SRB_REL3.1_LONGWAVE_DAILY - GEWEX Longwave Daily-Average Data Set README File 1.0 Introduction This README file provides information on the SRB_REL3.1_LONGWAVE_DAILY data set. The data set contains daily average global fields of six longwave (LW) surface and Top of Atmosphere (TOA) radiative parameters derived with the Longwave algorithm of the NASA World Climate Research Programme/Global Energy and Water-Cycle Experiment (WCRP/GEWEX) Surface Radiation Budget (SRB) Project. If users have questions, please contact the Langley Atmospheric Science Data Center (ASDC) User and Data Services Office at: Atmospheric Science Data Center User and Data Services Office Mail Stop 157D NASA Langley Research Center Hampton, Virginia 23681-2199 U.S.A. E-mail: support-asdc@earthdata.nasa.gov Phone: (757)864-8656 FAX: (757)864-8807 URL: http://eosweb.larc.nasa.gov This readme includes the following sections: 1.0 Introduction 2.0 Data Set Description 2.1 Data Quality 2.2 Input Information 2.3 Grid Description 2.4 Points of Contact 3.0 Format and Packaging 4.0 Science Parameters Information 5.0 Sample Read Software Description 6.0 Implementing the Sample Read Software 7.0 Sample Output 8.0 Additional Derivable Parameters References 2.0 Data Set Description There are a total of six parameters in these files as follows: 1. TOA Upward Clear-Sky Flux/Clear-sky Outgoing Longwave Radiation (OLR) (clr_toa_up) 2. Surface Clear-sky Upward Longwave Flux (clr_sfc_up) 3. Surface Clear-sky Downward Longwave Flux (clr_sfc_down) 4. TOA All-Sky Upward Longwave Flux/OLR (toa_up) 5. Surface All-Sky Upward Longwave Flux (sfc_up) 6. Surface All-Sky Downward Longwave Flux (sfc_down) These parameters are derived originally on a 3-hourly temporal resolution. The 3-hourly values are averaged into the daily values given in these files. The current version of the data sets is identified as Release 3.1. The GEWEX LW algorithm uses the Fu et al. (1997) thermal infrared radiative transfer code requiring atmospheric profile information, cloud, and surface properties. The sources for these inputs are briefly described below. A detailed description of the algorithm is currently being prepared for publication. Please contact the Dr. Paul W. Stackhouse Jr. at the address below for further details. Version History: Release 2.1: 12 year data set (July 1983-October 1995), on nested grid (described in Section 2.3), using GEOS-1 meteorological data. Release 2.5: 22 year data set (July 1983-June 2005). Using GEOS-4 meteorological inputs for the data set in place of GEOS-1. Release 3.0: 24.5 year data set (July 1983-December 2007). This version includes improved cloud properties in areas in missing and sun glint regions where ISCCP cloud retrievals aren't performed. Additionally, the IR radiative parameterization of ice clouds has been updated (Fu et al. 1998). The water vapor continuum has been updated (Kratz and Rose, 1999). An error in the ozone profile assignment is corrected. The surface vegetation type maps have been updated. This affects the surface emissivity values (Rutan et al. 2009). CO2 concentration value now varies month to month, based on monthly trend values from ftp://ftp.cmdl.noaa.gov/ccg/co2/trends/co2_mm_gl.txt Release 3.1: Corrections to nonphysical fluxes have been made to Rel. 3.0. Negative TOA fluxes in the 3-hourly data files were found to occur about 7 grid box times per month (out of 44016 grid boxes x 248 hours per month), with an additional 5 to 10 values per month identified as being unphysically low. The problem was found to be an numerical instability occurring due to an optimization switch in the Fortran compiler. The downwelling fluxes were also affected. So, the nonphysical values were replaced with a recomputation of those grid boxes using the same code but built without an optimization. Additionally, 3-hourly values of NaN's were detected and were traced to en error in the temperature profiles. The frequency of this occurrence was far more rare and was found to be mostly clustered for three months in 1987 and July 1990. After correcting the temperature profiles, the grid boxes were recomputed and replaced in the 3-hourly files using the same "non optimized" version of the code. The daily, monthly and 3-hourly monthly files were reprocessed using the improved 3-hourly files. Differences in the monthly averages proved to be small and mostly < 2 W m^-2 on a grid box level and < 0.01 W m^-2 on a global mean. Differences on the grid box level in the 3-hourly monthly and daily averages were mostly < 10 W m^-2. 