Curation of the Pathfinder AVHRR Land Data Set at CSIRO · by the Client’s contractors and...

Curation of the Pathfinder AVHRR Land Data Set at CSIRO Edward A. King CSIRO Marine and Atmospheric Research Internal Report 002 12 July 2006

King, Edward A. Curation of the Pathfinder AVHRR Land Data set at CSIRO. ISBN 1 921232 12 9. 1. Digital preservation. 2. Advanced very high resolution radiometers - Computer network resources. 3. Vegetation and climate - Remote sensing. 4. Earth sciences - Remote sensing. I. CSIRO. Marine and Atmospheric Research. II. Title. (Series : CSIRO Marine and Atmospheric Research internal report 002). 025.84

Enquiries should be addressed to:

Edward King CSIRO Marine and Atmospheric Research GPO Box 3023, Canberra, ACT, 2601 +61-2-6246-5894 +61-2-6246-5988 [email protected] Important Notice

© Copyright Commonwealth Scientific and Industrial Research Organisation (‘CSIRO’) Australia 2005

All rights are reserved and no part of this publication covered by copyright may be reproduced or copied in any form or by any means except with the written permission of CSIRO.

The results and analyses contained in this Report are based on a number of technical, circumstantial or otherwise specified assumptions and parameters. The user must make its own assessment of the suitability for its use of the information or material contained in or generated from the Report. To the extent permitted by law, CSIRO excludes all liability to any party for expenses, losses, damages and costs arising directly or indirectly from using this Report.

Use of this Report

The use of this Report is subject to the terms on which it was prepared by CSIRO. In particular, the Report may only be used for the following purposes.

this Report may be copied for distribution within the Client’s organisation;

the information in this Report may be used by the entity for which it was prepared (“the Client”), or by the Client’s contractors and agents, for the Client’s internal business operations (but not licensing to third parties);

extracts of the Report distributed for these purposes must clearly note that the extract is part of a larger Report prepared by CSIRO for the Client.

The Report must not be used as a means of endorsement without the prior written consent of CSIRO.

The name, trade mark or logo of CSIRO must not be used without the prior written consent of CSIRO.

Contents

Abstract iii

1 Introduction 1

2 Original Data - Level-A 3

2.1 Tape Media . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

2.1.1 Content . . . . . . . . . . . . . . . . . . . . . . . . . . 3

2.1.2 Media . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

2.1.3 Format . . . . . . . . . . . . . . . . . . . . . . . . . . 4

2.1.4 Missing Data . . . . . . . . . . . . . . . . . . . . . . . 5

2.2 Internet Files . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

2.2.1 Content . . . . . . . . . . . . . . . . . . . . . . . . . . 5

2.2.2 Format . . . . . . . . . . . . . . . . . . . . . . . . . . 6

3 Common HDF Format - Level-B 7

3.1 Motivation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

3.2 Implementation . . . . . . . . . . . . . . . . . . . . . . . . . . 8

3.2.1 Level-A HDF Files . . . . . . . . . . . . . . . . . . . . 8

3.2.2 Level-A Binary Files (Supplementary Data) . . . . . . 9

4 Geographic Projection - Level-C 11

5 Summary 15

A Readme.PAL 19

B Level-A HDF Header 33

C Level-B HDF Header 55

D Level-C HDF Header 69

E File Names and Sizes 81

i

ii CONTENTS

F Scripts 97

F.1 convert a b.tcl . . . . . . . . . . . . . . . . . . . . . . . . . 97F.2 make b from flat.tcl . . . . . . . . . . . . . . . . . . . . . . 102F.3 reproject b-c.tcl . . . . . . . . . . . . . . . . . . . . . . . 107F.4 make reprojection.tcl . . . . . . . . . . . . . . . . . . . . . 110

Abstract

The Pathfinder AVHRR Land data set at CSIRO, covering the period 1981through 1994, has been recovered from a number of ageing tape media,documented and converted into an HDF format that is much easier to use.Data that could not be recovered from damaged media have been replaced ascompletely as possible with data available via the Internet. Since the datawere originally obtained in the late 1990s, NASA have released updatescovering the period 1995 through 2001. These have been incorporated intothe current dataset. A subset containing the Australian region has beenextracted to reduce the data volume handling requirements for local users.The entire data set has been rearchived on modern magnetic media and willbe made available via the CMAR web site.

iii

iv ABSTRACT

Chapter 1

Introduction

The Pathfinder AVHRR Land (PAL) data set was created under the auspicesof the NOAA/NASA Pathfinder program at the Goddard Global ChangeData Centre in Maryland, USA. Numerous papers documenting the sciencebasis and algorithms underlying the data set appear in the published liter-ature and it is not the intention here to duplicate that work. For detailedbackground about the data set readers are referred to the Readme.PALfile (and references therein) that is distributed with the copy of the datapresently available on the internet at

http://disc.gsfc.nasa.gov/interdisc/readmes/pal ftp.shtml.

A copy of that file is reproduced as Appendix A of this report.This report documents the local copy of the global 8km x 8km PAL

data set held by CSIRO Marine and Atmospheric Research. The PAL datais iconic in land use research by virtue of its coverage of the decade from1981–1991 before the USGS Global 1km Project commenced and continen-tal 1km resolution AVHRR data was regularly archived in Australia. Asa result considerable effort has been made to preserve the data for use infuture research. This work has involved documenting the original mediaand file format, replacing data lost due to damaged media, translating thedata into a more convenient and standards-consistent form, and finally re-projecting from the original Goode Interrupted Homolosine projection intothe widely used and supported rectilinear Geographic Latitude/Longitudeprojection. The data has also been re-archived at each of these three stagesusing contemporary media. The purpose of this report is to act as a user’sguide and reference for researchers wishing to utilise any of these data sets,and to document the processes in each of the above steps.

The original data set held by CSIRO covered the period 1981–1994.Since then NASA have released updates through to September 2001. Thesedata have been obtained and subjected to the same processing as the earlierdata, leading to a consistently formatted and documented 20-year globaldata set.

1

2 CHAPTER 1. INTRODUCTION

Throughout this report we consistently refer to the versions of the dataas follows:

• Level-A: Original files and supplementary data replacing those lost dueto media damage.

• Level-B: Files translated to a consistent HDF format.

• Level-C: Files in Geographic projection in consistent HDF format.

For convenient reference, dumps of headers from representative HDFfiles for each of these data versions are provided in the appendices. Togetherwith complete listings of file names and sizes (also in the appendices), theycontribute the overwhelming bulk of the volume of this report.

All computer code used to perform the data manipulation is presentedin Appendix F and is available together with the data sets themselves.

Chapter 2

Original Data - Level-A

The original data that form the basis for this report were obtained from theGoddard DAAC for CSIRO by Damian Barrett in 1997. With time, someof the original tape media, and/or their copies have become unreadable.Consequently it has become necessary to source substitute data from theDAAC via the Internet. Unfortunately the two data sets differ in the specificchannels available so it is not possible to make the replacement seamless.This chapter describes the two input level-A data sets and their differences.The next chapter explains the steps taken to harmonise them.

2.1 Tape Media

2.1.1 Content

The original data set consisted of three HDF files per month containing 12channels of global data, nominally at 8km resolution. The data begin on 13July 1981 and finish on 10 September 1994. Each month is divided into 10-day intervals beginning on the 1st, 11th and 20th of the month; the lengthof third interval is adjusted, depending on the month, so that it ends on thelast day of the month. The 12 channels are listed in Table 2.1.

The data is all stored in the Goode Interrupted Homolosine Projectionin a rectangular grid of 2168 lines of 5004 pixels. There is no informationrelating line and pixel to latitude and longitude other than the naming ofthe projection.

2.1.2 Media

The data were originally provided on 14 4mm magnetic tapes, each con-taining two files. The first file on each tape is an ASCII listing of the tapecontents and file sizes. The second file on each tape contains up to 36 HDFfiles (3 per month) for an entire year packed in a Unix tar file. Each HDF

3

4 CHAPTER 2. ORIGINAL DATA - LEVEL-A

HDF Data Set Name Content

Data-Set-2 NDVIData-Set-4 CLAVR flagsData-Set-6 Quality Control flagsData-Set-8 Scan Angle (radians)Data-Set-10 Solar Zenith Angle (radians)Data-Set-12 Relative Azimuth Angle (Radians)Data-Set-14 Channel 1 Normalized Reflectance (%)Data-Set-16 Channel 2 Normalized Reflectance (%)Data-Set-18 Channel 3 BB Temperature (K)Data-Set-20 Channel 4 BB Temperature (K)Data-Set-22 Channel 5 BB Temperature (K)Data-Set-24 Day of Year

Table 2.1: Data sets contained in original HDF files.

file is compressed with the Unix “compress” program. The file names are ofthe form:

PAL mmm dd-dd yyyy.HDF.Z

where mmm is a three letter month code (eg JAN, FEB, etc), dd-dd are thestart and end days of month for the interval, and yyyy is the year. A conse-quence of this naming scheme is that the file names naturally sort into orderby month name, rather than by year/month/day which can unnecessarilycomplicate processing systems. A complete list of the level-A files appearsin the first column of the table in Appendix-E. At some point the 4mmtapes were duplicated on 8mm (Exabyte) media which are stored with the4mm media.

2.1.3 Format

Within each HDF file, each scientific data set (SDS) is documented solely byits own HDF attributes. An example header, produced by the HDF utilityprogram hdp, for one of these HDF files appears in Appendix B. The dataare all stored in what HDF calls “calibrated” form, where quantities areoffset and scaled so that they can be stored using a reduced number of bits(and precision) as integers. In general the physical quantities are related tothe stored data according to:

Physical Value = (Stored Value - Add Offset) * Scale Factor

The scale factor and add offset are recorded as attributes of each SDS. Otherattributes describe the maximum and minimum valid values for the data,the coordinate system and the units. The name of the physical variable is

2.2. INTERNET FILES 5

actually stored as an attribute rather than as the name of the SDS. TheHDF convention for the valid max and valid min attributes is that theyspecify the limits of the physical variable, not the calibrated (stored) quan-tity. Unfortunately in these files they refer to stored values, and so someHDF readers have difficulty in extracting the data correctly. Each SDS hasonly default dummy coordinate variables that HDF uses to enumerate eachaxis; and these are different for each SDS. Several of these factors motivatethe reformatting of the data described in the next chapter.

2.1.4 Missing Data

As mentioned, some of the media have become unreadable. In particularthe 1985 4mm tape is truncated with only 13 of the potential 36 files beingcomplete (see listing in Appendix E). The 8mm copy of this tape exhibitsthe problem identically, suggesting that the 4mm tape was defective beforethe copy was made.

The 8mm copies of both the 1991 and 1992 4mm tapes are also unread-able, though the 4mm originals produced complete data. There is howevera problem with the file from the middle interval of May 1991. Although itappeared to read successfully from the tape, and apparently uncompressedcorrectly, no HDF reader is able to open it. A total of 24 files are missing,23 from 1985, and one from 1991.

2.2 Internet Files

To supplement the missing data we used the collection provided by theGSFC DAAC at:


These data are described in detail in Appendix A. The supplementarydata covering 1995–2001 are in the same format, and have presumably beensubjected to the same processing as the pre-1995 data, even though theonline documentation has not been updated to reflect their existence.

2.2.1 Content

The major difference between these data and the data distributed on thetape is that only the NDVI and four of the five AVHRR channels are provided(Channel 3 is absent). There is no angle data, day of year or flag data. In allother respects the data channels and NDVI are identical to that containedin the HDF files from the tapes.

On the other hand, two ancillary files are provided that give the mappingof each pixel and line in the Goode Interrupted Homolosine Projection to a

6 CHAPTER 2. ORIGINAL DATA - LEVEL-A

corresponding latitude and longitude, and because the grids are identical tothose in the HDF files, they are equally applicable to those data.

2.2.2 Format

The major format difference is that each channel is provided as a separateflat binary data set, in “calibrated” form like the HDF data, but without thedescribing attributes. The Readme file reproduced in Appendix A containsall the necessary extra information needed to decode these files and computethe physical quantities. We downloaded the subset of 144 of these files (24intervals of 6 files each) necessary to replace the missing HDF files.

Chapter 3

Common HDF Format -Level-B

3.1 Motivation

As discussed, the loss of some files due to media errors in the basic level-Adata set necessitated substitution of data from an alternative source in adifferent format. For ease of use, it is very desirable that all the data be in acommon format. The HDF format used for the majority of the level-A dataset is obviously a far richer and portable style than the flat binary files ofthe supplementary data.

The previous chapter however enumerated several issues with the exist-ing HDF format. Specifically these include:

• Data stored as calibrated (smaller but more complicated to read, moreattributes required);

• Incorrect use of valid min and valid max attributes (some HDF read-ers have difficulty);

• SDS name is not descriptive (inconvenient);

• Lack of explicit dimension coordinate variables (potential ambiguity).

We have chosen therefore to convert the data to a new format, designatedlevel-B, that addresses all these issues. A key goal was to not change thedata values or affect in any way the valid values of the actual physical vari-ables, nor to lose any metadata. The HDF file storage format was retainedbecause it is a self-documenting, self-contained, architecturally-independentand widely supported standard.

7

8 CHAPTER 3. COMMON HDF FORMAT - LEVEL-B

3.2 Implementation

3.2.1 Level-A HDF Files

The primary advantage of storing data in calibrated form in an HDF fileis to save storage volume. With disk storage capacities increasing, andthe ubiquitous availability of high performance compression tools (togetherwith the CPU power to run them) the significance of the space issue isdecreasing. The disadvantage of the calibrated form is the complication ofhaving to decode and type convert the data whenever it is accessed. Wetherefore decided to perform the conversions once and store each physicalvariable in a native format appropriate to its type; i.e. integer types forflags and floating point types for real numbers.

To eliminate the issues with the valid min and valid max attributes, aFillValue attribute was defined for every data set and all non-valid datapixels were set to that value. The process steps for converting the data aretherefore:

• Read calibrated data without conversion;

• Define FillValue;

• Use valid min and valid max attributes to set invalid data to FillValue;

• Promote data type (e.g. integer to floating point) and apply calibra-tion parameters

Physical Value = (Stored Value - Add Offset) *

Scale Factor

• Attach standard x and y coordinate variables

• Write data directly (uncalibrated) to new HDF file with an SDS namederived from the name attribute

• Copy any other attributes to new file

Two additional global attributes were added to the new file; one brieflydescribing the above translation process and where to find more informa-tion, and another giving the name of the original level-A HDF file (fortraceability). Table 3.1 gives the relationships between the attributes forthe conversion between the two HDF formats.

Note that the dimension variables x and y are defined only once and thenused for every SDS. An example HDF header for the new format (level-B)is provided in Appendix C. Comparison with the level-A header given inAppendix B illustrates the extent of the simplification of the format.

3.2. IMPLEMENTATION 9

Original Format (level-A) New Format (level-B)

Type Entity Type Entity

n/a Global Attr. descriptionn/a Global Attr. source file

Dimension Var. fakeDimi Dimension Var. yDimension Var. fakeDimj Dimension Var. xDimension Attr. fakeDimi Dimension Attr. yDimension Attr. fakeDimj Dimension Attr. x

Attr. coordsys Attr. coordsysAttr. valid max n/aAttr. valid min n/aAttr. scale factor n/aAttr. scale factor err n/aAttr. add offset n/aAttr. add offset err n/aAttr. calibrated nt n/aAttr. remarks-1 Attr. remarks 1Attr. long name n/aAttr. Units Attr. unitsAttr. Format n/a

Table 3.1: Relationships between the attributes in the level-A HDF files andthe new level-B HDF files.

Since new files were being written, we took the opportunity to rearrangethe file names to a form that would more conveniently sort into ascendingtime order:

yyyymm dd-dd.hdf

where mm is the two digit month number and the other symbols are asbefore. The script program written to perform the conversion is reproducedin Appendix F, Section F.1.

3.2.2 Level-A Binary Files (Supplementary Data)

For the channels present in both data sets (NDVI and four channels ofAVHRR), comparison shows these data are identical to those stored in theHDF files. The major difference is that each channel is in a separate flatbinary file with no header or other directly attached metadata. Since thenew HDF format described above stores the data more naturally in its nativetype without HDF calibration and with a reduced number of attributes, itis a simple task to convert the data in the binary files to their native types,applying mostly the same steps as above, and aggregate them into a single

10 CHAPTER 3. COMMON HDF FORMAT - LEVEL-B

HDF file per ten-day interval. The attributes listed in the above table aretrivially added using data from the documentation available over the WWW(Appendix A). To make the file naming scheme consistent with the filescreated from the original HDF files, yet retain an indication that the datais derived from the supplementary set, we adopted a similar convention:

yyyymm dd.hdf

where the symbols are as before and the only difference is that onlythe start day of each time interval is included, not the end day. Again therelevant script program appears in the Appendix, Section F.2.

Chapter 4

Geographic Projection -Level-C

The Goode Interrupted Homolosine projection1 in which the original dataare stored is not in particularly widespread use and so not all image manip-ulation packages support it. It was felt that it would make it much easierfor users if the data could be presented in a more straightforward rectilin-ear Geographic Latitude/Longitude projection. This would simplify directlookup of particular locations (pixels) or spatial subsets of interest withoutcomplicating projection computations. The distortion in polar regions thatwould result is unlikely to be important, at least for Southern users of theNDVI dataset. In any case it is intended to continue to make the level-Bdata in the Goode Interrupted Homolosine projection available for users.

The level-A files obtained from the Internet included two files containinglayers of latitude and longitude in degrees as a function of line and pixelnumber in the Goode Interrupted Homolosine projection. Pseudo-colourimages of these two layers appear in Figure 4.1. The programming languagechosen to implement this project (Tcl-NAP, see Appendix F) has facilitiesthat make resampling and reprojecting data very straightforward. In par-ticular there is a function to take a pair of grids such as these giving amapping between two projections, and create an “inverse grid” which mapsin the opposite direction (i.e. in this case, grids of line and pixel as func-tions of latitude and logitude). A second function can use this inverse gridto remap data layers into the destination projection.

A short program (Section F.4) was written to read the two binary filescontaining the latitude and longitude grids, attach coordinate variables,compute the inverse grids and save both pairs of grids in a single HDF

1Goode Interrupted Homolosine is a combination of Sinusoidal and Mollweide projec-

tions with clipping boundaries and central meridians to reduce distortion around features

of interest. See for example ftp://ftp.remotesensing.org/proj/PROJ.4.3.I2.pdf for

further details.

11

12 CHAPTER 4. GEOGRAPHIC PROJECTION - LEVEL-C

Figure 4.1: Grids of latitude and longitude (both in degrees) in the GoodeInterrupted Homolosine projection.

13

Figure 4.2: Grids of pixel and line number to convert from Goode Inter-rupted Homolosine to Geographic Latitude/Longitude projection.

14 CHAPTER 4. GEOGRAPHIC PROJECTION - LEVEL-C

file for subsequent use. Pseudo-colour images of the two inverse grids arepresented in Figure 4.2. The original grids had dimensions of 5004 pixels by2168 lines with longitudes running from −180.00◦ to +180.00◦ and latitudesfrom +89.72◦ to −89.59◦. This corresponds to a cell size of 0.072◦ × 0.08◦.The output grid was chosen with a uniform square cell size of 0.08◦ giving anominal 4500 samples in longitude and 2250 lines in latitude. Because theinput grids do not contain latitudes right to the poles, the number of pixelsin latitude is actually 2241.

