STUDY ON THE STABILITY AND ACCURACY OF BASELINE VALUES MEASURED DURING THE CALIBRATING TIME INTERVALS

In order to investigate quantitatively the accuracy of geomagnetic daily variation recorded by the FGE magnetometer, we analyzed the stability and precision of some groups’ baseline values continuously recorded one day at seven observatories in 2009. The results show that the standard deviation and variable amplitude of the baseline values are small in D, H, and Z components, the standard deviation values are δD≤ 0.03, δH≤ 0.3nT, δZ≤ 0.3nT respectively, and the variable amplitude values are ΔD≤│0.05│, ΔH≤│0.5│, ΔZ≤│0.5│ respectively. Then we selected the baseline values continuously recorded one day at CDP and KSH observatories in 2009 and 2010 and analyzed the influence of absolute measurement intervals on the stability of the baselines.


INTRODUCTION
The testing and calibrating of instruments is an essential part of all measurements and especially important at observatories (Jerzy & Christian, 1996).Errors exist between geomagnetic daily variations recorded by the FGE magnetometer and true values because of the influences of the orthogonality of the sensors, scale value, the accuracy and stability of the orientation of the sensor, and temperature coefficients of variometer sensors.In spite of the fact that the manufacturers of the instruments give values for the instrument parameters, these quantities may change through time and during the transportation of the instrument (Jerzy & Christian, 1996).The geomagnetic station should test the variometer regularly and investigate quantitatively the accuracy of geomagnetic daily variations recorded by the FGE magnetometer, in order to determine the suitable intervals for absolute measurement, research, and application of data.
Few researchers do this, in spite of the fact that IAGA highly recommends the calibrating of the variometers at all observatories, because it may take years before a suitable magnetic disturbance happens to occur at the same time that the observer has an opportunity to make the long series of observations needed.Lu (2008) analyzed the stability and precision of the baseline value at observatories based on 46 groups of continuous absolute measurements carried out at Zhaoqing Geomagnetic Observatory on September 18, 2007, and showed that the absolute measurements of different time periods have a small influence on the stability of the baseline value on a quiet day.
It was recommended that calibration of variometers should be done during a magnetic disturbance at the China Geomagnetic Network.Absolute values were measured twice each hour from 8:30 to 16:30 local time on that day, and lasted at least nine hours.Then the difference (baseline values) between absolute values and data produced by the variometer were compared.
The accuracy and stability of data from calibration of the variometers were analyzed at seven observatories during 2009, and the data from CDP and WHN observatories in 2009 and 2010 were analyzed to study the influence of absolute measurement intervals on the baseline values.
At these observatories, the absolute values D and I were measured with MINGEO DIM fluxgate theodolites made in Hungary; F was measured by an Overhauser magnetometer made in Canada, and magnetic field variations were recorded by an FGE magnetometer made in Denmark.The parameters of the instruments are listed in Table 1.

The accuracy and stability of baseline values during the calibrating time intervals
We drew the curves of D B , H B , and Z B in Figures 1 -3.Noteworthily, in order to make the figure clearer, we show every baseline value minus its average value.Table 3 shows the standard deviations of D B , H B , and Z B were small at all observatories, δD≤ 0.03 ′ , δH≤ 0.3nT, δZ≤ 0.3nT.The D B , H B , and Z B were stable during all calibration time intervals with no abrupt jumps; the variable amplitude of D B was no more than 0.09 ′ , and the variable amplitudes of H B and Z B were no more than 0.80nT and 0.60nT respectively.

Study on the influence of absolute measurement time intervals on baseline values
We used the baseline values during the calibration time intervals at the CDP and KSH observatories during 2009 and 2010 and the hourly mean values during the corresponding period, in order to study the influence of absolute measurement time intervals on baseline values (Figure 4).The left column shows baseline value curves, and the right column shows hourly mean value curves in Figure 4.

Figure 4 .
Figure 4. Baseline values and hourly mean values during the calibration time intervals at the CDP and WHN observatories

Table 1 .
The main parameters of the absolute instruments and variometer

Table 2
gives the calibrated dates and relevant K indices of seven observatories in 2009.

Table 2 .
Calibrated date and relevant K index of every observatory in 2009Because the geomagnetic field was quiet in 2009, it was difficult to select a disturbed day to calibrate the variometer.By contrast, the K index of WHN showed the largest fluctuation among all observatories.As a whole, D B , H B , and Z B clearly fluctuated at the WHN observatory, possibly because the geomagnetic field was active during the calibration time intervals at WHN.It is a problem worth considering why the D B and H B clearly fluctuated at JYG, but the calibration time intervals at JYG were very quiet.

Table 3 .
The standard deviation and variable amplitude of D B , H B , and Z B during the calibration time intervals at all observatories