NASA Home Jet Propulsion Laboratory California Institute of Technology NASA Portal Follow this link to skip to the main content
JPL Home Page JPL - Earth JPL - Solar System JPL - Stars & Galaxies JPL - Technology
Home Page Mission Gallery Education News About Us Internal

MISR Plume Height Climatology Project
David Nelson, Cecelia Lawshe, Dominic Mazzoni, David Diner, Ralph Kahn

Suggestions for Using Data
Home Page


Introduction

This document is designed to acquaint you with important information pertaining to the numerical data on this website should you decide to use them in science applications. There are two principal sets of numerical data: raw data files and summary files. There is one raw data file for each digitized smoke plume or other digitized region type in a project, and there are two summary files for each project. Data in both sets of files are downloadable and are recorded in ASCII format readable with your favorite text editor or programming language. You should expect to do some programming to read, filter and utilize these data.

Click the link on the main page titled "Documentation for this Website" for more information on website contents, for information on viewing and downloading files and for definitions of some terms used below. You can find additional information in the "MISR INteractive explorer" (MINX) documentation - the software used to generate the data found on this website. Go to the Open Channel website at http://www.openchannelfoundation.org/ and specify MINX in the "Quick Application Search" box to download the documentation and/or the MINX software.

Raw Data File Format

A raw data file records information pertaining to a single smoke plume region or smoke cloud region as well as detailed information for each data point within the region. You can find these files on the website in either of two ways as described in the "Documentation for this Website" link on the main page. See the sections titled "Raw Data Files" and "Orbits/Block Ranges Table".

The header portion of a sample raw data file, which contains general information about the region, is shown below. Following it is a description of each data field.

Sample header data
Orbit # : 12398
Path # : 29
Block # : 75
Date Acquired : 2002-04-17
UTC Time : 17:37:14
MINX Version : V1.0
User Name : Rachel

Terrain Radiance File : MISR_AM1_GRP_TERRAIN_GM_P029_O012398_AN_F03_0024.b30-84.hdf
Terrain Heights File : MISR_AM1_AGP_P029_F01_24.hdf
Cam/Sun Geometry File : MISR_AM1_GP_GMP_P029_O012398_F03_0013.b30-84.hdf
SVM Classifiers File : MISR_AM1_TC_CLASSIFIERS_P029_O012398_F05_0008.b30-84.hdf
Aerosol Product File : MISR_AM1_AS_AEROSOL_P029_O012398_F09_0019.hdf

Region Name : O12398-B75-P1
Region Type : Smoke plume
Region Data Quality : GOOD
First point longitude : -103.08281
First point latitude : 19.19359
Perimeter length (km) : 20
Area (sq km) : 27
Area per point (sq km) : 1.210
Total points in table : 12
Wind-corrected points : 5
Percent area covered : 22
Best Median Ht (m ASL) : 2662
Best Top Ht (m ASL) : 2718
StdDev metric, corrht : 92
Diff WindDir, AlongDir : 24
Power of fire in MW : 577.3

Header data field definitions

Orbit #
MISR orbits the earth about 15 times a day in a descending polar orbit. This field records the sequential orbit number this region belongs to measured from launch date.
Path #
There are 233 unique paths or day-side orbit tracks that all MISR orbits follow precisely. This field is the number of the path that this orbit belongs to.
Block #
MISR collects data only on the sunlit side of its orbit. This day-side part of the orbit is divided into 180 blocks starting in the north and proceeding generally toward the southwest. This field records the block number that the first digitized point in a smoke plume or smoke cloud falls in. For plumes, this should also be the fire location or smoke source region on the ground.
Date Acquired
This is the date this orbit was acquired by MISR.
UTC Time
The time of day this block of the orbit was acquired in Greenwich mean time. MISR crosses the equator on the day-side of the orbit at about 10:30 AM local time on every orbit.
MINX Version
The version number of MINX used to produce these data.
User Name
The username of the person who digitized these data.

Terrain Radiance File
MISR level 1 terrain-referenced radiance product file used by MINX for producing 275 meter resolution images and data retrievals. Only the nadir camera file (AN) is listed. The other eight camera files that are used should have the same version number. This file and the other MISR products listed below are all in HDF (Hierarchical Data Format).
Terrain Heights File
MISR Ancillary Geographic Products (AGP) file containing terrain heights at 1100 meters spatial resolution.
Cam/Sun Geometry File
MISR level 1 geometry product containing azimuth and zenith angles for cameras and sun.
SVM Classifiers File
Optional MISR level 2 product containing Support Vector Machine smoke, dust and cloud masks that can be used to assist in identifying smoke etc.
Aerosol Product File
Optional MISR level 2 product containing aerosol optical depth and other aerosol parameters whose values are copied into the raw data files.

