An Introduction to the Special Sensor Microwave Water Vapor Profiler Sensor Counts Data Set TABLE OF CONTENTS 1.0 Introduction 2.0 Special Sensor Microwave Water Vapor Profiler 3.0 SSM/T-2 Data Set Summary 3.1 File Naming Conventions 3.2 SSM/T-2 Header Record Format 3.3 SSM/T-2 Data Record Format 4.0 References 1.0 Introduction This README file contains information on the Special Sensor Microwave Water Vapor Profiler (SSM/T-2) sensor count data sets. Brief descriptions of the Defense Meteorological Satellite Program (DMSP) satellites, the SSM/T-2 instrument, the computation of antenna temperatures and the format of the data sets are included. Pertinent scientific references are also included. The Defense Meteorological Satellite Program (DMSP) is a Department of Defense program which is responsible for designing, building, launching and operating polar orbiting meteorological satellites. The satellites can broadcast visual, infrared and microwave imagery directly to transportable tactical sites around the world. The data is also stored for transmission to the Navy's Fleet Numerical Meteorology and Oceanography Center (FNMOC) and to the Air Force Global Weather Central (AFGWC). Each of the DMSP satellites flies in a sun-synchronous, near-polar orbit. For a satellite in sun synchronous orbit, the ascending equatorial crossing time remains relatively constant with respect to the local time throughout the lifetime of the satellite. The SSM/T-2 sensor is flown aboard the DMSP F-11 and F-12 satellites. 2.0 The Special Sensor Microwave Water Vapor Profiler The information in this section was taken from the DMSP SSM/T-2 Calibration and Validation Document (1992). The SSM/T-2 is the first operational microwave water vapor sounder to be placed in orbit. It is a scanning, five channel, passive and total power microwave radiometer system. The SSM/T-2 consists of five channels: three water vapor channels centered around the 183.31 GHz water vapor line, a 91 GHz channel, and a 150 GHz channel. Table 2 lists the channel characteristics. The SSM/T-2 observation rate is 7.5 scans per minute. There are 28 observations (beam positions) per scan for each of the five channels, with each observation having a spatial resolution of approximately 48 km. All five channels have coincident centers. The total swath width for the SSM/T-2 is approximately 1400 km. Table 2. Channel Frequency Nadir Beamwidth Peak Number (GHz) Field-of-View (km) (degrees) (hpa) 1 183.31 +/- 3 48 3.3 650 2 183.31 +/- 1 48 3.3 500 3 183.31 +/- 7 48 3.3 800 4 91.665 +/- 1.25 84 6.0 surface 5 150.0 +/- 1.25 54 3.7 1000 The SSM/T-2 employs a single offset parabolic reflector with a 2.6 inch diameter projected aperature. The reflector is shrouded to eliminate the possibility of rays from the sun striking either of the calibration paths and causing unwanted thermal gradients. The feedhorn is a corrugated pyramidal horn with a flare designed to minimize phase center separation over the bandwidth (91 to 183.3 GHz), while providing a spherical wave illumination of the reflector. For 183.3 GHz, the antenna efficiency is greater than 94%. Table 3 shows the antenna subassembly beam efficiency. A 3.3 degree beamwidth is achieved for the 183.3 GHz channels and larger beamwidths of approximately 3.7 degrees and 6.0 degrees for 150 and 91.665 GHz, respectively. These correspond to the Field-of-View (FOV) parameters given in Table 2. Table 3. Beam Position Beam Efficiency Specified Measured 1 >= 93% 94.29% 14 >= 93% 94.43% To achieve the cross-track scanning, the reflector alone rotates. The rotation of the reflector produces a rotation of the plane of polarization of the upwelling scene Brightness Temperatures which is permitted provided that the polarization remains identical for the two window channels and 183.3 +/- 7 GHz. These channels must have the same polarization characteristics because they measure contributions from both the atmosphere and the surface. Note that all SSM/T-2 channels possess the same polarization. The SSM/T-2 inflight warm-load calibration target is a derivative of the SSM/T-2 warm load calibration target. The warm load (~300K) is shrouded to improve RF coupling of energy to the reflector/feedhorn antenna. This minimizes potential calibration errors arising from the reception of extraneous energy due to scattering of earth or solar radiation off of the spacecraft. 