The final step in the conversion process was to apply the inverse grids toevery data layer in each of the level-B HDF files to reproject, using nearestneighbour resampling, the data into the latitude-longitude projection. Thesereprojected data were then written to a corresponding level-C file after atrivial resizing step that added the correct number of lines at the top andbottom so that there were 2250 lines corresponding to a latitude range of+90.00◦ to −90.00◦. The program to do this appears in Section F.3. Nochanges were made to the individual metadata attributes other than tochange the coordinate system name. The global description attribute wascopied to the output file as an attribute with name parent description, anda new description attribute referencing it and explaining that the data hadbeen reprojected was inserted.

The filenaming scheme described previously for the level-B data wasretained, modified only by the addition of a geo suffix to main part of thename.

Chapter 5

Summary

The work described in this report has resulted in the assembly of a con-sistently curated copy of the PAL data set from 1981 to 2001. The dataare in a common file format with uniform metadata and an informative andhelpful naming scheme, for both the data sets and the files. The data hasbeen manipulated through three stages of processing, designated levels A, Band C. In the final stage there are 474 files corresponding to every “10-day”epoch between 13 July 1981 and 10 September 1994. A further 243 filescover the period from 1 January 1995 through 30 September 2001. Thereare 11 epochs missing between 11 September 1994 and 31 December 1994.

The data for each stage of the processing amounts to a tree of between 20and 25 GB of compressed files. The compression achieved (via gzip) is abouta factor of ten so that uncompressed each global time series is nearly 200 GB.To facilitate use by local users, an Australian subset has been extracted fromthe level-C data tree. These data are in an identical format to the level-Cfiles except the latitude ranges from −10◦ to −45◦ and longitude from 110◦ to155◦. The compressed Australian data tree amounts to 1.2 GB. Each of thethree main trees has been separately archived in duplicate to DLT IV media(making a total of 6 tapes). Each tape also contains a copy of this report, thereprojection file and all the software necessary to generate the subsequentstage(s). The level-C Australian subset is also included separately on eachtape.

Example images of the NDVI channel for the first epoch of February1982 are shown here in Figures 5.1, 5.2 and 5.3. A complete listing of allthe filenames and file sizes in each processing stage is given in Appendix E.

15

16 CHAPTER 5. SUMMARY

Figure 5.1: Level-A raw NDVI data in Goodes Interrupted Homolosine pro-jection. Note the data values are integers between zero and 255 with zerodenoting missing data.

17

Figure 5.2: Level-B scaled NDVI data in Goodes Interrupted Homolosineprojection. Here the data are represented by actual physical values andmissing data are not represented by a valid data value.

18 CHAPTER 5. SUMMARY

Figure 5.3: Level-C NDVI data in Geographic Projection.

Appendix A

Readme.PAL

This appendix reproduces verbatim the Readme.PAL file available at the timeof writing from


Pathfinder AVHRR Land Data

An HTML version of this document is available at the following URL:

http://daac.gsfc.nasa.gov/

1. DATA SET OVERVIEW

1.1 Sponsor

1.2 Original Archive

1.3 Future Updates

2. THE DATA

2.1 Characteristics

2.2 Source

3. THE FILES

3.1 Format

3.2 Name and Directory Information

3.3 Companion Software

4. THE SCIENCE

4.1 Theoretical Basis of Data

19

20 APPENDIX A. README.PAL

4.2 Processing Sequence and Algorithms

4.3 Scientific Potential of Data

4.4 Validation of Data

5. DATA ACCESS AND CONTACTS

5.1 FTP Site

5.2 Points of Contact

6. REFERENCES

1. DATA SET OVERVIEW

This data set, produced as part of the NOAA/NASA Pathfinder AVHRR Land (PAL)

program, contains global and continental monthly and 10-day composites of

channels 1, 2, 4 and 5 and the Normalized Difference Vegetation Index (NDVI) at

8 km and 1 degree resolution. The data, derived from the Advanced Very High

Resolution Radiometers (AVHRR) on the "afternoon" NOAA operational

meteorological satellites (NOAA-7, -9, -11), cover the period from 1981 to

1994. The Pathfinder Program produces long-term data sets processed in a

consistent manner for global change research.

1.1 SPONSOR

The production and distribution of this data set are funded by NASA’s Mission

to Planet Earth program. The data are not copyrighted; however, we request

that, when you publish data or results using these data, please acknowledge as

follows:

The authors wish to thank the Distributed Active Archive Center

(Code 902.2) at the Goddard Space Flight Center, Greenbelt, MD,

20771, for producing the data in their present form and

distributing them. The original data products were produced

under the NOAA/NASA Pathfinder program, by a processing team

headed by Ms. Mary James of the Goddard Global Change Data

Center; and the science algorithms were established by the

AVHRR Land Science Working Group, chaired by Dr. John Townshend

of the University of Maryland. Goddard’s contributions to these

activities were sponsored by NASA’s Mission to Planet Earth

program.

1.2 ORIGINAL ARCHIVE

This data set is part of the Pathfinder Land data set archived at the Goddard

DAAC. It is derived from the PAL 8 km Daily data. The Daily data are derived

from the NOAA AVHRR Global Area Coverage (GAC) 1B data, available from NOAA’s

Satellite Active Archive.

21

1.3 FUTURE UPDATES

This data set will be updated as data from other years are processed.

2. THE DATA

2.1 CHARACTERISTICS

Parameters, Units, Range

______________________________________________________________________________

PARAMETER DESCRIPTION UNITS DATA RANGE

______________________________________________________________________________

NDVI Normalized Difference Unitless -1 to +1

Vegetation Index derived

from the visible and

near-infrared channel

reflectances (0.58 to

0.68 um and 0.73 to

1.10 um, respectively)

Ch1 AVHRR Channel 1 % 0 to 100

Reflectance

Ch2 AVHRR Channel 2 % 0 to 100

Reflectance

Ch4 AVHRR Channel 4 Kelvin 160 to 340

Brightness Temperature

Ch5 AVHRR Channel 5 Kelvin 160 to 340

Brightness Temperature

elev Ancillary elevation file Meters -15000 to 10000

lsm Ancillary Land Sea Mask Not Applicable 0 to 1

lat Ancillary Latitude file Degrees -90 to 90

lon Ancillary Longitude file Degrees -180 to 180

______________________________________________________________________________

Temporal Coverage: July 1981 to September 1994

Temporal Resolution: 10-day and monthly composites.

Spatial Coverage: Global and Continental

Spatial Resolution: 8 km x 8 km and 1 degree x 1 degree


Starting Points and End Points

___________________________________________________________________________

North South West East

Area Latitude Latitude Longitude Longitude

___________________________________________________________________________

Global 90.0 -90.0 -180.0 180.0

Africa 38.5 -37.8 -20.0 61.3

Asia 79.0 4.5 25.6 143.7

Australia 7.5 -48.0 93.5 179.5

Europe 76.7 23.4 -13.0 60.5

North America 72.5 9.0 -165.0 -60.0

South America 13.2 -57.0 -83.0 -33.0

___________________________________________________________________________

2.2 SOURCE

These data were collected by the Advanced Very High Resolution Radiometer

(AVHRR) flown on NOAA-series satellites.

Nominal orbit parameters for the NOAA-7, -9, and -11 are

Launch date: 6/23/81 (NOAA-7), 12/12/84 (NOAA-9), 9/24/88 (NOAA-11)

Orbit: Sun synchronous, near polar

Nominal altitude: 833 km

Inclination: 98.8 degrees

Orbital period: 102 minutes

Equatorial crossing times: 114.30 (NOAA-7), 14.20 (NOAA-9),

13.40 (NOAA-11) LST

Nodal Increment: 25.3 degrees

The orbital period of about 102 minutes produces 14.1 orbits per day. Because

the daily number of orbits is not an integer, the suborbital tracks do not

repeat daily, although the local solar time of the satellite’s passage is

essentially unchanged for any latitude. The 110.8 degrees cross-track scan

equates to a swath of about 2700 km. This swath width is greater than the 25.3

degrees separation between successive orbital tracks and provides overlapping

coverage (side-lap).

The spectral band widths and Instantaneous Field of View (IFOV) of the AVHRR

instrument are given in the following table.

Channel Wavelength (micrometer) IFOV (milliradian)

------- ------------------------ ------------------

1 0.58 - 0.68 1.39

2 0.73 - 1.10 1.41

3 3.55 - 3.93 1.51

4 10.3 - 11.3 1.41

5 11.5 - 12.5 1.30

23

A more detailed, comprehensive description of the NOAA series satellites, the

AVHRR instrument, and the AVHRR GAC 1B data can be found in the NOAA Polar

Orbiter Data User’s Guide (Kidwell 1991), which can be obtained from NOAA’s

National Environmental Satellite Data and Information Service (NESDIS) (see

DATA ACCESS AND CONTACTS).

3. THE FILES

3.1 FORMAT

Compressed format:

The Pathfinder data on the ftp site has been compressed using Gzip, which

reduces the size of the named files using Lempel-Ziv coding.

Uncompressed format:

The Pathfinder data are processed as 32-bit floating point numbers to maintain

maximum accuracy. In generating the output data, each layer is scaled to an

appropriate 8-bit (unsigned) or 16-bit (unsigned) interger value corresponding

to the ranges shown in the table below. Consiquently, to obtain the

geophysical values from the scaled data value, offsets must be subtracted from

the scaled data value and the result multiplied by the gain. Complete

information on scaling and bit representations are provided int the table

below.

________________________________________________________________________

Parm. bits Offset Gain Bin. min/max

________________________________________________________________________

NDVI 8-bit unsigned 128 .008 3 - 253

Ch1 16-bit unsigned 10 .002 10 - 50010

Ch2 16-bit unsigned 10 .002 10 - 50010

Ch4 16-bit unsigned -31990 .005 10 - 36010

Ch5 16-bit unsigned -31990 .005 10 - 36010

elev 16-bit unsigned 15010 1 10 - 25010

lat 16-bit unsigned 9010 .01 10 - 18010

lon 16-bit unsigned 18010 .01 10 - 36010

lsm 8-bit unsigned 0 1 0 - 2

________________________________________________________________________

File Size, File Demension

_________________________________________________________________________

File Type File Size (bytes) Columns Rows

uncompressed

_________________________________________________________________________

Global 1-degree NDVI 64800 360 180


Global 1-degree Channel’s 129600 360 180

Global 8-km NDVI 10848672 5004 2168

Global 8-km Channel’s 21697344 5004 2168

Africa NDVI 1166000 1100 1060

Afica Channel’s 2332000 1100 1060

Asia NDVI 1320500 1390 950

Asia Channel’s 2641000 1390 950

Australia NDVI 831600 1080 770

Australia Channel’s 1663200 1080 770

Europe NDVI 522600 780 670

Europe Channel’s 1045200 780 670

North America NDVI 893800 1090 820

North America Channel’s 1787600 1090 820

South America NDVI 669300 690 970

South America Channel’s 1338600 690 970

_________________________________________________________________________

The ancillary files elev, lat, and lon are the same size and dimension as the

channel files and lsm is the same size and dimension as the NDVI file, for

thier respective coverage’s.

Headers, Trailers, and Delimiters: None.

Fill Values: 0 is Missing Data Over Land, 1 is Ocean, 2 is Goode’s

Interrupted Space.

Image Orientation: North to South

Map Projection: Goode Interrupted Homolosine Projection (global 1

degree data is not projected)

3.2 NAME AND DIRECTORY INFORMATION

Naming Convention

The file naming convention for this data set is

xxxxxxxx.pppppp.lctgrr.yymm[dd].gz

xxxxxxxx.pppppp.lctgrr.gz

where:

xxxxxxxx: data product designator (i.e. avhrrpf)

pppppp: parameter name

ndvi = Normalize Difference Vegetation Index

ch1 = Channel 1

ch2 = Channel 2

ch4 = Channel 4

ch5 = Channel 5

elev = Elevation

lat = Latitude

25

lon = Longitude

lsm = Land/Sea Mask

lctgrr: code for spatial/temporal resolution & coverage

l = number of vertical levels (1 level for this data set)

c = vertical coordinate(n = not applicable for this data set)

t = temporal period, with the following definitions:

t = ten day composites

m = monthly composites

n = not applicable

g = horizontal grid resolution

e = 1 deg lat x 1 deg lon

f = 8 km x 8 km

rr= spatial coverage, with the following definitions:

gl global land

af africa

as asia

eu europe

sa south america

na north america

au australia

yy: year (e.g. 87=1987)

mm: month (e.g. 12=December)

dd: first day of ten day period

gz: designates gzip compressed file

Directory Path

The directory path for data files is:

/data/avhrr/kkkkkkkk[/cccccccc]/yyyy/mmm/

where:

kkkkkkkk is coverage (i.e. global 8-km, global 1-degree and continent)

cccccccc is the continent

yyyy is year

mmm is month (e.g. jan = January

The directory path for ancillary files is:

/data/avhrr/kkkkkkkk[/cccccccc]/ancillary/

3.3 COMPANION SOFTWARE

Read software for these data files are located in the /data/avhrr/utilities

directory. An explanation of this read software can be aquired at the following

URL:

http://daac.gsfc.nasa.gov/CAMPAIGN_DOCS/LAND_BIO/GLBDST_Misc_SW.html


4. THE SCIENCE

4.1 THEORETICAL BASIS OF DATA

Spectral Wavelengths

On the NOAA-7, NOAA-9, and NOAA-11 satellites, the AVHRR sensor measures

emitted and reflected radiation in five channels (bands) of the

electromagnetic spectrum: a visible (0.58 to 0.68 micrometer) band that is

used for daytime cloud and surface mapping; a near-infrared (0.725 to 1.1

micrometer) band used for surface water delineation and vegetation cover

mapping; a mid-infrared (3.55 to 3.93 micrometer) band used for sea surface

temperature and nighttime cloud mapping; a thermal infrared (10.5 to 11.5

micrometer) band used for surface temperature and day and night cloud mapping;

and another thermal infrared (11.5 to 12.5 micrometer) band used for surface

temperature mapping (Kidwell 1991).

Vegetation Index

The first AVHRR channel is in a part of the spectrum where chlorophyll causes

considerable absorption of incoming radiation, and the second channel is in a

spectral region where spongy mesophyll leaf structure leads to considerable

reflectance. This contrast between responses of the two bands can be shown by

a ratio transform; i.e., dividing one band by the other. Several ratio

transforms have been proposed for studying different land surfaces (Tucker,

1979). The Normalized Difference Vegetation Index (NDVI) is one such ratio,

which has been shown to be highly correlated with vegetation parameters such

as green-leaf biomass and green-leaf area and, hence, is of considerable value

for vegetation discrimination (Justice et al. 1985).

NDVI Relationships With Geophysical Variables

A ratio between bands is of considerable use in reducing variations caused by

surface topography (Holben and Justice 1981). It compensates for variations in

radiance as a function of Sun elevation for different parts of an image. The

ratios do not eliminate additive effects caused by atmospheric attenuation,

but the basis for the NDVI and vegetation relationship holds generally. The

soil background contributes a reflected signal apart from the vegetation, and

interacts with the overlying vegetation through multiple scattering of radiant

energy. Huete (1988) found the NDVI to be as sensitive to soil darkening

(moisture and soil type) as to plant density over partially vegetated areas.

4.2 PROCESSING SEQUENCE AND ALGORITHMS

Formulae

Derivation Techniques and Algorithms

27

Recalibrated radiances for the Pathfinder data are converted to surface

reflectance and brightness temperature using the following procedures and

information available from NOAA (Rao, 1993a,b).

Calibration for Channels 1 and 2:

R = A exp(B x d)*(C(10) - C(0))

where

R = is the radiance in [W][m-2][micrometer-1][steradian-1]

d = is the number of days since launch

C(10) = is the measured counts on the 10-bit scale

A, B, C(0) = are parameters supplied by the NOAA/NASA Pathfinder

Calibration Working Group (Rao 1993b).

For Channel 3, the calibration is determined by the gain and offsets provided

in the NOAA 1B data, using procedures described in Kidwell (1991).

For Channels 4 and 5, radiances are computed from the temperatures of the

Internal Calibration Target (ICT) and the laboratory blackbody as convolutions

of the Planck function over AVHRR’s spectral response functions. Details are

available in NOAA Technical Report (Rao 1993a).

Reflectance for Channels 1 and 2 is derived as follows:

Reflectance = (calibrated surface leaving radiance) / (calibrated

incoming radiance).

The Atmospheric correction scheme follows the algorithm of Gordon et. al

(1988)

Brightness temperatures for Channels 3, 4, and 5 are derived as follows:

The calibrated radiances are converted to brightness temperatures

using Planck function. NOAA provides look-up tables for each satellite

(Kidwell 1991, Brown et al. 1985, Weinreb et al. 1990).

NDVI is derived:

(Channel 2 reflectance - Channel 1 reflectance)

-------------------------------------------------------

(Channel 2 reflectance + Channel 1 reflectance)

Data Processing Sequence

Daily data are produced for a compositing period and quality controlled. The

10-day composite is created from Daily data and quality controlled. Climate

data are generated from the composite. The Monthly Composite data are derived


from the 10-day data.

Processing Steps (and data sets)

The Pathfinder AVHRR data are created by the following steps.

1.Ascending GAC orbit data are unpacked and staged.

2.Ancillary data needed in processing are retrieved. These include ozone

data from Nimbus-7 Total Ozone Mapping Spectrometer (TOMS), land

surface elevation from the Earth Topographic Five Minute Grid (ETOPO5)

data set, land or sea mask, and satellite ephemeris files.

3.Each scan is navigated using an orbital model.

4.Based on the precise navigation, latitudes, longitudes, solar zenith,

solar azimuth, scan, and relative azimuth angles are determined for

each pixel.

5.Calibration and atmospheric corrections are applied, and counts are

converted to radiances that are used to derive reflectance and

brightness temperatures.

6.Cloud flags are calculated and appended and the NDVI is calculated from

the surface reflectances.

7.The data are then resampled to 8 km x 8 km pixels in the output product

(Daily data) and all ocean data are masked out.

8.Once 10 days of daily data are processed, they are composited by

choosing values for each bin based on the day that has the highest

NDVI value. Only those pixels within 42 degrees of nadir are used in

the composite. The monthly data are composited from the 10 day data

using the same method.

9.The global one degree data are produced from the Composites.

10.Data products and their associated metadata are quality controlled

before archiving at the Goddard DAAC.

Processing Changes

The changes included with Pathfinder data beginning in 1988 are

Relative azimuth representation

Bug fixes to CLAVR

Bug fixes to thermal calibration

Modification of ozone input to atmospheric correction

The change made beginning in 1986 data is

Composite routine changes

Calibration

The Pathfinder Calibration Working Group recommended time- dependent

calibration coefficients that incorporate the slopes derived from several

29

different calibration investigations and tie these to offsets corresponding to

certain aircraft underflights (Staylor 1990). For channels 4 and 5, new

methods for calibration were recommended based in part on reanalysis of

preflight calibration data that take into consideration the nonlinear response

of the instrument and provide corrections to earlier gain and offset

adjustments.

