URSI IIWG Format For Archiving Monthly Ionospheric Characteristics "CHARS"

Introduction

The format "CHARS" is now used by the World Data Centers (WDC) for archival of the monthly ionospheric characteristics; it was first introduced in 1989 by Gamache and Reinisch [1989a]. The CHARS "database" is actually a collection of the flat ASCII files each storing a month of data. This design was essentially an inexpensive way to provide potential users of the data with a platform-independent access to the CHARS database contents. The FORTRAN source code for reading the CHARS files has been released in a scientific report [Gamache and Reinisch, 1994]. The software for automatic creation of CHARS files from the individual ionogram SAO-4 files is currently under development.

The Ionospheric Informatics Working Group (IIWG) of URSI Commission G had recommended the CHARS format as a standard for ionogram data dessimination and archival in 1989. This format was then accepted as the URSI standard at the URSI general Assmebly in Prague, 1990. This report present the general description of the CHARS format with the latest updates as of November 1997.

CHARS Format for Ionospheric Characteristics

The CHARS file is a flat ASCII text file containing all available data for a month, including time series of major characteristics and their hourly statistical features (medians, deciles, etc.) It is reasonable to expect that the sounding schedule would not be strictly regular within the observation month. To solve a problem of uneven time sampling, a special header record is introduced into CHAR files [Gamache and Reinisch, 1989b] to serve as a key to encoding/decoding the remainder of the file.

Historically, the maximum length of individual text lines in a CHARS file was set to 120 characters so that it still could be printed without wrapping. The number of lines in a file is determined by the number of days in the month, the number of measurements made each day, and the number of characteristics being archived.

The structure of the CHARS file is shown in Table 1. It consists of two headers, Station Header and Data Header, followed by the main Data Group.

Station Header

The Station Header is one line comprising informative and encoding data. It contains the Station Name, (A30) format, where the data were recorded and the Station Code in (A5) format; the meridian time used by the station to indicate if time is recorded in UT or LT on the records is given next in (I4) format followed by the station coordinates, Latitude N and Longitude E both in (F5.1) format; next there are two (A10) format variables describing the Scaling type, this takes the value Manual or Automatic, and the Data editing variable which can be Edited, Non-Edited, or Mixed.; last in this line is space for the Ionosonde system name in (A30) format. Total length of the Station Header is 99 characters plus CR and LF.

Data Header

Data Header contains information necessary to properly arrange and represent the Data Group. It starts with the Year, Month, Number of days in the month, M; the Number of characteristics archived in the CHARS file, K; the Total number of measurements reported for the CHARS file, and the Number of daily measurements made for each of the M days. Two lines of integers in (30I4) format are required to store this part.

Then, a repeating format of (12A10) is used to list the Names of the particular characteristics being archived. There are K of them, hence more that one line might be required to fit all names. For example if one were archiving only the critical frequencies foF2, foF1, and foE, K would be three, and the characteristics list would be ' foF2' ' foF1' ' foE'. A list of the names of the characteristics, the units, and URSI codes taken from UAG23 [Piggott and Rawer, 1978] are given in Table 2. The URSI list has been enhanced with characteristics that are scaled by the Digisonde ARTIST [Reinisch and Huang, 1983; Tang et al., 1989]. The Chebyshev coefficients [Huang and Reinisch. 1996] used to represent the electron profile of ionosphere are also given, also the best B0 and B1 parameters [Reinisch and Huang, 1997] for the IRI F2 profile. and the calculated ionospheric electron content.

Next, the Units corresponding to the characteristics list (see Table 2) are given in the file, these are in (12A10) repeating format. The last lines of the Data Header are for the URSI codes specified for each of the characteristics (see Table 2) and are written in (60A2) repeating format.

From the information in this Data Header one knows immediately how many data for the time or for each characteristic are to be read. From the number of measurements for each day the time data can be separated into the times for the individual days of the month and the measured characteristics can uniquely be associated with a given time on a given day.

