Документ взят из кэша поисковой машины. Адрес оригинального документа : http://www.wdcb.ru/stp/data/geo_hour.val/AIA/DIGITAL%20WDC%20FORMATS.doc
Дата изменения: Mon Jan 18 17:08:07 2010
Дата индексирования: Tue Oct 2 13:20:21 2012
Кодировка:

Поисковые слова: m 81

[pic]
DIGITAL WDC OBSERVATORY EXCHANGE FORMAT
[pic]
HOURLY VALUES

COLUMNS FMT DESCRIPTION

1-3 A3 OBSERVATORY 3-LETTER CODE, left adjused
4-5 I2 YEAR (last 2 digits, 82 = 1982)
6-7 I2 MONTH (01-12)
8 A1 ELEMENT(D,H,X,Y,Z,or F)
9-10 I2 DAY OF MONTH (01-31)
11-12 A2 Blanks
13-14 A2 Arbitrary
15 A1 INTERNATIONAL QUIET or DISTURBED DAYS,
Q=1,D=2
16 I1 Blank for data since 1900,8 for data before
17-20 I4 Tabular base, in degrees for D and I, hundreds of nanoTeslas
(gammas) for the intensity elements. The bases are right
adjusted
and signed if negative. Negative values are identified with a
minus sign either adjacent to the first significant digit or
in
the high-order position of the field (position 17). NOTE: A
blank
digit will not appear between a (-) sign and the first
significant
digit. For example, a base may appear as -050 or b-50 but not
as
-b50(b=blank).
21-116 24I4 Twenty-four 4-digit Hourly Values for the day. The values are
in
tenth-minutes for D and in nanoTeslas (gammas) for the
intensity
elements. The first hourly value represents the mean value
between
00:00 UT and 01:00 UT, ..., the 24th value represents the mean
between 23:00 UT and 24:00 UT. Rules for negative values are
the same as those described for tabular bases. A missing value
is
identified by 9999.
117-120 I4 Daily Mean. Rules for negative values are the same as those

described for tabular bases. If any of the hourly mean values
for
the day are missing 9999 will appear as the daily mean.

The 25 values in positions 21-120 will have the range -999 to 9998, with
9999
reserved for missing values. To avoid a 4-digit negative value in positions
21-116, the tabular base will be adjusted for that day; for example for D,
one
degree is subtracted from the base and 600 units are added to each of the
hourly values for the day - for the intensity elements, 500 nT are
subtracted
from the base and 500 nT are added to each of the hourly values for the
day.
Each tape block contains 20 records (2400 characters). A standard inter-
record
gap appears between tape blocks. When necessary, all nines are used as
padding
to complete the last block of data. Two or more tape marks follow the last
block of data. The records are sorted according to observatory code, year,
month, element, day (positions 1-10).
[pic]
1-MIN VALUES
The logical record length is 400 coded characters containing header
information, blank spaces, and data for one element for one hour.

Each logical record contains header information and data in the
following format:

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


Characters: 1 - 6 Geographic Co-Latitude in 0.001 degree
7 - 12 East Geographic Longitude in 0.001 degree
13 - 14 Year (98)
15 - 16 Month (01 - 12)
17 - 18 Day (01 - 31)
19 Component of the field (H,E,Z)
20 - 21 Hour (00 - 23)
22 - 24 Observatory's IAGA 3-letter code
25 Origin of data (D - digital, A - digitized)
26 - 34 Future Use (NOT USED HERE)
35 - 40 1-st 1 minute average
41 - 46 2-nd 1 minute average
... ... .....
388 - 394 60-th 1 minute average
394 - 400 Hourly mean value

FORTRAN format statement would look like:

format (2i6,2i2,a1,i2,a3,a1,9a1,61(i6))

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


Comments:

DATA-1 ... DATA-60 are 1-minute values of the given element for
that
hour. H, X, Y, Z, or F are given to the nearest nanoTesla (gamma).
D is
given to the nearest tenth-minute of arc (612 = l degree + 01.2
minutes
East). Each value is in a 6-character field.