In the processing stream, the satellite number and days since launch are used

to calculate a revised gain. This new gain, along with offsets provided in Rao

(1993b), is used to calculate radiance. For the thermal channels, the gains

and offsets provided in the NOAA 1B record are corrected using the Internal

Calibration Target (ICT) temperature and corrections provided in Rao (1993a).

They are then applied to calculate a top of the atmosphere radiance. This is

then converted into brightness temperature using a Planck function equivalent

lookup table based on the response curve of each channel. Channel 3 is

converted to brightness temperature following procedures described in Kidwell

(1991).

Atmospheric Correction A Rayleigh correction is calculated and applied

using a standard radiative transfer equation and methodology, which follows

the work of Gordon et al. (1988). This includes a correction for ozone

absorption and daily ozone data from the Total Ozone Mapping Spectrometer

(McPeters et al. 1993) used in the correction. In addition, the pixel

elevation as determined from the ETOPO5 data set (NGDC 1993) is used to

correct the pressure level used in the calculation of Rayleigh coefficients.

The Rayleigh correction terms are applied to the Channels 1 and 2 radiance,

and the resulting reflectances are normalized for solar illumination.

4.3 SCIENTIFIC POTENTIAL OF DATA

This data set product (NDVI) is particularly useful for studies of temporal

and interannual behavior of surface vegetation and for developing surface

background characteristics for use in climate modeling. Some uses of NDVI

include

Global land cover classification

Regional agricultural crop monitoring

Desertification studies and drought monitoring

Terrestrial environmental monitoring

Global water and energy balance studies.

4.4 VALIDATION OF DATA

A few validation checks have been built into the Pathfinder data processing

(Quality Control Flags). Automated quality checks are made for consistency in

fields such as date and satellite or scan times. Geophysical values are

checked to see that they are within a reasonable range. Certain anomalies may

exist in the data set because of conditions inherent in the input data, for

example, missing scan lines or orbits, incorrect or incomplete calibration


coefficients, and many of these data are flagged with the Quality control

indicator.

Errors in the Computation of Solar Zenith Angles have been found in the PAL

Data Set. This error in the SZA in the daily and composite data, varies

systematically with time and geographical location. An analysis of SZA from

the PAL data set showed that the errors are smaller in the recent data (e.g.,

2o in the 1992 data) and larger in the earlier data (e.g., 7o in the 1982

data). Errors in the solar zenith angle layer affect channel 1 and 2

reflectances, NDVI, CLAVR flags and the relative azimuth angles. A memo has

been written that discusses the impact of these errors on the PAL data set.

The memo is accesible at the following URL:

http://daac/CAMPAIGN_DOCS/LAND_BIO/zenith_angle_memo.html

5. DATA ACCESS AND CONTACTS

5.1 FTP SITE

The Pathfinder AVHRR Land Composite Data Set resides on the Goddard DAAC

anonymous FTP site. You may access the files directly via FTP at

ftp daac.gsfc.nasa.gov

login: anonymous

password: < your internet address >

cd /data/avhrr

When downloaded the PAL 8-bit and 16-bit unsigned compressed files

with a WWW browser, use the pop-up menu or click on a link with

the shift key held down (option key on Macintosh). This will

produce the dialog box enabling you to save the file.

5.2 POINTS OF CONTACT

For information about or assistance in using any DAAC data, contact

EOS Distributed Active Archive Center(DAAC)

Code 902.2

NASA Goddard Space Flight Center

Greenbelt, Maryland 20771

Internet: [email protected]

301-614-5224 (voice)

301-614-5268 (fax)

6. REFERENCES

31

Brown, O.W., J.W. Brown, and R.H. Evans. 1985. Calibration of Advanced Very

High Resolution Radiometer observations. Journal of Geophysical Research,

90:11667- 11677.

Gordon, H.R., J.W. Brown, and R.H. Evans. 1988. Exact Rayleigh scattering

calculations for use with the Nimbus- 7 coastal zone color scanner. Applied

Optics, 27:2111-2122.

Holben, B.N., and C.O. Justice. 1981. An examination of spectral band ratioing

to reduce the topographic effect on remotely sensed data, International

Journal of Remote Sensing, 2:115-133.

Huete, A.R. 1988. A soil adjusted vegetation index (SAVI), Remote Sensing of

the Environment, 25:295-309.

Justice, C.O., J.R.G. Townshend, B.N. Holben, and C.J. Tucker. 1985. Analysis

of the phenology of global vegetation using meteorological satellite data,

International Journal of Remote Sensing, 6:1271-1318.

Kidwell, K. 1991. NOAA Polar Orbiter Data User’s Guide. NCDC/SDSD. National

Climatic Data Center, Washington, DC.

McPeters, R.D., et al. 1993. Nimbus-7 Total Ozone Mapping Spectrometer (TOMS)

Data Products User’s Guide. NASA Reference Publication 1323.

NGDC. 1993. 5 Minute Gridded World Elevation. NGDC Data, Announcement DA

93-MGG-01. Boulder.

Rao, C.R.N. 1993a. Nonlinearity corrections for the thermal infrared channels

of the Advanced Very High Resolution Radiometer: assessment and

recommendations. NOAA Technical Report NESDIS-69. NOAA/NESDIS. Washington, DC.

Rao, C.R.N. 1993b. Degradation of the visible and near-infrared channels of

the Advanced Very High Resolution Radiometer on the NOAAP9 spacecraft:

assessment and recommendations for corrections. NOAA Technical Report NESDIS-

70. NOAA/NESDIS. Washington, DC.

Staylor, W.F. 1990. Degradation rates of the AVHRR visible channel from the

NOAA-6, -7, and -9 spacecraft. Journal of Atmospheric and Oceanic Technology,

7:411-423.

Tucker, C.J. 1979. Red and photographic infrared linear combinations for

monitoring vegetation. Remote Sensing of the Environment, 8:127-150.

Weinreb, M.P., G. Hamilton, S. Brown, and R.J. Koczor. 1990. Nonlinearity

corrections in calibration of Advanced Very High Resolution Radiometer

infrared channels. Journal of Geophysical Research, 95:381-7388.

Appendix B

Level-A HDF Header

Produced via the HDF utility hdp with command line:hdp -dumpsds -h PAL MAR 01-10 1988.HDF

File name: PAL_MAR_01-10_1988.HDF

Dimension Variable Name = fakeDim0

Index = 0

Scale Type= number-type not set

Ref. = 29

Rank = 1

Number of attributes = 0

Dim0: Name=fakeDim0

Size = 2168


Index = 1


Ref. = 30

Rank = 1


Dim0: Name=fakeDim1

Size = 5004

Variable Name = Data-Set-2

Index = 2

Type= 8-bit unsigned integer

Ref. = 2

Rank = 2


Dim0: Name=fakeDim0

Size = 2168

Scale Type = number-type not set


Dim1: Name=fakeDim1

33

34 APPENDIX B. LEVEL-A HDF HEADER

Size = 5004



Attr0: Name = coordsys

Type = 8-bit signed char

Count= 30

Value = Interrrupted Goode Homolosine

Attr1: Name = valid_max

Type = 8-bit unsigned integer

Count= 1

Value = 253

Attr2: Name = valid_min


Count= 1

Value = 3

Attr3: Name = scale_factor

Type = 64-bit floating point

Count= 1

Value = 0.008000

Attr4: Name = scale_factor_err


Count= 1

Value = -9.000000

Attr5: Name = add_offset


Count= 1

Value = 128.000000

Attr6: Name = add_offset_err


Count= 1

Value = -9.000000

Attr7: Name = calibrated_nt

Type = 32-bit signed integer

Count= 1

Value = 21

Attr8: Name = remarks-1


Count= 194

Value = NDVI (Normalized Difference Vegetation In

dex) is a difference \012ratio of the two

visible channels from the AVHRR. Values

range\012from -1 to +1, though all negat

ive values represent NO vegetation.\012

Attr9: Name = long_name


Count= 4

Value = NDVI

Attr10: Name = units

35


Count= 3

Value = n/a

Attr11: Name = format


Count= 1

Value =


Index = 3


Ref. = 31

Rank = 1


Dim0: Name=fakeDim2

Size = 2168


Index = 4


Ref. = 32

Rank = 1


Dim0: Name=fakeDim3

Size = 5004


Index = 5


Ref. = 4

Rank = 2


Dim0: Name=fakeDim2

Size = 2168



Dim1: Name=fakeDim3

Size = 5004





Count= 30




Count= 1

Value = 32




Count= 1

Value = 1



Count= 1

Value = 1.000000



Count= 1

Value = -9.000000



Count= 1

Value = 0.000000



Count= 1

Value = -9.000000



Count= 1

Value = 21



Count= 264

Value = \012CLAVR (CLouds from AVHRR) values repr

esent an estimation of the\012cloudiness

of the data. Values from 1-11 are cloudy

pixels, 12-22\012are mixed pixels (parti

al cloud) and values from 23-31 represent

\012clear data. Ref: Stowe, et al. Adv.

Space Res. 11, 511-554, 1990.\012



Count= 11

Value = CLAVR Flags



Count= 3

Value = n/a



Count= 1

Value =


Index = 6


37

Ref. = 33

Rank = 1


Dim0: Name=fakeDim4

Size = 2168


Index = 7


Ref. = 34

Rank = 1


Dim0: Name=fakeDim5

Size = 5004


Index = 8


Ref. = 6

Rank = 2


Dim0: Name=fakeDim4

Size = 2168



Dim1: Name=fakeDim5

Size = 5004





Count= 30




Count= 1

Value = 32



Count= 1

Value = 1



Count= 1

Value = 1.000000



Count= 1

Value = -9.000000




Count= 1

Value = 1.000000



Count= 1

Value = -9.000000



Count= 1

Value = 21



Count= 521

Value = \012The QC layer provides per bit QC flag

ging for automated checks or\012notations

on data quality. Values: 1= ch1/2 non-s

tandard\012(when daily ozone is not avail

able for atmospehric correction, a\012cli

matalogical mean is used), 2= ch3/4/5 non

-standard (generally\012only when fewer t

han 50 calibration target temps were avai

lable for\012a set of scan lines), 4= dat

a were interpolated to fill gaps,\0128=da

ta failed internal range checking (often

in thermal channel\012saturation), and 16

= NOAA QC bits were set. These values ar

e\012additive.\012



Count= 21

Value = Quality Control Flags



Count= 3

Value = n/a



Count= 1

Value =


Index = 9


Ref. = 35

Rank = 1


Dim0: Name=fakeDim6

39

Size = 2168


Index = 10


Ref. = 36

Rank = 1


Dim0: Name=fakeDim7

Size = 5004


Index = 11


Ref. = 8

Rank = 2


Dim0: Name=fakeDim6

Size = 2168



Dim1: Name=fakeDim7

Size = 5004





Count= 30




Count= 1

Value = 20954



Count= 1

Value = 10



Count= 1

Value = 0.000100



Count= 1

Value = -9.000000



Count= 1

Value = 10481.980000




Count= 1

Value = -9.000000



Count= 1

Value = 23



Count= 163

Value = \012Scan angle describes the look angle o

f the satellite. 0 degrees\012represents

nadir with negative values for back-look

ing and positive\012values for forward sc

anning.\012



Count= 10

Value = Scan Angle



Count= 7

Value = radians



Count= 1

Value =


Index = 12


Ref. = 37

Rank = 1


Dim0: Name=fakeDim8

Size = 2168


Index = 13


Ref. = 38

Rank = 1


Dim0: Name=fakeDim9

Size = 5004


41

Index = 14


Ref. = 10

Rank = 2


Dim0: Name=fakeDim8

Size = 2168



Dim1: Name=fakeDim9

Size = 5004





Count= 30




Count= 1

Value = 13973



Count= 1

Value = 10



Count= 1

Value = 0.000100



Count= 1

Value = -9.000000



Count= 1

Value = 10.000000



Count= 1

Value = -9.000000



Count= 1

Value = 23



Count= 198

Value = \012Solar Zenith is the sun’s elevation.


Nominal equator crossing is\01214:30 (lo

cal time) however due to seasonal differe

nces and satellite\012drift, the elevatio

n changes along orbit and from day to day

.\012



Count= 18

Value = Solar Zenith Angle



Count= 7

Value = radians



Count= 1

Value =


Index = 15


Ref. = 39

Rank = 1


Dim0: Name=fakeDim10

Size = 2168


Index = 16


Ref. = 40

Rank = 1



Size = 5004


Index = 17


Ref. = 12

Rank = 2



Size = 2168




Size = 5004

43





Count= 30




Count= 1

Value = 62842



Count= 1

Value = 10



Count= 1

Value = 0.000100



Count= 1

Value = -9.000000



Count= 1

Value = 10.000000



Count= 1

Value = -9.000000



Count= 1

Value = 23



Count= 121

Value = \012Relative Azimuth is the absolute valu

e of the difference between the\012solar

azimuth angle and the sensor azimuth angl

e. \012



Count= 22

Value = Relative Azimuth Angle



Count= 7

Value = radians




Count= 1

Value =


Index = 18


Ref. = 41

Rank = 1



Size = 2168


Index = 19


Ref. = 42

Rank = 1



Size = 5004


Index = 20


Ref. = 14

Rank = 2



Size = 2168




Size = 5004





Count= 30




Count= 1

Value = 50010



Count= 1

Value = 10

45



Count= 1

Value = 0.002000



Count= 1

Value = -9.000000



Count= 1

Value = 10.000000



Count= 1

Value = -9.000000



Count= 1

Value = 23



Count= 238

Value = \012Channel 1 normalized reflectance is t

he reflectance from the .58 -\012.68 micr

on channel of the AVHRR. Data have been

atmospherically \012corrected and normali

zed for solar illumination. Ref: Goward,

et \012al, Rem. Sens Env. 35,257-277, 19

91.\012



Count= 21

Value = Channel 1 Reflectance



Count= 13

Value = % reflectance



Count= 1

Value =


Index = 21


Ref. = 43

Rank = 1




Size = 2168


Index = 22


Ref. = 44

Rank = 1



Size = 5004


Index = 23


Ref. = 16

Rank = 2



Size = 2168




Size = 5004





Count= 30




Count= 1

Value = 50010



Count= 1

Value = 10



Count= 1

Value = 0.002000



Count= 1

Value = -9.000000



Count= 1

47

Value = 10.000000



Count= 1

Value = -9.000000



Count= 1

Value = 23



Count= 240


he reflectance from the .725 -\0121.10 mi

cron channel of the AVHRR. Data have bee

n atmospherically \012corrected and norma

lized for solar illumination. Ref: Gowar

d, et \012al, Rem. Sens Env. 35,257-277,

1991.\012



Count= 21

Value = Channel 2 Reflectance



Count= 13




Count= 1

Value =


Index = 24


Ref. = 45

Rank = 1



Size = 2168


Index = 25


Ref. = 46

Rank = 1



Size = 5004



Index = 26


Ref. = 18

Rank = 2



Size = 2168




Size = 5004





Count= 30




Count= 1

Value = 36010



Count= 1

Value = 10



Count= 1

Value = 0.005000



Count= 1

Value = -9.000000



Count= 1

Value = -31990.000000



Count= 1

Value = -9.000000



Count= 1

Value = 23



49

Count= 156

Value = \012Channel 3 is the calculated brightnes

s temperature from the 3.55 to\0123.93 mi

cron channel of the AVHRR. Ref: Polar O

rbiter Users Guide,\012NOAA,NESDIS, 1991.

\012



Count= 24

Value = Channel 3 BB Temperature



Count= 9

Value = degrees K



Count= 1

Value =


Index = 27


Ref. = 47

Rank = 1



Size = 2168


Index = 28


Ref. = 48

Rank = 1



Size = 5004


Index = 29


Ref. = 20

Rank = 2



Size = 2168




Size = 5004






Count= 30




Count= 1

Value = 36010



Count= 1

Value = 10



Count= 1

Value = 0.005000



Count= 1

Value = -9.000000



Count= 1

Value = -31990.000000



Count= 1

Value = -9.000000



Count= 1

Value = 23



Count= 461



cron channel of the AVHRR. Counts have b

een fully\012calibrated and corrections

for the non-linear instrument response\012

have been applied. Ref: Polar Orbiter Use

rs Guide,NOAA,NESDIS, 1991.\012Ref: Non-L

inearity Corrections for the Thermal Infr

ared Channels of \012the AVHRR: Assessmen

t and Recommendations,Report of the NOAA/

NASA\012AVHRR Pathfinder Calibration Work

51

ing Group,Chair C.R.Rao,1992.\012



Count= 24




Count= 9

Value = degrees K



Count= 1

Value =


Index = 30


Ref. = 49

Rank = 1



Size = 2168


Index = 31


Ref. = 50

Rank = 1



Size = 5004


Index = 32


Ref. = 22

Rank = 2



Size = 2168




Size = 5004





Count= 30





Count= 1

Value = 36010



Count= 1

Value = 10



Count= 1

Value = 0.005000



Count= 1

Value = -9.000000



Count= 1

Value = -31990.000000



Count= 1

Value = -9.000000



Count= 1

Value = 23



Count= 461















Count= 24



53


Count= 9

Value = degrees K



Count= 1

Value =


Index = 33


Ref. = 51

Rank = 1



Size = 2168


Index = 34


Ref. = 52

Rank = 1



Size = 5004


Index = 35


Ref. = 24

Rank = 2



Size = 2168




Size = 5004





Count= 30




Count= 1

Value = 36633




Count= 1

Value = 10



Count= 1

Value = 0.010000



Count= 1

Value = -9.000000



Count= 1

Value = 10.000000



Count= 1

Value = -9.000000



Count= 1

Value = 23



Count= 192

Value = \012Day of year contains the day and hour

(hour being found in the\012decimal plac

e) at which each pixel was observed. The

hour of\012observation can be used to re

ference back to the orbit of origin.\012



Count= 11

Value = Day of Year



Count= 3

Value = n/a



Count= 1

Value =

Appendix C

Level-B HDF Header

Produced via the HDF utility hdp with command line:hdp -dumpsds -h 198803 01-10.hdf

File name: 198803_01-10.hdf

File attributes:

Attr0: Name = description


Count= 1786

Value = This data file was created from HDF files

originally distributed by the GSFC\012DA

AC in 1997 containing the PAL Global Data

Set. Those files contained some\012feat

ures that did not conform with the HDF st

andard for scaled data that made\012them

difficult to read. The processing that h

as been done to create these \012files is

:\012\0121. Renaming the SDSs and rationa

lision of attributes and dimension variab

les\0122. Converting the scaled data (cal

ibrated in HDF parlance) to unscaled,\012

either in integer or floating point fo

rm. \0123. Missing data (eg oceans) have

been set to a consistent value which is\012

recorded in the _FillValue attribute f

or every SDS.\0124. The files have been r

enamed to make them easier to sort into t

ime-order\012\012No change of units or re

mapping has been undertaken and no inform

ation has\012been lost. A consequence of

the elimination of the scaled data has b

een an\012approximate doubling of the fil

e size though this dissappears if the fil

e is \012stored compressed. Given the ra

55

56 APPENDIX C. LEVEL-B HDF HEADER

pid growth in computer storage capacity,

this \012inconvenience is judged to be le

ss than that associated with dealing with

the \012scaled data.\012\012The most rel

evant documentation for this data set can

be found\012at: \012 http://disc1.