Finally, the Data Header contains the measurement times for the month. With uneven time spacing the measurement times must be recorded to associate with the reported characteristics. This requires that hours, minutes, and seconds of each measurement be entered into the database. To conserve space, the times are written once per month and the reported characteristics are written to correspond to these times. The measurement times are written in a (30(3I2)) repeating format corresponding to the hours, minutes, and seconds, HHMMSS, of the measurements. The number of lines needed for this is determined by the data sampling rate for the month.
 
Table 1. IIWG CHARS File Structure for Flexible Data Rates
File Section 
FORTRAN Format 
Description 
Station Header  A30  Station Name 
A5  Station code 
I4  Meridian time used by station on records 
F5.1  Latitude N 
F5.1  Longitude E 
A10  Scaling type: Manual/Automatic 
A10  Data editing: Edited/Non-edited/Mixed 
A30  Ionosonde system name 
Data Header  30I4... *  Year 
Month 
Number of days in the month, M 
Number of Characteristics, K 
Numbers of measurements total 
Numbers of measurements for each of the M days (Ni, i=1..M
12A10...*  Names 
12A10...*  Units 
12A10...*  List of corresponding URSI codes 
20(3I2)...*  Measurement times HH:MM:SS for each of M days, (Ni values) 
Data Group  24(I3,2A1)...*  N1 values of characteristic 1 (Day 1
N2 values of characteristic 1 (Day 2
... 
NM values of characteristic 1 (Day M
24(I3,2A1)  24 hourly medians for characteristic 1 
24(I2,I3)  24 x 2 hourly counts and ranges 
24(I3,2A1)  24 hourly upper quartiles 
24(I3,2A1)  24 hourly lower quartiles 
24(I3,2A1)  24 hourly upper deciles 
24(I3,2A1)  24 hourly lower deciles 
Repeated for characteristic 2 
... 
Repeated for characteristic K 
* The format is repeated for as many lines as needed to store the data
 
 

Table 2. List of Characteristics, URSI codes, Factors, and Units
  
  
GROUP
CHARACTERISTIC 

ARTIST

Name
 

 
    

URSI 

Name
  

  

  
  
