantities decrease rapidly with latitude and become zero at about 70°N. The latitude distributions of the both quantities were approximately expressed by Eq. (3). As Figure 4 shows, the latitude dependence of ‹ΣT_10°C› is quite similar to that for the annual mean air temperature (‹T_a›), and the following relation is obtained between them
‹ΣT_10°C›=1538 exp (0.073‹T_a›).
This may be useful for evaluating the value of ‹ΣT_10°C› from the data of ‹T_a›. Poleward of 25°N, the value of correlation coefficient between ‹ΣT_10°C› and ‹D_10°C› increases rapidly with latitude.
3. The value of coefficient of variance (C. V_ΣT) characterizing the magnitude of yearly variation of ΣT_10°C increases rapidly with latitude. The value of C. V_ΣT at a higher latitude is about ten times as larg

抄録全体を表示