Region Name
The unique name assigned to the digitized smoke plume or smoke cloud. This consists of the orbit number preceded by "O" plus the block number preceded by "B" plus a plume or cloud identifier, sequentially numbered within each block, preceded by the region type code (see next entry).
Region Type
The type of region that was digitized. The codes are "P" for smoke plume (origin of smoke can be traced to a source point on the ground and wind-correction of heights is attempted), "S" for smoke cloud (smoke is widespread, is not clearly associated with a ground source and heights are not amenable to wind-correction) and "L" for clear land.
Region Data Quality
A quality flag that can take the values "GOOD", "FAIR" or "POOR." See the section below on "How to Use the Data" for help in interpreting these values, and see the section on "Plume Quality Flag Derivation" for more information on how the quality flag is derived.
First point longitude
The longitude on the WGS84 spheroid of the first point digitized in a plume or smoke cloud. If a smoke plume, the first point should coincide with the source region of the smoke.
First point latitude
The latitude on the WGS84 spheroid of the first point digitized in a plume or smoke cloud.
Perimeter length (km)
The distance in kilometers around the perimeter of a region following the line digitized by the user.
Area (sq km)
The area of a region inside the line digitized by the user measured in square kilometers.
Area per point (sq km)
When heights and other parameters are retrieved, they are retrieved at specific points on a square grid inside the digitized perimeter line of a region. The point separation can be used to calculate the area that each point effectively represents. For example, a 1.1 km point spacing in each direction represents an effective area per point of 1.21 square kilometers. This can be used to compute the fractional area of a plume that is covered by valid retrievals and is used in computing the "Region Data Quality" flag. The point separation should always be a multiple of 275 meters, which is the resolution of the level 1 radiance data.
Total points in table
This is simply the total number of rows of point data in the table that follows this header.
Wind-corrected points (plumes) or Zero-wind points (clouds)
This is the number of points in the data table where wind-corrected heights and winds were successfully retrieved (for smoke plumes) or the number of points where uncorrected heights were successfully retrieved (for smoke clouds).
Percent area covered
The area where retrievals were successful can be computed by multiplying the "Area per Point" by the number of "Wind-corrected points" (or "Zero-wind points"). Dividing the result by the "Area" of the plume gives this value. It is used in computing the "Region Data Quality" flag.
Best Median Ht (m ASL)
Wind-corrected (plumes) or Zero-wind (clouds) median height of region in meters above sea level after a plane is fitted to the successful height points and all points more than 1.5 standard deviations from the plane are removed. This can be considered to be an estimate of injection height for plumes. If pyro-cumulus clouds were digitized, this may remove their contribution from the final result. Note that the stereoscopic method of height determination retrieves the height of the top of a smoke layer. It cannot provide information on the height of the base.
Best Top Ht (m ASL)
Wind-corrected (plumes) or Zero-wind (clouds) maximum height of region in meters above sea level after a plane is fitted to the height points and all points more than 1.5 standard deviations from the plane are removed. This could be considered to be another estimate of injection height for plumes.
StdDev metric, corrht (plumes) or StdDev metric, zeroht (clouds)
This is the standard deviation of the wind-corrected heights for all points with successful retrievals in the smoke plume (zero-wind heights for smoke clouds) divided by the mean height of those values and multiplied by 1000. It is used in computing the "Region Data Quality" flag.
Diff WindDir, AlongDir
The accuracy of wind-corrected height results is affected by the direction in which a plume is being driven by the wind. Wind directions nearly parallel to the direction of the spacecraft ground track (along-track direction) can produce highly unstable wind-corrected height solutions. The value in this field represents the smallest angular separation in degrees between the along-track direction and the direction of the plume. It is used in computing the "Region Data Quality" flag. A value less than 10 degrees is assumed to significantly affect data quality.
Power of fire in MW
If MODIS thermal anomaly data from the MOD14 product were loaded into MINX during digitizing, then this field may contain the power of the fire for the current smoke plume measured in megawatts. The value is computed by summing the power in all the MODIS thermal anomaly pixels that fall inside the digitized plume. This result is subject to many sources of error (not least that thermal anomalies may not be visible beneath dense smoke), so it should be used with caution.

