Langley DAAC Version Number: LD_008_021_011_00_00_0_19990407 SAGE II Cloud Occurrence HDF Read Program The program rdsag2cldhdf.f (FORTRAN Version) allows the user to output data and associated metadata to an ASCII file for a season of the SAGE II Cloud Occurrence data by designating a starting year and month and ending year and month. You can further subset the data selection by choosing a geographic region to output. This can be done by selecting latitude and longitude ranges. Since this data set is 3 dimensional, you may also subset by altitude. A menu of available parameters allows for the selection of particular data measurements (e.g., Cloud Occurrence and/or Points Per Bin). The name of the file is as follows: sag2_cloud_yymm_yymm yy - the year of the data (84,86,87,88,89,90). mm - the month of the data (01,02,03,04,05,06,07,08,09,10,11,12) The program also builds an output file: sag2_cloud_yymm_yymm.rpt In order to run the read software, you will need to obtain the HDF libraries which are available in the public domain from the National Center for Supercomputing Applications (NCSA) via FTP (ftp.ncsa.uiuc.edu). Further information on obtaining the HDF libraries can be found in the Langley DAAC. Once the HDF libraries have been installed on your machine, the programs can be compiled as follows: ************************************************************************** UNIX Example for Compiling the Read Programs A compile script comes with your read software. It gives you the commands for compiling these codes on specific operating systems. Just edit the file to compile the correct code on the correct machine. **************************************************************************** Included below is a sample run and output of the SAGE II Cloud Occurrence read program. Comments or questions should be directed to: EOSDIS Langley DAAC Science, Users and Data Services Office NASA Langley DAAC Research Center Mail Code 157D 2 South Wright Street Hampton, Virginia 23681-2199 U.S.A. E-mail: larc@eos.nasa.gov Phone: (757)864-8656 FAX: (757)864-8807 ********************************************************************** SAMPLE RUN ********************************************************************** sorcerer 61: rdsag2cldhdf **************************************************** * * * SAGE II Cloud read program * * * * Version 3.0 January 26,1999 * * * * This program reads the SAGE II Cloud occurrence * * HDF formatted files. The input file name is * * built according to the naming convention formed * * by the Langley DAAC. If you have changed the * * name of the input file or need to specify a * * path, you will need to modify this program. * * All data are expected to be in your current * * working directory. * * * * NOTE TO THE USER: * * The seasons are broken into the following data * * months. * * 12 thru 02 = Dec, Jan, and Feb * * 03 thru 05 = Mar, Apr, and May * * 06 thru 08 = Jun, Jul, and Aug * * 09 thru 11 = Sep, Oct, and Nov * * Data are available from December 1984 to * * November 1990. * * * **************************************************** Enter Q to quit or return to continue. Please enter the file to be read. sag2_cloud_8503_8505 Input file is sag2_cloud_8503_8505 Output file is sag2_cloud_8503_8505.rpt Item Number Label 1 Cloud Occurrence 2 Points per Bin Please choose the number of the data items you wish to output, one per line. Enter 0 to quit choosing items. Enter -1 for help 2 0 Do you wish to further subset by (1) latitude and longitude and altitude or (2) no subsetting 1 Enter beginning and ending longitude (from -180 to 165 deg., delta=15 deg.) -180 165 Enter beginning and ending latitude (from -80 to 80 deg., delta=5 deg.) -70 -70 Enter beginning and ending altitude (from 8.5 to 19.5 Km, delta=1 Km) 8.5 8.5 Getting data, Please wait ... NOTE:numtype is: 24 The SAGE II Cloud read program has ended. sorcerer 62: ********************************************************************** SAMPLE OUTPUT ********************************************************************** sorcerer 63: cat sag2_cloud_8503_8505.rpt File ID: Version 5.9 Cloud Occurrence for Mar - May, 1985 File description is as follows: The basic SAGE II measurement is of the total extinction due to both aero- sols and cloud along a horizontal path through the atmosphere from the sun to the SAGE II satellite instrument. In order to determine the cloud frequency, it is necessary to separate events containing aerosol only from those in which cloud occurs along the optical path. The method used, and the corresponding al- gorithm, is a modified version of that described by Kent et al, in "A Model for the Separation of Cloud and Aerosol" (J. Geophys. Res., 98, 20725-20735, 1993). After the cloud and aerosol points are separated, the cloud occurrence at a given location is found by determining the number of cloud points, within a specified area about that location, as a percentage of the total number of cloud and aerosol points within that same area. This statistic is calculated at 15 degree intervals from -180 degrees to 165 degrees longitude and at 5 degree intervals from -80 degrees to 80 degrees latitude. The area, called a bin, for which this statistic is calculated, extends +/- 22.5 degrees longitude and +/- 5 degrees latitude from each location. These overlapping bins have the effect of smoothing the cloud