Missing data spaces are padded with 99999. No alteration of logical
record length is required for different types of computers.


Codes for sources of digital magnetometer data in the WDC system
not
only indicate the source organization, but also show whether the
data
are average values or point data. For example, 1-minute point
values
scaled from analog magnetograms for the production of AE indices
are
coded with a "D" because they are "digitized". Typically, digital
1-minute values received by WDCs from organizations operating
automatic
magnetic observatory instruments are averages of more frequendy
sampled
values, e.g. 10-second point samples. Different organizations
process
their higher time resolution observations in different ways. Some
may
filter and smooth the observations. Some follow the practice recom-
mended by IAGA of averaging higher time resolution samples from
before and after the minute to obtain a 1-minute value centered
exactly on the minute. Others average values from the beginning of
a
given minute to the beginning of the next minute, effectively
centering
the mean on the half-minute, in similar fashion to the processing
of
1-minute values to obtain hourly means. If the method used to
obtain
1-minute average values is important to a user, the WDC will assist
in
determining the exact procedure applied.

In general, digital values from national networks are "absolute"
and
are tied to baselines determined by are operating institutions.
Often
only timely variations data are needed to support special research
campaigns and digital values may be transmitted from regular
observatory sites via satellite relay platforms. Such values are
"flagged" with a "V" as noted below and eventually are replaced by
the
standard digital observatory output. Values from special networks
such
as the IMS chains are variations only. Attempts are made to check
the
absolute output of these instruments but usually no systematic
absolute
observations are possible or they are later replaced by adopted
standard observatory digital values.

ORG (data origin codes)

A = Alaskan meridian magnetometer chain (includes Canadian sites)
for
lMS
C = Canadian standard observatory network
O = point samples digitized from analog magnetograms
F = France
G = USGS standard observatory network (one station operated by
NOAA)
J = Japan
K = US AFGL E-W sub-auroral zone magnetometer chain
R = Western Canadian meridian magnetometer chain operated for IMS
T = Lungping magnetic observatory, Taiwan.
U = E-W mid-latitude magnetometer chain operated for IMS
V = Variations only sent via NOAA GOES satellite relay
W = Eastern Canadian meridian magnetometer chain operated for IMS
[pic]





The updated standard formats for the international geomagnetic data
exchange
A proposal from the World Data Center C2 for Geomagnetism

