Documentation for IGRA Station History Information Version 2.1 ------------------------------------------------------------------------ I. QUICK START ------------------------------------------------------------------------ Start by downloading "igra-metadata.txt", which has station history information for all stations in IGRA for which such information is available. You can find this file in the same location where this description file resides, i.e., by going to directory /pub/data/igra at ftp.ncdc.noaa.gov or pointing your browser to http://www1.ncdc.noaa.gov/pub/data/igra. Then, to manually locate the history information for a specific station, proceed as follows: - Open "igra-metadata.txt" with your favorite text editor and adjust the window size to accommodate the rather long lines in the file. - Scroll or search through the file to find the station's identification number or name (e.g., 72201 or Key West). Note that station records are indexed by the station identification number used in IGRA (columns 1-5) which represents the most recent World Meteorological Organization (WMO) identification number for a particular location (e.g., 71600 for Sable Island). Original WMO identification numbers are provided in columns 7-11 (e.g., 72600 for pre-1977 records at Sable Island). - Look at the date and event type fields to identify the station history information you are interested in. Alternatively, to read the metadata file with a computer program, refer to the section in this description file that defines the metadata fields and identify the fields of interest. ------------------------------------------------------------------------ II. FORMAT OF METADATA RECORDS ------------------------------------------------------------------------ Each Record in the "igra-metadata.txt" file contains information for one "event" as identified by the station identification number, date, and event type. Event types are either dynamic (i.e., reporting a change) or static (i.e., reporting a practice or equipment in use at the time of the event date). The format of the fields is described below. Variable Columns Type ------------------------------ ID 1- 5 Integer ORIGID 7- 11 Integer NAME 13- 42 Character NAMFLAG 43- 43 Character FIPSCODE 45- 46 Character FIPSFLAG 47- 47 Character LATITUDE 49- 57 Real LATFLAG 59- 59 Character LONGITUDE 61- 69 Real LONFLAG 71- 71 Character ELEVATION 73- 76 Integer ELVFLAG 78- 78 Character YEAR 80- 83 Integer MONTH 85- 86 Integer DAY 88- 89 Integer HOUR 91- 92 Integer DATEIND 94- 94 Integer EVENT 96-114 Character ALTIND 116-117 Character BEFINFO 119-158 Character BEFFLAG 159-159 Character LINK 161-162 Character AFTINFO 164-203 Character AFTFLAG 204-204 Character REFERENCE 206-230 Character COMMENT 232-309 Character UPDCOM 311-341 Character UPDDATE 343-349 Character ------------------------------ These variables have the following definitions: ID is the IGRA station identification number [i.e., most recent WMO identification number for a site, e.g., 71600 for Sable Island]. ORIGID is the original WMO station identification number (e.g., 72600 for Sable Island before 1977, 71600 from 1977 on). NAME is the station name (often a city name; includes the state abbreviation for U.S. stations). NAMEFLAG is the quality flag for the station name. There are three possible values: Blank = no indication of questionable quality ? = questionable C = corrected FIPSCODE is the Federal Information Processing Standard (FIPS) country code. For a complete list of possible values, see "igra-countries.txt" in the IGRA FTP directory. FIPSFLAG is the quality flag for the FIPS country code. There are three possible values: Blank = no indication of questionable quality ? = questionable C = corrected LATITUDE is the latitude of the station (in decimal degrees, missing = 9999.0000). LATFLAG is the quality flag for the latitude. There are three possible values: Blank = no indication of questionable quality ? = questionable C = corrected LONGITUDE is the longitude of the station (in decimal degrees, missing = 9999.0000). LONFLAG is the quality flag for the longitude. There are three possible values: Blank = no indication of questionable quality ? = questionable C = corrected ELEVATION is the elevation of the station (in meters above sea level, missing = 9999). ELVFLAG is the quality flag for the elevation. There are three possible values: Blank = no indication of questionable quality ? = questionable C = corrected YEAR is the year of the event. MONTH is the month of the event (99 = missing/unknown). DAY is the day of the event (99 = missing/unknown). HOUR is the hour of the event [in Universal Coordinated Time (UTC), 99 = missing/unknown]. DATEIND is the uncertain date indicator. There are two possible values: 0 = reasonably certain. 