TARFOX_UWC131A_SUNP Readme 1.0 Introduction This file contains information about files in the TARFOX_UWC131A_SUNP data set and the sample read software provided for those files. This data set is available through the NASA Langley Distributed Active Archive Center (DAAC). This data set contains measurements made by the NASA Ames Research Center 6-channel Sunphotometer flown on the University of Washington C-131A aircraft during the Tropospheric Aerosol Radiative Forcing Observational eXperiment (TARFOX) mission in July, 1996. The file names indicate the flight date and flight number for that date and are of the form ames_sunp_960710_1. The files are ASCII data, with a format following the "Format Specification for Data Exchange" developed by Gaines and Hipskind of NASA Ames. A PostScript version of that document is included with the README files for this data set; the files are named readme_tarfox_uwc131a_sunp_gaines_hipskind_1.ps and readme_tarfox_uwc131a_sunp_gaines_hipskind_2.ps. The data files can be read with the DAAC-provided read software, read_sunp.f. This Readme file includes five other sections: Section 2.0 - describes the available read software. Section 3.0 - discusses how to create an executable program from read software source code. Section 4.0 - demonstrates how to invoke the executable program. Section 5.0 - provides general information for the data set. Section 6.0 - provides information about the measurements and processing algorithms. Additional information about the TARFOX project may be obtained from: http://geo.arc.nasa.gov/sgg/tarfox/ If there are questions about using the read software, please contact the Langley Science User and Data Services (SUDS) office. The SUDS office may be reached by E-Mail: support-asdc@earthdata.nasa.gov, by telephone at (757)864-8656, or by FAX at (757)864-8807. The SUDS mailing address is: Langley DAAC Science User and Data Services Office NASA Langley Research Center Mail Stop 157D, 2 S. Wright St. Hampton, VA 23681-2199 USA 2.0 Read Software Files Currently, there is one sample read program for this data set, read_sunp.f. It is written in Fortran 77. This program has been tested on the following computers and operating systems: Computer Operating System ------------------- ---------------- Sun Sparc Solaris 2.6 SGI Origin 2000 IRIX 6.5 DEC Alpha Digital UNIX 4.0A This program is written as an example of how to read in the TARFOX Ames Sunphotometer data. As delivered, it reads in all of the information in a data file writes a portion of the data to an output file. 3.0 Creating Executable Program Files 3.1 Create with Fortran 77 compiler To compile the program, use the Unix f77 ccommand to invoke the Fortran 77 compiler: f77 read_sunp.f -o read_sunp Use of the -o option above creates an executable program file named read_sunp. If this option is omitted, the executable program file will have the default name of a.out. If the appropriate compile command is not found on your system, check to be sure that your PATH environment variable includes the directory that contains the compiler or specify the entire path to the compiler in your command. 4.0 Invoking Read Software 4.1 Running the executable version of the Ames Sunphotometer Fortran Read Program The executable program can be invoked at the command line by entering the name of the executable program file. If the -o option was used as shown in Section 3.1, the command would be read_sunp The program prompts for the name of the Ames Sunphotometer data file to be read. If the file is not in the directory from which the program is invoked, it is necessary to specify the full directory path to the data file along with its name. The program allows for path/file names up to 36 characters long. If your file name is longer, you can modify the length of the byte variable used to contain the input file name in the line BYTE FILESUNREAD(36) changing 36 to the desired length. The program writes to a file named REWRSUNARC.WRT. If that file already exists in the current directory, the program will exit with an error when it tries to open that file. During execution, all parameters from the file are read into arrays, but as delivered, not all are written to output. It is expected that the user will modify the program to perform their desired analysis and output needed parameters. As delivered, the program writes a partial list of auxiliary and primary values to the output file in ASCII in a format suitable for import into a spreadsheet or other program. The following is a sample sesson showing compilation and execution of the read_sunp program on a machine named darrin. darrin 4% f77 read_sunp.f -o read_sunp darrin 5% read_sunp Input name of Ames Sunphotometer file to be read: ames_sunp_960710_1 END OF FILE REACHED AFTER READING 2420 OBSERVATIONS 5.0 Data Set Information This data set is written in ASCII following the Gaines-Hipskind Format for Data Exchange (see Section 1.0). LONG NAME FOR THIS DATA SET: Tropospheric Aerosol Radiative Forcing Observational eXperiment Ames Sunphotometer Measurements made from University of Washington C-131A Aircraft FILE NAMING CONVENTION: TAROX_UWC131A_SUNP data set files are prefixed with "ames_sunp_" to