Data Analysis Center for Geomagnetism and Space Magnetism

Graduate School of Science, Kyoto University
Toyo Kamei
The draft of this proposal was written in 1996 and then discussed between
various data centers in a few occasions. The most concerns were expressed
on the suggestion to erase Character 15 in the hourly and 1-min WDC formats
and then use this character to add the millenium digit. Currently, the
Character 15 is used for marking the quiet and disturbed days. Some
magnetic observatories insist that it is local Q or D at that station if
that column should be used by a observatory. Others (including the WDC
system staff) think that this position is for the international quiet and
disturbed days. The question arises that because these days are
international (Q or D), it is unnecessary to put that information in each
observatory data.
PROPOSED UPDATE OF THE GEOMAGNETIC HOURLY VALUE FORMAT
ONE RECORD CONTAINS 120 CHARACTERS
*** RECORD FORMAT ***
COLUMN CHAR/NUM DESCRIPTION
1 3 A3 Station code, or index NAME (AE AL AO AU)
4 5 N2 THE LAST TWO DIGITS OF THE YEAR
6 7 N2 MONTH '01' TO '12' (some maybe leading space eg ' 1')
8 COMPONENT * FOR INDEX, H,D,Z,I,X,Y
9 10 N2 DATE '01' TO '31' (some maybe leading space eg ' 1')
(11 16 used to be RESERVED AREA)
11 12 C2 space (for compatibility with 2.5 minute format)
13 14 C2 quality, source and version identification (see attached table for
version)
15 C1 first digit of year (e.g., 1 for 1995, 2 for 2010, if character or
space is used, it will be treated as 1). Q, 1(=Q), C(=Q), D, 2(=D) were
used for quiet/disturbed days, but new format does not care for this
qualification of day.
16 C1 second digit of year (e.g., 0 for 2010, 9 for 1957, 8 for 1898, space
is regarded as 9)
17 20 N4 TABULAR VALUE OR BASE BALUE, UNIT 100 NT FOR HZXY*, DEGREE FOR D,I
21 116 N4 24 HOURLY VALUES, 4 DIGIT NUMBER, UNIT 1 NT FOR HZXY,1/10 MIN FOR
D,I; FIRST DATA IS FOR THE FIRST HOUR OF THE DAY, N4 LAST DATA IS FOR THE
LAST HOUR OF THE DAY
117 120 DAILY MEAN VALUE 4 DIGIT NUMBER SAME UNIT AS HOURLY VALUES
Notes:
. To get absolute hourly (daily) value add the tabular value to each
value.
. Explanation of the year positions: currently data look like following
:STNYYMMCDDbbbbQCBASE....
. KAK9601H01 310 for data 1896 this should be KAK9601H01 8 310 or if QD
day is included KAK9601H01 D 310 in this new format, above data are
permitted but new data will be written like following for 1996
KAK9601H01 19 310, for 1896 KAK9601H01 18 310, for 2096 KAK9601H01 20
310. Program to handle these information can be written by considering
the two character QC field and if this field is not " 8","18",
"18","D8","Q8" regards this field as "19".
. For user and stations who used QD qualifier as local or planetary QD
information: Usually QD information is decided after the month is
over. This new format does not accommodate this information because of
possible difference of the timing for the hourly (mean) values and QD
information. The QD information should be taken care by a separate
format.
. Two character source/quality/version information this field is the
first two characters of the four character source/quality/version
field for one minute W0, W1, W2 formats. For example, V0 variation
only real-time data, (not for analyses), M2 derived from 1-min data,
final quality (can be used for analyses, D2 digitized (manually
scaled) final quality (can be used for analyses), X0 index, real-time
quality (not for analyses), X1 index, provisional version, may be used
for analyses, but final version should be used when they become
available * this format accommodates only public domain data, for
restricted data use H0 H1 H2 formats (short version of W0 W1 W2) with
more information.
PROPOSED UPDATE OF THE 1-MIN GEOMAGNETIC DATA FORMAT
(KYOTO/W1/W0/W2/Wn FORMAT)
RECORD SIZE 400 CHARACTERS, FIXED LENGTH
Record separator(s) such as C/R, L/F may be added to each record depending
on device, Operating system, but one file should be in the same separating
method (no mixture of different file types)
*** RECORD FORMAT ***
COLUMN FORMAT SHORT DESCRIPTION
FROM TO FORTRAN/COBOL
1 2 A2 X(2) FORMAT IDENTIFIER 'W1' for W1 format 'Wn' for Wn format, others
WDC-A old format
3 4 2X X(2) FILLER (IDENTIFIER) VALUE ' ' reserved for update
5 10 A6 X(6) IDENTIFIER for Station usually GG code, (co-latitude 3 digits,
longitude 3 digits)
11 14 I4 9(4) 4 digit year e.g. '1903' '2015'
15 16 I2 9(2) MONTH '01' TO '12'
17 18 I2 9(2) DAY OF THE MONTH '01' TO '31'
19 A1 X(1) COMPONENT 'HDZFXYIAB1-90' and others for index
20 21 I2 9(2) HOUR UT, '00' TO '23'
22 24 A3 X(3) IAGA station code, or name code of index
25 28 A4 X(4) quality, version, source identifiers value ' ' reserved for
future use (see attached table)
29 34 I6 -99999 base value (in 1000*unit) (comment for old the WDC-A
format)
35 394 60I6 -99999 TIMES 60; 60 ONE MINUTE VALUES, UNIT 0.1 nT for W1, 0.01
nT for W2, 1 nT for W0 and for the old WDC-A format; 0.01 arc_min for W1,
0.001 arc_min for W2, 0.1 arc_min for W0 and old WDC-A formats. VALUE
'999999' FOR THE MISSING DATA, ' 99999' maybe used in the old WDC-A format.
( 35 40 ONE MINUTE VALUE FOR THE FIRST MINUTE OF THE HOUR)
(389 394 ONE MINUTE VALUE FOR THE LAST MINUTE OF THE HOUR)
395 400 I6 -99999 HOURLY MEAN VALUE (rounded in absolute value)
This field is used for rotated angle in 0.1 arc_min for the A and B
component.
. Note on mixture of OLD WDC-A format, W1 W0 W2, etc. Wn formats are
physically compatible with other Wn and OLD WDC-A format, and may be
mixed in a file. However, the filename or notes should be attached if
the data is mixed.
. Note on sequence of records (sorting strategy). Some data centers
store data by station and month in different files, so users will
receive data in many files if request is not one station or not one
month. Data centers used to distribute data sorted by co-latitude,
longitude and time for OLD format. However data center will distribute
data sorted by Station code, then time if multiple stations are in one
file in mixture of Wn formats. The simplest way is to convert all the
data into one of Wn formats. If users know which format is suitable
for their uses, users may ask for a specific format of Wn or OLD
formats to one of the data centers.
. Note that Wn formats have latitude and longitude information in
different length from the OLD format, so simple sorting by latitude
and longitude fields is not applicable when these formats are mixed.
As OLD format cannot handle year 2000 and later, simple sorting by
time fields will not give correct sequence when the format is mixed.
However, for data centers which wish to keep the input as it, the most
successful way is to convert 400 byte format to a longer record work
file with sorting key fields, it is done by a simple software. The
software is obtainable from WDC-C2 Kyoto.

