Version 1.0, February 1, 2002
EORC, NASDA / RESTEC

Table of Contents

1. Introduction
   
   
2. Format of Raw Data
   
   
3. Data Division and File
   • Data Division
   • File
   
   
4. Data Processing
   • Data Processing Procedure
   • Radiometric Correction
   • Geometric Correction
   • Composite
   • References
   
   
5. Data Utilization and Redistribution
   • Copyright, exemption from responsibility, and redistribution
   • Credit
   
   

1. Introduction

This global data set processed from NASDA's Ocean Color and Temperature Scanner (OCTS) onboard the Advanced Earth Observing Satellite (ADEOS). The data set was produced in the research project "Development of a precise geometric correction technique and the production of a satellite data set," which is part of the "Global Carbon Cycle and Related Mapping Based on Satellite Imagery (GCMAPS) program" founded by the Ministry of Education, Culture, Sports, Science and Technology of Japan.

To Contents

2. Format of Raw Data

• Image size

  10000 (pixels) x 5000 (lines) x 16 (ch) x 2 (byte integer)

• Channel

  ch.		Content		Unit		Slope*		Intercept*
  01 - 08		Visible and Near Infrared Radiance		mW/cm2/sr/micrometer		0.01		0.0000
  09		Thermal infrared Radiance		mW/cm2/sr		0.000001		0.0000
  10 - 12		Thermal infrared Radiance		mW/cm2/sr		0.0001		0.0000
  13		Solar Zenith Angle		Radian		0.0001		0.0000
  14		Solar Azimuth Angle		Radian		0.0001		0.0000
  15		Satellite Zenith Angle		Radian		0.0001		0.0000
  16		Satellite Azimuth Angle		Radian		0.0001		0.0000

  * : [Physical value] = Slope x [Digital Number] + Intercept

To Contents

3. Data Division and File Name

3.1 Data Division

It is set to 1.6GB per file in the carbon circulation database. Therefore, data was divided on Web to the size (about several MB) which can be downloaded. First, divided for every channel. Furthermore, divided the earth into 16, divided into 2 in the direction of north and south and divided into 8 in the direction of east and west (Figure.1). The data size after division is as follows.

1.6 (GB) / 16 (ch.) / 2 (byte integer) / 8

To Contents

3.2 File Name

The file name of each data after division is as follows.

com_xxdy.aabcc

xx	:	10 or 30	(Composite period,Refer to Table1)
y	:	1 - 20	(The Serial Number of a Composite Image, Refer to Table.1)
aa	:	01 - 16	(Channel Number)
b	:	1 or 2	(North and South division, Refer to Figure.1)
cc	:	01 - 08	(East and West division, Refer to Figure.1)

To Contents

4. Data Processing

4.1 Data Processing Procedure

This data set was generated based on the following procedure.

(1) Radiometric Correction

Calculation of coefficients Conversion from digital number to radiance or brightness temperature Elimination of stripe noise

(2) Geometric Correction

Automatic GCP collection Modification of satellite position and attitude Calculation of geolocation of observed points Resampling image data onto map coordinates

(3) Composite

Separation to four areas Selection of cloud-free data

To Contents

4.2 Radiometric Correction

The band image in OCTS Level 1A data is described as an observed digital number (DN), and OCTS has an Ocean/Land gain mode and Tilt/Non-tilt mode. Therefore, data should be corrected radiometrically. For visible and near infrared bands, The DN was converted to radiance (mW/cm²/sr/micrometer) by using coefficients calculated for each condition of the observation (tilt angle, gain, temperature) line by line. In thermal infrared bands, a black body was observed for each scan. The DN could be converted to radiance (mW/cm²/sr) by using blackbody data. The radiance was converted to brightness temperature (K) using a look-up table included in the level 1A data. A de-striping process was applied for both visible and thermal bands.