occurrence information over data-sparse regions. A minimum of five data points per bin is required to produce a statistic for each location. This cloud occurrence map is produced for altitudes of 8.5 km to 19.5 km, at 1 km intervals for three-month seasons. Each season of data occu- pies a 3-D array with the dimensions 25 by 33 by 12 (longitude by latitude by altitude). The corresponding number of data points per bin is provided in an array of similar dimensions. The cloud occurrence values for bins which contain fewer than 5 data points have been set to minus one and the corresponding number of data points per bin set to zero. Using binomial statistics, the confidence intervals (error estimates) can be calculated from the cloud occurrence value and the number of points for each bin. The cloud occurrence data is graphically represented in the browse images which were created by a program written in Interactive Data Language (IDL) from Research Systems, Inc. The browse images are grouped by season and year, with twelve altitudes, 8.5 through 19.5 km, in each browse image file. Each image shows a cylindrical projection of the globe where the data covers -180 to 180 degrees longitude and, nominally, -80 to 80 degrees latitude, depending on SAGE II satellite coverage for that season. Color contours are used to repre- sent the cloud occurrence values at the locations described in the previous paragraph and show gradients in between data locations. White areas are shown where fewer than 5 data points are available. Lower altitudes have more white space, due to a lack of data points, as a result of obscuration by clouds at higher altitudes. The most significant problem occurring with the data set is the inability of the algorithm, under certain circumstances, to distinguish between the ex- tinction due to volcanic aerosol and that due to cloud. The method used to sep- arate aerosol from cloud is based on the variation in extinction between the measurement wavelengths of 0.525 and 1.02 micrometers, where little or no varia- tion in extinction is expected for cloud particles. The termination of an event profile at a given altitude has also been taken to indicate the presence of a cloud just below the altitude of the last useful data point. There is normally no difficulty in separating cloud from background aerosol, but newly injected volcanic aerosol may consist of sufficiently large particles such that there is little or no wavelength variation in the measured extinction. Such aerosol dis- tributions will appear on the browse images as cloud. Time periods for the browse images where this is particularly in evidence are December 1985 - Febru- ary 1986, following the eruption of Ruiz, and December 1989 - May 1990, fol- lowing the eruption of Kelut. No data are included for altitudes greater then 19.5 Km, and any events that terminate at altitudes greater than two kilometers above the tropopause have had the cloud, that would normally be associated with the bottom data point, eliminated from the data set. The latter step has been taken because of the occasional erroneous event that begins or terminates at an altitude above the top cloud layer. This filter also has the effect of removing some high altitude volcanic layers as well as some opaque polar stratospheric clouds. The current version of the browse images and associated data arrays re- places an earlier version, which was withdrawn from the Langley DAAC in late 1993. The original version was withdrawn in order to examine ambiguities existing between the effects of volcanic aerosol and those of cloud. These were found to be much more common than had been previously anticipated. At the time of the withdrawal, it was hoped that the processing algorithm could be revised in such a manner as to eliminate such ambiguities. This has not been possible, although the number of cases in which this occurs has been much reduced. The algorithm for separating the extinction due to aerosol and cloud has been modi- fied, and various other steps have been taken, as noted in the previous para- graph. Dimensions are 24 longitudes by 33 latitudes by 12 altitudes. Dimension 1 attributes: Label : Longitude Unit : Degrees Format : f6.1 Dimension 2 attributes: Label : Latitude Unit : Degrees Format : f5.1 Dimension 3 attributes: Label : Altitude Unit : Km Format : f4.1 Dimension 1 scales: -180.0 -165.0 -150.0 -135.0 -120.0 -105.0 -90.0 -75.0 -60.0 -45.0 -30.0 -15.0 0.0 15.0 30.0 45.0 60.0 75.0 90.0 105.0 120.0 135.0 150.0 165.0 Dimension 2 scales: -80.0 -75.0 -70.0 -65.0 -60.0 -55.0 -50.0 -45.0 -40.0 -35.0 -30.0 -25.0 -20.0 -15.0 -10.0 -5.0 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 45.0 50.0 55.0 60.0 65.0 70.0 75.0 80.0 Dimension 3 scales: 8.5 9.5 10.5 11.5 12.5 13.5 14.5 15.5 16.5 17.5 18.5 19.5 Slice encompasses: longitude -180.000 to 165.000 latitude -70.0000 to -70.0000 altitude 8.50000 to 8.50000 Data Label : Number of Data Points per Bin, Mar - May, 1985 Data Units : Unitless Data Format: i3 Data Max: 64 Data Min: 1 Data Values varying by longitude, latitude, and altitude (dimension 1,2,3) order are as follows : (Please note that a given section contains values for all requested longitudes for a given latitude and altitude.) 12 13 13 14 12 13 12 11 11 11 11 12 13 13 13 12 12 12 12 13 12 11 12 11 sorcerer 64: Updated: April 07, 1999 LD_008_021_011_00_00_0_19990407