Factor
  
  
Units
 
 
UAG23
ref.#
 
DEFINITION 
 F2 foF2  foF2  00  .1  MHz  1.11  The ordinary wave critical frequency of the highest stratification in the F region 
fxF2  01  .1  MHz  1.11  The extraordinary wave critical frequency 
fzF2  02  .1  MHz  1.11  The z-mode wave critical frequency 
M(D)  M3000F2  03  .01    1.50  The maximum usable frequency at a defined distance divided by the critical frequency of that layer 
h'F2  12  h'F2  04    km  1.33  The minimum virtual height of the ordinary wave trace for the highest stable stratification in the F region 
hpF2  05    km  1.41  The virtual height of the ordinary wave mode at the frequency given by 0.834 of foF2 (or other 7.34) 
h'Ox  06    km  1.39  The virtual height of the x trace at foF2 
MUF(D)  MUF3000F2  07  .1  MHz  1.5C  The standard transmission curve for 3000 km 
hc  08    km  1.42  The height of the maximum obtained by fitting a theoretical h'F curve for the parabola of best fit to the observed ordinary wave trace near foF2 and correcting for underlying ionization 
ScaleF2 40 qc  09    km  7.34  Scale height 
 F1 foF1  foF1  10  .01  MHz  1.13  The ordinary wave F1 critical frequency 
fxF1  11  .01  MHz  1.13  The extraordinary wave F1 critical frequency 
12 
M3000F1  13  .01  MHz  1.50  See Code 03 
h'F1  14    km  1.30  The minimum virtual height of reflection at a point where the trace is horizontal 
15 
h'F  11  h'F  16    km  1.32  The minimum virtual height of the ordinary wave trace taken as a whole 
MUF3000F1  17  .1  MHz  1.5C  See Code 07 
18 
19 
 E  foE  foE  20  .01  MHz  1.14  The ordinary wave critical frequency of the lowest thick layer which causes a discontinuity 
fxE  fxE  21  .01  MHz  1.14  The extraordinary wave critical frequency of the lowest thick layer which causes a discontinuity 
foE2  22  .01  MHz  1.16  The critical frequency of an occulting thick layer which sometimes appears between the normal E and F1 layers 
foEa 44  foEa  23  .01  MHz  The critical frequency of night time auroral E layer 
h'E  13  h'E  24    km  1.34  The minimum virtual height of the normal E layer trace 
25 
h'E2  26    km  1.36  The minimum virtual height of the E2 layer trace 
h'Ea  45  h'Ea  27    km  The minimum virtual height of the night time auroral E layer trace 
28 
29 
Es  foEs  foEs  30  .1  MHz  1.17  The highest ordinary wave frequency at which a mainly continuous Es trace is observed 
fxEs  31  .1  MHz  1.17  The highest extraordinary wave frequency at which a mainly continuous Es trace is observed 
fbEs  48  fbEs  32  .1  MHz  1.18  The blanketing frequency of the Es layer 
ftEs  33  .1  MHz  Top frequency Es any mode. 
h'Es  14  h'Es  34    km  1.35  The minimum height of the trace used to give foEs 
fminEs  34  fminEs  35  .1  MHz  The minimum frequency of the trace used to give foEs 
Type Es  49 Type Es  36      7.26  A characterization of the shape of the Es trace 
37 
38 
39 
Other 1  foF1.5  40  .01  MHz  1.12  The ordinary wave critical frequency of the intermediate stratification between F1 and F2 
41 
fmin  fmin  42  .1  MHz  1.19  The lowest frequency at which echo traces are observed on the ionogram 
M3000F1.5  43  .01  MHz  1.50  See Code O3 
h'F1.5  44    km  1.38  The minimum virtual height of the ordinary wave trace between foF1 and foF1.5 (equals h'F2 7.34) 
45 
46 
fm2  47  .1  MHz  1.14  The minimum frequency of the second order trace 
hm  48    km  7.34  The height of the maximum density of the F2 layer calculated by the Titheridge method 
fm2  47  .1  MHz  1.25  The minimum frequency of the third order trace 
Spread F,   Oblique foI  50  .1  MHz  1.26  The top ordinary wave frequency of spread F traces 
fxI  10  fxI  51  .1  MHz  1.21  The top frequency of spread F traces 
fmI  52  .1  MHz  1.23  The lowest frequency of spread F traces 
M3000I  53  .01  MHz  1.50  See Code 03 
h'I  54    km  1.37  The minimum slant range of the spread F traces 
foP  46  foP  55  .1  MHz  Highest ordinary wave critical frequency of F region patch trace 
h'P  47  h'P  56    km  Minimum virtual height of the trace used to determine foP 
dfs  57  .1  MHz  1.22  The frequency spread of the scatter pattern 
  58      7.34  Frequency range of spread fxI-foF2 
59 
N(h)   Titheridge fhpF2  30  fh'F2  60  .1  MHz  7.34  The frequency at which h'F2 is measured 