To Contents

4.3 Geometric Correction

An OCTS product contains the geolocation data (Latitude/Longitude) of observed points derived by a systematic correction method. A comparison of these data with corresponding locations on the image revealed an error of a few kilometers on the ground. Therefore, accurate geometric correction utilizing ground control points (GCP) was applied to make composite images.

The GCPs are automatically extracted by a image-matching method using image data and coastal line data prepared in advance. The satellite position and attitude are modified based on the principle of photogrammetry. The subsequent procedure is the same as systematic correction.

To Contents

4.4 Composite

Cloud-free composite data were generated from radiometrically and geometrically corrected data. Basically, the period for composite data is 30 days. However, since OCTS was operated in the non-tilt mode for three months from Dec.19, 1996, we set the composite period flexibly. The eight generated images of the 30-day composite data as shown in table below.

The cloud-free composite algorithms are different for four areas: ocean, high latitude; ocean, mid-low latitude; land, high latitude; and land, mid-low latitude. The distinction between ocean and land was made based on the land mask data. Latitude was distinguished by whether solar zenith angle (SZA) exceeded the threshold of 1.35 radians or not. The algorithm for each area is shown below.

• Ocean, high latitude
  Select the pixel with the minimum value in band 5.

• Ocean, mid - low latitude
  Select minimum band 5 within the scenes that have SZA below the threshold.

• Land, high latitude
  Select minimum band 11 from scenes with saturated band 7, and select the minimum SZA scene from scenes with unsaturated band 7. Select the larger band 11 from these two scenes.

• Land, mid - low latitude
  Select the minimum band 5 scene from saturated band 7 scenes, and select the maximum Normalized Difference Vegetation Index (NDVI) from the scenes with unsaturated band 7. Finally select the larger band 11 scene.

To Contents

4.5 References

• NOAA, 1997, NOAA GVI GUIDE, on http://www2.ncdc.noaa.gov/docs/gviug/
• T. Hashimoto, "Precise geometric correction of ADEOS/OCTS imagery," Journal of the Japan Society of Photogrammetry and Remote Sensing, vol.36, no.5, pp.42-51, 1997. (in Japanese)
• T. Hashimoto, "The estimation of motion and attitude of ADEOS satellite utilizing the principle of exterior orientation," Journal of the Japan Society of Photogrammetry and Remote Sensing, vol.37, no.6, pp.4-13, 1998. (in Japanese)
• T. Hashimoto, T. Igarashi, A. Mukaida, R. Higuchi, "Global Image Composite of ADEOS/OCTS GAC Data," Proc. of 21st Asian Conference on Remote Sensing, pp.152-157, 2000.

To Contents

5. Data Utilization and Redistribution

5.1 Copyright, exemption from responsibility, and redistribution

The National Space Development Agency of Japan (NASDA) reserves copyrights of all data downloaded from this homepage. NASDA is not responsible for any trouble caused by using this data set. The redistribution of this data set (in whole or part) is permitted only if this README file is attached.

5.2 Credit

The data downloaded from this homepage is for scientific use only. Anybody wishing to publish any results using the data downloaded from this homepage should clearly specify the credit ((c) National Space Development Agency of Japan or (c) NASDA) beside the image or acknowledge the source of the data in the publication.

Example: The OCTS data utilized in this study is provided by National Space Development Agency of Japan (NASDA) as part of the "Global Carbon Cycle and Related Mapping Based on Satellite Imagery: GCMAPS program" of "Research Project by the Special Coordination Funds for Promoting Science and Technology" funded by the Ministry of Education, Culture, Sports, Science and Technology of Japan (MEXT).

To Contents

Contact Point

Earth Observation Research Center
National Space Development Agency of Japan

Office Tower X-23F, Harumi Island Triton Square,
1-8-10, Harumi, Chuo-ku, Tokyo 104-6023 Japan

Phone: +81-3-6221-9000 (Rep.)
Fax: +81-3-6221-9191
E-mail: helpdesk@eorc.nasda.go.jp
URL: http://www.eorc.nasda.go.jp

To Contents