An effect of the urban heat island o n modification of the land and sea breeze circulation was discussed by making a differential diurnal temperature change between urban and suburban areas. Here a two dimensional numerical model was applied, and the following assumption was used: higher daily mean temperature and smaller diurnal change in urban area and colder mean temperature and larger diurnal change in suburban area. Under these conditions invasion of sea breeze was slowed down and land breeze was intensified. Double vertical circulation cells were formed; one was around the coastline and another along the border between the urban and suburban areas. A three-dimensional experiment of the land and sea breeze circulation over the Kanto district of Japan was integrated in a domain of 204 km × 204 km × 1.8 km for two simulated days. In this experiment the actual coastline was outlined in the grid of 6 km meshpoint interval, and a uniform diurnal temperature change was assumed on the land area. The computed land breeze and sea breeze were converged or diverged according to a concave or convex coastline. Internal gravity oscillations were found in the front of sea breeze.
It is explained by the law of the io n ic dissociation equilibrium that in the carbon dioxide system in sea water not only the bicarbonate ion content prevails among others ( >90%), but also it is nearly constant especially in deep layers (94%). In the next place, in order to clarify the carbon dioxide system in the ocean, taking into consideration the effects of biological production and decay as well as dissolution and precipitation of calcium carbonate, the theoretical relations between the carbon dioxide system and biological processes are deduced. As to the ratio between calcium content and carbonate alkalinity, it is concluded from the theoretical view point that it is not constant but varies widely. The discussion is done by using the observed date obtained during the Antipode 15Expedition and others. The results show that there is a good agreement between observation and theory.
Newly devised calibration methods t o improve accuracy of calibration are discussed together with the conventional methods. Our special transducers which must undergo hard vibrations and shocks before actual operation were calibrated with these methods. It was confirmed that these transducers can stand up against the presumed vibrations and shocks. The advantage of the MKSA unit system in the calculation of natural frequency, damping constant and sensitivity is shown in the course of the discussion. These calibration methods are as follows: (1) Natural frequency. Two methods, named the phase method and the oscillation method, were devised. The circuit diagram of the phase method is shown in Fig.2. As can be derived from Eq. (17), when the Lissajous figure shows agreement in phase as between coil output and oscillator output, the frequency of the oscillator fits the natural frequency of the transducer. When the damping constant is less than three, the error is at most one percent. The circuit diagram of the oscillation method is shown in Fig.4. As shown in Eq. (23), the pendulum oscillates just at its natural frequency when the negative resistance, defined in Eq. (20), satisfies the condition of Eq. (21). Fig.6 shows the record of free oscillation of the pendulum and the record obtained with the oscillation method. As can be seen from Fig.6, the two frequencies completely agree with each other. Even if the damping constant is greater than three, the error can be lowered less than one percent. (2) Damping constant. There are no d i fficulties in an accurate measurement of the damping constant, if it is very small or very great, even with the conventional methods described by HAGIwARA (1945). But with transducers with nearly critical damping, there appear some problems that make it difficult to measure damping constant accurately with the conventional methods. For such transducers, the so-called resonance method, the circuit diagram of which is shown in Fig.2, is available. The damping constant can be calculated from the sharpness of the resonance characteristics. The theoretical characteristics and the calculation method are shown in Fig.7 drawn upon Eq. (31). Owing to the accurate natural frequency measured with the previous methods, even if the damping is nearly critical, one can measure the damping constant accurately with this method. (3) Sensitivity. Sensiti v ity of a transducer can be measured with various methods. The sensitivities of transducers A and B were measured with the following methods in which newly devised methods are included: damping constant method, condenser method, DC current method, weight method, AC current method, test coil method and vibration table method, The measurement values of sensitivities of transducers A and B are summarized in Table 1. Measurement values are consistent with each other except for the vibration table method. So far, without reasonable foundation, some of these methods have been regarded as not so accurate. Nevertheless Table 1 shows that every one of them except the vibration table method is available for accurate measurement of sensitivity.