3.0 SSM/T-2 Data Set Summary 3.1 File Naming Conventions The SSM/T-2 orbit files are stored as one logical daily file by employing the Unix tape archive (tar) procedure. Each daily file is then compressed using the standard Unix compression routine. The SSM/T-2 daily files are named using the convention: scxxt2yy.ddd_nesdis_swath.tar.Z where sc = sensor counts xx = satellite number (11 or 12) yy = year of the century ddd = julian day tar = Unix tar format Z = Unix compress format When an SSM/T-2 data file is uncompressed and untarred, the results will be one day's worth of data in individual files. These files will follow the naming convention: NSS.SMT2.Sn.Dyyjjj.Shhmm.Ehhmm.Axxxxxxx.NS.A where the upper case characters remain fixed and the lower case characters vary from orbit to orbit. The lower case characters correspond to the following variables: n = Spacecraft identification (5 for F11 and 6 for F12) yy = Year of the century jjj = day of the year hhmm = Time in hours and Minutes xxxxxxx = Orbit Number The upper case characters represent the following: NSS.SMT2 = NESDIS SSM/T-2 D = Day S = Start Time E = End Time 3.2 SSM/T-2 Header Record Format The information in this section was taken from the SSM/T-2 Level 1b Interface Control Document (1994). Each SSM/T-2 level 1b data set contains a header record. A header record contains the following data groups: Idenfication Block 48 Bytes Preflight Calibration Data 108 Bytes Antenna Pattern Correction Data 280 Bytes QC Summary 14 Bytes Spare 242 Bytes Each of these data groups are described in 3.2.1 thru 3.2.4. 3.2.1 Identification Block The identification block is contained in Bytes 1 - 48 in the header record. This data group contains the data set name, the number of scans and the number gaps in the data. The data set name provides the spacecraft identification, the orbit start day, orbit start and end times, and the processing block identification. The breakdown of the identification block can be found in Table 4. The spacecraft identification is a numerical ID assigned to a spacecraft (i.e. ID=6 corresponds to spacecraft F12). The start and end times of the level 1b orbit are rounded off to the nearest five minutes. Therefore, the first scan may not necessarily be at the start time and the last scan may not necessarily be at the end time. The processing block identification contains a letter code followed by a five digit starting rev number and a two digit ending orbit number. The ending orbit number is obtained by incrementing the last two digits of the starting orbit number by one. The total number of scan records in the data set is given by the number of scans parameter. The number of data gaps parameter corresponds to missing data. If the scene data block is empty or missing, then it is considered as a data gap. If a data gap covers one or more consecutive scans then it is counted as one data gap. The data gap parameter can be used to determine the completeness of the data. 3.2.2 Preflight Calibration Data The preflight calibration data group is contained in bytes 49 - 156 in the header record. This data group includes the coefficents used to convert warm load calibration counts to the corresponding temperatures, as well as the correction terms used to compute the slope and intercepts. To compute the temperatures, a set of eleven counts and the corresponding temperatures are provided for each thermistor. Included in the correction terms are a cold path temperature correction term and a warm path temperature correction term for each channel. The preflight calibration data are retreived from the SSM/T-2 constants file. The breakdown of the preflight calibration data group is provided in Table 4. Note: Preflight calibration data have no utility to the user. These data are intended for troubleshooting calibration related problems. 3.2.3 Antenna Pattern Correction Data The antenna pattern correction data group is contained in bytes 157 - 436 in the header record. This data group provides the coefficients required to perform antenna pattern correction. No correction to antenna pattern is performed by the Level 1b software. Therefore, the coefficients to perform this correction as a post-processing step are supplied in this data group. The correction coefficients are stored in the following order: Words 1 - 5 Beam Postion #1 Channels 1 - 5 Words 6 - 10 Beam Postion #2 Channels 1 - 5 . . . . . . Words 136 - 140 Beam Postion #28 Channels 1 - 5 3.2.4 Quality Control Summary The quality control (QC) summary data group is contained in bytes 437 - 450 in the header record. This data group provides a summary of the quality of the earth locations, scene data and calibration data at the orbit level. The overall quality of the level 1b data can be determined from the QC summary. The QC summary is reported as a percentage of the total number of samples upon which the quality control is performed. The criteria used to assess the quality of a scan or channel are as follows. While computing earth locations, the two ephemeris minute vectors are verified for valid data. The scan time is verified to determine if it lies between the ephemeris times. The earth locations are not computed if the ephemeris data are deemed invalid or the scan time is not bounded by the ephemeris times and a filler value of 0 is used for the earth locations. If a scene data block is empty or missing, the corresponding scan is treated as bad as far as the scene data are concerned. No filler values are provided as substitutes for missing data. The calibration algorithm uses the averages of calibration data taken from 8 scans (four preceding scans, the scan being calibrated, and three succeeding scans). A scan or channel is treated as bad, as far as the calibration data are concerned, when on of the following conditions is satisfied: 1. Warm load counts failed the limit check or the difference of warm load temperatures computed from the two thermistors exceeded the pre-defined limit or less than four good scans were obtained to perform the averages. 