The header portion of a raw data file is followed by a wide table containing one region data point per row and 37 columns of parameters. Each region data point represents a location within the perimeter of the digitized region where data were retrieved. A sample table containing only three rows of data and broken into four sections for ease of display is shown below. Following it is a list of the column parameters with brief descriptions.

Sample point data table 

      Long-     Lat-                 Km to   Dg Cw   Terr  Feature Ht (m)  
Pt#   itude    itude  Blk Samp Line   Pt 1  From N   Elev  NoWind  W/Wind
---  -------  ------  --- ---- ----  -----  ------  -----  --------------
  1 -103.083  19.194   75 1206  340    0.0    164    1223     -99     -99  
  2 -103.076  19.178   75 1209  346    1.8    164    1187    2927     -99  
  3 -103.077  19.168   75 1209  350    2.9    164    1161    2851    2662   

Windspeed (m/s)        Albedo by Band        BB TOA     Optical Depth by Band     
across   along   blue   green   red    NIR   Albedo   blue   green   red    NIR
--------------  --------------------------  ------  ---------------------------
 -99.9    0.17   0.13   0.13   0.20   0.13  -9.999  -9.999 -9.999 -9.999 -9.999
 -99.9  -99.9    0.15   0.11   0.13   0.16    0.10  -9.999 -9.999 -9.999 -9.999
  -0.1    0.2    0.15   0.11   0.12   0.16    0.10  -9.999 -9.999 -9.999 -9.999  

  Single-Scattering Albedo    Tau Fraction by Particle Type   Ang   
 blue   green   red    NIR    small  medium   large   spher   Exp
---------------------------  ------------------------------  -----
-9.999 -9.999 -9.999 -9.999  -9.999  -9.999  -9.999  -9.999 -9.999
-9.999 -9.999 -9.999 -9.999  -9.999  -9.999  -9.999  -9.999 -9.999
-9.999 -9.999 -9.999 -9.999  -9.999  -9.999  -9.999  -9.999 -9.999
 
 Power  Refl   BT21   BT31   BBT21  BBT31
 MWatt  0->1   deg K  deg K  deg K  deg K
 -----  ----   -----  -----  -----  -----
 -99.9 -9.999 -999.9 -999.9 -999.9 -999.9
  78.2  0.180  359.2  303.9  316.0  305.4
   9.5  0.196  324.0  303.1  316.2  305.8