gsfc.nasa.gov/FTP_DATA/avhrr/Readme.pal\012

\012though users are cautioned that that

document relates to a later but very\012s

imilar version of the data set which lac

ks several of the ancillary data\012chann

els present in these files (angles, CLAVR

flags etc).\012\012The data are global i

n Goode Interrupted Homolosine Projection

which is a\012pseudo-cylindrical project

ion comprising Sinusoidal Projection for

the \012equatorial regions where |latitud

e| < 44 deg. 44 min. and Mollwiede Projec

tion\012for the polar regions.\012\012The

bounding box is:\012 Latitude: -90.0

- +90.0\012 Longitude: -180.0 - + 180.0\012

Attr1: Name = source_file


Count= 22

Value = PAL_MAR_01-10_1988.HDF

Variable Name = ndvi

Index = 0

Type= 32-bit floating point

Ref. = 7

Rank = 2


Dim0: Name=y

Size = 2168

Scale Type = 32-bit signed integer


Dim1: Name=x

Size = 5004



Attr0: Name = _FillValue


Count= 1

Value = nan

Attr1: Name = units


Count= 3

Value = n/a

57



Count= 30


Attr3: Name = remarks_1


Count= 194






Attr4: Name = input_sds


Count= 10

Value = Data-Set-2

Dimension Variable Name = y

Index = 1

Scale Type= 32-bit signed integer

Ref. = 8

Rank = 1


Dim0: Name=y

Size = 2168

Dimension Variable Name = x

Index = 2

Scale Type= 32-bit signed integer

Ref. = 10

Rank = 1


Dim0: Name=x

Size = 5004

Variable Name = clavr_flags

Index = 3


Ref. = 93

Rank = 2


Dim0: Name=y

Size = 2168



Dim1: Name=x

Size = 5004






Count= 1

Value = 255

Attr1: Name = units


Count= 3

Value = n/a



Count= 30




Count= 264







Space Res. 11, 511-554, 1990.\012



Count= 10

Value = Data-Set-4

Variable Name = quality_control_flags

Index = 4


Ref. = 210

Rank = 2


Dim0: Name=y

Size = 2168



Dim1: Name=x

Size = 5004





Count= 1

Value = 255

Attr1: Name = units


Count= 3

Value = n/a

59



Count= 30




Count= 521














e\012additive.\012



Count= 10

Value = Data-Set-6

Variable Name = scan_angle

Index = 5


Ref. = 362

Rank = 2


Dim0: Name=y

Size = 2168



Dim1: Name=x

Size = 5004





Count= 1

Value = nan

Attr1: Name = units


Count= 7

Value = radians




Count= 30




Count= 163





anning.\012



Count= 10

Value = Data-Set-8

Variable Name = solar_zenith_angle

Index = 6


Ref. = 549

Rank = 2


Dim0: Name=y

Size = 2168



Dim1: Name=x

Size = 5004





Count= 1

Value = nan

Attr1: Name = units


Count= 7

Value = radians



Count= 30




Count= 198




61



.\012



Count= 11

Value = Data-Set-10

Variable Name = relative_azimuth_angle

Index = 7


Ref. = 771

Rank = 2


Dim0: Name=y

Size = 2168



Dim1: Name=x

Size = 5004





Count= 1

Value = nan

Attr1: Name = units


Count= 7

Value = radians



Count= 30




Count= 121




e. \012



Count= 11

Value = Data-Set-12

Variable Name = channel_1_reflectance

Index = 8



Ref. = 1028

Rank = 2


Dim0: Name=y

Size = 2168



Dim1: Name=x

Size = 5004





Count= 1

Value = nan

Attr1: Name = units


Count= 13




Count= 30




Count= 238






et \012al, Rem. Sens Env. 35,257-277, 19

91.\012



Count= 11

Value = Data-Set-14


Index = 9


Ref. = 1320

Rank = 2


Dim0: Name=y

Size = 2168



63

Dim1: Name=x

Size = 5004





Count= 1

Value = nan

Attr1: Name = units


Count= 13




Count= 30




Count= 240







1991.\012



Count= 11

Value = Data-Set-16

Variable Name = channel_3_bb_temperature

Index = 10


Ref. = 1647

Rank = 2


Dim0: Name=y

Size = 2168



Dim1: Name=x

Size = 5004





Count= 1


Value = nan

Attr1: Name = units


Count= 9

Value = degrees K



Count= 30




Count= 156





\012



Count= 11

Value = Data-Set-18


Index = 11


Ref. = 2009

Rank = 2


Dim0: Name=y

Size = 2168



Dim1: Name=x

Size = 5004





Count= 1

Value = nan

Attr1: Name = units


Count= 9

Value = degrees K



Count= 30



65


Count= 461















Count= 11

Value = Data-Set-20


Index = 12


Ref. = 2406

Rank = 2


Dim0: Name=y

Size = 2168



Dim1: Name=x

Size = 5004





Count= 1

Value = nan

Attr1: Name = units


Count= 9

Value = degrees K



Count= 30




Count= 461
















Count= 11

Value = Data-Set-22

Variable Name = day_of_year

Index = 13


Ref. = 2838

Rank = 2


Dim0: Name=y

Size = 2168



Dim1: Name=x

Size = 5004





Count= 1

Value = nan

Attr1: Name = units


Count= 3

Value = n/a



Count= 30




Count= 192





67




Count= 11

Value = Data-Set-24

Appendix D

Level-C HDF Header

Produced via the HDF utility hdp with command line:hdp -dumpsds -h 198803 01-10 geo.hdf

File name: 198803_01-10_geo.hdf

File attributes:

Attr0: Name = description


Count= 375

Value = These data are global in Geographic Proje

ction (rectilinear latitude and\012longit

ude). The pixel size is 0.08 deg. in bot

h directions (~8km).\012They have been cr

eated by using nearest neighbour resampli

ng of the data\012in the parent file (see

:source_file and :parent_description glo

bal\012attributes).\012\012The bounding b

ox is:\012 Latitude: -90.0 - +90.0\012

Longitude: -180.0 - + 180.0\012

Attr1: Name = source_file


Count= 16

Value = 198803_01-10.hdf

Attr2: Name = parent_description


Count= 1786

Value = This data file was created from HDF files

originally distributed by the GSFC\012DA

AC in 1997 containing the PAL Global Data

Set. Those files contained some\012feat

ures that did not conform with the HDF st

andard for scaled data that made\012them

difficult to read. The processing that h

as been done to create these \012files is

69

70 APPENDIX D. LEVEL-C HDF HEADER

:\012\0121. Renaming the SDSs and rationa

lision of attributes and dimension variab

les\0122. Converting the scaled data (cal

ibrated in HDF parlance) to unscaled,\012

either in integer or floating point fo

rm. \0123. Missing data (eg oceans) have

been set to a consistent value which is\012

recorded in the _FillValue attribute f

or every SDS.\0124. The files have been r

enamed to make them easier to sort into t

ime-order\012\012No change of units or re

mapping has been undertaken and no inform

ation has\012been lost. A consequence of

the elimination of the scaled data has b

een an\012approximate doubling of the fil

e size though this dissappears if the fil

e is \012stored compressed. Given the ra

pid growth in computer storage capacity,

this \012inconvenience is judged to be le

ss than that associated with dealing with

the \012scaled data.\012\012The most rel

evant documentation for this data set can

be found\012at: \012 http://disc1.

gsfc.nasa.gov/FTP_DATA/avhrr/Readme.pal\012

\012though users are cautioned that that

document relates to a later but very\012s

imilar version of the data set which lac

ks several of the ancillary data\012chann

els present in these files (angles, CLAVR

flags etc).\012\012The data are global i

n Goode Interrupted Homolosine Projection

which is a\012pseudo-cylindrical project

ion comprising Sinusoidal Projection for

the \012equatorial regions where |latitud

e| < 44 deg. 44 min. and Mollwiede Projec

tion\012for the polar regions.\012\012The

bounding box is:\012 Latitude: -90.0

- +90.0\012 Longitude: -180.0 - + 180.0\012

Variable Name = ndvi

Index = 0


Ref. = 11

Rank = 2


Dim0: Name=latitude

Size = 2241

71

Scale Type = 32-bit floating point


Dim1: Name=longitude

Size = 4500





Count= 1

Value = nan

Attr1: Name = units


Count= 3

Value = n/a



Count= 18

Value = Geographic Lat/Lon



Count= 194






Dimension Variable Name = latitude

Index = 1

Scale Type= 32-bit floating point

Ref. = 12

Rank = 1


Dim0: Name=latitude

Size = 2241

Dimension Variable Name = longitude

Index = 2

Scale Type= 32-bit floating point

Ref. = 14

Rank = 1



Size = 4500

Variable Name = clavr_flags

Index = 3


Ref. = 84


Rank = 2


Dim0: Name=latitude

Size = 2241




Size = 4500





Count= 1

Value = nan

Attr1: Name = units


Count= 3

Value = n/a



Count= 18




Count= 264







Space Res. 11, 511-554, 1990.\012

Variable Name = quality_control_flags

Index = 4


Ref. = 177

Rank = 2


Dim0: Name=latitude

Size = 2241




Size = 4500





73

Count= 1

Value = nan

Attr1: Name = units


Count= 3

Value = n/a



Count= 18




Count= 521














e\012additive.\012

Variable Name = scan_angle

Index = 5


Ref. = 294

Rank = 2


Dim0: Name=latitude

Size = 2241




Size = 4500





Count= 1

Value = nan

Attr1: Name = units


Count= 7


Value = radians



Count= 18




Count= 163





anning.\012

Variable Name = solar_zenith_angle

Index = 6


Ref. = 435

Rank = 2


Dim0: Name=latitude

Size = 2241




Size = 4500





Count= 1

Value = nan

Attr1: Name = units


Count= 7

Value = radians



Count= 18




Count= 198






.\012

75

Variable Name = relative_azimuth_angle

Index = 7


Ref. = 600

Rank = 2


Dim0: Name=latitude

Size = 2241




Size = 4500





Count= 1

Value = nan

Attr1: Name = units


Count= 7

Value = radians



Count= 18




Count= 121




e. \012


Index = 8


Ref. = 789

Rank = 2


Dim0: Name=latitude

Size = 2241




Size = 4500






Count= 1

Value = nan

Attr1: Name = units


Count= 13




Count= 18




Count= 238






et \012al, Rem. Sens Env. 35,257-277, 19

91.\012


Index = 9


Ref. = 1002

Rank = 2


Dim0: Name=latitude

Size = 2241




Size = 4500





Count= 1

Value = nan

Attr1: Name = units


Count= 13




Count= 18


77



Count= 240







1991.\012


Index = 10


Ref. = 1239

Rank = 2


Dim0: Name=latitude

Size = 2241




Size = 4500





Count= 1

Value = nan

Attr1: Name = units


Count= 9

Value = degrees K



Count= 18




Count= 156





\012


Index = 11



Ref. = 1500

Rank = 2


Dim0: Name=latitude

Size = 2241




Size = 4500





Count= 1

Value = nan

Attr1: Name = units


Count= 9

Value = degrees K



Count= 18




Count= 461














Index = 12


Ref. = 1785

Rank = 2


Dim0: Name=latitude

Size = 2241



79


Size = 4500





Count= 1

Value = nan

Attr1: Name = units


Count= 9

Value = degrees K



Count= 18




Count= 461













Variable Name = day_of_year

Index = 13


Ref. = 2094

Rank = 2


Dim0: Name=latitude

Size = 2241




Size = 4500





Count= 1


Value = nan

Attr1: Name = units


Count= 3

Value = n/a



Count= 18




Count= 192






Appendix E

File Names and Sizes

The table below lists the names and uncompressed size in bytes of every file

existing for each of the three processing levels for every 10-day interval

from 1981-07-13 through 1994-09-01.

For conciseness the ‘‘.HDF’’ and ‘‘.hdf’’ suffixes are not shown.