2.1 Data Quality An assessment of the quality of these monthly average fluxes was accomplished by comparisons with corresponding ground-measured fluxes over a period from January 1992 to December 2007 from a number of sites of the Baseline Surface Radiation Network (BSRN). From the aggregate data set for all sites and years, mean bias was determined to be about 0.52 W/m**2 (0.16%, model fluxes higher), and the root mean square difference is 21.8 W/m**2 (7.1%). Uncertainties associated with operational BSRN measurements during this period are believed to be about +/- 3-5 W/m**2 (1-1.5%, Ellsworth Dutton, NOAA, BSRN Manager). Thus, the mean bias for the present results is within the uncertainty for BSRN measurements. Errors for individual daily values may be different from the above values because those are subject to bias and random errors due to local meteorological conditions. 2.1.2. Indian Ocean Gap Artifact There is a visible and common artifact in much of the data set period, due to a lack of coverage from geostationary satellites over an area centered on 70 degrees east longitude. This situation, commonly called the Indian Ocean gap, occurs for all of the July 1983 - June 1998 time period, except for April 1988 - March 1989, when data from the INSAT satellite is available to cover the gap. In July of 1998, Meteosat-5 was moved over the gap area, eliminating the gap. When the Indian Ocean gap occurs, the gap area is covered by polar orbiting satellites, which can result in only one or two daytime overpasses per day. Geosynchronous temporal sampling during the daytime is 3-5 times per day depending upon the latitude (between 55 degrees North and South) and the time or year. In addition, the limbs of the geostationary satellites which bound the gap may suffer from spuriously high cloud amounts, due to large view angles. This results in an abrupt drop-off of cloud fraction in the gap as compared to the gap boundary. Downward longwave radiation is lower in the gap, creating an appearance of a flux discontinuity. For Daily averaged fluxes any discontinuity in instantaneous fluxes will be exacerbated by the temporal gaps of coverage in the Indian Ocean gap region. LW and LWQC daily averages are less affected by the temporal gap because the 2 night time observations of the region are also used in determining the daily average. For these daily averaged fluxes, a discontinuity of magnitude less than 20 W/m**2 for TOA fluxes and less than 5 W/m**2 for surface fluxes may appear in the Indian Ocean gap region. 2.2 Input Information Inputs to the algorithm were obtained from the following sources: Cloud parameters were derived from the International Satellite Cloud Climatology Project (Rossow and Schiffer, 1999) DX data product. The cloud pixels were separated into categories of high, middle and low where middle and low clouds could be composed of ice or water, while high clouds were composed of ice only. Cloud fractions and cloud optical depths were determined within these categories. Cloud particle sizes were assumed and cloud physical thicknesses were also assigned based upon information from literature. Random overlap is used between the high, middle and low layers to better approximate undercast conditions.. Temperature and moisture profiles were obtained from the 4-D data assimilation Goddard EOS Data Assimilation System, level-4 (GEOS-4) obtained from the Global Modeling and Assimilation Office (GMAO) at NASA Goddard Space Flight Center (GSFC) (Bloom et al., 2005) Column ozone values for most of the duration of this dataset (July 1983 to December 2004) were obtained primarily from the Total Ozone Mapping Spectrometer (TOMS) archive. For the early period (July 1983-November 1994), TOMS data came from NIMBUS-7 and Meteor-3 satellites. There was an interruption of about 20 months (December 1994-July 1996) after which TOMS data from EP-TOMS became available in August 1996 and continued until December 2004. All gaps in TOMS data, including those over the polar night areas every year, were filled with column ozone values from TIROS Operational Vertical Sounder (TOVS) data. Column ozone data continued to be available beyond December 2004 from OMI instrument aboard Aura satellite but TOVS data, which is essential for filling the gaps in OMI data, developed some unexplained gaps of its own and became unusable. Beginning in January 2005, GEWEX/SRB started using a daily analysis ozone product from NOAA Climate Predictions Center (CPC), known as the Stratospheric Monitoring-group Ozone Blended Analysis (SMOBA). Surface emissivities were taken from a map developed at NASA LaRC (Wilber et al. 1999). 