Point data table column definitions

Pt#
A number corresponding to the order in which this point was placed into the grid inside the perimeter of the digitized region. Note that there may be missing numbers in the earliest version of MINX results.
Longitude
The longitude of this point on the WGS84 spheroid.
Latitude
The longitude of this point on the WGS84 spheroid.
Blk
The MISR block number in which this point falls. Refer to the section "Header Data Field Definitions" above for more information.
Samp
The MISR sample number or across-track pixel coordinate of this point referenced to a SOM (Space-Oblique Mercator) projection. The resolution of the MISR radiance data used in the height analysis is 275 meters, and there are 2048 of these pixels across-track in a MISR block. Therefore the value in this field can range from 0 to 2047.
Line
The MISR line number or along-track pixel coordinate of this point referenced to a SOM projection. There are 512 pixels along-track in a MISR block. Therefore the value in this field can range from 0 to 511.
Km to Pt 1
The distance in kilometers measured from the first digitized point in a region to the current point. This is the metric used to plot points against the x-axis on the "Height and Wind Profile" plots on this website.
Dg Cw From N
This represents the azimuthal orientation of a smoke plume, measured clockwise from true north. When a smoke plume is digitized, the direction in which the plume is driven by the wind is indicated by creating a line connecting two or more points along the wind direction. If there are multiple line segments on this wind direction line, and they bend to represent a change in direction, then the wind direction at any point within the plume is equal to the wind direction at the nearest point on the wind direction line. End points on a direction line take on the direction of the single line segment they touch. Interior points on a direction line take on a direction that is the average of the two line segments they touch.
Terr Elev
The elevation of the terrain above sea level at this point as defined in the MISR AGP product.
Feature Ht (m) NoWind
The height above sea level of the feature measured at this point. The feature could be a small patch on a smoke plume or cloud or on the ground. No attempt at correcting the height for feature movement due to wind is made for this field. Therefore retrieval coverage is generally better, but height accuracy is in error by an amount that depends on wind direction and wind speed. Note that the stereoscopic method of height determination retrieves the height of the top of a smoke layer. It cannot provide information on the height of the base.
Feature Ht (m) W/Wind
The height above sea level of the feature measured at this point corrected for wind. These values are generally more accurate than zero-wind heights, but they often suffer from poorer coverage, especially if smoke is homogeneous or optically thin or if smoke advection does not obey the assumption of linear, horizontal flow. For plumes that satisfy the assumption of linear, horizontal flow, that are digitized using the correct wind direction and whose wind direction is not nearly along-track, the accuracy of these heights is generally about +/- 200 meters.
Windspeed (m/s) across
This is the retrieved component of wind-speed in the across-track direction.
Windspeed (m/s) along
This is the retrieved component of wind-speed in the along-track direction.
Albedo by Band - blue, green, red, NIR
An estimate of top-of-atmosphere spectral albedo is made during height retrievals. Four values representing the four MISR spectral bands are included here. These values are not considered accurate enough to be used for detailed analytical work.
BB TOA Albedo
A rough estimate of top-of-atmosphere broad-band albedo is made during height retrievals. This value is generated using an obsolete algorithm. It is not considered reliable and should not be used.
Optical Depth by Band - blue, green, red, NIR
Optical depth in each of MISR's four spectral bands is reported for each point by reading the values from the MISR standard aerosol product. Note that the standard aerosol product reports these values on 17.6 kilometer centers, so the same values can be expected at numerous nearby points. This parameter and the following aerosol parameters will not have valid values when the optical depth is very high or when there is cloud contamination within the 17.6 km retrieval region of the standard product.
Single-Scattering Albedo - blue, green, red, NIR
Single-scattering albedo in each of MISR's four spectral bands is reported for each point by reading the values from the MISR standard aerosol product. Note that the standard aerosol product reports these values on 17.6 kilometer centers, so the same values can be expected at numerous nearby points.
Tau Fraction by Particle Type - small, medium, large, spher
These values represent the fraction of the green-band optical depth value that is attributable to small, medium and large particle sizes and the fraction of the green-band optical depth value that is attributable to spherical particles. Note that these are read from the MISR standard aerosol product which reports the values on 17.6 kilometer centers, so the same values can be expected at numerous nearby points.
Ang Exp
The angstrom exponent for the aerosol retrieval at this point. Note that these are read from the MISR standard aerosol product which reports the values on 17.6 kilometer centers, so the same values can be expected at numerous nearby points.
Power MWatt
This field contains the power of the fire at the current point measured in megawatts. Only those points which closely correspond to a MODIS thermal anomaly pixel will contain a valid value. The power in all the valid points in this table are summed to give the value in the header (refer to the field "Power of fire in MW").
Refl 0->1
The MODIS MOD14 near-IR (band 2) reflectance at this point.
BT21 - deg K
The MODIS MOD14 channel 21/22 brightness temperature at this point.
BT31 - deg K
The MODIS MOD14 channel 31 brightness temperature at this point.
BBT21 - deg K
The MODIS MOD14 background channel 21/22 brightness temperature at this point.
BBT31 - deg K
The MODIS MOD14 background channel 31 brightness temperature at this point.
Summary File Format

There are two summary files in each project directory. These record information about smoke plumes and smoke clouds with each row of the file corresponding to one smoke plume or smoke cloud region. You can find the summary files for a project by clicking the link on each project page labeled "Project Summary Files." On the summary file page there will be two additional links under the "Databases:" heading. Click either "Smoke Plume Summary File" or Smoke Cloud Summary File" to access the data of your choice.

Summary files are internally documented. Refer to the header section of any summary file to obtain a complete description of the contents of the file. Since summary files are constructed by compiling data from the raw smoke plume or smoke cloud data files in a project, you may also find it helpful to refer to the descriptions in "Raw Data File Format" above.

How to Use the Data

Choosing smoke plumes vs. smoke clouds

The digitized regions represented on this website are comprised mainly of smoke plumes and smoke clouds. These are defined in the MINX documentation as follows:

Smoke plume: This refers to the region of dense smoke above an active fire that often has well-defined edges and is connected to the fire source so that the direction in which it is being blown by wind can be determined.