level-A (.HDF) bytes level-B (.hdf) bytes level-C (.hdf) bytes

------------------------------------------------------------------------------

PAL_JUL_13-20_1981 227830487 198107_13-20 456085504 198107_13-20_geo 484410045

PAL_JUL_21-31_1981 227830419 198107_21-31 456085504 198107_21-31_geo 484410045

PAL_AUG_01-10_1981 227830364 198108_01-10 456085504 198108_01-10_geo 484410045

PAL_AUG_11-20_1981 227830335 198108_11-20 456085504 198108_11-20_geo 484410045

PAL_AUG_21-31_1981 227830448 198108_21-31 456085504 198108_21-31_geo 484410045

PAL_SEP_01-10_1981 227830432 198109_01-10 456085504 198109_01-10_geo 484410045

PAL_SEP_11-20_1981 227830419 198109_11-20 456085504 198109_11-20_geo 484410045

PAL_SEP_21-30_1981 227830361 198109_21-30 456085504 198109_21-30_geo 484410045

PAL_OCT_01-10_1981 227830405 198110_01-10 456085504 198110_01-10_geo 484410045

PAL_OCT_11-20_1981 227830333 198110_11-20 456085504 198110_11-20_geo 484410045

PAL_OCT_21-31_1981 227830443 198110_21-31 456085504 198110_21-31_geo 484410045

PAL_NOV_01-10_1981 227830388 198111_01-10 456085504 198111_01-10_geo 484410045

PAL_NOV_11-20_1981 227830408 198111_11-20 456085504 198111_11-20_geo 484410045

PAL_NOV_21-30_1981 227830408 198111_21-30 456085504 198111_21-30_geo 484410045

PAL_DEC_01-10_1981 227830405 198112_01-10 456085504 198112_01-10_geo 484410045

PAL_DEC_11-20_1981 227830413 198112_11-20 456085504 198112_11-20_geo 484410045

PAL_DEC_21-31_1981 227830485 198112_21-31 456085504 198112_21-31_geo 484410045

PAL_JAN_01-10_1982 227830327 198201_01-10 456085504 198201_01-10_geo 484410045

PAL_JAN_11-20_1982 227830356 198201_11-20 456085504 198201_11-20_geo 484410045

PAL_JAN_21-31_1982 227830437 198201_21-31 456085504 198201_21-31_geo 484410045

PAL_FEB_01-10_1982 227830472 198202_01-10 456085504 198202_01-10_geo 484410045

PAL_FEB_11-20_1982 227830465 198202_11-20 456085504 198202_11-20_geo 484410045

PAL_FEB_21-28_1982 227830455 198202_21-28 456085504 198202_21-28_geo 484410045

PAL_MAR_01-10_1982 227830371 198203_01-10 456085504 198203_01-10_geo 484410045

PAL_MAR_11-20_1982 227830399 198203_11-20 456085504 198203_11-20_geo 484410045

PAL_MAR_21-31_1982 227830503 198203_21-31 456085504 198203_21-31_geo 484410045

81

82 APPENDIX E. FILE NAMES AND SIZES

PAL_APR_01-10_1982 227830469 198204_01-10 456085504 198204_01-10_geo 484410045

PAL_APR_11-20_1982 227830459 198204_11-20 456085504 198204_11-20_geo 484410045

PAL_APR_21-30_1982 227830415 198204_21-30 456085504 198204_21-30_geo 484410045

PAL_MAY_01-10_1982 227830548 198205_01-10 456085504 198205_01-10_geo 484410045

PAL_MAY_11-20_1982 227830455 198205_11-20 456085504 198205_11-20_geo 484410045

PAL_MAY_21-31_1982 227830441 198205_21-31 456085504 198205_21-31_geo 484410045

PAL_JUN_01-10_1982 227830404 198206_01-10 456085504 198206_01-10_geo 484410045

PAL_JUN_11-20_1982 227830411 198206_11-20 456085504 198206_11-20_geo 484410045

PAL_JUN_21-30_1982 227830421 198206_21-30 456085504 198206_21-30_geo 484410045

PAL_JUL_01-10_1982 227830373 198207_01-10 456085504 198207_01-10_geo 484410045

PAL_JUL_11-20_1982 227830399 198207_11-20 456085504 198207_11-20_geo 484410045

PAL_JUL_21-31_1982 227830458 198207_21-31 456085504 198207_21-31_geo 484410045

PAL_AUG_01-10_1982 227830437 198208_01-10 456085504 198208_01-10_geo 484410045

PAL_AUG_11-20_1982 227830484 198208_11-20 456085504 198208_11-20_geo 484410045

PAL_AUG_21-31_1982 227830381 198208_21-31 456085504 198208_21-31_geo 484410045

PAL_SEP_01-10_1982 227830468 198209_01-10 456085504 198209_01-10_geo 484410045

PAL_SEP_11-20_1982 227830417 198209_11-20 456085504 198209_11-20_geo 484410045

PAL_SEP_21-30_1982 227830491 198209_21-30 456085504 198209_21-30_geo 484410045

PAL_OCT_01-10_1982 227830416 198210_01-10 456085504 198210_01-10_geo 484410045

PAL_OCT_11-20_1982 227830438 198210_11-20 456085504 198210_11-20_geo 484410045

PAL_OCT_21-31_1982 227830393 198210_21-31 456085504 198210_21-31_geo 484410045

PAL_NOV_01-10_1982 227830382 198211_01-10 456085504 198211_01-10_geo 484410045

PAL_NOV_11-20_1982 227830448 198211_11-20 456085504 198211_11-20_geo 484410045

PAL_NOV_21-30_1982 227830419 198211_21-30 456085504 198211_21-30_geo 484410045

PAL_DEC_01-10_1982 227830357 198212_01-10 456085504 198212_01-10_geo 484410045

PAL_DEC_11-20_1982 227830410 198212_11-20 456085504 198212_11-20_geo 484410045

PAL_DEC_21-27_1982 227830410 198212_21-27 456085504 198212_21-27_geo 484410045

PAL_JAN_01-08_1983 227830405 198301_01-08 456085504 198301_01-08_geo 484410045

PAL_JAN_14-20_1983 227830436 198301_14-20 456085504 198301_14-20_geo 484410045

PAL_JAN_21-31_1983 227830371 198301_21-31 456085504 198301_21-31_geo 484410045

PAL_FEB_01-10_1983 227830404 198302_01-10 456085504 198302_01-10_geo 484410045

PAL_FEB_11-20_1983 227830449 198302_11-20 456085504 198302_11-20_geo 484410045

PAL_FEB_21-28_1983 227830404 198302_21-28 456085504 198302_21-28_geo 484410045

PAL_MAR_01-10_1983 227830386 198303_01-10 456085504 198303_01-10_geo 484410045

PAL_MAR_11-20_1983 227830423 198303_11-20 456085504 198303_11-20_geo 484410045

PAL_MAR_21-31_1983 227830495 198303_21-31 456085504 198303_21-31_geo 484410045

PAL_APR_01-10_1983 227830445 198304_01-10 456085504 198304_01-10_geo 484410045

PAL_APR_11-20_1983 227830464 198304_11-20 456085504 198304_11-20_geo 484410045

PAL_APR_21-30_1983 227830366 198304_21-30 456085504 198304_21-30_geo 484410045

PAL_MAY_01-10_1983 227830369 198305_01-10 456085504 198305_01-10_geo 484410045

PAL_MAY_11-20_1983 227830478 198305_11-20 456085504 198305_11-20_geo 484410045

PAL_MAY_21-31_1983 227830407 198305_21-31 456085504 198305_21-31_geo 484410045

PAL_JUN_01-10_1983 227830468 198306_01-10 456085504 198306_01-10_geo 484410045

PAL_JUN_11-20_1983 227830475 198306_11-20 456085504 198306_11-20_geo 484410045

PAL_JUN_21-30_1983 227830376 198306_21-30 456085504 198306_21-30_geo 484410045

PAL_JUL_01-10_1983 227830448 198307_01-10 456085504 198307_01-10_geo 484410045

PAL_JUL_11-20_1983 227830459 198307_11-20 456085504 198307_11-20_geo 484410045

PAL_JUL_21-31_1983 227830422 198307_21-31 456085504 198307_21-31_geo 484410045

PAL_AUG_01-10_1983 227830390 198308_01-10 456085504 198308_01-10_geo 484410045

83

PAL_AUG_11-20_1983 227830403 198308_11-20 456085504 198308_11-20_geo 484410045

PAL_AUG_21-31_1983 227830430 198308_21-31 456085504 198308_21-31_geo 484410045

PAL_SEP_01-10_1983 227830468 198309_01-10 456085504 198309_01-10_geo 484410045

PAL_SEP_11-20_1983 227830458 198309_11-20 456085504 198309_11-20_geo 484410045

PAL_SEP_21-30_1983 227830637 198309_21-30 456085504 198309_21-30_geo 484410045

PAL_OCT_01-10_1983 227830369 198310_01-10 456085504 198310_01-10_geo 484410045

PAL_OCT_11-20_1983 227830364 198310_11-20 456085504 198310_11-20_geo 484410045

PAL_OCT_21-31_1983 227830477 198310_21-31 456085504 198310_21-31_geo 484410045

PAL_NOV_01-10_1983 227830458 198311_01-10 456085504 198311_01-10_geo 484410045

PAL_NOV_11-20_1983 227830394 198311_11-20 456085504 198311_11-20_geo 484410045

PAL_NOV_21-30_1983 227830380 198311_21-30 456085504 198311_21-30_geo 484410045

PAL_DEC_01-10_1983 227830399 198312_01-10 456085504 198312_01-10_geo 484410045

PAL_DEC_11-20_1983 227830333 198312_11-20 456085504 198312_11-20_geo 484410045

PAL_DEC_21-31_1983 227830420 198312_21-31 456085504 198312_21-31_geo 484410045

PAL_JAN_01-10_1984 227830337 198401_01-10 456085504 198401_01-10_geo 484410045

PAL_JAN_11-20_1984 227830436 198401_11-20 456085504 198401_11-20_geo 484410045

PAL_JAN_21-31_1984 227830373 198401_21-31 456085504 198401_21-31_geo 484410045

PAL_FEB_01-10_1984 227830389 198402_01-10 456085504 198402_01-10_geo 484410045

PAL_FEB_11-20_1984 227830444 198402_11-20 456085504 198402_11-20_geo 484410045

PAL_FEB_21-29_1984 227830487 198402_21-29 456085504 198402_21-29_geo 484410045

PAL_MAR_01-10_1984 227830366 198403_01-10 456085504 198403_01-10_geo 484410045

PAL_MAR_11-20_1984 227830451 198403_11-20 456085504 198403_11-20_geo 484410045

PAL_MAR_21-31_1984 227830452 198403_21-31 456085504 198403_21-31_geo 484410045

PAL_APR_01-09_1984 227830469 198404_01-09 456085504 198404_01-09_geo 484410045

PAL_APR_11-20_1984 227830561 198404_11-20 456085504 198404_11-20_geo 484410045

PAL_APR_21-30_1984 227830432 198404_21-30 456085504 198404_21-30_geo 484410045

PAL_MAY_01-10_1984 227830381 198405_01-10 456085504 198405_01-10_geo 484410045

PAL_MAY_11-20_1984 227830448 198405_11-20 456085504 198405_11-20_geo 484410045

PAL_MAY_21-31_1984 227830411 198405_21-31 456085504 198405_21-31_geo 484410045

PAL_JUN_01-10_1984 227830458 198406_01-10 456085504 198406_01-10_geo 484410045

PAL_JUN_11-20_1984 227830468 198406_11-20 456085504 198406_11-20_geo 484410045

PAL_JUN_21-30_1984 227830399 198406_21-30 456085504 198406_21-30_geo 484410045

PAL_JUL_01-10_1984 227830457 198407_01-10 456085504 198407_01-10_geo 484410045

PAL_JUL_11-20_1984 227830443 198407_11-20 456085504 198407_11-20_geo 484410045

PAL_JUL_21-31_1984 227830349 198407_21-31 456085504 198407_21-31_geo 484410045

PAL_AUG_01-10_1984 227830425 198408_01-10 456085504 198408_01-10_geo 484410045

PAL_AUG_11-20_1984 227830464 198408_11-20 456085504 198408_11-20_geo 484410045

PAL_AUG_21-31_1984 227830399 198408_21-31 456085504 198408_21-31_geo 484410045

PAL_SEP_01-10_1984 227830431 198409_01-10 456085504 198409_01-10_geo 484410045

PAL_SEP_11-20_1984 227830366 198409_11-20 456085504 198409_11-20_geo 484410045

PAL_SEP_21-30_1984 227830417 198409_21-30 456085504 198409_21-30_geo 484410045

PAL_OCT_01-10_1984 227830449 198410_01-10 456085504 198410_01-10_geo 484410045

PAL_OCT_11-20_1984 227830421 198410_11-20 456085504 198410_11-20_geo 484410045

PAL_OCT_21-31_1984 227830438 198410_21-31 456085504 198410_21-31_geo 484410045

PAL_NOV_01-10_1984 227830374 198411_01-10 456085504 198411_01-10_geo 484410045

PAL_NOV_11-20_1984 227830389 198411_11-20 456085504 198411_11-20_geo 484410045

PAL_NOV_21-30_1984 227830370 198411_21-30 456085504 198411_21-30_geo 484410045

PAL_DEC_01-10_1984 227830547 198412_01-10 456085504 198412_01-10_geo 484410045

PAL_DEC_11-20_1984 227830333 198412_11-20 456085504 198412_11-20_geo 484410045


PAL_DEC_21-31_1984 227830376 198412_21-31 456085504 198412_21-31_geo 484410045

198501_01 217114996 198501_01_geo 201815807

198501_11 217114996 198501_11_geo 201815807

198501_21 217114996 198501_21_geo 201815807

198502_01 217114996 198502_01_geo 201815807

PAL_FEB_12-20_1985 227830415 198502_12-20 456085504 198502_12-20_geo 484410045

PAL_FEB_21-28_1985 227830511 198502_21-28 456085504 198502_21-28_geo 484410045

PAL_MAR_01-10_1985 227830453 198503_01-10 456085504 198503_01-10_geo 484410045

PAL_MAR_12-20_1985 227830489 198503_12-20 456085504 198503_12-20_geo 484410045

PAL_MAR_21-31_1985 227830425 198503_21-31 456085504 198503_21-31_geo 484410045

PAL_APR_01-10_1985 227830460 198504_01-10 456085504 198504_01-10_geo 484410045

PAL_APR_11-20_1985 227830441 198504_11-20 456085504 198504_11-20_geo 484410045

PAL_APR_21-30_1985 227830460 198504_21-30 456085504 198504_21-30_geo 484410045

PAL_MAY_01-10_1985 227830449 198505_01-10 456085504 198505_01-10_geo 484410045

PAL_MAY_11-20_1985 227830411 198505_11-20 456085504 198505_11-20_geo 484410045

PAL_MAY_21-31_1985 227830434 198505_21-31 456085504 198505_21-31_geo 484410045

198506_01 217114996 198506_01_geo 201815807

198506_11 217114996 198506_11_geo 201815807

198506_21 217114996 198506_21_geo 201815807

PAL_JUL_01-10_1985 227830438 198507_01-10 456085504 198507_01-10_geo 484410045

PAL_JUL_11-20_1985 227830361 198507_11-20 456085504 198507_11-20_geo 484410045

198507_21 217114996 198507_21_geo 201815807

198508_01 217114996 198508_01_geo 201815807

198508_11 217114996 198508_11_geo 201815807

198508_21 217114996 198508_21_geo 201815807

198509_01 217114996 198509_01_geo 201815807

198509_11 217114996 198509_11_geo 201815807

198509_21 217114996 198509_21_geo 201815807

198510_01 217114996 198510_01_geo 201815807

198510_11 217114996 198510_11_geo 201815807

198510_21 217114996 198510_21_geo 201815807

198511_01 217114996 198511_01_geo 201815807

198511_11 217114996 198511_11_geo 201815807

198511_21 217114996 198511_21_geo 201815807

198512_01 217114996 198512_01_geo 201815807

198512_11 217114996 198512_11_geo 201815807

198512_21 217114996 198512_21_geo 201815807

PAL_JAN_01-10_1986 227830329 198601_01-10 456085504 198601_01-10_geo 484410045

PAL_JAN_11-20_1986 227830473 198601_11-20 456085504 198601_11-20_geo 484410045

PAL_JAN_21-31_1986 227830538 198601_21-31 456085504 198601_21-31_geo 484410045

PAL_FEB_01-10_1986 227830407 198602_01-10 456085504 198602_01-10_geo 484410045

PAL_FEB_11-20_1986 227830406 198602_11-20 456085504 198602_11-20_geo 484410045

PAL_FEB_21-28_1986 227830370 198602_21-28 456085504 198602_21-28_geo 484410045

PAL_MAR_01-10_1986 227830359 198603_01-10 456085504 198603_01-10_geo 484410045

PAL_MAR_11-20_1986 227830628 198603_11-20 456085504 198603_11-20_geo 484410045

PAL_MAR_21-31_1986 227830541 198603_21-31 456085504 198603_21-31_geo 484410045

PAL_APR_01-10_1986 227830459 198604_01-10 456085504 198604_01-10_geo 484410045

PAL_APR_11-20_1986 227830456 198604_11-20 456085504 198604_11-20_geo 484410045

PAL_APR_21-30_1986 227830426 198604_21-30 456085504 198604_21-30_geo 484410045

85

PAL_MAY_01-10_1986 227830471 198605_01-10 456085504 198605_01-10_geo 484410045

PAL_MAY_11-20_1986 227830405 198605_11-20 456085504 198605_11-20_geo 484410045

PAL_MAY_21-31_1986 227830543 198605_21-31 456085504 198605_21-31_geo 484410045

PAL_JUN_01-10_1986 227830407 198606_01-10 456085504 198606_01-10_geo 484410045

PAL_JUN_11-20_1986 227830427 198606_11-20 456085504 198606_11-20_geo 484410045

PAL_JUN_21-30_1986 227830416 198606_21-30 456085504 198606_21-30_geo 484410045

PAL_JUL_01-10_1986 227830560 198607_01-10 456085504 198607_01-10_geo 484410045

PAL_JUL_11-20_1986 227830577 198607_11-20 456085504 198607_11-20_geo 484410045

PAL_JUL_21-31_1986 227830572 198607_21-31 456085504 198607_21-31_geo 484410045

PAL_AUG_01-10_1986 227830510 198608_01-10 456085504 198608_01-10_geo 484410045

PAL_AUG_11-20_1986 227830518 198608_11-20 456085504 198608_11-20_geo 484410045

PAL_AUG_21-31_1986 227830489 198608_21-31 456085504 198608_21-31_geo 484410045

PAL_SEP_01-10_1986 227830441 198609_01-10 456085504 198609_01-10_geo 484410045

PAL_SEP_11-20_1986 227830494 198609_11-20 456085504 198609_11-20_geo 484410045

PAL_SEP_21-30_1986 227830500 198609_21-30 456085504 198609_21-30_geo 484410045

PAL_OCT_01-10_1986 227830579 198610_01-10 456085504 198610_01-10_geo 484410045

PAL_OCT_11-20_1986 227830526 198610_11-20 456085504 198610_11-20_geo 484410045

PAL_OCT_21-31_1986 227830494 198610_21-31 456085504 198610_21-31_geo 484410045

PAL_NOV_01-10_1986 227830536 198611_01-10 456085504 198611_01-10_geo 484410045

PAL_NOV_11-20_1986 227830448 198611_11-20 456085504 198611_11-20_geo 484410045

PAL_NOV_21-30_1986 227830533 198611_21-30 456085504 198611_21-30_geo 484410045

PAL_DEC_01-10_1986 227830377 198612_01-10 456085504 198612_01-10_geo 484410045

PAL_DEC_11-20_1986 227830368 198612_11-20 456085504 198612_11-20_geo 484410045

PAL_DEC_21-31_1986 227830400 198612_21-31 456085504 198612_21-31_geo 484410045

PAL_JAN_01-10_1987 227830362 198701_01-10 456085504 198701_01-10_geo 484410045

PAL_JAN_11-20_1987 227830374 198701_11-20 456085504 198701_11-20_geo 484410045

PAL_JAN_21-31_1987 227830443 198701_21-31 456085504 198701_21-31_geo 484410045

PAL_FEB_01-10_1987 227830403 198702_01-10 456085504 198702_01-10_geo 484410045

PAL_FEB_11-20_1987 227830396 198702_11-20 456085504 198702_11-20_geo 484410045

PAL_FEB_21-28_1987 227830434 198702_21-28 456085504 198702_21-28_geo 484410045

PAL_MAR_01-10_1987 227830418 198703_01-10 456085504 198703_01-10_geo 484410045

PAL_MAR_11-20_1987 227830473 198703_11-20 456085504 198703_11-20_geo 484410045

PAL_MAR_21-31_1987 227830465 198703_21-31 456085504 198703_21-31_geo 484410045

PAL_APR_01-10_1987 227830466 198704_01-10 456085504 198704_01-10_geo 484410045

PAL_APR_11-20_1987 227830426 198704_11-20 456085504 198704_11-20_geo 484410045

PAL_APR_21-30_1987 227830374 198704_21-30 456085504 198704_21-30_geo 484410045

PAL_MAY_01-10_1987 227830354 198705_01-10 456085504 198705_01-10_geo 484410045

PAL_MAY_11-20_1987 227830355 198705_11-20 456085504 198705_11-20_geo 484410045

PAL_MAY_21-31_1987 227830378 198705_21-31 456085504 198705_21-31_geo 484410045

PAL_JUN_01-10_1987 227830388 198706_01-10 456085504 198706_01-10_geo 484410045

PAL_JUN_11-20_1987 227830387 198706_11-20 456085504 198706_11-20_geo 484410045

PAL_JUN_21-30_1987 227830394 198706_21-30 456085504 198706_21-30_geo 484410045

PAL_JUL_01-10_1987 227830379 198707_01-10 456085504 198707_01-10_geo 484410045

PAL_JUL_11-20_1987 227830436 198707_11-20 456085504 198707_11-20_geo 484410045

PAL_JUL_21-31_1987 227830454 198707_21-31 456085504 198707_21-31_geo 484410045

PAL_AUG_01-10_1987 227830363 198708_01-10 456085504 198708_01-10_geo 484410045

PAL_AUG_11-20_1987 227830412 198708_11-20 456085504 198708_11-20_geo 484410045

PAL_AUG_21-31_1987 227830383 198708_21-31 456085504 198708_21-31_geo 484410045

PAL_SEP_01-10_1987 227830368 198709_01-10 456085504 198709_01-10_geo 484410045


PAL_SEP_11-20_1987 227830357 198709_11-20 456085504 198709_11-20_geo 484410045

