modis_color - MODIS ocean color retrieval

This function requires a TeraScan MODIS license.

SYNOPSIS

modis_color  [ parameter=value ]  [ inputfile outputfile ]

modis_color  [ parameter=value ]  [ inputfile ... directory ]

Parameters are: climatology, met_file, interp_met, delta_x, delta_mod, stray_lit, out_of_band, compute_ndvi, fill_bad.

DESCRIPTION

modis_color implements the SeaWiFS-type ocean color algorithms to derive chlorophyll concentrations and aerosol optical depth (see REFERENCES, Esaias et al, 1998). The algorithm is designed to work with calibrated MODIS ocean color bands (see IMAPP) which are 8 channels in the visible and near-infrared with center frequencies at 411, 442, 487, 530, 547, 665, 746, and 866 nm. The bandwidths of these channels are 15, 10, 10, 10, 10, 10, 10 and 15 nm, respectively.

The function requires both input and output TDF filenames. The input to modis_color is the dataset output by the function get_modis_color.

The most fundamental task of modis_color is to derive water-leaving radiance values from radiance values measured at the sensor. Here radiance is measured in milliWatts * centimeter^-1 * steradian^-1 * micrometer^-1. For details of water-leaving computations, see swcolor.

The derived quantity approximates the radiance that would exit the ocean surface in the absence of the atmosphere when the sun is at its zenith. From these normalized radiance values and the information obtained in performing the atmospheric correction, the following variables are derived and written to the output dataset:

  Variable   Units       Description

  nLw_412    radiance    Normalized water-leaving radiance.
  nLw_443    radiance    Normalized water-leaving radiance.
  nLw_490    radiance    Normalized water-leaving radiance.
  nLw_530    radiance    Normalized water-leaving radiance.
  nLw_550    radiance    Normalized water-leaving radiance.
  La_670     radiance    Aerosol radiance contribution.
  La_865     radiance    Aerosol radiance contribution.
  epsilon    none        Epsilon value for channel 1.
  tau_865    none        Aerosol optical thickness at 865 nm.

Chlorophyll concentrations are computed using mod_chlor, a function that implements the MODIS Case 1 bio-optical algorithms.

PARAMETERS

climatology

Specifies whether or not to use climatology for the meteorological data. Climatology files are expected to reside in the $REFDATA/climate directory and be in TeraScan data format (see datasets). The climatology should contain the following variables with the specified names and units:

Variable                  Type    Units

surface_zonal_wind        float   meters/second
surface_meridional_wind   float   meters/second
surface_pressure          float   Pa
precip_water              float   kg/m^2
ozone                     float   dobson

Each variable is expected to be three dimensional, with dimensions line, sample and month (the same for each variable), with month being of length 12. The data set is expected to have a valid Earth transform (see etx). Two climatologies are available with TeraScan, one from European Center for Medium-Range Weather Forecasting (ECMWF) analyses, named ecmwfclim.tdf, and one based on COADS data (i.e., ship and buoy reports), named coadsclim.tdf. The latter tends to have more missing areas. If climatology=no, the user must specify the name of a met_file containing real-time meteorological data. The default is yes.

met_file: Specifies the name of the file containing the necessary meteorological data.; If climatology=yes, the file must conform to the requirements listed under climatology. The default is ecmwfclim.tdf.; If climatology=no, the file must be from a real-time weather forecasting analysis (see NOTES). The data should conform to the same specifications given for the climatological file except the variables are expected to only be two dimensional, with dimensions line and sample. In this case, the ozone variable can have a different set of line and sample dimensions, and different Earth transform. Note that only one met_file is specified for any given number of input files. Thus, if real-time meteorological data is a requirement, as opposed to using climatology, and a number of input files from different days/periods require processing, they should each be processed with a different invocation of modis_color so that the associated meteorological file for that day/period can be specified for met_file. If climatology=no, there is no default.

interp_met: If interp_met=yes, then each anchor point (see delta_x) has its meteorological information (see met_file and climatology) assigned using a weighted linear interpolation of the nearest four points based on their distance from the latitude and longitude of the anchor point. Note if any of the four nearest points contains a missing value, then the retrieved value is set to missing. If interp_met=no, the meteorological information is retrieved based on the values of grid with its center latitude and longitude nearest that of the anchor points (i.e., nearest neighbor). Note, there is a very minor computation cost incurred in the interpolation. The default value is yes.

delta_x: Specifies the distance (in samples) across the scan line between full calculations of auxiliary quantities such as zenith and azimuth angles, rayleigh scattering, ozone absorption, white cap correction, etc. These "anchor" points are also the locations at which meteorological information is retrieved from the met_file. Values for auxiliary and meteorological quantities in between the "anchor" points are determined by linear interpolation. Valid responses are between 1 and 1000. The default value is 8.

delta_mod: Specifies the distance (in samples) across the scan line between which to force full atmospheric model calculations regardless of the epsilon check carried out within the atmospheric correction model. Valid responses are between 1 and 1000. The default value is 48.

stray_lit

Not currently implemented; future implementation when information and algorithms from the MODIS science team are available. The default is no.

out_of_band

Not currently implemented; future implementation when information and algorithms from the MODIS science team are available. The default is no.

compute_ndvi

Not currently implemented; future implementation when information and algorithms from the MODIS science team are available. The default is no.

fill_bad

Specifies whether or not to replace the masked pixels in the water-leaving radiances with the original channel radiance. The default is no.

EXAMPLES

% modis_color MOD021KM.2001255.160317.clr_input MOD021KM.2001255.160317.COLOR 
climatology  : char(  3) ? [yes]
met_file     : char(255) ? [ecmwfclim.tdf]
interp_met   : char(  3) ? [yes]
delta_x      : int       ? [8]
delta_mod    : int       ? [48]
stray_lit    : char (3)  ? [no]
out_of_band  : char (3)  ? [no]
compute_ndvi : char (3)  ? [no]
fill_bad     : char (3)  ? [no] yes

Processing:MOD021KM.2001255.160317.clr_input 16:03:17 2001/09/12: 
Lines: 1020 Samples: 1354

Real-time met_files must be customer furnished and converted to TeraScan data format. Sources for the required meteorological and ozone data include private companies and the National Weather Service, both of which can provide the data for a fee. These files are in WMO's Gridded Binary (GRIB) format (Stackpole, 1994).

The shell script:

$TSCANROOT/share/scripts/seawifs_ancillary_tdf

can be used to convert the surface meteorology and total ozone from these files into a single TDF dataset for use with modis_color:

% $TSCANROOT/share/scripts/seawifs_ancillary_tdf grib_file ozone_file

The names of the files on the command line need to be specified in the order shown above. The 2-D file output by the script will have the name of the GRIB file specified. modis_color expects to find the real-time met file in the directory $REFDATA/climate.

Note: The script calls the program readgrib to read and extract data from the files. readgrib is a freeware FORTRAN program that you can download via anonymous FTP from ncardata.ucar.edu from the directory libraries/grib. Download the README file first for information on the appropriate file(s) to download for your operating system. Then, once the program file is downloaded, compiled, and linked, rename the program readgrib, put it in your path, and make the script executable.

REFERENCES

Esaias et al., 1998: An Overview of MODIS capabilities for Ocean science observations. IEEE Trans. Geosci Remote sensing. Vol. 36, 4 pp 1250-1265.

Stackpole, J. D., 1994: The WMO format for the storage of weather product information and the exchange of weather product messages in gridded binary form. Office Note 388. National Meteorological Center, National Oceanic and Atmospheric Administration.

Last Update: $Date: 2001/12/14 18:54:32 $