1 = original metadata source uncertain or unclear about the date or the date was judged to be dubious. EVENT is the event type. Event types are either dynamic (i.e., reporting change) or static (i.e., reporting a practice or equipment in use at the time of the event date). The following 43 event types are currently used in the file: CHANGE BALLOON TYPE = change in balloon material, size, or weight. CHANGE COMPUTER = change in computing system. CHANGE CORD LENGTH = change in length of suspension court between balloon and instrument. CHANGE CORRECTION = change in data correction. CHANGE DATA CUTOFF = change in data cutoff (i.e., temperature or humidity threshold below or above which measurements are not reported. CHANGE FREQUENCY = change in radio frequency of operations. CHANGE GRAVITY VAL. = change in gravitational constant. CHANGE GROUND EQUIP = change in ground equipment. CHANGE ID NUMBER = change in station identification number. CHANGE OBS. TIME = change in observation time. CHANGE OPERATOR = change in entity responsible for station. CHANGE P SENSOR = change in pressure sensor. CHANGE RADIAT. CORR = change in radiation correction. CHANGE REPOR. PRAC. = change in reporting practice. CHANGE RH ALGORITHM = change in relative humidity algorithm. CHANGE RH DUCT = change in type of duct for relative humidity sensor. CHANGE RH SENSOR = change in relative humidity sensor. CHANGE SONDE MODEL = change in radiosonde model. CHANGE T DUCT = change in type of duct for the temperature sensor. CHANGE WIND EQUIP. = change in equipment used for determining winds. END OF PROBLEM = end of a problem. MISCELLANEOUS = miscellaneous metadata, including national independence dates. OBS. PROG. REDUCED = observation program reduced. OBS. RESUMED = observations resumed after suspension. OBS. SUSPENDED = observations suspended. OPERATED BY = entity responsible for station on event date. START OF PROBLEM = start of a problem. STATION CLOSED = observations discontinued at station. STATION MOVED = change in station location. STATION OPENED = observations initiated at station. USING BALLOON TYPE = balloon material, size, or weight in use on the event date. USING COMPUTER = computing system in use on the event date. USING DATA CUTOFF = data cutoff in use on the event date. USING GROUND EQUIP. = ground equipment in use on the event date. USING OBS. TIME = observation time(s) in use on the event date. USING P SENSOR = pressure sensor in use on the event date. USING RADIAT. CORR. = radiation correction in use on the event date. USING REPOR. PRAC. = reporting practice in use on the event date. USING RH ALGORITHM = relative humidity algorithm in use on the event date. USING RH SENSOR = relative humidity sensor in use on the event date. USING SONDE MODEL = radiosonde model in use on the event date. USING T SENSOR = temperature sensor in use on the event date. USING WIND EQUIP. = equipment used for determining winds on the event date. ALTIND is the alternates indicator. This two-character indicator is used in cases in which an event type appears more than once for a particular station and date, and the information provided in the before/after fields differs among instances. The first character of the indicator represents the type of event that is recurring. It has 11 possible values: Blank = not used. B = balloon Type. C = computing system. G = ground equipment. H = Relative humidity algorithm. I = instrument. P = reporting practice. R = radiation correction. S = sensor. W = wind equipment. X = other. The second character is a digit representing the index of recurrence (e.g., 1 = first instance, 2 = second instance, etc.). When an event type only appears once for a specific station and date, the indicator may be set to the appropriate characters for the first instance, or both characters are set to blank. BEFINFO is the before information. For events indicating a change, this is the practice/equipment in use prior to the event date. For "static" events, it reflects the practice/equipment in use around the time of the event date. Note that for the STATION MOVED event type, this field has historically been left blank since the coordinates for the period prior to the event date are contained in the previous record for the station. See Tables 4-16 in Gaffen (1996) for many examples of the entries in this field. BEFFLAG is the quality flag for the before information. There are three possible values: Blank = no indication of questionable