PAL_SEP_21-30_1987 227830430 198709_21-30 456085504 198709_21-30_geo 484410045

PAL_OCT_01-10_1987 227830415 198710_01-10 456085504 198710_01-10_geo 484410045

PAL_OCT_11-20_1987 227830577 198710_11-20 456085504 198710_11-20_geo 484410045

PAL_OCT_21-31_1987 227830328 198710_21-31 456085504 198710_21-31_geo 484410045

PAL_NOV_01-10_1987 227830344 198711_01-10 456085504 198711_01-10_geo 484410045

PAL_NOV_11-20_1987 227830338 198711_11-20 456085504 198711_11-20_geo 484410045

PAL_NOV_21-30_1987 227830340 198711_21-30 456085504 198711_21-30_geo 484410045

PAL_DEC_01-10_1987 227830347 198712_01-10 456085504 198712_01-10_geo 484410045

PAL_DEC_11-20_1987 227830433 198712_11-20 456085504 198712_11-20_geo 484410045

PAL_DEC_21-31_1987 227830364 198712_21-31 456085504 198712_21-31_geo 484410045

PAL_JAN_01-10_1988 227830333 198801_01-10 456085504 198801_01-10_geo 484410045

PAL_JAN_11-20_1988 227830334 198801_11-20 456085504 198801_11-20_geo 484410045

PAL_JAN_21-31_1988 227830432 198801_21-31 456085504 198801_21-31_geo 484410045

PAL_FEB_01-10_1988 227830359 198802_01-10 456085504 198802_01-10_geo 484410045

PAL_FEB_11-20_1988 227830347 198802_11-20 456085504 198802_11-20_geo 484410045

PAL_FEB_21-29_1988 227830362 198802_21-29 456085504 198802_21-29_geo 484410045

PAL_MAR_01-10_1988 227830327 198803_01-10 456085504 198803_01-10_geo 484410045

PAL_MAR_11-20_1988 227830329 198803_11-20 456085504 198803_11-20_geo 484410045

PAL_MAR_21-31_1988 227830350 198803_21-31 456085504 198803_21-31_geo 484410045

PAL_APR_01-10_1988 227830333 198804_01-10 456085504 198804_01-10_geo 484410045

PAL_APR_11-20_1988 227830333 198804_11-20 456085504 198804_11-20_geo 484410045

PAL_APR_21-30_1988 227830333 198804_21-30 456085504 198804_21-30_geo 484410045

PAL_MAY_01-10_1988 227830367 198805_01-10 456085504 198805_01-10_geo 484410045

PAL_MAY_11-20_1988 227830333 198805_11-20 456085504 198805_11-20_geo 484410045

PAL_MAY_21-31_1988 227830341 198805_21-31 456085504 198805_21-31_geo 484410045

PAL_JUN_01-10_1988 227830333 198806_01-10 456085504 198806_01-10_geo 484410045

PAL_JUN_11-20_1988 227830373 198806_11-20 456085504 198806_11-20_geo 484410045

PAL_JUN_21-30_1988 227830333 198806_21-30 456085504 198806_21-30_geo 484410045

PAL_JUL_01-10_1988 227830333 198807_01-10 456085504 198807_01-10_geo 484410045

PAL_JUL_11-20_1988 227830321 198807_11-20 456085504 198807_11-20_geo 484410045

PAL_JUL_21-31_1988 227830331 198807_21-31 456085504 198807_21-31_geo 484410045

PAL_AUG_01-10_1988 227830410 198808_01-10 456085504 198808_01-10_geo 484410045

PAL_AUG_11-20_1988 227830375 198808_11-20 456085504 198808_11-20_geo 484410045

PAL_AUG_21-31_1988 227830389 198808_21-31 456085504 198808_21-31_geo 484410045

PAL_SEP_01-10_1988 227830351 198809_01-10 456085504 198809_01-10_geo 484410045

PAL_SEP_11-20_1988 227830333 198809_11-20 456085504 198809_11-20_geo 484410045

PAL_SEP_21-30_1988 227830341 198809_21-30 456085504 198809_21-30_geo 484410045

PAL_OCT_01-10_1988 227830327 198810_01-10 456085504 198810_01-10_geo 484410045

PAL_OCT_11-20_1988 227830332 198810_11-20 456085504 198810_11-20_geo 484410045

PAL_OCT_21-31_1988 227830349 198810_21-31 456085504 198810_21-31_geo 484410045

PAL_NOV_01-10_1988 227830591 198811_01-10 456085504 198811_01-10_geo 484410045

PAL_NOV_11-20_1988 227830366 198811_11-20 456085504 198811_11-20_geo 484410045

PAL_NOV_21-30_1988 227830487 198811_21-30 456085504 198811_21-30_geo 484410045

PAL_DEC_01-10_1988 227830405 198812_01-10 456085504 198812_01-10_geo 484410045

PAL_DEC_11-20_1988 227830396 198812_11-20 456085504 198812_11-20_geo 484410045

PAL_DEC_21-31_1988 227830381 198812_21-31 456085504 198812_21-31_geo 484410045

PAL_JAN_01-10_1989 227830452 198901_01-10 456085504 198901_01-10_geo 484410045

PAL_JAN_11-20_1989 227830368 198901_11-20 456085504 198901_11-20_geo 484410045

87

PAL_JAN_21-31_1989 227830444 198901_21-31 456085504 198901_21-31_geo 484410045

PAL_FEB_01-10_1989 227830380 198902_01-10 456085504 198902_01-10_geo 484410045

PAL_FEB_11-20_1989 227830362 198902_11-20 456085504 198902_11-20_geo 484410045

PAL_FEB_21-28_1989 227830427 198902_21-28 456085504 198902_21-28_geo 484410045

PAL_MAR_01-10_1989 227830467 198903_01-10 456085504 198903_01-10_geo 484410045

PAL_MAR_11-20_1989 227830377 198903_11-20 456085504 198903_11-20_geo 484410045

PAL_MAR_21-31_1989 227830451 198903_21-31 456085504 198903_21-31_geo 484410045

PAL_APR_01-10_1989 227830608 198904_01-10 456085504 198904_01-10_geo 484410045

PAL_APR_11-20_1989 227830400 198904_11-20 456085504 198904_11-20_geo 484410045

PAL_APR_21-30_1989 227830454 198904_21-30 456085504 198904_21-30_geo 484410045

PAL_MAY_01-10_1989 227830544 198905_01-10 456085504 198905_01-10_geo 484410045

PAL_MAY_11-20_1989 227830497 198905_11-20 456085504 198905_11-20_geo 484410045

PAL_MAY_21-31_1989 227830396 198905_21-31 456085504 198905_21-31_geo 484410045

PAL_JUN_01-10_1989 227830629 198906_01-10 456085504 198906_01-10_geo 484410045

PAL_JUN_11-20_1989 227830529 198906_11-20 456085504 198906_11-20_geo 484410045

PAL_JUN_21-30_1989 227830412 198906_21-30 456085504 198906_21-30_geo 484410045

PAL_JUL_01-10_1989 227830441 198907_01-10 456085504 198907_01-10_geo 484410045

PAL_JUL_11-20_1989 227830429 198907_11-20 456085504 198907_11-20_geo 484410045

PAL_JUL_21-31_1989 227830401 198907_21-31 456085504 198907_21-31_geo 484410045

PAL_AUG_01-10_1989 227830391 198908_01-10 456085504 198908_01-10_geo 484410045

PAL_AUG_11-20_1989 227830432 198908_11-20 456085504 198908_11-20_geo 484410045

PAL_AUG_21-31_1989 227830453 198908_21-31 456085504 198908_21-31_geo 484410045

PAL_SEP_01-10_1989 227830523 198909_01-10 456085504 198909_01-10_geo 484410045

PAL_SEP_11-20_1989 227830345 198909_11-20 456085504 198909_11-20_geo 484410045

PAL_SEP_21-30_1989 227830345 198909_21-30 456085504 198909_21-30_geo 484410045

PAL_OCT_01-10_1989 227830569 198910_01-10 456085504 198910_01-10_geo 484410045

PAL_OCT_11-20_1989 227830439 198910_11-20 456085504 198910_11-20_geo 484410045

PAL_OCT_21-31_1989 227830512 198910_21-31 456085504 198910_21-31_geo 484410045

PAL_NOV_01-10_1989 227830405 198911_01-10 456085504 198911_01-10_geo 484410045

PAL_NOV_11-20_1989 227830431 198911_11-20 456085504 198911_11-20_geo 484410045

PAL_NOV_21-30_1989 227830366 198911_21-30 456085504 198911_21-30_geo 484410045

PAL_DEC_01-10_1989 227830395 198912_01-10 456085504 198912_01-10_geo 484410045

PAL_DEC_11-20_1989 227830453 198912_11-20 456085504 198912_11-20_geo 484410045

PAL_DEC_21-31_1989 227830449 198912_21-31 456085504 198912_21-31_geo 484410045

PAL_JAN_01-10_1990 227830602 199001_01-10 456085504 199001_01-10_geo 484410045

PAL_JAN_11-20_1990 227830534 199001_11-20 456085504 199001_11-20_geo 484410045

PAL_JAN_21-31_1990 227830680 199001_21-31 456085504 199001_21-31_geo 484410045

PAL_FEB_01-10_1990 227830519 199002_01-10 456085504 199002_01-10_geo 484410045

PAL_FEB_11-20_1990 227830624 199002_11-20 456085504 199002_11-20_geo 484410045

PAL_FEB_21-28_1990 227830421 199002_21-28 456085504 199002_21-28_geo 484410045

PAL_MAR_01-10_1990 227830610 199003_01-10 456085504 199003_01-10_geo 484410045

PAL_MAR_11-20_1990 227830442 199003_11-20 456085504 199003_11-20_geo 484410045

PAL_MAR_21-31_1990 227830490 199003_21-31 456085504 199003_21-31_geo 484410045

PAL_APR_01-10_1990 227830348 199004_01-10 456085504 199004_01-10_geo 484410045

PAL_APR_11-20_1990 227830472 199004_11-20 456085504 199004_11-20_geo 484410045

PAL_APR_21-30_1990 227830443 199004_21-30 456085504 199004_21-30_geo 484410045

PAL_MAY_01-10_1990 227830447 199005_01-10 456085504 199005_01-10_geo 484410045

PAL_MAY_11-20_1990 227830473 199005_11-20 456085504 199005_11-20_geo 484410045

PAL_MAY_21-31_1990 227830475 199005_21-31 456085504 199005_21-31_geo 484410045


PAL_JUN_01-10_1990 227830362 199006_01-10 456085504 199006_01-10_geo 484410045

PAL_JUN_11-20_1990 227830334 199006_11-20 456085504 199006_11-20_geo 484410045

PAL_JUN_21-30_1990 227830334 199006_21-30 456085504 199006_21-30_geo 484410045

PAL_JUL_01-10_1990 227830422 199007_01-10 456085504 199007_01-10_geo 484410045

PAL_JUL_11-20_1990 227830616 199007_11-20 456085504 199007_11-20_geo 484410045

PAL_JUL_21-31_1990 227830509 199007_21-31 456085504 199007_21-31_geo 484410045

PAL_AUG_01-10_1990 227830492 199008_01-10 456085504 199008_01-10_geo 484410045

PAL_AUG_11-20_1990 227830504 199008_11-20 456085504 199008_11-20_geo 484410045

PAL_AUG_21-31_1990 227830505 199008_21-31 456085504 199008_21-31_geo 484410045

PAL_SEP_01-10_1990 227830373 199009_01-10 456085504 199009_01-10_geo 484410045

PAL_SEP_11-20_1990 227830330 199009_11-20 456085504 199009_11-20_geo 484410045

PAL_SEP_21-30_1990 227830467 199009_21-30 456085504 199009_21-30_geo 484410045

PAL_OCT_01-10_1990 227830355 199010_01-10 456085504 199010_01-10_geo 484410045

PAL_OCT_11-20_1990 227830334 199010_11-20 456085504 199010_11-20_geo 484410045

PAL_OCT_21-31_1990 227830338 199010_21-31 456085504 199010_21-31_geo 484410045

PAL_NOV_01-10_1990 227830334 199011_01-10 456085504 199011_01-10_geo 484410045

PAL_NOV_11-20_1990 227830506 199011_11-20 456085504 199011_11-20_geo 484410045

PAL_NOV_21-30_1990 227830551 199011_21-30 456085504 199011_21-30_geo 484410045

PAL_DEC_01-10_1990 227830376 199012_01-10 456085504 199012_01-10_geo 484410045

PAL_DEC_11-20_1990 227830459 199012_11-20 456085504 199012_11-20_geo 484410045

PAL_DEC_21-31_1990 227830474 199012_21-31 456085504 199012_21-31_geo 484410045

PAL_JAN_01-10_1991 227830334 199101_01-10 456085504 199101_01-10_geo 484410045

PAL_JAN_11-20_1991 227830480 199101_11-20 456085504 199101_11-20_geo 484410045

PAL_JAN_21-31_1991 227830420 199101_21-31 456085504 199101_21-31_geo 484410045

PAL_FEB_01-10_1991 227830334 199102_01-10 456085504 199102_01-10_geo 484410045

PAL_FEB_11-20_1991 227830485 199102_11-20 456085504 199102_11-20_geo 484410045

PAL_FEB_21-28_1991 227830334 199102_21-28 456085504 199102_21-28_geo 484410045

PAL_MAR_01-10_1991 227830418 199103_01-10 456085504 199103_01-10_geo 484410045

PAL_MAR_11-20_1991 227830358 199103_11-20 456085504 199103_11-20_geo 484410045

PAL_MAR_21-31_1991 227830374 199103_21-31 456085504 199103_21-31_geo 484410045

PAL_APR_01-10_1991 227830338 199104_01-10 456085504 199104_01-10_geo 484410045

PAL_APR_11-20_1991 227830398 199104_11-20 456085504 199104_11-20_geo 484410045

PAL_APR_21-30_1991 227830361 199104_21-30 456085504 199104_21-30_geo 484410045

PAL_MAY_01-10_1991 227830364 199105_01-10 456085504 199105_01-10_geo 484410045

199105_11 217114996 199105_11_geo 201815807

PAL_MAY_21-31_1991 227830504 199105_21-31 456085504 199105_21-31_geo 484410045

PAL_JUN_01-10_1991 227830390 199106_01-10 456085504 199106_01-10_geo 484410045

PAL_JUN_11-20_1991 227830338 199106_11-20 456085504 199106_11-20_geo 484410045

PAL_JUN_21-30_1991 227830402 199106_21-30 456085504 199106_21-30_geo 484410045

PAL_JUL_01-10_1991 227830436 199107_01-10 456085504 199107_01-10_geo 484410045

PAL_JUL_11-20_1991 227830334 199107_11-20 456085504 199107_11-20_geo 484410045

PAL_JUL_21-31_1991 227830352 199107_21-31 456085504 199107_21-31_geo 484410045

PAL_AUG_01-10_1991 227830398 199108_01-10 456085504 199108_01-10_geo 484410045

PAL_AUG_11-20_1991 227830459 199108_11-20 456085504 199108_11-20_geo 484410045

PAL_AUG_21-31_1991 227830418 199108_21-31 456085504 199108_21-31_geo 484410045

PAL_SEP_01-10_1991 227830445 199109_01-10 456085504 199109_01-10_geo 484410045

PAL_SEP_11-20_1991 227830524 199109_11-20 456085504 199109_11-20_geo 484410045

PAL_SEP_21-30_1991 227830334 199109_21-30 456085504 199109_21-30_geo 484410045

PAL_OCT_01-10_1991 227830409 199110_01-10 456085504 199110_01-10_geo 484410045

89

PAL_OCT_11-20_1991 227830443 199110_11-20 456085504 199110_11-20_geo 484410045

PAL_OCT_21-31_1991 227830396 199110_21-31 456085504 199110_21-31_geo 484410045

PAL_NOV_01-10_1991 227830370 199111_01-10 456085504 199111_01-10_geo 484410045

PAL_NOV_11-20_1991 227830407 199111_11-20 456085504 199111_11-20_geo 484410045

PAL_NOV_21-30_1991 227830366 199111_21-30 456085504 199111_21-30_geo 484410045

PAL_DEC_01-10_1991 227830404 199112_01-10 456085504 199112_01-10_geo 484410045

PAL_DEC_11-20_1991 227830436 199112_11-20 456085504 199112_11-20_geo 484410045

PAL_DEC_21-31_1991 227830390 199112_21-31 456085504 199112_21-31_geo 484410045

PAL_JAN_01-10_1992 227830411 199201_01-10 456085504 199201_01-10_geo 484410045

PAL_JAN_11-20_1992 227830317 199201_11-20 456085504 199201_11-20_geo 484410045

PAL_JAN_21-31_1992 227830411 199201_21-31 456085504 199201_21-31_geo 484410045

PAL_FEB_01-10_1992 227830362 199202_01-10 456085504 199202_01-10_geo 484410045

PAL_FEB_11-20_1992 227830362 199202_11-20 456085504 199202_11-20_geo 484410045

PAL_FEB_21-29_1992 227830407 199202_21-29 456085504 199202_21-29_geo 484410045

PAL_MAR_01-10_1992 227830465 199203_01-10 456085504 199203_01-10_geo 484410045

PAL_MAR_11-20_1992 227830454 199203_11-20 456085504 199203_11-20_geo 484410045

PAL_MAR_21-31_1992 227830451 199203_21-31 456085504 199203_21-31_geo 484410045

PAL_APR_01-10_1992 227830472 199204_01-10 456085504 199204_01-10_geo 484410045

PAL_APR_11-20_1992 227830481 199204_11-20 456085504 199204_11-20_geo 484410045

PAL_APR_21-30_1992 227830363 199204_21-30 456085504 199204_21-30_geo 484410045

PAL_MAY_01-10_1992 227830342 199205_01-10 456085504 199205_01-10_geo 484410045

PAL_MAY_11-20_1992 227830456 199205_11-20 456085504 199205_11-20_geo 484410045

PAL_MAY_21-31_1992 227830431 199205_21-31 456085504 199205_21-31_geo 484410045

PAL_JUN_01-10_1992 227830391 199206_01-10 456085504 199206_01-10_geo 484410045

PAL_JUN_11-20_1992 227830419 199206_11-20 456085504 199206_11-20_geo 484410045

PAL_JUN_21-30_1992 227830336 199206_21-30 456085504 199206_21-30_geo 484410045

PAL_JUL_01-10_1992 227830361 199207_01-10 456085504 199207_01-10_geo 484410045

PAL_JUL_11-20_1992 227830352 199207_11-20 456085504 199207_11-20_geo 484410045

PAL_JUL_21-31_1992 227830379 199207_21-31 456085504 199207_21-31_geo 484410045

PAL_AUG_01-10_1992 227830359 199208_01-10 456085504 199208_01-10_geo 484410045

PAL_AUG_11-20_1992 227830334 199208_11-20 456085504 199208_11-20_geo 484410045

PAL_AUG_21-31_1992 227830373 199208_21-31 456085504 199208_21-31_geo 484410045

PAL_SEP_01-10_1992 227830334 199209_01-10 456085504 199209_01-10_geo 484410045

PAL_SEP_11-20_1992 227830347 199209_11-20 456085504 199209_11-20_geo 484410045

PAL_SEP_21-30_1992 227830380 199209_21-30 456085504 199209_21-30_geo 484410045

PAL_OCT_01-10_1992 227830352 199210_01-10 456085504 199210_01-10_geo 484410045

PAL_OCT_11-20_1992 227830365 199210_11-20 456085504 199210_11-20_geo 484410045

PAL_OCT_21-31_1992 227830365 199210_21-31 456085504 199210_21-31_geo 484410045

PAL_NOV_01-10_1992 227830334 199211_01-10 456085504 199211_01-10_geo 484410045

PAL_NOV_11-20_1992 227830334 199211_11-20 456085504 199211_11-20_geo 484410045

PAL_NOV_21-30_1992 227830397 199211_21-30 456085504 199211_21-30_geo 484410045

PAL_DEC_01-10_1992 227830360 199212_01-10 456085504 199212_01-10_geo 484410045

PAL_DEC_11-20_1992 227830349 199212_11-20 456085504 199212_11-20_geo 484410045

PAL_DEC_21-31_1992 227830391 199212_21-31 456085504 199212_21-31_geo 484410045

PAL_JAN_01-10_1993 227830362 199301_01-10 456085504 199301_01-10_geo 484410045

PAL_JAN_11-20_1993 227830384 199301_11-20 456085504 199301_11-20_geo 484410045

PAL_JAN_21-31_1993 227830452 199301_21-31 456085504 199301_21-31_geo 484410045

PAL_FEB_01-10_1993 227830373 199302_01-10 456085504 199302_01-10_geo 484410045

PAL_FEB_11-20_1993 227830462 199302_11-20 456085504 199302_11-20_geo 484410045


PAL_FEB_21-28_1993 227830378 199302_21-28 456085504 199302_21-28_geo 484410045

PAL_MAR_01-10_1993 227830384 199303_01-10 456085504 199303_01-10_geo 484410045

PAL_MAR_11-20_1993 227830368 199303_11-20 456085504 199303_11-20_geo 484410045

PAL_MAR_21-31_1993 227830393 199303_21-31 456085504 199303_21-31_geo 484410045

PAL_APR_01-10_1993 227830397 199304_01-10 456085504 199304_01-10_geo 484410045

PAL_APR_11-20_1993 227830365 199304_11-20 456085504 199304_11-20_geo 484410045

PAL_APR_22-30_1993 227830509 199304_22-30 456085504 199304_22-30_geo 484410045

PAL_MAY_01-10_1993 227830389 199305_01-10 456085504 199305_01-10_geo 484410045

PAL_MAY_11-20_1993 227830383 199305_11-20 456085504 199305_11-20_geo 484410045

PAL_MAY_21-31_1993 227830400 199305_21-31 456085504 199305_21-31_geo 484410045

PAL_JUN_01-10_1993 227830334 199306_01-10 456085504 199306_01-10_geo 484410045

PAL_JUN_11-20_1993 227830333 199306_11-20 456085504 199306_11-20_geo 484410045

PAL_JUN_21-30_1993 227830360 199306_21-30 456085504 199306_21-30_geo 484410045

PAL_JUL_01-10_1993 227830408 199307_01-10 456085504 199307_01-10_geo 484410045

PAL_JUL_11-20_1993 227830499 199307_11-20 456085504 199307_11-20_geo 484410045

PAL_JUL_21-31_1993 227830358 199307_21-31 456085504 199307_21-31_geo 484410045

PAL_AUG_01-10_1993 227830363 199308_01-10 456085504 199308_01-10_geo 484410045

PAL_AUG_11-20_1993 227830364 199308_11-20 456085504 199308_11-20_geo 484410045

PAL_AUG_21-31_1993 227830394 199308_21-31 456085504 199308_21-31_geo 484410045

PAL_SEP_01-10_1993 227830334 199309_01-10 456085504 199309_01-10_geo 484410045