2. All cold view or warm view counts from a given channel failed the limit check. 3. The averages of cold view and warm view counts were identical in a given channel. 4. Slope or intercept failed the limit check in a given channel. When the first condition is satisfied, the entire scan is not calibrated and the individual channels are treated as bad. When conditions 2, 3 or 4 are satisfied, the corresponding channel is treated as bad. In all other instances, the quality of earth location, scene data and calibration counts is validated to be good and identified as such. The level 1b software compiles the data accumulated over the entire orbit, computes statistics and stores the results in the header record in terms of percentages. Table 4: Format of the SSM/T-2 Level 1b Header Record Contents Count*Bytes Position Scaling Factor Level 1b Data Set Name 1*44 1-44 Number of Scans 1*2 45-46 Number of Data Groups 1*2 47-48 Warm Load Thermistor #1 Counts 11*2 49-70 Warm Load Thermistor #1 Temperatures (in degrees C) 11*2 71-92 100 Warm Load Thermistor #2 Counts 11*2 93-114 Warm Load Thermistor #2 Temperatures (in degrees C) 11*2 115-136 100 Cold Path Temperature Correction (ch 1-5) (in degrees K) 5*2 137-146 100 Warm Path Temperature Correction (ch 1-5) (in degrees K) 5*2 147-156 100 Antenna Pattern Correction (5 channels X 28 beam positions) 140*2 157-436 100 QC Summary (Earth loc, scene data, cal-ch. 1-5) 7*2 437-450 Spares 121*2 451-692 3.3 SSM/T-2 Data Record Format The information in this section is quoted from SSM/T-2 Level 1b Interface Control Document (1994). Each SSM/T-2 level 1b data set contains up to 900 data records. Each data record contains one scan of SSM/T-2 data. The data record contains the following data groups: Scan Information 20 bytes Earth Location Data 112 bytes Scene Data 336 bytes Calibration Data 192 bytes Quality Control Data 14 bytes Spare 18 bytes. Table 5 shows the format of the SSM/T-2 Level 1b data records and each of the data groups are described in 3.3.1 thru 3.3.5. Table 5: Format of the SSM/T-2 Level 1b Data Record Contents Count*Bytes Position Scaling Factor Orbit Number 1*4 1-4 Scan Number 1*2 5-6 Scan Index to SSM/T1 1*2 7-8 SSM/T-2 Scan Time (Year and Day, OLS, Scan start time) 3*4 9-20 Earth Locations (28 lat-long pairs) 56*2 21-132 128 Time at Beam Position #1 (Milliseconds to be added to OLS) 1*2 133-134 Raw Counts (ch. 1-5 raw data at beam position #1) 5*2 135-144 . . . Time at Beam Position #28 (Milliseconds to be added to OLS) 1*2 457-458 Raw Counts (ch. 1-5 raw data at beam position #1) 5*2 459-468 Warm Calibration Counts (5 channels X 4 views) 20*2 469-508 Cold Calibration Counts (5 channels X 4 views) 20*2 509-548 Stepped Automatic Gain Control (ch. 1-5) 5*2 549-558 Thermal Reference Voltage 1*2 559-560 Thermistor Counts (TEMP 1-TEMP 18) 18*2 561-596 Warm Load Temperatures (Thermistors 1-2)(in degrees Kelvin) 2*2 597-600 100 Warm Load Temperature Average (Ch. 1-5)(in degrees Kelvin) 5*2 601-610 100 Warm Load Count Average (Ch 1-5) 5*2 611-620 4 Cold Load Count Average (Ch 1-5) 5*2 621-630 4 Cold Load Temperatures (Ch 1-5)(in degress Kelvin) 5*2 631-630 100 Slopes (Ch. 1-5) 5*2 641-650 10,000 Intercepts (Ch. 1-5) 5*2 651-660 100 QC Flags (Earth Location, Scene Data, Calibration) 7*2 661-674 Spares 9*2 675-692 3.3.1 Scan Information The scan information data group is contained in bytes 1 - 20 in the data record. This data group includes the orbit number, scan number, scan index and scan time. The orbit number corresponds to the rev number provided in the readout header of the raw data. The scan number is a sequence number assigned to a scan. It also coincides with the data record number, that is, the scan number in the first data record is 1, the scan number in the second data record is 2, and so on. The scan index is used for collocation with T1 data. The index is a composite of scan group and scan sequence number. The scan group is a sequence number assigned to a set of 4 T-2 scans and the scan sequence number is a number which identifies individual scans within that group. The scan index is stored as scan group*10+scan sequence. For example, if the scan number is 75, the corresponding index is 193 (scan numbers 73-76 become group 19; scan number 75 is the third scan in that group, that is, 19*10+3). The scan time contains the year of the century and the day, OLS time in seconds and TS in milliseconds. The OLS time is the time extracted from the first subframe (data ident = 0, 8, 16 or 24) of 8 subframes which make up the T-2 scan. TS is the time between beam position 1 and the following readout enable. TS is used as an additive term in the earth location algorithm and has no utility to the user. The scan time is stored in the following format: Bytes 9 - 12 Year of century and day of year in YYJJJ format Bytes 13-16 OLS time in seconds Bytes 17-20 TS in milliseconds Note: The day parameter is not reset for orbits which are on a day boundary. 