Smoke cloud: This refers to a region of dispersed smoke from a nearby fire that has no obvious connection to a source on the ground and for which no wind direction can be determined.

Smoke plumes actively pump combustion products into the atmosphere, and they have observable, dynamic behavior, a snapshot of which can be observed in near-3D by MISR. If you are studying plume dynamics or processes related to the injection of smoke above the boundary layer, then you should choose to use smoke plume data from this website.

Smoke clouds represent a later stage in the life of smoke plumes, consisting of smoke that has drifted from its source area and has perhaps merged with smoke from other plumes. Smoke clouds should not be ignored, even though their heights are not corrected for the effects of wind. See the section on "Other considerations" below for cautionary information on using smoke cloud data.

Choosing raw data files vs. summary data files

Science users of this website must decide whether to use raw data files or the summary data files. The choice depends on what you expect to accomplish. For example, if you wish to investigate plume dynamics, then you will need to use the raw data files, because they contain the detailed 3D spatial information that can be related to meteorological conditions, fuel availability, fire power, etc. If you are creating histograms of injection heights for plumes within a spatial or temporal range, then the summary files may contain all the information you need.

Anything between the above extremes requires a decision on your part. The summary files impose a somewhat arbitrary definition of plume or cloud heights (see "Best Median Ht (m ASL)" in the section "Header data field definitions" above). So you may want to choose the raw data files, provide your own filter to select the points to use and devise your own estimate of height. Or rather than employing a single estimate of height per plume or cloud, you may want to use all the raw points you consider valid in your analysis.

If you plan to use summary files to investigate the relative amount of smoke that is injected above the boundary layer, you should be aware of a potential bias. Considering all plumes on an equivalent basis can bias toward lower injection heights, because there are generally more small plumes than large ones, and smaller plumes typically have lower power and lower injection heights. In this case, you might consider weighting the contribution from each plume by its digitized area (see the parameter "Area of digitized plume in square kilometers" in the summary files).

If you use wind-corrected height points from the raw data files to investigate injection heights, a similar bias can occur. A large plume might have only a few successful wind-corrected height retrievals, while a small plume could have a greater number of successful heights. The relative contribution of the large plume will be understated. In this case you can weight the contribution of each plume by the inverse fraction of plume area that is covered by successful heights (see the "Percent area covered" parameter in the "Header data field definitions" section).

If you decide to use raw data files, then you will probably want to download those you need from the "Raw Data Files" page of the website accessed from the link on each project page. For instructions on downloading files, see the "Documentation for this Website" link on the main website page.

Choosing regions by quality

The quality of height retrievals for plumes and clouds can be quite variable. To assist in choosing which regions to accept for use in your application, each raw data file and each line in the summary files contains a flag called "Region Data Quality". This flag can assume any of three values, "GOOD", "FAIR" or "POOR". The value assigned to any region is a function of the number of heights successfully retrieved for a region, the percentage of the region area that is filled with successful heights, the standard deviation/mean of the successful heights and the wind direction. The derivation of this flag is discussed in the last section of this document under "Plume Quality Flag Derivation".

You should be able to use regions that are classified as "GOOD" and "FAIR" for most applications. Note that this does not guarantee that the wind-corrected heights are accurate to the +/- 200 meters standard referred to in the definition of "Feature Ht (m) W/Wind" above. If the plume was digitized with an incorrect wind direction or the plume exhibits turbulence that results in rapidly changing wind directions, then the heights will be less accurate the greater the error in wind direction and the greater the wind speed.

Regions classified as "POOR" should not be used for most applications unless you personally inspect them and are comfortable with their quality. Many of them possess too few points to provide a valid estimate of height, and some have along-track wind directions that render their accuracy suspect. These regions have been included in the database for completeness.

The "FAIR" classification pertains to regions that are generally poorer in quality than "GOOD" regions. You may want to inspect the images for several regions that have this classification to determine whether or not you want to use them.

Other considerations

When using height information from the summary files or raw data files, be aware that they are measured relative to sea level. If you need heights above terrain, then subtract the terrain heights that are also included in both sets of files (parameter "Terr Elev" in raw data files and "Median terrain elevation in meters ASL" in summary files).