PAL_SEP_11-20_1993 227830360 199309_11-20 456085504 199309_11-20_geo 484410045

PAL_SEP_21-30_1993 227830398 199309_21-30 456085504 199309_21-30_geo 484410045

PAL_OCT_01-10_1993 227830337 199310_01-10 456085504 199310_01-10_geo 484410045

PAL_OCT_11-20_1993 227830339 199310_11-20 456085504 199310_11-20_geo 484410045

PAL_OCT_21-31_1993 227830339 199310_21-31 456085504 199310_21-31_geo 484410045

PAL_NOV_01-10_1993 227830359 199311_01-10 456085504 199311_01-10_geo 484410045

PAL_NOV_11-20_1993 227830334 199311_11-20 456085504 199311_11-20_geo 484410045

PAL_NOV_21-30_1993 227830334 199311_21-30 456085504 199311_21-30_geo 484410045

PAL_DEC_01-10_1993 227830324 199312_01-10 456085504 199312_01-10_geo 484410045

PAL_DEC_11-20_1993 227830380 199312_11-20 456085504 199312_11-20_geo 484410045

PAL_DEC_21-31_1993 227830336 199312_21-31 456085504 199312_21-31_geo 484410045

PAL_JAN_01-10_1994 227830375 199401_01-10 456085504 199401_01-10_geo 484410045

PAL_JAN_11-20_1994 227830340 199401_11-20 456085504 199401_11-20_geo 484410045

PAL_JAN_21-31_1994 227830337 199401_21-31 456085504 199401_21-31_geo 484410045

PAL_FEB_01-10_1994 227830451 199402_01-10 456085504 199402_01-10_geo 484410045

PAL_FEB_11-20_1994 227830340 199402_11-20 456085504 199402_11-20_geo 484410045

PAL_FEB_21-28_1994 227830383 199402_21-28 456085504 199402_21-28_geo 484410045

PAL_MAR_01-10_1994 227830350 199403_01-10 456085504 199403_01-10_geo 484410045

PAL_MAR_11-20_1994 227830360 199403_11-20 456085504 199403_11-20_geo 484410045

PAL_MAR_21-31_1994 227830377 199403_21-31 456085504 199403_21-31_geo 484410045

PAL_APR_01-10_1994 227830464 199404_01-10 456085504 199404_01-10_geo 484410045

PAL_APR_11-20_1994 227830384 199404_11-20 456085504 199404_11-20_geo 484410045

PAL_APR_21-30_1994 227830404 199404_21-30 456085504 199404_21-30_geo 484410045

PAL_MAY_01-10_1994 227830339 199405_01-10 456085504 199405_01-10_geo 484410045

PAL_MAY_11-20_1994 227830341 199405_11-20 456085504 199405_11-20_geo 484410045

PAL_MAY_21-31_1994 227830363 199405_21-31 456085504 199405_21-31_geo 484410045

PAL_JUN_01-10_1994 227830334 199406_01-10 456085504 199406_01-10_geo 484410045

PAL_JUN_11-20_1994 227830342 199406_11-20 456085504 199406_11-20_geo 484410045

PAL_JUN_21-30_1994 227830407 199406_21-30 456085504 199406_21-30_geo 484410045

91

PAL_JUL_01-10_1994 227830334 199407_01-10 456085504 199407_01-10_geo 484410045

PAL_JUL_11-20_1994 227830354 199407_11-20 456085504 199407_11-20_geo 484410045

PAL_JUL_21-31_1994 227830372 199407_21-31 456085504 199407_21-31_geo 484410045

PAL_AUG_01-10_1994 227830358 199408_01-10 456085504 199408_01-10_geo 484410045

PAL_AUG_11-20_1994 227830326 199408_11-20 456085504 199408_11-20_geo 484410045

PAL_AUG_21-31_1994 227830414 199408_21-31 456085504 199408_21-31_geo 484410045

PAL_SEP_01-10_1994 227830383 199409_01-10 456085504 199409_01-10_geo 484410045

199501_01 217114948 199501_01_geo 202787497

199501_11 217114948 199501_11_geo 202787497

199501_21 217114948 199501_21_geo 202787497

199502_01 217114948 199502_01_geo 202787497

199502_11 217114948 199502_11_geo 202787497

199502_21 217114948 199502_21_geo 202787497

199503_01 217114948 199503_01_geo 202787497

199503_11 217114948 199503_11_geo 202787497

199503_21 217114948 199503_21_geo 202787497

199504_01 217114948 199504_01_geo 202787497

199504_11 217114948 199504_11_geo 202787497

199504_21 217114948 199504_21_geo 202787497

199505_01 217114948 199505_01_geo 202787497

199505_11 217114948 199505_11_geo 202787497

199505_21 217114948 199505_21_geo 202787497

199506_01 217114948 199506_01_geo 202787497

199506_11 217114948 199506_11_geo 202787497

199506_21 217114948 199506_21_geo 202787497

199507_01 217114948 199507_01_geo 202787497

199507_11 217114948 199507_11_geo 202787497

199507_21 217114948 199507_21_geo 202787497

199508_01 217114948 199508_01_geo 202787497

199508_11 217114948 199508_11_geo 202787497

199508_21 217114948 199508_21_geo 202787497

199509_01 217114948 199509_01_geo 202787497

199509_11 217114948 199509_11_geo 202787497

199509_21 217114948 199509_21_geo 202787497

199510_01 217114948 199510_01_geo 202787497

199510_11 217114948 199510_11_geo 202787497

199510_21 217114948 199510_21_geo 202787497

199511_01 217114948 199511_01_geo 202787497


199511_11 217114948 199511_11_geo 202787497

199511_21 217114948 199511_21_geo 202787497

199512_01 217114948 199512_01_geo 202787497

199512_11 217114948 199512_11_geo 202787497

199512_21 217114948 199512_21_geo 202787497

199601_01 217114948 199601_01_geo 202787497

199601_11 217114948 199601_11_geo 202787497

199601_21 217114948 199601_21_geo 202787497

199602_01 217114948 199602_01_geo 202787497

199602_11 217114948 199602_11_geo 202787497

199602_21 217114948 199602_21_geo 202787497

199603_01 217114948 199603_01_geo 202787497

199603_11 217114948 199603_11_geo 202787497

199603_21 217114948 199603_21_geo 202787497

199604_01 217114948 199604_01_geo 202787497

199604_11 217114948 199604_11_geo 202787497

199604_21 217114948 199604_21_geo 202787497

199605_01 217114948 199605_01_geo 202787497

199605_11 217114948 199605_11_geo 202787497

199605_21 217114948 199605_21_geo 202787497

199606_01 217114948 199606_01_geo 202787497

199606_11 217114948 199606_11_geo 202787497

199606_21 217114948 199606_21_geo 202787497

199607_01 217114948 199607_01_geo 202787497

199607_11 217114948 199607_11_geo 202787497

199607_21 217114948 199607_21_geo 202787497

199608_01 217114948 199608_01_geo 202787497

199608_11 217114948 199608_11_geo 202787497

199608_21 217114948 199608_21_geo 202787497

199609_01 217114948 199609_01_geo 202787497

199609_11 217114948 199609_11_geo 202787497

199609_21 217114948 199609_21_geo 202787497

199610_01 217114948 199610_01_geo 202787497

199610_11 217114948 199610_11_geo 202787497

199610_21 217114948 199610_21_geo 202787497

199611_01 217114948 199611_01_geo 202787497

199611_11 217114948 199611_11_geo 202787497

199611_21 217114948 199611_21_geo 202787497

199612_01 217114948 199612_01_geo 202787497

199612_11 217114948 199612_11_geo 202787497

199612_21 217114948 199612_21_geo 202787497

199701_01 217114948 199701_01_geo 202787497

199701_11 217114948 199701_11_geo 202787497

199701_21 217114948 199701_21_geo 202787497

199702_01 217114948 199702_01_geo 202787497

199702_11 217114948 199702_11_geo 202787497

199702_21 217114948 199702_21_geo 202787497

199703_01 217114948 199703_01_geo 202787497

199703_11 217114948 199703_11_geo 202787497

93

199703_21 217114948 199703_21_geo 202787497

199704_01 217114948 199704_01_geo 202787497

199704_11 217114948 199704_11_geo 202787497

199704_21 217114948 199704_21_geo 202787497

199705_01 217114948 199705_01_geo 202787497

199705_11 217114948 199705_11_geo 202787497

199705_21 217114948 199705_21_geo 202787497

199706_01 217114948 199706_01_geo 202787497

199706_11 217114948 199706_11_geo 202787497

199706_21 217114948 199706_21_geo 202787497

199707_01 217114948 199707_01_geo 202787497

199707_11 217114948 199707_11_geo 202787497

199707_21 217114948 199707_21_geo 202787497

199708_01 217114948 199708_01_geo 202787497

199708_11 217114948 199708_11_geo 202787497

199708_21 217114948 199708_21_geo 202787497

199709_01 217114948 199709_01_geo 202787497

199709_11 217114948 199709_11_geo 202787497

199709_21 217114948 199709_21_geo 202787497

199710_01 217114948 199710_01_geo 202787497

199710_11 217114948 199710_11_geo 202787497

199710_21 217114948 199710_21_geo 202787497

199711_01 217114948 199711_01_geo 202787497

199711_11 217114948 199711_11_geo 202787497

199711_21 217114948 199711_21_geo 202787497

199712_01 217114948 199712_01_geo 202787497

199712_11 217114948 199712_11_geo 202787497

199712_21 217114948 199712_21_geo 202787497

199801_01 217114948 199801_01_geo 202787497

199801_11 217114948 199801_11_geo 202787497

199801_21 217114948 199801_21_geo 202787497

199802_01 217114948 199802_01_geo 202787497

199802_11 217114948 199802_11_geo 202787497

199802_21 217114948 199802_21_geo 202787497

199803_01 217114948 199803_01_geo 202787497

199803_11 217114948 199803_11_geo 202787497

199803_21 217114948 199803_21_geo 202787497

199804_01 217114948 199804_01_geo 202787497

199804_11 217114948 199804_11_geo 202787497

199804_21 217114948 199804_21_geo 202787497

199805_01 217114948 199805_01_geo 202787497

199805_11 217114948 199805_11_geo 202787497

199805_21 217114948 199805_21_geo 202787497

199806_01 217114948 199806_01_geo 202787497

199806_11 217114948 199806_11_geo 202787497

199806_21 217114948 199806_21_geo 202787497

199807_01 217114948 199807_01_geo 202787497

199807_11 217114948 199807_11_geo 202787497

199807_21 217114948 199807_21_geo 202787497


199808_01 217114948 199808_01_geo 202787497

199808_11 217114948 199808_11_geo 202787497

199808_21 217114948 199808_21_geo 202787497

199809_01 217114948 199809_01_geo 202787497

199809_11 217114948 199809_11_geo 202787497

199809_21 217114948 199809_21_geo 202787497

199810_01 217114948 199810_01_geo 202787497

199810_11 217114948 199810_11_geo 202787497

199810_21 217114948 199810_21_geo 202787497

199811_01 217114948 199811_01_geo 202787497

199811_11 217114948 199811_11_geo 202787497

199811_21 217114948 199811_21_geo 202787497

199812_01 217114948 199812_01_geo 202787497

199812_11 217114948 199812_11_geo 202787497

199812_21 217114948 199812_21_geo 202787497

199901_01 217114948 199901_01_geo 202787497

199901_11 217114948 199901_11_geo 202787497

199901_21 217114948 199901_21_geo 202787497

199902_01 217114948 199902_01_geo 202787497

199902_11 217114948 199902_11_geo 202787497

199902_21 217114948 199902_21_geo 202787497

199903_01 217114948 199903_01_geo 202787497

199903_11 217114948 199903_11_geo 202787497

199903_21 217114948 199903_21_geo 202787497

199904_01 217114948 199904_01_geo 202787497

199904_11 217114948 199904_11_geo 202787497

199904_21 217114948 199904_21_geo 202787497

199905_01 217114948 199905_01_geo 202787497

199905_11 217114948 199905_11_geo 202787497

199905_21 217114948 199905_21_geo 202787497

199906_01 217114948 199906_01_geo 202787497

199906_11 217114948 199906_11_geo 202787497

199906_21 217114948 199906_21_geo 202787497

199907_01 217114948 199907_01_geo 202787497

199907_11 217114948 199907_11_geo 202787497

199907_21 217114948 199907_21_geo 202787497

199908_01 217114948 199908_01_geo 202787497

199908_11 217114948 199908_11_geo 202787497

199908_21 217114948 199908_21_geo 202787497

199909_01 217114948 199909_01_geo 202787497

199909_11 217114948 199909_11_geo 202787497

199909_21 217114948 199909_21_geo 202787497

199910_01 217114948 199910_01_geo 202787497

199910_11 217114948 199910_11_geo 202787497

199910_21 217114948 199910_21_geo 202787497

199911_01 217114948 199911_01_geo 202787497

199911_11 217114948 199911_11_geo 202787497

199911_21 217114948 199911_21_geo 202787497

199912_01 217114948 199912_01_geo 202787497

95

199912_11 217114948 199912_11_geo 202787497

199912_21 217114948 199912_21_geo 202787497

200001_01 217114948 200001_01_geo 202787497

200001_11 217114948 200001_11_geo 202787497

200001_21 217114948 200001_21_geo 202787497

200002_01 217114948 200002_01_geo 202787497

200002_11 217114948 200002_11_geo 202787497

200002_21 217114948 200002_21_geo 202787497

200003_01 217114948 200003_01_geo 202787497

200003_11 217114948 200003_11_geo 202787497

200003_21 217114948 200003_21_geo 202787497

200004_01 217114948 200004_01_geo 202787497

200004_11 217114948 200004_11_geo 202787497

200004_21 217114948 200004_21_geo 202787497

200005_01 217114948 200005_01_geo 202787497

200005_11 217114948 200005_11_geo 202787497

200005_21 217114948 200005_21_geo 202787497

200006_01 217114948 200006_01_geo 202787497

200006_11 217114948 200006_11_geo 202787497

200006_21 217114948 200006_21_geo 202787497

200007_01 217114948 200007_01_geo 202787497

200007_11 217114948 200007_11_geo 202787497

200007_21 217114948 200007_21_geo 202787497

200008_01 217114948 200008_01_geo 202787497

200008_11 217114948 200008_11_geo 202787497

200008_21 217114948 200008_21_geo 202787497

200009_01 217114948 200009_01_geo 202787497

200009_11 217114948 200009_11_geo 202787497

200009_21 217114948 200009_21_geo 202787497

200010_01 217114948 200010_01_geo 202787497

200010_11 217114948 200010_11_geo 202787497

200010_21 217114948 200010_21_geo 202787497

200011_01 217114948 200011_01_geo 202787497

200011_11 217114948 200011_11_geo 202787497

200011_21 217114948 200011_21_geo 202787497

200012_01 217114948 200012_01_geo 202787497

200012_11 217114948 200012_11_geo 202787497

200012_21 217114948 200012_21_geo 202787497

200101_01 217114948 200101_01_geo 202787497

200101_11 217114948 200101_11_geo 202787497

200101_21 217114948 200101_21_geo 202787497

200102_01 217114948 200102_01_geo 202787497

200102_11 217114948 200102_11_geo 202787497

200102_21 217114948 200102_21_geo 202787497

200103_01 217114948 200103_01_geo 202787497

200103_11 217114948 200103_11_geo 202787497

200103_21 217114948 200103_21_geo 202787497

200104_01 217114948 200104_01_geo 202787497

200104_11 217114948 200104_11_geo 202787497


200104_21 217114948 200104_21_geo 202787497

200105_01 217114948 200105_01_geo 202787497

200105_11 217114948 200105_11_geo 202787497

200105_21 217114948 200105_21_geo 202787497

200106_01 217114948 200106_01_geo 202787497

200106_11 217114948 200106_11_geo 202787497

200106_21 217114948 200106_21_geo 202787497

200107_01 217114948 200107_01_geo 202787497

200107_11 217114948 200107_11_geo 202787497

200107_21 217114948 200107_21_geo 202787497

200108_01 217114948 200108_01_geo 202787497

200108_11 217114948 200108_11_geo 202787497

200108_21 217114948 200108_21_geo 202787497

200109_01 217114948 200109_01_geo 202787497

200109_11 217114948 200109_11_geo 202787497

200109_21 217114948 200109_21_geo 202787497

Appendix F

Scripts

The four scripts presented here are written in the Tcl/Tk language with theNAP extension package available from http://tcl-nap.sourceforge.net/.The NAP extension provides a succinct HDF file interface, together with fa-cilities for managing missing data, data type conversions, reprojections andcoordinate variables.

The three data conversion scripts are all written to expect the files laidout in the following directory structure:

|

|

|--------level-a/YYYY/<PAL_MMM_DD-DD_YYYY.HDF>

| |

| |*/<avhrrpf.ndvi.1ntfgl.YYMMDD.gz>

|

|--------level-b/YYYY/<YYYYMM_DD(-DD).hdf>

|

|--------level-c/YYYY/<YYYYMM_DD(-DD)_geo.hdf>

where the symbols are explained in the text. See the in-code documentationfor further details.

F.1 convert a b.tcl

# Script to convert original PAL HDF files (from tape media) to

# new CSIRO format.

#

# Assumes input files in a subtree with names

# level-a/YYYY/PAL_MMM_DD-DD_YYYY.HDF

# where YYYY is the 4-digit year

# MMM is the first 3 letters of the month name (capitalised)

# DD-DD are the start and end days in the month of the ten day interval

#

97

98 APPENDIX F. SCRIPTS

# Outputs level-b files with names

# level-b/YYYY/YYYYMM_DD-DD.hdf

# where MM is the two digit month number

# and all other symbols as above.

#----------------------------------------------------------------------------

# procedure to convert single file

#----------------------------------------------------------------------------

proc convert_a_b { file out } {

# Set up coordinate variables, description and source_file attributes

nap x=0..5003

nap y=0..2167

set words \

"This data file was created from HDF files originally distributed by the GSFC

DAAC in 1997 containing the PAL Global Data Set. Those files contained some

features that did not conform with the HDF standard for scaled data that made

them difficult to read. The processing that has been done to create these

files is:

1. Renaming the SDSs and rationalision of attributes and dimension variables

2. Converting the scaled data (calibrated in HDF parlance) to unscaled,

either in integer or floating point form.

3. Missing data (eg oceans) have been set to a consistent value which is

recorded in the _FillValue attribute for every SDS.

4. The files have been renamed to make them easier to sort into time-order

No change of units or remapping has been undertaken and no information has

been lost. A consequence of the elimination of the scaled data has been an

approximate doubling of the file size though this dissappears if the file is

stored compressed. Given the rapid growth in computer storage capacity, this

inconvenience is judged to be less than that associated with dealing with the

scaled data.

The most relevant documentation for this data set can be found

at:

http://disc1.gsfc.nasa.gov/FTP_DATA/avhrr/Readme.pal

though users are cautioned that that document relates to a later but very

similar version of the data set which lacks several of the ancillary data

channels present in these files (angles, CLAVR flags etc).

The data are global in Goode Interrupted Homolosine Projection which is a

pseudo-cylindrical projection comprising Sinusoidal Projection for the

equatorial regions where |latitude| < 44 deg. 44 min. and Mollwiede Projection

F.1. CONVERT A B.TCL 99

for the polar regions.