WDC-A Digital Exchange Format
for
One-Minute Geomagnetic Data
(Update August 1993)

1. The logical record length is 400 coded characters containing header
information, blank spaces, and data for one element for one hour. Twelve
such
logical records, containing a total of 4800 coded characters, are combined
into a single data block on magnetic media. There are 12 hours of data for
one
component for one station in each block. The data are sequenced according
to
positions 1-24 of each record. Therefore, 24 hours of D are followed by 24
hours of the next element for the same OBS, YR, MO, DA.

2. Each observatory month is written as one file on the media.

3. Blocks are padded with (9's) internally to fill missing data segments
and to complete a data set.

4. Each logical record contains header information and data in the
following format: North polar distance, longitude, year, month, day, hour,
element, observatory code, blank spaces, 60 data values, and an hourly
mean.

NPD LONG YR MO DA E HR OBS ORG BLANK DATA-1 DATA-60 HR MEAN
1-6 7-12 13-14 15-16 17-18 19 20-21 22-24 25 26-34 35-40 389-394 395-400

5. NPD is the observatory's North Polar Distance (0 to 180) from the
north geographic pole in thousandths of a degree and is allotted 6
characters. A decimal point is implied between positions 3 an 4.

LONG is the geographic longitude (0 to 360) measured east from
Greenwich in thousandths of a degree and also has a 6 character
field.
A decimal points is implied between positions 9 and 10.

YR, MO, DA, HR are each 2-digit numbers giving the year, month, day
and hour (GMT).

E is the element symbol in 1 character; it may be D, F, H, P, R, X,
Y, or Z.

OBS is the 2 or 3 letter abbreviation for the observatory.

ORG is the origin of the data.
A = Alaska Chain.
C = Canadian Standard Network.
D = Digitized from analog records.
F = France
G = U.S.G.S. Network.
K = AFGL Chain.
M = McClay Chain.
R = Rostoker Chain.
S = S.U.N.Y. Chain.
T = Tiawan
U = U.C.L.A. Chain.
V = Variation Values via SMS-GOES Satellite.
W = Walker Chain.

BLANK are 10 character spaces reserved for future additions.