MISR has an effective swath width of about 360 kilometers and an orbital period of 99 minutes. It takes from two to nine days, depending on latitude, for MISR to pass over the same spot on the earth twice. This means that MISR samples only a small percentage of the smoke plumes and clouds that may be present at any time. Also, MISR's sampling of smoke is almost instantaneous (9 snapshots at different angles over a period of 7 minutes) relative to temporal variations in fire behavior, so this can further limit the effective coverage of fires. These factors should be taken into consideration when using the data on this website.

Because MINX uses a stereoscopic height determination algorithm employing red band data, the heights that are retrieved are necessarily smoke top heights. There is no way to determine heights at the base of smoke plumes or clouds.

Note that heights of smoke clouds are not as amenable to quality determination as for smoke plumes, because they are based on zero-wind estimates. Zero-wind heights are not filtered to remove points at or near the terrain elevation as are wind-corrected heights. Therefore, you may need to do a filtering pass on the raw zero-wind heights of smoke clouds to remove those points that are near the terrain level, before you use these data. You can inspect the "Height Profile" image for individual smoke clouds to get a good idea of whether the smoke cloud needs to be processed. The brown "Median Height Est." and "Maximum Height Est." lines on these plots reflect the values recorded in the summary files for smoke clouds (and smoke plumes).

Plume Quality Flag Derivation

The "Region Data Quality" flag is determined using the logic outlined below. The logic and thresholds were determined empirically by examining the effects of different values on individual regions and by considering the detrimental effect of wind when directed along-track.

Quality Flag values (QF = Quality Flag)
GOOD - region satisfies threshold tests for number of points retrieved, for percentage coverage of region area and for wind direction if applicable
POOR - region fails threshold tests for number of points retrieved, for percentage of plume area covered, for height standard deviation metric or for wind direction as applicable
FAIR - covers conditions intermediate between GOOD and POOR
NOTES - The standard deviation metric is: StdDevMet = StdDev(Hts) / Mean[Hts] * 1000. IsPlume is TRUE if the region is a smoke plume and FALSE otherwise.

Data parameters per region
NumHtPnts - number of successful height point retrievals
StdDevMet - standard deviation metric for successful heights
PcntHtArea - percent of plume area covered by successful height points
WindDir_AT - wind direction relative to along-track direction (degrees)

Thresholds
NUM_HT_PNTS_GOOD = 15 - upper and lower thresholds on number of successful
NUM_HT_PNTS_POOR = 4 - height points retrieved
SD_HT_PNTS_GOOD = 100 - upper and lower thresholds on standard deviation
SD_HT_PNTS_POOR = 450 - metric for successful heights
PC_HT_AREA_GOOD = 25 - upper and lower thresholds on percent of plume
PC_HT_AREA_POOR = 10 - area covered by successful height points
WIND_DIR_THRESH = 10 - threshold on minimum difference in degrees between plume direction and along-track direction

Quality Flag logic

if NumHtPnts <= NUM_HT_PNTS_POOR
    QualityFlag = 'POOR'
else 
    if NumHtPnts >= NUM_HT_PNTS_GOOD
        if PcntHtArea >= PC_HT_AREA_GOOD or StdDevHts  <= SD_HT_PNTS_GOOD
            QualityFlag = 'GOOD'
        else
            QualityFlag = 'FAIR'
    else
        if PcntHtArea >= PC_HT_AREA_GOOD and StdDevHts  <= SD_HT_PNTS_GOOD
            QualityFlag = 'GOOD'
        else
            if PcntHtArea >= PC_HT_AREA_GOOD or StdDevHts  <= SD_HT_PNTS_GOOD
               QualityFlag = 'FAIR'
            else
               if PcntHtArea <= PC_HT_AREA_POOR and StdDevHts  >= SD_HT_PNTS_POOR
                   QualityFlag = 'POOR'
               else
                   QualityFlag = 'FAIR'
            
    if IsPlume and DiffSOMalong <= WIND_DIR_THRESH
        if QualityFlag == 'GOOD'
            QualityFlag = 'FAIR'
        else
            QualityFlag = 'POOR'
   
    if NOT IsPlume and StdDevHts >= SD_HT_PNTS_POOR
        if QualityFlag == 'GOOD'
            QualityFlag = 'FAIR'
        else
            QualityFlag = 'POOR'


USA GOV website - Your first click to the U.S. Government.
Privacy/Copyright		 NASA Contact Us: Ask MISR
Last Updated:

Copyright © 1996-2006, California Institute of Technology
ALL RIGHTS RESERVED
Based on U.S. Government sponsored research