| Date: | 2002-12-23 17:44:00 | |
| Sender: | Van S. Husson, ILRS Central Bureau <Husson, Van <Van.Husson@honeywell-tsi.com>> | |
| Subject: | [SLR-Mail] No. 1020: SLR Data Corrections File | |
| Author: | Van S. Husson, ILRS Central Bureau | |
| Content: | ******************************************************************************** SLR Electronic Mail 2002-12-23 17:44:00 UTC Message No. 1020 ******************************************************************************** Author: Van S. Husson, ILRS Central Bureau Subject: SLR Data Corrections File Dear Colleagues, INTRODUCTION ------------ SLR stations are capable of measuring ranges very precisely and accurately, but sometimes; due to the complexity of our systems, biases can be induced. Engineering analysis can uncover the error sources and frequently bias problems are recoverable. In the past, SLR stations have made announcements via SLRMail to document problem data. However; there has not been a common format for the reporting of this information nor has there been a centralized location for this information. Realizing these weaknesses and the value in this information to the ILRS analyst community, the ILRS CB has established the 1st data corrections file in the Software INdepedent EXchange (SINEX) format. The file format is described in Appendix A & B. The SINEX format is available at: ftp://alpha.fesg.tu-muenchen.de/iers/sinex/format/ In our effort to further enhance ILRS services and to improve our science products as we push toward millimeter/millimetre accuracy, a new product has been produced, the ”SLR_data_corrections file”. This file will provide our data analysts the necessary information to remove known data errors and should prove to be invaluable to analysts who re-analyze historical data (e.g. back to LAGEOS-1 launch). The cause of a bias can be very diverse. Biases that are routinely recoverable through post-engineering analysis (i.e. not via orbit determination) fall into the following categories: range bias - a fixed, constant value that may exist for a pass or series of passes. This correction is applied to the time of flight measurement. time bias - a fixed, constant value that may exist for a pass or series of passes. This correction is applied to the epoch measurement. pressure bias - a fixed, constant value that may exist for a single normal point or a set of normal points or a series of passes. This correction is applied to the barometric measurement. non-linear bias - a set of values which change as a function of the time-of-flight due to non-linearities in the timing equipment. These biases are usually sub-centimeter and can be closely approximated by using a fixed range bias. Please note that a scale bias is not listed above, because it can not be easily measured on-site, and it is caused by an error in frequency (i.e. the time base to the counter). Frequency errors can be large and will drift, therefore, when this problem is known to exist, it is highly recommended to edit or down weight the data. ACCESSIBILITY ------------- The first version of the SLR data corrections file was an attempt to document any known biases in the year 1999 in support of the Analysis Working Group Pilot Project on Positioning and Earth Orientation Parameters. This file will be maintained to bring the corrections current and also to include historical problems (i.e. prior to 1999). The file is available at the ILRS Data Centers: CDDIS at: ftp://cddisa.gsfc.nasa.gov/pub/slr/slrql/slr_data_corrections.snx EDC at: ftp://ftp.dgfi.badw-muenchen.de/pub/laser/qldata/slr_data_corrections.snx USERS GUIDE ----------- In addition to data corrections, this file also documents ”BAD” data in 1999. The CSR and other weekly quick-look reports were very helpful in identifying potential problem normal points and passes. A special thanks to my colleague, Mark Davis, who provided LAGEOS-1 and LAGEOS-2 residuals for 1999. ”BAD” data are usually blunders and are usually unrecoverable. ”BAD” data can be individual normal points, a pass segment, an entire pass, or a series of passes. The four most common causes of these type errors are: 1. Data smoothing errors caused by sparse data usually near the start or end of a pass or a gap in the data. 2. Equipment malfunctions 3. Procedural error (i.e. improper calibration) 4. Incorrect date due to a software error (e.g. mid-night crossing or new year). Description of BIAS/EPOCHS BLOCK -------------------------------- The bias file contains 2 SINEX blocks (BIAS/EPOCHS and SOLUTION/APRIORI). These blocks of data are sorted by the 4-digit CDP site identifiers and the start epoch. The BIAS/EPOCHS block contains the following fields (see Appendix A for more detail): 1. The 4-digit site code (i.e. the 4-digit CDP site identifier) 2. The 1-character