A nationwide survey for measuring field strengths at 40 and 60 kHz was carried out in the winter months of 2004 to estimate the service area of the standard frequency and time signals. The data obtained were analyzed in comparison with the resultant field strengths of sky waves and the ground wave which were predicted using a method developed by one of the authors. Good agreement was seen between the measured and predicted field strengths.
 The standard frequency and time signals (SFTS) are transmitted from two stations in eastern and western Japan at frequencies of 40 and 60 kHz since 1999 by the National Institute of Information and Communications Technology (NICT). They are widely used for calibrating frequency standard oscillators as well as radio control clocks, which are popular in Japan.
 Field strengths of the signals, however, to be received in areas exceeding a distance of about 700 km from the transmitter where the sky wave component predominates over the ground wave have not yet been evaluated.
 The resultant field strengths of the sky wave and the ground wave vary mainly with the change of the sky wave component reflected from the ionospheric D region. The ionization distribution in the region is controlled seasonally and diurnally by the solar zenith angle as well as secularly by solar activity [Belrose, 1968]. In addition, since the sky wave component attenuates more slowly with the propagation distance than the ground wave does, fluctuations of resultant field strengths known as the Hollingworth interference pattern [Bracewell et al., 1951] appear progressively with distance, exhibiting a maximum interference dip (MID) at a distance where both components have an equal strength and opposite phase.
 The main purpose of the survey was to estimate the level of the SFTS signals at 40 and 60 kHz in areas all over Japan. In practice, measurements were made to obtain the diurnal change of resultant field strength for 24 hours at various distances from the transmitter. Mobile measurement was also performed aboard a cruising vehicle along the course away from or toward the transmitter to confirm the MID effect.
2. Outline of Survey
2.1. Objective Transmitters and Period of Survey
 The field strength of signals transmitted from the two stations of SFTS service were measured from January to February in 2004 in areas covering Japan, including Okinawa. Particulars of stations the call sign of which is JJY are listed in Table 1.
Table 1. LF Standard Frequency Wave Transmitting Stations
JJY 40 kHz
JJY 60 kHz
10 June 1999
1 October 2001
Mt. Ohtakadoya, Fukushima
Mt. Hagane, Saga/Fukuoka
Latitude and longitude
Stayed monopole of umbrella type
Stayed monopole of umbrella type
Antenna height, m
Transmitter power, kW
Radiation efficiency, %
Carrier frequency, kHz
Type of emission
±1 × 10–12
±1 × 10–12
 The transmitter power is 50 kW for both JJY 40 and 60 kHz stations, while the radiated powers necessary for the analysis of received field strengths are estimated as 12.5 kW and 25 kW for the respective transmitter by multiplying the transmitter power by the radiation efficiency.
2.2. Receiving Equipment
 Measuring systems were installed in three vehicles, each equipped with a field strength meter whose loop antenna was fixed on the vehicle's roof, each driven over distinctly different courses, but during the same time frame. The field strength meters were calibrated in dB above 1 μV/m by use of the standard magnetic field generator before and after the survey.
2.3. Fixed Point Measurements
 Continuous measurements for 24 hours were made at 31 fixed points which were carefully selected to minimize the disturbance of electric fields due to surrounding facilities such as power lines, factories and high buildings. The measurement points are classified into the following five groups (Figure 1):
 1. Nine points from n1 to n9, one about every 100 km along the northbound course starting from the JJY 40 kHz transmitter at Mt. Ohtakadoya.
 2. Eleven points from e1 to e11, one about every 100 km along eastbound course starting from the JJY 60 kHz at Mt. Hagane.
 3. Five points from i1 to i5 where deep interference dips for the transmitters might be observed.
 4. Four points from u1 to u4 in urban areas.
 5. Two remote points from r1 to r2 in Okinawa separated from the mainland by sea.
2.4. Mobile Measurements
 Continuous recordings of the field strength were obtained along two courses between m1 and m4, using the receivers aboard cruising vehicles to observe the interference pattern, which are expected to appear by theory. One course covered a distance range from 600 to 800 km for JJY 60 kHz, while the other covered a range from 470 to 600 km for JJY 40 kHz.
2.5. Data Acquisition and Processing
 Since the SFTS signals are pulse modulated in accordance with the time code given by International Telecommunication Union , the data sampling had to be made in synchronization with the code. In the measurement, one hundred samples in a 10 second time block were taken. Then the upper decile values of samples were taken as the representative values to remove the effects of impulsive interference.
 The field strengths representing the daytime are the average of 5 values at 10 to 14 hours, while the nighttime field strengths are the average of 5 values at 22 to 02 hours in 135 degrees East Longitude Mean Time.
3. Analysis of Data Obtained
3.1. Theoretical Background
 Recommendation ITU-R P.684 [International Telecommunication Union, 2002] provides the method for calculating sky wave field strengths below about 150 kHz; it recommends that the waveguide method be applied for VLF waves and that the wave hop method be applied for LF waves. The wave hop method based on research in Canada [Belrose, 1968] has been modified recently to a computer-based method [Wakai et al., 2004]. This was followed by a further revision that introduced a reflection height model derived from the parabolic distribution of electron densities in the D and E layers [Wakai, 2005]. This prediction method is used in the analysis of the data obtained.