indicate that they belong to this data set. This is followed by a six digit date indicating the date the measurements were obtained, an underscore, and a flight number for that date. Here is an example file name: ames_sunp_960725_2 data taken from the second flight on July 25, 1996 5.1 Data File Format The organization of the data file is shown below, using the same variable names used in the Gaines-Hipskind specification (which are explained following the file layout). NLHEAD FFI ONAME ORG SNAME MNAME IVOL NVOL DYEAR DMON DDAY RYEAR RMON RDAY DX (1..2) NX NXDEF X (1..NXDEF) XNAME (1..2) NPV PVSCALE (1..NPV) PVMISS (1..NPV) PVNAME (1) ... PVNAME (NPV) NAUXV AUXVSCALE (1..NAUXV) AUXVMISS (1..NAUXV) AUXVNAME (1) ... AUXVNAME (NAUXV) NSCOML SCOM (1) ... SCOM (NSCOML) NNCOML NCOM (1) ... NCOM (NNCOML) IV (1,1..2) AUXV (1,1..NAUXV) PV (1,1..NPV) IV (2) AUXV (2,1..NAUXV) PV (2,1..NPV) ... IV, AUXV and PV are repeated for each (monotonically increasing) value of IV nlhead - number of lines in the file header ffi - file format index, a number that (loosely) describes the particular format used in the data file The ffi for these files is 1010, indicating one independent variable oname - originator of data file org - organization or affiliation of the data file originator sname - source of the measurements (instrument name, measurement platform, etc. mname - mission which the data is supporting ivol - volume number out of total number of volumes (not used?) see nvol below nvol - total number of volumes required to store the complete data set (not used?) dyear - UT date at which the data in the file begins. For dmon aircraft data, this it the UT date of takeoff. Format dday is YYYY MM DD, all numeric. This code stores the date in three separate variables listed at left. ryear - date of data reduction or revision. Format is YYYY MM DD. rmon This code stores in three variables listed at left. rday dx - interval between values of the independent variable nx - number of values for the independent variable nxdef - number of values of the independent variable which are explicitly recorded in the file header xname - name and/or description of the independent variable npv - number of primary (dependent measured) variables in the data file pvscales- array of scale factors, one for each primary variable, by which that primary variable is mutiplied to obtain data values in the units specified for that variable pvmiss - array of values used to indicate missing or erroneous data, one for each primary variable pvnames - array of names/descriptions, one for each primary variable nauxv - number of auxiliary variables, variables associated with a value of of the independent variable additional related measurements auxvscales - array of scale factors, one for each auxiliary variable, by which that auxiliary variable is multiplied to obtain data values in the units specified for that variable auxvmiss- array of values used to indicate missing or erroneous data, one for each auxiliary variable auxvnames- array of names/descriptions, one for each auxiliary variable nscoml - number of special comment lines (used to note special problems or circumstances concerning the data) scom - array of all special comment lines nncoml - number of normal comment lines ncom - array of all normal comment lines iv - a single independent variable value auxv - array of auxiliary variable values for a given independent variable value pv - array of primary variable values for a given independent variable value Questions about this data may be addressed to: Philip B. Russell NASA Ames Research Center Mail Stop 245-5 Moffett Field, CA 94035-1000 U.S.A. Telephone: +1 415 604-5404 Fax: +1 415 604-3625 Internet: philip_russell@qmgate.arc.nasa.gov or to: Langley DAAC User Services NASA Langley Research Center MS 157D Hampton, VA 23681-2199 U.S.A. Telephone: +1 757 864-8656 Fax: +1 757 864-8807 Internet: support-asdc@earthdata.nasa.gov Web URL: http://eosweb.larc.nasa.gov 5.2 Sample Data File Here are the first 80 lines from the data file ames_sunp_960710_1: 56 2010 Russell, Philip B., John M. Livingston, Jens Redemann NASA Ames Research Center Ames 6-Channel Tracking Sunphotometer TARFOX 1 1 1996 7 10 2000 3 7 0.0 0.0 4 4 380.1 450.7 525.3 1020.7 Wavelengths (nm) Elapsed UT seconds from 0 hours on day given by DATE 5 0.001 0.001 0.001 0.001 0.001 99999 99999 99999 99999 99999 Detector output voltage Total optical depth Rayleigh optical depth Particulate optical depth Absolute uncertainty in particulate optical depth 9 0.001 0.001 0.1 1.0 1.0 0.001 0.001 0.01 0.01 99999 99999 99999 99999 99999 99999 99999 99999 99999 Aircraft latitude (deg) at the indicated time Aircraft longitude (deg) at the indicated time Atmospheric pressure (hPa) at the indicated time Aircraft pressure altitude (m) at the indicated time Aircraft geometric altitude (m) at the indicated time Airmass for uniformly distributed gases at the indicated time Detector output voltage in 941.4-nm channel Water vapor column content (cm) derived from measurement in 941.4-nm channel Uncertainty in water vapor column content (cm) 0 21 University of Washington C-131A Flight No: 1723 Ozone column content [matm-cm]: 310.0 Nitrogen dioxide