Ionospheric Absorption Effects of the Solar Eclipse of 24 October 1995

The solar eclipse of 24 October 1995 was observed in Wuhan. Field strength equipment was used to detect the radio wave absorption in the D­ region and absorption data on the eclipse and control days were obtained. The comparison of these curves shows that the absorption values during the eclipse are less than the normal values. It was found that absorption effects began before the first contact and ended after the last contact. (


INTRODUCTION
High frequency radio waves propagate by reflecting in the ionosphere. Their reflecting points are generally in the E and the F-region, but their energy loss mainly takes place in the ionospheric D-region. That is to say, the D-region's electron density is responsible for the absorption of HF radio waves. Sometimes we use the absorption measurement methods to observe the ionospheric variation including the electron density, collision frequency, long periodic fluctuation in the mesosphere and so on.
During a solar eclipse, the variation of radio absorption comprises of two parts when reflection takes place in the E-region, namely, the nondeviative absorption and the deviative absorption (Zhang et al., 1985). In general the critical frequency in the E-region decreases, the height of the reflection increases and the deviative absorption may increase. Moreover, be cause of the ionospheric tilt produced by the solar eclipse, radio wave energy undergoes focus ing and defocusing; it also influences the measurement of absorption.
Much research has been done on the D-region ionosphere. The sources of ionization in the D-region are: (l)Ionization of the atmospheric constituents by cosmic rays. This is the predominant source of ionization in the polar region. The contribution of cosmic rays to D region ionization decreases with latitude, becoming negligible at low latitudes; (2)1onization of mesospheric nitric oxide by hydrogen Lymana emission from the sun; (3) Ionization of 234 TAO, Vol. 8, No. 2, June 1997 0 molecular oxygen and nitrogen b J solar X-rays of wavelengths less than 100 A; (4) Ionization of 02 ( �g) by 1027<A<l l 18 A UV radiation. The dominant source in a nonpolar latitude changes from cosmic rays to Lyman-a to X-rays as one moves from the lower to the higher mesosphere, so X-rays are responsible for the variation in electron density below the reflect ing height in the altitude range where most of the absorption takes place ( Sengupta, 1980). But

OBSERVATION
A solar eclipse may be viewed as a vast geophysical experiment where the rapid but predictable change in solar ionizing radiation presents a unique opportunity to study the changes that occur within the middle and upper atmosphere. Such an opportunity was anticipated for 24 October 1995 in southern China.
During the solar eclipse, we did our observation experiments in the Wuhan Institute of Physics and Mathematics (30°38'N, 1l4°l7'E). We used Field Strength Equipment to detect the absorption of HF radio waves. This is the so-called A3 method (Schwentek.H., 1966). This method has the advantage of being both simple and sensitive. Though the calibration is some what difficult, we are usually interested in changes in absorption, so accurate knowledge of the zero absorption level is not critical.
The main parameters of our instrument are listed as follows: In Wuhan we received the short radio wave signals transmitted from Taipei (2 5°02'N, 12 1°3 l 'E). Since the distance between Wuhan and Taipei is about 950 kilometer, the circuit should be a one-hop path. The reflecting point's latitude and longitude are 27°50'N and 1 l 7°54'E, respectively, and is located in the northeast of Jiangxi province. This arrangement moves the reflecting point near to the zone of solar eclipse. There the totality is about 0.50. The fre quency of the radio waves is 9.6MHz.
An absorption in the ionosphere is usually divided into nondeviative and deviative ab sorption. The absorption coefficient in the ionosphere can be expressed as (Davies K., 1989) /3= In nondeviative absorption regions,µ= 1 , and for HF radio waves, m 2 )) V 2 , so (1) can be simplified (2) This is the type of absorption to HF and VHF waves that occurs in the D-region, where f3 is in decibels per meter. In deviative absorption regions,µ is close to zero, and the absorption coefficient is written approximately as /3 u ' :::: :: -µ

2c
( 3 ) where µ ' is the group refractive index and f3 is in decibels per meter.

RESULTS AND DISCUSSIONS
The received energy can be divided into three parts: 1. the transmitted power; 2. the path attenuation; 3. the ionospheric absorption. Since we know the transmitted power and can cal culate the attenuation along the path, we can draw the absorption curves according to the received field strength.
In our experiment, we also used LF radio receivers to receive Loran-C signals transmitted from Niijima, Japan(34°24'N, 139°16'E). We studied the phase and amplitude ofLF(lOOKHz) and the ionospheric critical frequency around the day of the eclipse, and found that .none of these changed significantly. So we can conclude that the model of propagation didn't vary significantly during the solar eclipse.     (Rottman, 1987). During a given period, each EUV flux is keyed to a 0 0 chromospheric(e.g. Lyman-a) or coronal (e.g. 50 A <A<100 A) emission (Tobiska, 1990).
We know that the chromosphere is partially, and the corona completely, invisible. So, a radia tion eclipse may last longer than an optical eclipse, beginning before first contact and ending after last contact. Perhaps this is one of the reasons for the above phenomena.
We also used a fractal technique (Essex, 1991) to calculate correlation dimensions for the data in our experiment. The fractal dimensions of the data for 24, 25 and 26 October are 2.231, 2.212, and 2.236, respectively. No obvious difference exists among their correlation dimen sions. We consider the reason is that the totality of the reflecting points is only about 0.50. The influence of the solar eclipse is too weak to make a difference to the correlation dimension.