2.3 Grid Description The fluxes are generated on a nested grid, which contains 44016 cells. The grid has a resolution of 1 degree latitude globally, and longitudinal resolution ranging from 1 degree in the tropics and subtropics to 120 degrees at the poles. The first cell is Latitude 89-90 degrees South, Longitude 0-120 degrees East. The cells start at the Greenwich meridian and proceed east around the globe, then shift one degree to the north. The number of cells per latitude band starting at the South Pole are: 3, 45, 45, 45, 45, 45, 45, 45, 45, 45, 90, 90, 90, 90, 90, 90, 90, 90, 90, 90, 180, 180, 180, 180, 180, 180, 180, 180, 180, 180, 180, 180, 180, 180, 180, 180, 180, 180, 180, 180, 180, 180, 180, 180, 180, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 360, 180, 180, 180, 180, 180, 180, 180, 180, 180, 180, 180, 180, 180, 180, 180, 180, 180, 180, 180, 180, 180, 180, 180, 180, 180, 90, 90, 90, 90, 90, 90, 90, 90, 90, 90, 45, 45, 45, 45, 45, 45, 45, 45, 45, 3 The read software described below contains a subroutine to regrid the fluxes to 1 degree latitude by 1 degree longitude equal-angle grid using replication. 2.4 Points of Contact Scientific contact: Dr. Paul W. Stackhouse Jr. Mail Stop 420 21 Langley Boulevard NASA Langley Research Center Hampton, VA 23681-2199 U.S.A. E-mail: Paul.W.Stackhouse@nasa.gov Production Contact: Atmospheric Science Data Center User and Data Services Office Mail Stop 157D NASA Langley Research Center Hampton, VA 23681-2199 U.S.A. E-mail: support-asdc@earthdata.nasa.gov 3.0 Format and Packaging Each data file contains an entire month of daily global fields of the parameters described in Section 4.0 on an approximately 1 deg x 1 deg equal-area grid described in Section 2.3. The files contain binary data and are named according to the following convention: srb_rel3.1_longwave_daily_yyyymm.binary, where srb Project name, Surface Radiation Budget rel3.1 Release number for these data (Release 3.1) longwave Name of the algorithm, GEWEX Longwave daily Time resolution of the data file yyyy 4-digit year mm 2-digit month binary file format 4.0 Science Parameters Information The files contain global fields of daily values of the six radiative parameters on the nested grid. Each file has 6 records, containing one global field for every time period in each record. The parameters are: Name: Top-of-Atmosphere Clear-sky Upward LW Flux Units: Watts per square meter Type: Real Range: 50 to 600 Fill Values: -999.0 Scale Factor: None Name: Surface Clear-sky Upward LW Flux Units: Watts per square meter Type: Real Range: 50 to 800 Fill Values: -999.0 Scale Factor: None Name: Surface Clear-sky Downward LW Flux Units: Watts per square meter Type: Real Range: 50 to 600 Fill Values: -999.0 Scale Factor: None Name: Top-of-Atmosphere All-sky Upward LW Flux Units: Watts per square meter Type: Real Range: 50 to 600 Fill Values: -999.0 Scale Factor: None Name: Surface All-sky Upward LW Flux Units: Watts per square meter Type: Real Range: 50 to 800 Fill Values: -999.0 Scale Factor: None Name: Surface All-sky Downward LW Flux Units: Watts per square meter Type: Real Range: 50 to 600 Fill Values: -999.0 Scale Factor: None 5.0 Sample Read Software Description Sample read software written in Fortran-90, read_longwave_daily.f90, was developed for reading these data. The software constitutes the name of the input data file, accesses and reads it, using the information provided in the namelist file (longwave_daily.nml). The input files are read as direct-access binary on the nested (44016 box) grid. The software reads one or more of the 6 parameter fields, regrids them to an equal-angle 1 deg x 1 deg grid, and writes the output as ascii or binary format. The choice of file format (ascii or binary) and of the location of the output files is also made through the namelist file. A sample namelist file that would be used to read the July 1992 data file and write all parameters to an ascii format output file is presented below: &time_vars yr=1992 mon=7 