columnar number density [molec/cm2]: 5.000e+015 Earth-sun distance [au]: 1.0166652 Relative uncertainty in airmass [%]: 0.25 Relative uncertainty in Rayleigh optical depth [%]: 1.0 Relative uncertainty in ozone optical depth [%]: 15.0 Relative uncertainty in nitrogen dioxide optical depth [%]: 30.0 Absolute uncertainty in instantaneous measured voltage [mV]: 10.0 Cloud screening relative standard deviation criterion: 0.0030 Wavelength [nm]: 380.1 450.7 525.3 1020.7 Ozone opt depth: 0.000 0.001 0.018 0.000 NO2 opt depth: 0.003 0.002 0.001 0.000 VZERO [V]: 5.494 5.213 7.628 6.967 Rel Unc in VZERO [%]: 1.0 0.8 0.7 0.6 Special Notes: Possible thin clouds indistinguishable from haze after 2100 UTC, but standard deviation in aerosol optical depth <= 0.02 for this time period. 69440 38657 -75242 9033 958 1026 1184 4436 110 14 2723 3410 5689 6374 565 330 220 47 399 197 105 7 163 130 96 40 10 8 7 5 69716 38853 -75112 10014 99 157 1198 3457 184 16 2453 3223 5447 6357 645 374 254 49 442 218 116 8 200 152 118 41 10 8 7 5 69719 38854 -75111 10020 94 152 1198 3449 185 16 2444 3213 5433 6355 649 376 256 49 442 218 116 8 203 154 120 41 10 8 7 5 69722 38857 -75109 10014 99 157 1198 3432 186 16 2437 3204 5425 6350 651 379 257 50 442 218 116 8 206 157 122 42 10 8 7 5 6.0 Data Provider Documentation The following information was supplied by the data provider. Readme File for Version 2.0 AATS-6 TARFOX Data Archive John Livingston (jlivingston@mail.arc.nasa.gov), Jens Redemann (jredemann@mail.arc.nasa.gov) NASA Ames Research Center, MS 245-5 Moffett Field, CA 94025 21 March 2000 During TARFOX, the six-channel NASA Ames Airborne Tracking Sunphotometer (AATS-6) was operated aboard the University of Washington C-131A aircraft. AATS-6 detector voltages were sampled at a digitization frequency of three Hz and then averaged in software over three-second time intervals prior to being saved to disk. For each three-second (nine-sample) average measurement the corresponding sample standard deviation was also calculated in software and saved. Measurements acquired during fifteen flights on thirteen days during the period 10-31 July 1996 were archived (AATS-6 Version 1.0) in the NASA Langley DAAC in 1997. Archived parameters included time, aircraft latitude, longitude and pressure altitude, calculated relative optical airmass, measured sunphotometer detector voltages, and calculated total and aerosol optical depths and uncertainties in four wavelength bands centered at 380.1, 450.7, 525.3 and 1020.7 nm. The measured voltages in the 941.4-nm channel were also archived, but only placeholder values of "99999" corresponding to missing water vapor column content (derived from the 941.4-nm measurements) values were included in the Version 1.0 files. The pressure altitudes were calculated from aircraft measurements of static atmospheric pressure by assuming a standard atmospheric model of temperature, pressure, and density. No true geometric altitudes were archived. This summary describes the Version 2.0 archive of the AATS-6 TARFOX data set. This version was generated in March 2000 and represents a reduced data set due to application of a more objective cloud screening procedure. In addition to the calculated values of pressure altitude that were included in the Version 1.0 files, this version also includes the measured atmospheric pressure readings and geometric altitudes calculated by integration of the hypsometric equation (see below) using measurements and/or estimates of the pressure profile below 30 m (the lowest in-flight altitude of the C-131A above the sea surface). It also includes values of columnar water vapor (cm) that have been calculated from measurements in the AATS-6 941.4-nm channel by using the methodology described in Schmid et al. (1996). Total optical depths were calculated from the average signal voltages measured in the channels centered at 380.1, 450.7, 525.3 and 1020.7 nm by applying the sunphotometer analysis methodology as described by Russell et al. (1993). This methodology, which is widely used in the analysis of sunphotometer measurements, assumes that attenuation of the measured incoming direct beam solar radiation in channels unaffected by water vapor absorption follows the Bouguer-Lambert-Beer extinction relationship. Aerosol optical depths were then derived by subtracting the contributions due to molecular (Rayleigh) scattering and gaseous (ozone and nitrogen dioxide) absorption from the calculated total optical depths. The Version 1.0 AATS-6 data archive set was generated by subjectively screening the original measurements to remove all records contaminated by partial obscuration of the solar disk by an overhanging aircraft antenna wire and to remove some, but not all, records affected by cloud attenuation. Measurements were objectively screened to eliminate all records for which aircraft position (latitude, longitude) or static atmospheric pressure was missing or obviously incorrect, as well as those records for which the measured AATS-6 detector plate temperature fell outside an acceptable range of 44.0-46.5 deg C. In the Version 2.0 AATS-6 data archive, an objective screening procedure that utilizes the sample standard deviations