observation code (for ILRS, the BIAS code) 3. The 1-character solution ID number (for ILRS the data release flag) 4. The 2-character point code (for ILRS, the satellite identifier) 5. The start and end epochs The mean epoch was intentionally left blank. Lines which begin with a ”*” are comments. NOTES: Any bias that is recoverable will HAVE a corresponding entry in the SOLUTION/APRIORI blocks. In some cases, two different biases may apply to a site during the same time interval. In this case, all biases should be applied to the data. Unrecoverable errors (i.e. BAD data) will NOT HAVE a corresponding entry in the SOLUTION/APRIORI blocks, because the data is to be edited. Spans of BAD data may overlap periods where there is a fixed bias. In cases where the time spans for ”X” and real corrections overlap, periods of BAD data should take precedence. Description of SOLUTION/APRIORI Block ------------------------------------- The SOLUTION/APRIORI block contains the following fields (see Appendix B for more detail): 1. The 6-character parameter type (for ILRS, the bias type) 2. The 4-digit site code (i.e. the 4-digit CDP site identifier) 3. The 2-character point code (for ILRS, the satellite identifier) 4. The 1-character solution number (for ILRS the data release flag) 5. The start epochs 6. The parameter units 7. The constraint code (for biases, this has been set to ”0” indicating a fixed/tight constraint) 8. The parameter apriori (i.e. the actual bias) 9. The parameter standard deviation (i.e. the engineering uncertainty in the data correction) The Apriori Parameter Index is non-applicable in this situation and therefore left blank. Lines which begin with a ”*” are comments. NOTES: Any bias that is recoverable will HAVE a corresponding entry in the BIAS/EPOCHS blocks with exactly the same start epoch. The convention for the sign (i.e. ”+” or ”-”) of the corrections are as follows: A range bias of +1 meter means that the observation(s) is(are) 1 meter long. (this is a correction to the normal point time of flight converted into a 1-way range by the equation) range = C * normal point time of flight (in secs) / 2 where C is the speed of light in a vacuum (2.99792458 m/sec) new (corrected) range = range - range bias For example, for a +1m range bias and a initial range of 7,000,000m: new range = old range(7,000,000m) - range bias(+1m) ”” = 7,000,000m - 1m ”” = 6,999,999m A time bias of +1 millisecond means that the observation(s) is(are) 1 millisecond late. A pressure bias of +1 millibar means that the observation(s) is(are) 1 millibar too high. NOTE: The current SINEX format does not support non-linear counter corrections that Zimmerwald (7810) and Herstmonceux (7840) have reported. However, a simple fixed bias will remove most of the error for a given set of altitude satellites (i.e. LEO, LAGEOS, high). MAINTENANCE ----------- 1. File Name - The name will always be slr_data_corrections.snx for ease of accessibility. A date of the last modification will be documented within the file. 2. Version Control - Previous file versions will be renamed to: slr_data_corrections_yyyymmdd.snx and a difference file will be generated to let users know what was changed between versions. The difference file will be named: slr_data_corrections_yyyymmdd.dif . where yyyymmdd is the year, month and day of the previous file. 3. Stations are strongly encouraged to review their historical records and document any known errors or periods of poor performance and to provide this information via e-mail in the format described to the ILRS Central Bureau (CB) at cb@ilrs.gsfc.nasa.gov. The CB will work with an stations who have problems with the SINEX format. 4. Quality Control - Prior to a file update the CB will review all corrections provided by the ILRS network for format, data integrity, and completeness. Comments/suggestions are welcome. Best regards, Van Husson, ILRS Central Bureau (van.husson@honeywell-tsi.com) APPENDIX A: BIAS/EPOCHS FORMAT Description: List of epochs of data corrections for each Site Code/Point Code/Solution Number/Bias Type combination. Contents: ___________________________________________________________________ |____B_I_A_S___E_P_O_C_H_S___D_A_T_A___L_I_N_E______________________| | | | | |__Field_________|______Description__________________|___Format_____| | | | | | [Site Code] | Site code for which some | 1X,A4 | | | parameters are estimated. | | | |(ILRS: 4 digit CDP#, i.e. marker#) | | |________________|___________________________________|______________| | | | | | [Point Code] | Point Code at a site for which | 1X,A2 | | | some parameters are estimated. | | | | (ILRS: Satellite abbr.) | | | | Satellite abbreviations are: | | | | | | | | L1: LAGEOS-1 | | | | L2: LAGEOS-2 | | | | LC: LAGEOS-1,2 combined | | | | E1: Etalon-1 | | | | E2: Etalon-2 | | | | EC: Etalon-1,2 combined | | | | --: Wildcard (all satellites) | | |________________|___________________________________|______________| | | | | | [Solution ID] | Solution Number at a Site/Point | 1X,A4 | | | code for which some parameters | | | | are estimated. | | | | | | | | For ILRS only, the solution | | | | number is the data release flag | | | | (i.e. byte 48 in the normal point | | | | data record). The 1st release of | | | | data is indicated with ´0´. The | | | | 2nd release of data is indicated | | | | with a ´1´ and .... | | |________________|___________________________________|______________| | | | | | [Observation | Identification of the type of | | | Code] | observation bias. | 1X,A1 | | | | | | | Observation bias codes are: | | | | R: range bias | | | | T: time bias | | | | P: pressure bias | | | | S: scale bias | | | | Z: tropospheric bias at zenith | | | | X: blunder, please edit | | |________________|___________________________________|______________| | | | | | [Time] | Start time for which the solution | 1X,I2.2, | | | identified (SPNO) has observations| 1H:,I3.3, | | | | 1H:,I5.5 | |________________|___________________________________|______________| | | | | | [Time] | End time for which the solution | 1X,I2.2, | | | identified (SPNO) has observations| 1H:,I3.3, | | | | 1H:,I5.5 | |________________|___________________________________|______________| | | | | | [Time] | Mean time of the observations for | 1X,I2.2, | | | which the solution (SPNO) is | 1H:,I3.3, | | | derived. | 1H:,I5.5 | |________________|___________________________________|______________| | | | 54 | |______________| APPENDIX B: SOLUTION/APRIORI FORMAT Description: Apriori data corrections for SLR. Contents: ___________________________________________________________________ |______S_O_L_U_T_I_O_N___A_P_R_I_O_R_I_____D_A_T_A___L_I_N_E________| | | | | |__Field_________|______Description__________________|___Format_____| | | | | | Apriori Parame-| Index of apriori parameters. | 1X,I5 | | ters Index | values from 1 to the number of | | | | parameters. | | |________________|___________________________________|______________| | | | | |[Parameter Type]| Identification of the type of | 1X,A6 | | | parameter. Typical parameters | | | | are: | | | | | | | | RBIAS: range bias | | | | TBIAS: time bias | | | | SBIAS: scale bias | | | | PBIAS: pressure bias | | | | ZBIAS: tropospheric bias (zenith) | | |________________|___________________________________|______________| | | | | | [Site Code] | Site code with apriori parameter | 1X,A4 | | | estimate. | | | |(ILRS: 4 digit CDP#, i.e. marker#) | | |________________|___________________________________|______________| | | | | | [Point Code] | Point Code with apriori parameter | 1X,A2 | | | estimate. | | | | (for ILRS: satellite abbr.) | | | | | | | | Satellite abbreviations are: | | | | | | | | L1: LAGEOS-1 | | | | L2: LAGEOS-2 | | | | LC: LAGEOS-1,2 combined | | | | E1: Etalon-1 | | | | E2: Etalon-2 | | | | EC: Etalon-1,2 combined | | | | --: Wildcard (all satellites) | | |________________|___________________________________|______________| | | | | | [Solution ID] | Solution ID at a Site/Point | 1X,A4 | | | code with apriori parameter | | | | estimate. | | | | | | | | For ILRS only, the solution | | | | number is the data release flag | | | | (i.e. byte 48 in the normal point | | | | data record). The 1st release of | | | | data is indicated with ´0´. The | | | | 2nd release of data is indicated | | | | with a ´1´ and .... | | |________________|___________________________________|______________| | | | | | [Time] | Epoch at which the apriori | 1X,I2.2, | | | parameter is valid. | 1H:,I3.3, | | | | 1H:,I5.5 | |________________|___________________________________|______________| | | | | | Parameter Units| Units used for the aprioris and | 1X,A4 | | | sigmas. Typical units are: | | | | | | | | m: meters | | | | ms: milliseconds | | | | us: microseconds | | | | mas: milli-arc-seconds | | | | mb: millibars |________________|___________________________________|______________| | | | | | [Constraint | Constraint applied to the parame- | 1X,A1 | | Code] | ter. | | | | 0: fixed/tight constraints | | | | 1: significant constraints | | | | 2: unconstrained | | |________________|___________________________________|______________| | | | | | Parameter | Apriori value of the parameter. | 1X,E21.15 | | Apriori | | | |________________|___________________________________|______________| | | | | | Parameter | Apriori standard deviation for | 1X,E11.6 | | Standard | the parameter. | | | Deviation | | | |________________|___________________________________|______________| | | | 80 | |______________| From: ”Husson, Van” ******************************************************************************** |