quality ? = questionable C = corrected LINK is the link between the before and after information fields. It has the following two possible values: Blank = no after information follows. TO = after information follows. AFTINFO is the after information. For events indicating a change, this is the information for the period beginning on the event date. For "static" events, this field is blank. Note that for the STATION MOVED event type, this field has historically been left blank since the coordinates for the period beginning on the event date are contained in the same record for the station. See Tables 4-16 in Gaffen (1996) for many examples of the entries in this field. AFTFLAG is the quality flag for the after information. There are three possible values: Blank = no indication of questionable quality ? = questionable C = corrected REFERENCE is the citation of a document, website, organization, or individual from which the information was obtained. COMMENT is a text field for additional notes. UPDCOM is a text field for notes referring to changes made to a previously existing record. UPDDATE is a field containing the month and year (MM/YYYY) during which the record was last updated. ------------------------------------------------------------------------ III. Version History ------------------------------------------------------------------------ Version 2.1 (October 5, 2010) - Thanks to Liangying Zhang of NCAR, the metadata for U.S. and Australian stations were updated during the summer of 2010. The 475 added or edited records are identified with an update date of 06/2010 and a reference of "Updated by NCAR/EOL". Updates are based on information received by NCAR from the U.S. National Weather Service and the Australian Bureau of Meteorology. Version 2.0 (November 2009): - A total of 2763 events supplied by Steve Schroeder in February 2007 and some additional events received from WMO member countries during 2006 and 2007 were added. - The lengths of the "before information" and "after information" fields were increased to 40 characters each, and the formatting and spelling of entries in those fields (e.g., instrument names) were standardized. - Two sets of records that appear to have been inserted automatically but do not add to the value of the station history information were removed. The first set includes records identifying a change in a station's geopolitical affiliation that carried a year prior to 1900 or whose year was equal to 9999. The second set consists of all "Station opened" and "Station closed" events attribute it to "NOAA NCDC" since they appear to be based on often spurious changes in coordinates. - Event types whose spelling did not correspond with Table 1 of Gaffen (1996) were corrected. - A variety of additional minor format inconsistencies were corrected. - The documentation was expanded considerably to include a more detailed description of the fields, additional background information, a version history, and a list of all references corresponding to citations of publications that appear in the metadata file. Version 1.0 (released February 2006): - The original Gaffen (1996) digital file of station history information was reformatted to include "day," "hour," "update date," and "update comment" field and to convert longitude to decimal degrees ranging from -180 to 180. - Updates to records for GUAN and RATPAC stations were added wherever possible. - Some of the original Gaffen (1996) records were removed if they were unnecessary given previous records or given the newly received information. ------------------------------------------------------------------------ IV. Background ------------------------------------------------------------------------ The IGRA metadata consist of three principal types of radiosonde station history information. 1. The primary source is Gaffen (1996): Records from this source are identified by an update date of 99/1996. 2. Updates to records for GUAN and RATPAC stations (Free et al. 2005) for the period since Gaffen (1996): Most of these records are identified by a source attribution of Hammer (personal communication) and are based on information received since 2004 through communication between the NCDC GCOS Lead Center and GCOS focal points at Hydrological and Meteorological Service Centers (HMSCs) in WMO member countries. Additional sources of updated information include Joyce (personal communication) and NOAA/NWS. 3. Events supplied by Steve Schroeder of Texas A&M University: Identified by a source attribution of Schroeder (2007), these events cover stations in the Russian Federation, India, Japan, China (including Hong Kong and Taiwan), and Antarctic stations operated by those countries. Cautionary notes: 1. Radiosonde station history information is generally received in a variety of formats and to varying degrees of specificity. Consequently, the completeness, precision, and accuracy of IGRA metadata records also vary among stations and over time. 2. Gaffen (1996) aimed to reveal inconsistencies and errors in sources of metadata but not to correct them, and the IGRA metadata are a reflection of this approach. 