of measured detector voltages has been applied to the original unscreened AATS-6 measurements. This procedure rejects all data records for which the relative magnitude of the measured detector voltage sample standard deviation in any channel exceeds a critical threshold (0.0015 to 0.01) that was varied for data collected during separate flights and, for a few cases, was varied for data taken within a particular flight. Because each AATS-6 data point represents a nine-sample average over a three-second time interval, this screening procedure is especially well-suited for removing short duration solar obscuration events that resulted from shadowing of the instrument by the aircraft antenna wire or by attenuation of the solar beam by clouds with an optical thickness large enough to cause an abrupt increase in the calculated aerosol plus cloud optical thickness. This procedure is ineffective at removing those data points for which the incoming solar beam was attenuated for longer than three seconds by a cloud of relatively constant optical thickness or by a cloud with optical thickness much smaller than that of the aerosol along the sunphotometer viewing path. Although the application of this cloud screening technique is objective, the critical threshold of voltage standard deviation that was used for data rejection is inherently subjective because it was determined by analysis of the measured voltages and the calculated aerosol optical depths for each flight. Nevertheless, we feel confident in claiming that this technique successfully removed at least 95% of those data records affected by clouds with an optical thickness of at least 0.02. One exception is the aircraft ascent of Flight 1731 (~1845-1900 UTC), when cirrus clouds of apparently uniform optical thickness contaminated nearly all measurements. Specific comments pertaining to the level of cloud removal and data quality concerns for specific flights have been added to the comments section at the beginning of each archived data file. In addition to application of this new screening procedure, the objective procedures used in the initial archival for removing data points based on missing aircraft location, spurious atmospheric pressure readings, or unacceptable detector plate temperature readings have been retained. Also, minimal subjective screening has been applied to remove data of questionable quality not removed by the objective screening procedures. Geometric altitudes have been calculated from the static atmospheric pressure and temperature measured aboard the UW C-131A. For each flight, a relationship between temperature and pressure was established using all of the available p-T data. In a second step, the pressure-altitude integral of this temperature profile was calculated, thus yielding a relationship between pressure and altitude-averaged temperature. This altitude-averaged temperature was then used to calculate the geometric altitude via the hypsometric equation (Holton, 1992): z(p) = -(R_d / g_0) * T_bar(p) * ln(p/p_surf) , where z is geometric altitude, R_d is the specific gas constant for dry air, g_0 is the gravitational constant, p is pressure, and p_surf is the assumed surface pressure (see below). T_bar(p) is calculated using T_bar(p) = integral(T dlnp) / integral(dlnp). The surface pressure was assumed to vary in time. Surface pressures at take-off and landing were assumed to be equal to the maximum pressures measured on the ground during those periods, although considerable pressure differences were encountered while the plane was located on the ground. Hence, for pressures smaller than the maximum pressure during taxiing, this method yields geometric altitudes greater than zero m, although the aircraft was situated on the ground. Surface pressures in flight were estimated during low-altitude transects over the ocean where available. Here we assumed that the surface pressure was 4 mb greater than the maximum pressure measured aboard the C-131A during any given low-level transect. This assumption was necessitated by the fact that the radar altimeter of the C-131 frequently measured zero-m altitude during these low-level transects and could not be used for an accurate estimate of instantaneous altitude. This methodology resulted in surface pressures over the ocean which were up to 8 mb greater than those measured upon take-off at Wallops Island, which is conceivable given the synoptic scale meteorology. REFERENCES Holton, J. R., An Introduction to Dynamic Meteorology, 511 pp., Academic Press, San Diego, 1992. Russell, P. B., J. M. Livingston, E. G. Dutton, R. F. Pueschel, J. A. Reagan, T. E. DeFoor, M. A. Box, D. Allen, P. Pilewskie, B. M. Herman, S. A. Kinne and D. J. Hofmann, 1993. Pinatubo and pre-Pinatubo optical-depth spectra: Mauna Loa measurements, comparisons, inferred particle size distributions, radiative effects, and relationship to lidar data. J. Geophys. Res., 98, 22,969-22,985. Schmid, B., Thome, K. J., Demoulin, P., Peter, R., Matzler, C. and Sekler, J., 1996. Comparison of modeled and empirical approaches for retrieving columnar water vapor from solar transmittance measurements in the 0.94-mm region. J. Geophys. Res., 101, 9,345-9358.