DATA-1..DATA-60 are one-minute values of the given element for that
data hour.

H, Z, X, Y, or F to the nearest nanotesla and D in the tenth-minutes
(e.g. 612 = 1 degree 1.2 minutes), each in a six character field.

HR MEAN is the average of the proceeding 60 one-minute values.

6. Each element value and the hourly mean is given a six-digit field
including a minus sign for negative values or a blank for positive values.

7. Missing data-value spaces are padded with 99999. No alteration of
logical record length is required for different computer types.

8. Data on this CD are written in monthly ASCII files as (code)YYMM
(i.e. BOU9003 would be Boulder for March 1990).
WDC-A Digital Exchange Format
for
One-Minute Geomagnetic Data
(Update August 1993)

1. The logical record length is 400 coded characters containing header
information, blank spaces, and data for one element for one hour. Twelve
such
logical records, containing a total of 4800 coded characters, are combined
into a single data block on magnetic media. There are 12 hours of data for
one
component for one station in each block. The data are sequenced according
to
positions 1-24 of each record. Therefore, 24 hours of D are followed by 24
hours of the next element for the same OBS, YR, MO, DA.

2. Each observatory month is written as one file on the media.

3. Blocks are padded with (9's) internally to fill missing data segments
and to complete a data set.

4. Each logical record contains header information and data in the
following format: North polar distance, longitude, year, month, day, hour,
element, observatory code, blank spaces, 60 data values, and an hourly
mean.

NPD LONG YR MO DA E HR OBS ORG BLANK DATA-1 DATA-60 HR MEAN
1-6 7-12 13-14 15-16 17-18 19 20-21 22-24 25 26-34 35-40 389-394 395-400

5. NPD is the observatory's North Polar Distance (0 to 180) from the
north geographic pole in thousandths of a degree and is allotted 6
characters. A decimal point is implied between positions 3 an 4.

LONG is the geographic longitude (0 to 360) measured east from
Greenwich in thousandths of a degree and also has a 6 character
field.
A decimal points is implied between positions 9 and 10.

YR, MO, DA, HR are each 2-digit numbers giving the year, month, day
and hour (GMT).

E is the element symbol in 1 character; it may be D, F, H, P, R, X,
Y, or Z.

OBS is the 2 or 3 letter abbreviation for the observatory.

ORG is the origin of the data.
A = Alaska Chain.
C = Canadian Standard Network.
D = Digitized from analog records.
F = France
G = U.S.G.S. Network.
K = AFGL Chain.
M = McClay Chain.
R = Rostoker Chain.
S = S.U.N.Y. Chain.
T = Tiawan
U = U.C.L.A. Chain.
V = Variation Values via SMS-GOES Satellite.
W = Walker Chain.

BLANK are 10 character spaces reserved for future additions.

DATA-1..DATA-60 are one-minute values of the given element for that
data hour.

H, Z, X, Y, or F to the nearest nanotesla and D in the tenth-minutes
(e.g. 612 = 1 degree 1.2 minutes), each in a six character field.

HR MEAN is the average of the proceeding 60 one-minute values.

6. Each element value and the hourly mean is given a six-digit field
including a minus sign for negative values or a blank for positive values.

7. Missing data-value spaces are padded with 99999. No alteration of
logical record length is required for different computer types.

8. Data on this CD are written in monthly ASCII files as (code)YYMM
(i.e. BOU9003 would be Boulder for March 1990).
IMFV1.22 INTERMAGNET GIN Dissemination Format for Minute Values
Magnetic data, with tenth-nanotesla resolution, are organized on a day file
basis. One file contains 24 one-hour blocks, each containing 60 minutes
worth of values. Blocks of 60 minutes of data are transmitted. Blocks are
padded with 9's if incomplete. Information is coded in ASCII.
File name: To remain compatible with all operating systems, the file name
is limited to 8 characters and will contain the date and the three-letter
code as an extension. eg: MAR1591.BOU for Boulder, March 15, 1991; and
JUN2391.OTT for June 23, 1991 at Ottawa.
Description of the block header (64 characters including CrLf)