3.3.2 Earth Location Data The earth location data group is contained in bytes 21-132 of the data record. This data group includes the earth locations for each of the 28 beam positions. The earth locations (latitude-longitude pairs) appear in the following order: First 2 bytes Latitude of beam position #1 Next 2 bytes Longitude of beam position #1 Next 2 bytes Latitude of beam position #2 Next 2 bytes Longitude of beam position #2 and so on. The earth locations are specified in degrees and the following convention is used for latitude-longitudes: Latitudes: North > 0 and South < 0; (-90 <= lat <= 90) Longitudes: East > 0 and West < 0; (-180 <= long <= 180). Note: The earth locations provided in the data record must be descaled to determine the approriate latitude and longitudes. If the earth locations of all beam positions are identically equal to zero, it indicates missing earth locations and these locations must not be used. 3.3.3 Scene Data The scene data group is contained in bytes 133-468 of the data record. This data group provides the time at each beam position followed by raw counts of all channels. Scene data appear in the following order: Word 1 Time at beam position #1 Word 2-6 Raw counts of channels 1-5 Word 7 Time at beam position #2 Word 8-12 Raw counts of channels 1-5 . . . . . . Word 163 Time at beam position #28 Word 164-168 Raw counts of channels 1-5 The raw counts range from 0 to 4095. These counts are used to compute channel brightness temperatures using the following relationship: BTn = Bn * Cn + An where n is the SSM/T-2 channel number (1-5), BT, B, C and A are the brightness temperatures, slopes, raw counts and offsets, respectively. The time given at each beam position is the time measured relative to the OLS time. The beam position time is stored in milliseconds and can be converted to an absolute time by adding these milliseconds to the OLS time (byte 13-16). Since the beam position times are derived as a function of ephemeris time, occasionally, the beam position times could lag behind the OLS time by a few seconds. This is a normal condition. 3.3.4 Calibration Data The calibration data group is contained in bytes 469-660 of the data record. Calibration data consists of the following parameters: Warm calibration counts (5 channels X 4 views) Cold calibration counts (5 channels X 4 views) Stepped automatic gain controls (channels 1-5) Thermal reference voltage Thermistor counts (18) Warm load temperatures (2 thermistors) Warm load temperature averages (channels 1-5) Warm count averages (channels 1-5) Cold count averages (channels 1-5) Cold load temperatures (channels 1-5) Slopes (channels 1-5) Intercepts (channels 1-5) Note: Only the channel slop and offset values have utility to the user. The other parameters provided under the calibration data group are intended for troubleshooting calibration related problems. 3.3.5 Quality Control Data The quality control data group is contained in bytes 661-674 of the data record. This data group indicates the quality of earth locations, scene data and calibrationat the scan level. The criteria used for quality control is outlined in section 3.2.4 above. The QC data appears in the following order: Word 1 Earth locations QC flag Word 2 Scene data QC flag Word 3-9 Calibration QC flags (channels 1-5) If the earth locations QC flag contains a non-zero value, the earth locations from that scan must not be used. If the scene data QC flag contains a non-zero value, it indicates missing scene data and the scene data from that scan should be used with caution. If a calibration QC flag contains a non-zero value, the slopes and intercepts computed for that channel are probably erroneous and are not usable. Note: The QC data provided in this data group are only applicable to the data record in which they are reported. 4.0 References Falcone, V.J., M.K. Griffin, R.G. Isaacs, J.D. Pickle, J.F. Morrissey, A.J. Jackson, A. Bussey, R. Kakar, J. Wang, P. Racette, D.J. Boucher, B.H. Thomas, A.M. Kishi: DMSP F11 SSM/T-2 Calibration and Validation Data Analysis, Phillips Laboratory, Hanscom Air Force Base, MA, 29 October 1992. Tadepalli, K.: SSM/T-2 Level 1b Interface Control Document, NOAA/NESDIS, Suitland, MD, 11 July, 1994.