**********************************************
Some notes 
* On July 5, 2010, the one-min ascii files of 1990-2010 was re-made from 
one-second files, in order to correct the element symbol E (X,Y,Z-->H,D,Z) 
in the one-min ascii files, and to consistently calculate the 
1-min averages.  We found that the data copied before July 5, 2010 may 
contain mixture of one-min average and one-min sampled data from the 
original one-second sampled data.  

***********************************************
Time Tag Errors

interval             stn  error
Oct.30-Nov.19, 1991  LEM  Time tag data are 2 hours behind the correct time.
Nov.2-16, 1991       BRV  Time tag data are 1 hour ahead of the correct time.
Nov.24-Dec.7, 1992   BRV  Time tag data are 10 hours ahead of the correct time.
Nov.25-Dec.6, 1993   ADL  Time tag data are 1 hour behind the correct time.
Dec.7,'93-Jan.17,'94 KOT  Time tag data are 2 hours behind the correct time.
Jan.18-22, 1994      CBI  Time tag data are 9 hours ahead of the correct time.
Mar.18-Apr.27, 1994  CBI  Time tag data are 1 day behind the correct time.
Mar.17-Apr.18, 1994  WEP  Time tag data are 6 hours behind the correct time.
Apr.21-Jul.29, 1994  WEP  Time tag data are 6 hours behind the correct time.
Mar.24-May 6, 1995   BRV  Time tag data are 1 day behind the correct time.
Apr.7-9, 1995        ADL  Time tag data are 1 hour behind the correct time.
Apr.27-29, 1995      ADL  Time tag data are 1 hour behind the correct time.
May 23-25, 1995      ADL  Time tag data are 1 hour behind the correct time.
Nov.1-Dec.9, 1995    CBI  Time tag data are 12 hours behind the correct time.
Mar.9-28, 1996       ADL  Time tag data are 1 day ahead the correct time.
Feb.15-Mar.7, 1997   ONW  Time tag data are 1 day behind the correct time.  
Feb.28, 1997         BIK  Time tag data are 1 hour behind the correct time.  
Mar.5-12, 1997       YAP  Time tag data are incorrect. 
May 22-29, 1997      YAP  Time tag data are incorrect.  
Jul. 1-Dec. 31, 1997 YMK  Time tag data are 9 hours ahead the correct time.  
Apr.29-May 22, 1998  WEP  Time tag data are 12 hours ahead the correct time. 
Apr.24-May 22, 1998  DAL  Time tag data are 1 hour behind the correct time. 
Aug.29-Sep.17, 1998  DAL  Time tag data are 1.5 hour behind the correct time.
Sep.24-Oct.22, 1998  DAL  Time tag data are 1.5 hour ahead the correct time. 
Nov.16, 1998 - 
       Jun 19, 1999  DAL  Time tag data are 1.5 hour ahead the correct time.
Apr.14-May 7, 1999   KAT  Time tag data are 9 hours ahead the correct time.
Apr.4-19, 2001       MCQ  Time tag data are 12 hours ahead of correct time. 
Apr.20-Sep.22, 2001  MCQ  Time tag data are 1 day ahead the correct time.   
Apr.12-Jun.15, 2002  MCQ  Time tag data are 2 days behind the correct time.
Jun.16-Jul.4, 2002   MCQ  Time tag data are 1 day behind the correct time.
Oct.4-26, 2002       ONW  Time tag data are 3 hours ahead the correct time. 
Aug.8-Oct.27, 2003   BIK  H and Z are exchanged, time tag may be incorrect in max 13 sec due to GPS receiver error. 
Feb.4-July 10, 2005  KTB  Time tag data may be delayed for maximum 52 sec from the correct time (network NTP stop in this interval). 
Sep.8, 2007-Jan.28, 2009  KAG  H and D components are short-circuited in the cable before the data recorder. 
Sept.7, 2009-June 2 0110UT, 2010: RIK  Time tag data was shifted a few min to 2-3 hours due to the lack of clock correction. 


*************
some more notes for the data

BSV(BRV): 
some data jump in the first half of 2008 
Z-component was stopped on Aug-Sept.2008 
sometimes GPS clock fails and the time may have offset of ~14 sec

CHD: 
large noise on March, August-Sept, 2008 

KTN:
large drift in H and D components for Jun-Oct 2007. 

LMT:
080223-080304: output satulation 
091110-091112: H-component satulation
091110-091119: D-component satulation 
091112-091114: Z-component satulation

MGD:
output drift on Nov.2007 and after May, 2008 
abnormal output after July 2007 in D-component 
output satulation in August 2008 

PPI:
output drift for Feb-Mar, 2007, Jan-Apr, 2008, and Dec 2008-Feb 2009 

TIK:
large noise up to Sept 2008

ZYK:
large noise 
no GPS correction before Sept.2007 

MSR: 
Sinusoidal noise with a period of tens of seconds sometimes 
appears in all three components after January 2010.  

******************

Channel Correspondence: 

a) The each channel corresponds to each value as follows.

 ch.1:H        ch.2:D       ch.3:Z      ch.4:time pulse
 ch.5:dH/dt    ch.6:dD/dt   ch.7:dZ/dt  ch.8:meaningless

 (for the (new) data from CBI)
 ch.1:H        ch.2:Z       ch.3:D       ch.4:time pulse
 ch.5:dH/dt    ch.6:dZ/dt   ch.7:dD/dt   ch.8:meaningless

b) Directions of magnetic sensors are as follows.

  (1) for the data from the stations in the northern hemisphere,
      for data from WEP before 05UT on Jan. 31, 1991
      and for data from ADL before July 14, 1993
      [H:+ = northward]  [D:+ = eastward]  [Z:+ = downward]
  (2) for the data from the stations in the southern hemisphere,
      [H:+ = northward]  [D:+ = eastward]  [Z:+ = upward]
  (3) for data from CBI
      [H:+ = northward]  [D:+ = westward]  [Z:+ = downward]
  (4) for data from WEW
      [H:+ = northward]  [D:+ = westward]  [Z:+ = upward]

c) Some other remarks

* The Z-component of BSV data sometimes becomes zero due to a
      temporal disconnection of cable.
* The Z-component of ADL data sometimes shows noisy fluctuations 
      due to an equipment problem. 
* The Z-component of DAL data is meaningless after Jan. 1, 1993.
* The D- and Z-components of MSR data have a DC-level, artificial 
      fluctuation on a few days/year.
* ADL is the Adelaide station (ADE).
* The H-component of EWA data sometimes shows noisy fluctuations 
      due to an equipment problem
* The D-component of DAL data satulate for April 13, 1992 - July 
      10, 1992





Geomagnetic Observation by Magnetometer Network.