The bounding box is:

Latitude: -90.0 - +90.0

Longitude: -180.0 - + 180.0

"

nap description=’$words’

nap input_file=’[file tail $file]’

puts "PROCESSING: $file"

puts " CREATING: $out"

$description hdf $out :description

$input_file hdf $out :source_file

set list [nap_get hdf -list $file]

# Make a list of the data SDSs (not fakeDims - which we ignore)

set sdslist {}

foreach v $list {

if [regexp {^[^:]*$} $v] {

if [regexp {^fakeDim} $v] {

# Fake dimension - just remember its name and size

nap dim=[nap_get hdf $file $v]

set dimsize($v) [[nap nels(dim)]]

unset dim

} else {

lappend sdslist $v

}

}

}

# loop over all SDSs in the file

foreach sds $sdslist {

puts -nonewline "$sds"

# Start by reading all the attributes

foreach v $list {

if [regsub (^$sds:) $v "" aname] {

# Note that we don’t like minus signs in attribute

# names so we convert them all to underscores.

regsub {\-} $aname "_" aname

set attr($aname) $aname

nap $aname = [nap_get hdf $file $v]

}

}

# Figure out a name from the ’long_name’. The algorithm is:

# 1. convert to lowercase


# 2. replace space with ’_’

regsub -all { } [string tolower [$long_name]] "_" name

puts " $name"

# Read the variable using RAW mode so that scales and

# offsets are not used. Apply the valid_min and valid_max

# attribute values to determine which data should be set missing.

# Note that all attributes are vectors in NAP by default, hence the

# subscript (0)

nap var = [nap_get hdf $file $sds "" 1]

nap var = ((valid_min(0)<=var) && (var<=valid_max(0))) ? var : _

nap scale = scale_factor(0)

nap offset = add_offset(0)

if { [$scale] != 1.0 || [$offset] != 0.0 } {

# This data needs a conversion applied. If it is only an

# offset, we do that separately to avoid unnecessarily

# promoting from integer to floating point.

if { [$scale] != 1.0 } {

nap var = f32((var-offset)*scale)

} else {

nap var = var-offset

}

}

# Special treatment for particular SDSs. Everything gets a mention

# here to trap any surprise SDSs in odd files

switch -regexp $name {

^ndvi$ {

}

^clavr_flags$ {

nap var = u8(var)

}

^quality_control_flags$ {

nap var = u8(var)

}

.*_angle$ {

}

^channel_(1|2)_reflectance$ {

}

^channel_(3|4|5)_bb_temperature$ {

}

^day_of_year$ {

}

default { puts "Unrecognised SDS: $name" }

}

# Attach coordinate variables and write to the new file

F.1. CONVERT A B.TCL 101

$var set coordinate y x

$var set dimension y x

$var hdf $out $name

# Write the relevant SDS-specific attributes

$units hdf $out $name:units

$coordsys hdf $out $name:coordsys

$remarks_1 hdf $out $name:remarks_1

nap input_sds=’$sds’

$input_sds hdf $out $name:input_sds

# Make sure no attributes survive from one SDS to the next

foreach a [array names attr] {

unset $a

}

unset attr

}

}

#----------------------------------------------------------------------------

# Code to drive level-a -> level-b conversion process.

#----------------------------------------------------------------------------

array set month { JAN 01 FEB 02 MAR 03 APR 04 MAY 05 JUN 06

JUL 07 AUG 08 SEP 09 OCT 10 NOV 11 DEC 12}

# loop over all PAL files in each year - they are gzipped

foreach gzfile [lsort [glob "level-a/????/P*gz"]] {

# Figure out ungzipped name

set file [regsub {\.gz$} $gzfile ""]

# Use regexp match to extract date range, month and year from filename

if { [regexp {PAL_([A-Z][A-Z][A-Z])_(\d\d-\d\d)_(\d\d\d\d).HDF$} $file short_name

mon days year] != 1} {

puts "Unrecognisable filename: $file"

exit 1

}

# Make new name, and make sure output directory exists

set out "$year$month($mon)_$days.hdf"

set outdir [file dirname $file]

regsub {level-a} $outdir level-b outdir

file mkdir $outdir

set out [file join $outdir $out]

# Test to make sure we haven’t already done this file

if [file exists "$out.gz"] {


puts "Skipping $file"

} else {

# We’re doing it:

# 1. Unzip to temporary file (saves rezipping input file)

# 2. Make new level-b

# 3. gzip level-b

# 4. Remove temporary file

exec gunzip -c $gzfile > $file

convert_a_b $file $out

exec gzip $out

file delete $file

}

}

F.2 make b from flat.tcl

# Script to convert original PAL binary files (from internet) to

# new CSIRO format.

#


# level-a/*/avhrrpf.ndvi.1ntfgl.YYMMDD.gz

# where YY is the 2-digit year

# MM is the two digit month number

# DD is the start day in the month of the ten day interval

# Only the path to the NDVI channel file is needed. The other channels

# are found by substituting their names in place of ’ndvi’ (so they’d better

# be in the same place!)

#


# level-b/YYYY/YYYYMM_DD.hdf

# where YYYY is the four digit year number

# and all other symbols as above.

#

# Typically the path to the input files would be something like:

# level-a/YYYY/disc1.sci.gsfc.nasa.gov/avhrr/global_8km/YYYY/MMM/

# where YYYY is the four digit year and MMM is the first three letters

# of the month name (in lower case).

#----------------------------------------------------------------------------

# procedure to convert files from single epoch to single HDF file

#----------------------------------------------------------------------------

proc make_b_from_flat { stem out } {

F.2. MAKE B FROM FLAT.TCL 103

# Set up coordinate variables, description and source_file attributes

nap x=0..5003

nap y=0..2167

set words \

"This data file was created from flat files distributed by the GSFC

DAAC containing the PAL Global Data Set. These data are intended to

complement a gap in our existing HDF PAL data set caused by a partially

unreadable tape for the 1985 data, though due to differences in distributed

channels, this substitute data is not as complete (see below). The processing

that has been done to create these files is:

1. Merging the multiple files (1 per channel) into this single HDF file with

attributes and dimension variables consistent with the HDF files for other

years.

2. Converting the scaled data (stored as byte or short integer) to unscaled,

in floating point form.

3. Missing data (eg oceans) have been set to a consistent value which is

recorded in the _FillValue attribute for every SDS.

4. The files have been renamed to make them easier to sort into time-order

No change of units or remapping has been undertaken and no information has

been lost. A consequence of the elimination of the scaled data has been an

approximate doubling of the file size though this disappears if the file is

stored compressed. Given the rapid growth in computer storage capacity, this

inconvenience is judged to be less than that associated with dealing with the

scaled data.

The relevant documentation for this data set can be found at:

http://disc1.gsfc.nasa.gov/FTP_DATA/avhrr/Readme.pal

Users should note that these data do not include channel-3 brightness

temperatures or any of the ancillary data channels (angles, CLAVR flags etc)

that were present in the originally distributed HDF files (apparently

no longer available).

The data are global in Goode Interrupted Homolosine Projection which is a

pseudo-cylindrical projection comprising Sinusoidal Projection for the

equatorial regions where |latitude| < 44 deg. 44 min. and Mollwiede Projection

for the polar regions.


Latitude: -90.0 - +90.0

Longitude: -180.0 - + 180.0

"


#nap input_file=’[file tail $stem]’


nap coordsys=’Interrupted Goode Homolosine’

# Lists channel data types, calibration factors and offsets, and valid

# ranges. Normally these would come from HDF attributes but here they

# come from the accompanying document Readme.PAL

array set type { ndvi u8 ch1 u16 ch2 u16 ch4 u16 ch5 u16 }

array set scale { ndvi 0.008 ch1 0.002 ch2 0.002 ch4 0.005 ch5 0.005 }

array set offset { ndvi 128 ch1 10 ch2 10 ch4 -31990 ch5 -31990 }

array set valid_min { ndvi 3 ch1 10 ch2 10 ch4 10 ch5 10 }

array set valid_max { ndvi 253 ch1 50010 ch2 50010 ch4 36010 ch5 36010 }

puts "PROCESSING: $stem"



#$input_file hdf $out :source_file

# Loop over each channel and process separately

foreach chan {ndvi ch1 ch2 ch4 ch5} {

puts -nonewline "$chan "

# Figure out the file name

set infile [regsub {.ndvi.} $stem ".$chan."]

# If it’s gzipped, unzip it to temporary

if [regexp {\.gz$} $infile] {

set file [regsub {\.gz} $infile ""]

exec gunzip -c $infile > $file

} else {

set file $infile

}

# Read it

puts "Reading: [file tail $file]"

set f [open $file]

nap "var = reshape([nap_get swap $f $type($chan)], {2168 5004})"

close $f

# And delete temporary if original was gzipped

if [regexp {\.gz$} $infile] {

file delete $file

}

# Read the variable using RAW mode so that scales and

# offsets are not used. Apply the valid_min and valid_max

# attribute values to determine which data should be set missing

# and promote to floating point.

nap var = (($valid_min($chan)<=var)&&(var<=$valid_max($chan))) ? var : _

nap var = f32((var-$offset($chan))*$scale($chan))

F.2. MAKE B FROM FLAT.TCL 105

# Handle each data set separately. There is plenty of work to be done

# since no attributes are available.

switch -regexp $chan {

^ndvi$ {

nap units=’n/a’

set name "ndvi"

set words \

"NDVI (Normalized Difference Vegetation Index) is a difference

ratio of the two visible channels from the AVHRR. Values range

from -1 to +1, though all negative values represent NO vegetation."

}

^ch1$ {

nap units=’% reflectance’

set name "channel_1_reflectance"

set words \

"Channel 1 normalized reflectance is the reflectance from the .58 - .68

micron channel of the AVHRR. Data have been atmospherically corrected

and normalized for solar illumination.

Ref: Goward, et al, Rem. Sens Env. 35,257-277, 1991."

}

^ch2$ {

nap units=’% reflectance’

set name "channel_2_reflectance"

set words \

"Channel 2 normalized reflectance is the reflectance from the .725 - 1.10

micron channel of the AVHRR. Data have been atmospherically corrected

and normalized for solar illumination.

Ref: Goward, et al, Rem. Sens Env. 35,257-277, 1991."

}

^ch4$ {

nap units=’degrees K’

set name "channel_4_bb_temperature"

set words \

"Channel 4 is the calculated brightness temperature from the 10.3 to 11.3

micron channel of the AVHRR. Counts have been fully calibrated and

corrections for the non-linear instrument response have been applied.

Ref: Polar Orbiter Users Guide,NOAA,NESDIS, 1991. Ref: Non-Linearity

Corrections for the Thermal Infrared Channels of the AVHRR: Assessment

and Recommendations,Report of the NOAA/NASA AVHRR Pathfinder Calibration

Working Group,Chair C.R.Rao,1992."

}

^ch5$ {

nap units=’degrees K’

set name "channel_5_bb_temperature"

set words \

"Channel 5 is the calculated brightness temperature from the 11.5 to 12.5

micron channel of the AVHRR. Counts have been fully calibrated and

corrections for the non-linear instrument response have been applied.

Ref: Polar Orbiter Users Guide,NOAA,NESDIS, 1991. Ref: Non-Linearity


Corrections for the Thermal Infrared Channels of the AVHRR: Assessment

and Recommendations,Report of the NOAA/NASA AVHRR Pathfinder Calibration

Working Group,Chair C.R.Rao,1992."

}

default { puts "Unrecognised channel: $chan" }

}

# Attach coordinate variables and write to the new file

$var set coordinate y x

$var set dimension y x

$var hdf $out $name

# Write the relevant SDS-specific attributes

$units hdf $out $name:units

$coordsys hdf $out $name:coordsys

nap remarks_1=’$words’

$remarks_1 hdf $out $name:remarks_1

nap input_sds=’[file tail $file]’

$input_sds hdf $out $name:input_sds

}

}

#----------------------------------------------------------------------------

# Code to drive level-a -> level-b conversion process.

#----------------------------------------------------------------------------

# wild is a wildcarded description of the data path for the NDVI channels.

# It is assumed that the files for the other channels are in the same directory

# as the NDVI data and can be found by substituting their names for ’ndvi’.

# There are two possible wildcard descriptions here, one for 1985 and one for

# 1991.

set wild "level-

a/1985/disc1.sci.gsfc.nasa.gov/avhrr/global_8km/1985/???/avhrrpf.ndvi.1ntfgl.*.gz"

set wild "level-

a/1991/disc1.sci.gsfc.nasa.gov/avhrr/global_8km/1991/may/avhrrpf.ndvi.1ntfgl.*.gz"

# Loop over each file in the wildcard list

foreach gzfile [lsort [glob "$wild"]] {


if { [regexp {gl\.(\d\d)(\d\d)(\d\d)\.gz$} $gzfile junk year mon days] != 1} {

puts "Unrecognisable filename: $gzfile"

exit 1

}

set year [expr $year + 1900]


set out "$year${mon}_$days.hdf"

set outdir "level-b/$year/"

F.3. REPROJECT B-C.TCL 107

regsub {level-a} $outdir level-b outdir

file mkdir $outdir


# perform the conversion and gzip the result

make_b_from_flat $gzfile $out

exec gzip $out

}

F.3 reproject b-c.tcl

# Script to convert new PAL HDF files in Goode Interrupted Homolosine

# projection to geographic lat/lon.

#


# level-b/YYYY/YYYYMM_DD-DD.hdf (created from tape media)

# or level-b/YYYY/YYYYMM_DD.hdf (created from Internet distribution)

# where YYYY is the 4-digit year

# MM is the two digit month number

# DD-DD are the start and end days in the month of the ten day interval

#


# level-c/YYYY/YYYYMM_DD-DD_geo.hdf

# or level-c/YYYY/YYYYMM_DD_geo.hdf

# respectively, all symbols as above.

#

# The inverse grid used to map the x,y -> lon,lat is assumed to be

# stored in reprojection_data.hdf created by script ’make_reprojection.tcl’

#----------------------------------------------------------------------------

# procedure to reproject single file

#----------------------------------------------------------------------------

proc reproject { file out inverseGrid olat olon } {

# Set up description and input_file attributes

set words \

"These data are global in Geographic Projection (rectilinear latitude and

longitude). The pixel size is 0.08 deg. in both directions (~8km).

They have been created by using nearest neighbour resampling of the data

in the parent file (see :source_file and :parent_description global

attributes).



Latitude: -90.0 - +90.0

Longitude: -180.0 - + 180.0

"


nap input_file=’[file tail $file]’

puts "PROCESSING: $file"



$input_file hdf $out :source_file

# Read the description of the input file and write it to the

# output as :parent_description

nap description=[nap_get hdf $file :description]

$description hdf $out :parent_description

nap coordsys = ’Geographic Lat/Lon’

set list [nap_get hdf -list $file]

# Make a list of the data SDSs (not coordinate variables)

set sdslist {}

foreach v $list {

if [regexp {^[^:]*$} $v] {

if [regexp {^(x|y)$} $v] {

# coordinate variable - skip

} else {

lappend sdslist $v

}

}

}

# Process each SDS

foreach sds $sdslist {

puts "$sds"

# Start by reading the data and reprojecting it.

nap var = [nap_get hdf $file $sds]

caps warp warpedNearest_var

# Because the inverse_grid doesn’t go right to the limits of

# Lat and Lon, we resample onto the correct size grid before writing

# the output file. Use nearest neighbour in case of round-off

# issues; technically exact match should do.

nap v=warpedNearest_var(@@olat,@@olon)

$v hdf $out $sds

F.3. REPROJECT B-C.TCL 109

# And now do the atributes we’re interested in

foreach v $list {

if [regsub (^$sds:) $v "" aname] {

# Read the attribute and then massage and write as approp.

nap attr = [nap_get hdf $file $v]

switch -regexp $aname {

^units$ { $attr hdf $out $sds:units }

^coordsys$ { $coordsys hdf $out $sds:coordsys }

^remarks_1$ { $attr hdf $out $sds:remarks_1 }

^input_sds$ { }

^_FillValue$ { }

default { puts "Unrecognised attribute $v" }

}

}

}

}

}

#----------------------------------------------------------------------------

# Code to drive level-b -> level-c conversion process.

#----------------------------------------------------------------------------

# Read inverse grid

nap inverseGrid = [nap_get hdf reprojection_data.hdf inverseGrid]

# Make coordinate variables/dimension names for output latitude/longitude axes

# Very important to do this as integer arithmetic multiplied by a scale factor,

# otherwise we end up with short CVs due to round-off.

nap olat=0.08*(1125..-1125...-1)

$olat set dim latitude

nap olon=0.08*(-2250..2250...1)

$olon set dim longitude

# loop over all gzipped HDF files in the level-b tree

foreach gzfile [lsort [glob "level-b/????/*hdf.gz"]] {

# Figure out ungzipped name

set file [regsub {\.gz$} $gzfile ""]


if { [regexp {(\d\d\d\d)(\d\d)_((\d\d)|(\d\d-\d\d)).hdf$} $file junk year mon days] != 1} {

puts "Unrecognisable filename: $file"

exit 1

}


set out "$year${mon}_${days}_geo.hdf"


set outdir level-c/$year

file mkdir $outdir


# Test to make sure we haven’t already done this file

if [file exists "$out.gz"] {

puts "Skipping $file"

} else {

# We’re doing it:

# 1. Unzip to temporary file (saves rezipping input file)

# 2. Reproject to new level-c

# 3. gzip level-c

# 4. Remove temporary file

exec gunzip -c $gzfile > $file

reproject $file $out $inverseGrid $olat $olon

exec gzip $out

file delete $file

}

}

F.4 make reprojection.tcl

#

# Create an HDF file with latitude and longitude grids and inverseGrid

# (for reprojecting) using the data provided with the "Internet version"

# of the PAL data set.

#

# Read latitude data, apply valid-data limits, and "uncalibration"

# parameters to convert to physical values and promote to floating point.

set f [open ancillary/avhrrpf.lat.1ctfgl]

nap "lat=reshape([nap_get swap $f u16], {2168 5004})"

close $f

nap lat = (10<=lat) && (lat<=18010) ? lat : _

nap lat = f32((lat-9010)*0.01)

# Repat for longitude data

set f [open ancillary/avhrrpf.lon.1ctfgl]

nap "lon=reshape([nap_get swap $f u16], {2168 5004})"

close $f

nap lon = (10<=lon) && (lon<=36010) ? lon : _

nap lon = f32((lon-18010)*0.01)

F.4. MAKE REPROJECTION.TCL 111

# Create and attach trivial, but named, coordinate variables

nap x=0..5003

nap y=0..2167

$lat set coordinate y x

$lat set dimension y x

$lon set coordinate y x

$lon set dimension y x

# Make coordinate variables/dimension names for output latitude/longitude axes

# Very important to do this as integer arithmetic multiplied by a scale factor,

# otherwise we end up with short CVs due to round-off.

nap glat=0.08*(1125..-1125...-1)

$glat set dim latitude

nap glon=0.08*(-2250..2250...1)

$glon set dim longitude

# Use NAP procedure to create two layer inverseGrid which permits mapping

# from lat/lon space to y/x pixel number

nap inverseGrid=invert_grid(lat,glat,lon,glon)

# Save latitude, longitude and inverse grids to HDF file

$lat hdf reprojection_data.hdf latitude_grid

$lon hdf reprojection_data.hdf longitude_grid

$inverseGrid hdf reprojection_data.hdf inverseGrid

Curation of the Pathfinder AVHRR Land Data Set at CSIRO · by the Client’s contractors and...

Documents

Transcript of Curation of the Pathfinder AVHRR Land Data Set at CSIRO · by the Client’s contractors and...