IDC_DDDDDDD_DOY_HH_COMP_T_GIN_COLALONG_DECBAS_RRRRRRRRRRRRRRRRCrLf
|IDC |:|Indicates the IAGA three letter observatory |
| | |identification(ID) code eg: BOU for Boulder, OTT for Ottawa, |
| | |LER for Lerwick, etc. |
|DDDDDDD |:|Indicates the date, eg: FEB1591 for February 15, 1991. |
|DOY |:|Indicates day of the year (1-366) |
|HH |:|Indicates the Hour (0-23). The first line following the |
| | |header will contain the values corresponding to minute 0 and |
| | |1 of this hour. The first value of the day file is hour 0 |
| | |minute 0. |
|COMP |:|Order in which the components are listed, can be HDZF, XYZF. |
| | |All components excluding D must be in tenths of nT. D must be|
| | |in hundredths of minutes, east. |
|T |:|One-letter code for data type. R=Reported, A=Adjusted, |
| | |D=Definitive data. |
| | |Reported data are defined as: the raw data obtained from the |
| | |IMO, either by satellite, computer link, or other means. It |
| | |will be formatted in either version IMFV2.8N (binary) or |
| | |IMFV1.2N (ASCII,) without any BRM (Baseline Reference |
| | |Measurements), or other modifications applied to it. |
| | |Adjusted data are defined as: the Reported data with BRM, |
| | |spike removal, timeshifts, and/or other modifications applied|
| | |to it. It is emphasized that only one (1) adjusted version of|
| | |the data would be allowed, to be completed within 7 days of |
| | |receipt of the Reported data to prevent the proliferation of |
| | |multiple versions of the Adjusted data. |
| | |Definitive data are defined as the final adopted data values.|
| | |Definitive data will only be distributed by the institution |
| | |responsible for the observatory. |
|GIN |:|Three-letter code for GIN responsible for processing the |
| | |station (IMO) data eg: EDI(Edinburgh), GOL(Golden), |
| | |OTT(Ottawa), PAR(Paris). |
|COLALONG |:|Colatitude and east longitude of the observatory in tenths of|
| | |degrees. |
|DECBAS |:|Baseline declination value in tenths of minutes East |
| | |(0-216,000). Declination baseline values to be provided |
| | |annually. If components are X,Y,Z then DECBAS=000000. |
|RRR..RRR |:|Reserved 16 bytes of R-characters for future use. |
|_ |:|Indicates a space character. |
|CrLf |:|Indicates a Carriage return, Line feed. |


Description of data space (64 characters per line including CrLf)
Component values are coded as signed integers, right-justified with a field
width of 7. Total field (F) values are coded as unsigned integers, right-
justified with a field width of 6. The field widths must be maintained,
either through zero-filling or space-filling. The '+' sign for positive
values is optional.
Two (2) minutes of data are concatenated on the same line

AAAAAAA_BBBBBBB_CCCCCCC_FFFFFF__AAAAAAA_BBBBBBB_CCCCCCC_FFFFFFCrLf

(values for minute 0) (values for minute 1)

. .

. .

. .

AAAAAAA_BBBBBBB_CCCCCCC_FFFFFF__AAAAAAA_BBBBBBB_CCCCCCC_FFFFFFCrLf

(values for minute 58) (values for minute 59)

|AAAAAAA |:|Indicates Component 1 data field (H,X, etc.).|
|BBBBBBB |:|Indicates Component 2 data field (D,Y, etc.).|
|CCCCCCC |:|Indicates Component 3 data field (Z,I, etc.).|
|FFFFFFF |:|Indicates Total Field data field. |
|_ |:|Indicates space character. |
|CrLf |:|Indicates Carriage Return and Line Feed. |


[pic]
|[pic] |

http://www.intermagnet.org/FormatData_e.html

|Comments:web|
|master |
|2001-11-14 |
|Ss |
|ї2001 |
|INTERMAGNET |