fhpF  29  fh'F  61  .1  MHz  7.34  The frequency at which h'F is measured 
62 
h'mF1  63    km  7.34  The maximum virtual height in the o-mode F1 cusp 
h1  64    km  7.34  True height at f1 Titheridge method 
h2  65    km  7.34  True height at f2 Titheridge method 
h3  66    km  7.34  True height at f3 Titheridge method 
h4  67    km  7.34  True height at f4 Titheridge method 
h5  68    km  7.34  True height at f5 Titheridge method 
69    km  7.34  Effective scale height at hmF2 Titheridge method 
T.E.C.  I2000  70  1016  e/m2  7.34  Ionospheric electron content Faraday technique 
71  1016  e/m2  7.34  Total electron content to geostationary satellite 
ITEC  39  I1000  72  1016  e/m2  7.34  Ionospheric electron content to height 1000 km using Digisonde technique 
73 
74 
75 
76 
77 
78 
79  1016  e/m2  7.34  Total sub-peak content Titheridge method 
Other 2  fminF  FMINF  80  .1  MHz  Minimum frequency of F trace (50 kHz increments) Equals fbEs when E present 
fminE  FMINE  81  .1  MHz  Minimum frequency of E trace (50 kHz increments). 
zE  15  HOM  82    km  Parabolic E layer peak height 
yE  16  yE  83    km  Parabolic E layer semi-thickness 
QF  17  QF  84    km  Average range spread of F trace 
QE  18  QE  85    km  Average range spread of E trace 
FF  22  FF  86  .01  MHz  Frequency spread between fxF2 and fxI 
FE  23  FE  87  .01  MHz  As FF but considered beyond foE 
fMUF(D)  25  fMUF3000  88  .01  MHz  MUF(D)/obliquity factor 
hpMUF(D)  26  h'MUF3000  89    km  Virtual height at fMUF 
N(h)  zmE  15  zmE  90    km  Peak height E layer 
zmF1  33  zmF1  91    km  Peak height F1 layer 
zmF2  32  zmF2  92    km  Peak height F2 layer 
zhalfNm  34  zhalfNm  93    km  True height at half peak electron density 
yF2  37  yF2  94    km  Parabolic F2 layer semi-thickness 
yF1  38  yF1  95    km  Parabolic F1 layer semi-thickness 
96 
97 
98 
99 
Digisonde   profile,   F2 layer [A0F2]  A0    km  Coefficient A0, truncated to integer km 
<A0F2>  A1    A0 - [A0], truncation remainder 
[A1F2]  A2    km  Coefficient A1, truncated 
<A1F2>  A3    A1 - [A1] 
[A2F2]  A4    km  Coefficient A2, truncated 
<A2F2>  A5    A2 - [A2] 
[A3F2]  A6    km  Coefficient A3, truncated 
<A3F2>  A7    A3 - [A3] 
[A4F2]  A8    km  Coefficient A4, truncated 
<A4F2>  A9    A4 - [A4] 
[fsF2]  AA    MHz  starting frequency, truncated 
<fsF2>  AB    kHz  fs - [fs] 
[fmF2]  AC    MHz  ending frequency, truncated 
<fmF2>  AD    kHz  fm - [fm] 
[hmF2]  AE    km  peak height, truncated 
<hmF2>  AF    hm - [hm] 
EppF2  AG  0.1  km  error per point, an average mismatch of original h'(f) trace and the trace reconstructed from the calculated profile 
Digisonde   profile,   F1 layer [A0F1]  B0    km  Coefficient A0, truncated to integer km 
<A0F1>  B1    A0 - [A0], truncation remainder 
[A1F1]  B2    km  Coefficient A1 
<A1F1>  B3    A1 - [A1] 
[A2F1]  B4    km  Coefficient A2 
<A2F1>  B5    A2 - [A2] 
[A3F1]  B6    km  Coefficient A3 
<A3F1>  B7    A3 - [A3] 
[A4F1]  B8    km  Coefficient A4 
<A4F1>  B9    A4 - [A4] 
[fsF1]  BA    MHz  starting frequency of the layer, truncate 
<fsF1>  BB    kHz  fs - [fs] 
[fmF1]  BC    MHz  ending frequency fm 
<fmF1>  BD    kHz  fm - [fm] 
[hmF1]  BE    km  peak height 
<hmF1>  BF    hm - [hm] 
EppF1  BG  0.1  km  error per point, an average mismatch of original h'(f) trace and the trace reconstructed from the calculated profile 
Digisonde   profile,   E layer [A0E]  C0    km  Coefficient A0, truncated to integer km 
<A0E>  C1    A0 - [A0], truncation remainder 
[A1E]  C2    km  Coefficient A1 
<A1E>  C3    A1 - [A1] 
[A2E]  C4    km  Coefficient A2 
<A2E>  C5    A2 - [A2] 
[W]  C6    km  Valley width [W], truncated 
<W>  C7    W - [W] 
[D]  C8    km  Valley depth [D], truncated 
<D>  C9    D - [D] 
[fsE]  CA    MHz  starting frequency 
<fsE>  CB    kHz  fs - [fs] 
[fmE]  CC    MHz  ending frequency 
<fmE>  CD    kHz  fm - [fm] 
[hmE]  CE    km  peak height 
<hmE>  CF    hm - [hm] 
EppE  CG  0.1  km  error per point, an average mismatch of original h'(f) trace and the trace reconstructed from the calculated profile 
ValleyID  CH      Valley Model ID 
IRI  B0  41  B0  D0    km  IRI Thickness parameter 
B1  42  B1  D1  0.1    IRI Profile Shape parameter 
D1  43  D1  D2  0.1    IRI Profile Shape parameter, F1 layer 
D3 
D4 
D5 
D6 
D7 
D8 
D9 