ascii=.true. binary=.false. path_in='**** input file path here****' path_out='**** output file path here****' little_endian=.false. clr_toa_up=.true. clr_sfc_up=.true. clr_sfc_down=.true. toa_up=.true. sfc_up=.true. sfc_down=.true. / There is a choice to convert the input fields from big endian to little endian byte order with the logical variable "little_endian" in the namelist. This applies to operating systems where byte order is stored opposite that of the machines used to create the data set, such as Linux and Macs with the Intel chip. If possible, a better choice for doing the conversion in these cases would be to use a compiler option. If using a compiler option, do not set little_endian to true. Both input and output fields have the same orientation: they start at the Greenwich meridian-south pole and go east and north from there. A limitation of this code is that it provides a complete global field of the specified parameters in the above orientation. The user should be easily able to extract values for any box or lat-lon region from these fields. 6.0 Implementing the Sample Read Software The sample read software can be compiled with any Fortran 90 or 95 compiler. To compile: % f90 -o run_longwave_daily read_longwave_daily.f90 The providers used a gfortran compiler but any F90/F95 compiler should work. Edit the namelist file to select month and year to be processed, choose the parameters to be read and the format of the output file. Run the software: % run_longwave_daily 6.1 Read Software Incompatibilities With some Fortran compilers the RECL keyword in the OPEN statement assumes record lengths are specified in 4 byte increments. If that is the case, then the following statement in the read program: open (10,file=infile,status='old',form='unformatted',access='direct',recl=4*nreg) should be modified to: open (10,file=infile,status='old',form='unformatted',access='direct',recl=nreg) The same should be done with the output binary file: open (15,file=outfile, form='unformatted', access='direct', & recl=nlon*nlat*4, status='replace') should be modified to: open (15,file=outfile, form='unformatted', access='direct', & recl=nlon*nlat, status='replace') 7.0 Sample Output The six tables of numbers below show the values of the parameters contained in these files for latitude bands 45-51 (starting at the south pole) and longitude boxes 100-104 (starting at the Greenwich meridian) for day 14 of the month. Values for only a small lat-lon box for a single time are printed to the screen. When the is code run, the following information appears on the screen: ***************************************************************** * * * * * Data Set srb_rel3.1_longwave_daily Read Software * * * * Version: 1.0 * * * * Contact: Atmospheric Science Data Center * * User and Data Services Office * * Mail Stop 157D * * NASA Langley Research Center * * Hampton, Virginia 23681-2199 * * U.S.A. * * * * E-mail: support-asdc@earthdata.nasa.gov * * Phone: (757)864-8656 * * FAX: (757)864-8807 * * * ***************************************************************** srb_rel3.1_longwave_daily_199207.binary input file is opened Variable clr_toa_up_Day = 14 lon # = 100 101 102 103 104 lat band # 45 245.327 244.915 244.915 244.801 244.801 lat band # 46 247.520 247.680 248.006 248.343 248.896 lat band # 47 250.532 250.871 251.111 251.925 252.439 lat band # 48 253.046 253.233 254.051 254.887 255.141 lat band # 49 254.765 255.675 256.743 257.401 257.755 lat band # 50 257.504 258.671 259.672 260.471 261.323 lat band # 51 260.901 262.123 263.274 264.491 265.498 file clr_toa_up_daily_199207.ascii has been written Variable clr_sfc_up_Day = 14 lon # = 100 101 102 103 104 lat band # 45 353.569 354.206 354.206 354.565 354.565 lat band # 46 358.028 358.095 358.281 358.352 357.961 lat band # 47 362.000 361.768 361.654 361.386 360.668 lat band # 48 364.541 364.190 363.842 363.287 362.441 lat band # 49 366.082 365.849 365.418 364.756 364.043 lat band # 50 367.851 367.934 367.660 367.124 366.692 lat band # 51 370.105 370.577 370.641 370.465 370.437 file clr_sfc_up_daily_199207.ascii has been written Variable clr_sfc_down_Day = 14 lon # = 100 101 102 103 104 lat band # 45 268.886 274.565 274.565 277.481 277.481 lat band # 46 276.740 278.562 279.635 280.197 280.477 lat band # 47 280.638 281.595 282.141 282.577 282.598 lat band # 48 283.647 284.436 284.888 285.058 284.859 lat band # 49 286.992 287.558 287.377 287.487 