3.2. Field Strength Variations With Propagation Distance
 The field strength variations with propagation distance are described by classifying fixed point measurement data into three ranges of distance from the transmitter: short (less than about 500 km), intermediate (less than about 1000 km), and long (more than about 1000 km).
 The daytime field strengths measured are shown by dots in Figure 2 for JJY 40 kHz. Figure 2 also shows the predictions obtained from International Telecommunication Union  for the ground wave, and from the method developed by Wakai  for the resultant of one- and two-hop sky waves and the ground wave. Since in the short range the level of the ground wave exceeds that of sky waves, the resultant field strengths fluctuated increasingly with the distance around the level of the ground wave because of the interference between the two components. In the intermediate range, where the strength of the wave components is comparable, the MID of the resultant field strengths appeared in the range of about 630 km for JJY 40 kHz as a result of the interference between the two components having equal strength and opposite phase. In the long range, where the sky wave components exceed the level of the ground wave, the resultant field strengths slowly attenuated in both measured data and the predicted curve.
 The nighttime field strengths showed basically the same trend of decrease with the distance as in the daytime, having no distinct MID in the intermediate range for JJY 60 kHz (Figure 3). Since the ionospheric absorption decreases in the nighttime, the two-hop sky wave suffered from less absorption than it did in the daytime, resulting in rather rapid fluctuations of resultant field strength over the whole range in Figure 3. The measured and predicted field strengths agreed fairly well up to about 1800 km for both frequencies and for daytime and nighttime.
3.3. Field Strength Variations With Time
 The SFTS signal propagated via the ionosphere has a limited stability of frequency, since it is degraded because of variations in the reflection height, particularly in sunrise and sunset hours. The diurnal variations measured at the 31 fixed points provide useful information to users concerning the sufficiency of field strength and stability of signals.
Figure 4 shows the diurnal variations typical in the short range for JJY 40 kHz. In addition to the measured values denoted by dots, the field strengths predicted for the ground wave and the resultant wave are shown in the upper portion of Figure 4, while the phase delay angle of the resultant of one- and two-hop sky wave components to the ground wave is shown in the lower portion of Figure 4 as expressed in the unit of 1/10 of the angle in degrees.
Figure 5 shows the diurnal variation of field strengths measured at points close to the MID distance in the intermediate range for JJY 60 kHz. The low values of these field strengths at midday, coinciding with those predicted, are the result of the interference between two components having almost equal strength and opposite phase, in reference to the phase delay curve Figure 5.
Figure 6 shows the diurnal variation of field strengths measured and predicted at a distance of about 900 km from the transmitter of JJY 40 kHz. The variation in resultant field strength predicted coincides well with that in the values measured for the 24 hours excepting the period from 19 to 22 hours, when the measurement was disturbed by man-made noises.
3.4. MID Evidenced by Mobile Measurements
Belrose  reported mobile experiments with airplanes in Canada by which the MID in field strengths of 80 kHz signals was observed at a distance of about 650 km from the transmitter. Since the MID effect was also expected theoretically before the survey, mobile measurements with cruising vehicles were performed along a course from 470 to 800 km for JJY 40 kHz and one from 600 to 800 km for JJY 60 kHz.
Figure 7 shows the recorded field strength variation of the JJY 60 kHz transmission. Along with these recordings, Figure 7 shows the ground wave curve and three curves of resultant field strengths calculated for reflection heights of 69, 69.4 and 70 km. These reflection heights correspond to semithicknesses of 31, 30.6 and 30 km for the reflecting layer. The MID distance of 692 km where the maximum dip was observed agrees closely with that predicted for 69.4 km of reflection height. Lowering the reflection height by 1 km caused a decrease of 25 km in the MID distance. Furthermore, the phase delay curve in Figure 7 clearly shows that the MID effect occurred at a distance of the phase delay of 180 degrees. The MID distance is very sensitive to variations in the reflection height, the local time, the season and the geographic latitude.
 Unfortunately, a sharp drop of field strengths as that observed for JJY 60 kHz could not be observed in the mobile measurement for JJY 40 kHz, since the measurement was impaired by high noise levels when passing urban areas along the course.
4. Concluding Remarks
 The fixed point measurements made at points every 100 km from the transmitters revealed resultant field strengths of about 10 dB higher than the ground wave level over propagation distances from 1000 to 1800 km. The diurnal variations of received field strengths obtained at the 31 points of observation also agreed fairly well with those predicted.
 Mobile measurement data obtained with field strength meters on board clarified the relation of the interference dip of resultant field strengths to the reflection height as a function of the solar zenith angle.
 The measurement data obtained at 40 and 60 kHz in winter for the solar activity maximum period are sufficient to verify the accuracy of the revised prediction method which was developed for predicting received field strengths and phases of signals in lower LF band.
 The solar activity effect on the LF sky wave propagation would be appreciably smaller than that on the HF propagation, while the solar zenith angle dependence of the D region is quite remarkable. Therefore it is desirable to conduct a similar survey in the summer or in other geographic locations, not only to confirm the service area of the SFTS signal, but also to validate the prediction method of LF propagation.
 The authors wish to express their sincere thanks to John Goodman, Kenneth Davies, George Lane, and John Wang for their valuable comments and suggestions for our paper.