Data Group

The Data Group contains the actual values of the characteristics and the corresponding hourly medians and statistics. The group is comprised of a number of lines per each archived characteristic which are repeated for each characteristic. The order of the characteristics follows that given in the "List of characteristics". On a per characteristic basis, for each characteristic one has the N1 values of the characteristic for day 1 corresponding to the reported measurement times for day 1. These are followed by the values for day 2, day 3, ... for each of the M days of the month. The characteristics are written in a repeating (24(I3,2A1)) format corresponding to the integer value (I3) of the characteristic and the qualifying and descriptive letters [see UAG 23]. The actual values of the characteristics can be obtained by multiplying the integer value by the value found in the corresponding Units list (group 2) of the database record (see Table 2). Thus a value of 86 reported for foF2 is multiplied by the Units factor 0.1 MHz to give a foF2 value of 8.6 MHz.

The IIWG Workshop suggested the use of two slashes, //, in place of the qualifying and descriptive letters for monthly characteristics data that were autoscaled but not validated or "edited", i.e. where no quality control procedure has been applied. This code has been extended to consider data that have been edited but no descriptive or qualifying letters introduced. With two positions to fill and the use of a single or double slash there are four codes which can be defined. The first is no slashes implying the use of the descriptive or qualifying letters. The next is the use of two slashes which signifies no editing. The third choice is to put a slash in the first position followed by a blank. This is used to signify autoscaled data that have been edited but no descriptive or qualifying letters are used. The last possibility is a blank in the first position followed by the slash. This is not currently used thus it leaves the possibility for future extension of the code The codes are summarized in Table 3.

 
Table 3. IIWG Codes for the Descriptive and Qualifying Fields of the Characteristics.
Symbolic code  Description 
Q D  Qualifying and descriptive letters used according to UAG #23A. 
Data, edited but no qualifying and descriptive letters used. 
  /  No current meaning, for future extension. 
/ /  Autoscaled data, no editing, no qualifying and descriptive letters used. 
 
Immediately following the characteristics data are the hourly medians given in a (24(I3,2A1)) format; the counts for the hourly medians and the range in (24(I2,I3)) format; the upper quartiles in a (24(I3,2A1)) format; the lower quartiles in a (24(I3,2A1)) format; then the upper deciles in a (24(I3,2A1)) format; and finally the lower quartiles again in a (24(I3,2A1)) format.

The above sections are repeated for each characteristic given in the "characteristics list." This completes the CHARS file, i.e. a month of characteristics data.

References

Gamache R. R. and B. W. Reinisch, Proceedings from the International Workshop on "Digital Ionogram Data Formats for World Data Center Archiving," University of Lowell Center for Atmospheric Research, November 1989a.

Gamache R. R. and B. W. Reinisch, "Ionogram Characteristics at Uneven Data Rates," Presented at URSI Working Group G.4 Ionospheric Informatics International Workshop, July, 1989., University of Lowell Center for Atmospheric Research, 1989b.

Gamache R. R. and B. W. Reinisch, "Ionospheric Characteristics Data Format for Archiving at the World Data Centers", University of Lowell Center for Atmospheric Research, Sci.Report 467, 1994.

Reinisch B.W., and X. Huang, Automatic calculation of electron density profiles from digital ionograms, 3, Processing of bottomside ionograms, Radio Sci., 18, 472-492, 1983.

Tang, J., R.R. Gamache, and B.W. Reinisch, Progress on ARTIST improvements, Sci. Rep. No. 14, GL-TR-89-0185, Air Force Geophys. Lab., Hanscom AFB, Mass., 1989.