287.150 lat band # 50 289.850 290.039 290.042 289.872 289.106 lat band # 51 292.100 292.030 291.915 291.501 289.917 file clr_sfc_down_daily_199207.ascii has been written Variable toa_up_Day = 14 lon # = 100 101 102 103 104 lat band # 45 180.671 178.472 178.472 168.542 168.542 lat band # 46 181.359 180.805 167.129 176.950 176.185 lat band # 47 190.173 189.124 179.695 183.332 181.039 lat band # 48 181.612 176.374 176.100 176.507 181.200 lat band # 49 187.283 176.242 185.380 206.484 200.462 lat band # 50 203.121 200.613 219.523 204.988 216.215 lat band # 51 222.810 230.272 227.199 230.204 235.574 file toa_up_daily_199207.ascii has been written Variable sfc_up_Day = 14 lon # = 100 101 102 103 104 lat band # 45 354.181 354.740 354.740 355.175 355.175 lat band # 46 358.488 358.555 358.818 358.990 358.548 lat band # 47 362.574 362.454 362.231 361.997 361.235 lat band # 48 365.124 364.782 364.339 363.817 363.066 lat band # 49 366.746 366.451 366.104 365.445 364.696 lat band # 50 368.486 368.686 368.380 367.866 367.238 lat band # 51 370.883 371.203 371.230 371.134 371.102 file sfc_up_daily_199207.ascii has been written Variable sfc_down_Day = 14 lon # = 100 101 102 103 104 lat band # 45 310.619 311.337 311.337 319.686 319.686 lat band # 46 308.493 310.400 316.692 324.216 321.018 lat band # 47 320.300 329.086 321.991 324.808 321.747 lat band # 48 323.953 325.432 319.310 321.733 328.154 lat band # 49 333.156 329.375 334.968 335.287 332.478 lat band # 50 334.128 342.390 340.112 341.416 327.033 lat band # 51 346.269 335.561 332.821 337.970 336.035 file sfc_down_daily_199207.ascii has been written 8.0 Additional Derivable Parameters The net LW flux at the top-of-atmosphere (TOA) is simply the TOA upward LW flux. The net LW flux at the surface can be defined as: Net LW Flux = Downward LW Flux - Upward LW Flux and is, therefore, generally a negative number. Net fluxes can be computed for the clear-sky and all-sky conditions. The estimates of clear-sky and all-sky fluxes also allow the estimation of the contribution by clouds to the all-sky fluxes. This is commonly referred to as the cloud radiative forcing (CRF) and is computed according to: CRF = Flux (all-sky) - Flux (clear-sky) Thus, the cloud radiative forcing on the downward longwave flux is generally positive because clouds act to increase the emission to the surface. In this way, the effect of the cloud emission on the fluxes can be estimated for each flux component. Lastly, providing TOA and surface fluxes allows one to derive the net radiative flux of the atmosphere. This is given by the relation Net Atmos. Flux = Net TOA Flux - Net Surface Flux For the LW, this flux is negative meaning that the atmosphere is cooling over the LW wavelengths. References: Bloom, Stephen, A. daSilva. D. Dee, M. Bosilovich, J-D. Chern, S. Pawson, S. Schubert, M. Sienkiewicz, I. Stajner, W-W. Tan, and M-L Wu, 2005: Documentation and Validation of the Goddard Earth Observing System (GEOS) Data Assimilation System, Version 4, NASA Technical Report,Report Number: NASA/TM-2005104606/ VOL26/VER4, Rept- 2005-01264-0/VOL26/VER4 Fu, Qiang, K. N. Liou, M. C. Cribb, T. P. Charlock, and A. Grossman, 1997: Multiple Scattering Parameterization in Thermal Infrared Radiative Transfer. J. Atmos. Sci. , Vol. 54, 2799-2812, doi: 10.1175/1520-0469(1997)054<2799:MSPITI>2.0.CO;2 Fu, Qiang, P. Yang, and W. B. Sun, 1998: An Accurate Parameterization of the Infrared Radiative Properties of Cirrus Clouds for Climate Models. J. Climate, Vol. 11, 2223-2237, doi: 10.1175/1520-0442(1998)011<2223:AAPOTI>2.0.CO;2 Kratz, David P. and Rose, Fred G., 1999: Accounting for Molecular Absorption Within the Spectral Range of the CERES Window Channel. J. Quant. Spectrosc. Radiat. Transfer, Vol. 61, 83-95. Rossow, William B. and R. A. Schiffer, 1999: Advances in Understanding Clouds from ISCCP. BAMS, Vol. 80, 2261-2287, doi: 10.1175/1520-0477(1999)080<2261:AIUCFI>2.0.CO;2. Rutan, D., F. Rose, M. Roman, N. Manalo-Smith, C. Schaaf, and T. Charlock (2009), Development and assessment of broadband surface albedo from Clouds and the Earth's Radiant Energy System Clouds and Radiation Swath data product, J. Geophys. Res., 114, D08125, doi:10.1029/2008JD010669. Wilber, Anne C., Kratz, D. P., Gupta, S. K., 1999: Surface Emissivity Maps for Use in Satellite Retrievals of Longwave Radiation, NASA Technical Report, Report Number: L-17861, NAS 1.60:209362, NASA/TP-1999-209362.