3. Different station catalogs often disagree about the exact locations of stations, and the number of small location or elevation discrepancies is too large to hope to resolve. While a large number of discrepancies can be traced to differences in rounding methods, it is probable that many small location errors arise from interpreting a surface observing location as an upper air location or vice versa. 4. The Definition of ground equipment is often ambiguous. The ground system and radar may carry separate names, or the entire ground system may be referred to by the name of the radar, or the radar may be referred to by the name of the ground system. In addition, a radiosonde is sometimes referred to by the name of the ground system it uses (such as an AN/GMD-1A radiosonde), implying that any radiosonde compatible with that type of ground equipment may have been used. 5. The current country code is used in all of a station's metadata records even if the station has been under the control of different countries throughout its history. Significant changes in control are noted in station names or comments. ------------------------------------------------------------------------ V. References ------------------------------------------------------------------------ The following are full literature citations for the abbreviated references used in the IGRA metadata file as well as in this documentation: CIA (Central Intelligence Agency), 1993: The World Factbook 1993. 489 pp. plus maps. [Available from National Technical Information Service, 5285 Port Royal Road, Springfield, VA 22161, Tel. (703) 487-4650.] Dobesch, H., E. Rudel, and K. Zimmermann, 1992: Radiosounding in Austria: A historical review of instruments and data. Papers presented at the WMO Technical Conference on Instruments and Methods of Observation (TECO-92), Vienna, Austria, 11-15 May 1992, WMO/TD - No. 462. Instruments and Observing Methods Report No. 49, pp. 431-435. World Meteorological Organization, Geneva, Switzerland. Free, M., D.J. Seidel, J.K. Angell, J. Lanzante, I. Durre, and T.C. Peterson, 2005: Radiosonde Atmospheric Temperature Products for Assessing Climate (RATPAC): A new data set of large-area anomaly time series. J. Geophys. Res., 110, D22101, doi:10.1029/2005JD006169. Gaffen, D.J., 1993: Historical changes in radiosonde instruments and practices. WMO/TD-No. 541. Instruments and Observing Methods Report No. 50. World Meteorological Organization, Geneva, Switzerland, 123 pp. Gaffen, D.J., 1996: A digitized metadata set of global upper-air station histories. NOAA Technical Memorandum ERL ARL-211. [Available online from http://www.ncdc.noaa.gov/oa/climate/igra/index.php?name=metadata.] Miyagawa,K, 1990: A quantitative analysis of day-night temperature differences observed by Japanese rawinsondes. Tenki, 2, 38-48. [NASA Tech. Translation NASA-TT-21268, available from SCITRAN, P.O. Box 5456, Santa Barbara, CA 93150, tel. (805) 969-2413.] Oakley, T., 1993: Report by the rapporteur on radiosonde compatibility: Part A, WMO catalogue of radiosondes and upper-air wind systems in use by members (1993). WMO/TD-No. 587, Instruments and Observing Methods Report No. 56. World Meteorological Organization, Geneva, Switzerland, 89 pp. Raj, Y.E.A., V. Mathew, and J.C. Natu, 1987: Discontinuities in temperature and contour heights resulting from change of instruments at Indian radiosonde stations. Mausam, 38, 407-410. Schwartz, B.E., and M. Govett, 1992: A hydrostatically consistent North American radiosonde data base at the Forecast Systems Laboratory, 1946-present. NOAA Technical Memorandum ERL FSL-4, NOAA/ERL/FSL, 325 Broadway, Boulder, CO 80303, 81 pp. Weather Bureau, 1979: Climate of South Africa: Part 13, Upper-air statistics. WB 39, U.D.C/U.D.K. 551-587(68), Research Division, Weather Bureau, Department of Transport, Pretoria, South Africa, 99 pp. Weather Bureau, 1990: Climate of South Africa: Part 14, Upper-air statistics 1968-1987. ------------------------------------------------------------------------- -------------------------------------------------------------------------