NNDVIM
NDVI(Normalized Difference Vegetation Index) is an index which has been devised to represent the activity and/or
quantity of vegetation from spectral reflectance characteristics. NDVI data to be provided in this system has been calculated by the following formula, which is used most commonly.
NDVI = (NIR - RED) / (NIR + RED)
NIR is the reflectance in the near infrared, and is calculated using the channel 2 of the AVHRR.
RED is the reflectance in the wavelength range in red of visible light, and is calculated using the channel 1 of the AVHRR.
NDVI is as index that is determined by the ratio of the NIR(AVHRR_CH2) and RED(AVHRR_CH1). Because, in vagetation,
this ratio is higher than that of the water and the soil, it will be a large value as the more quantity and as the
higher activity.
The value of NDVI is a real number in the range of -1 to 1. Because the value of -0.5 or less can be determined as a cloud, it has been converted as follows to fit into an integer value of 1 byte as the range is -0.5 to 1 .
INDVI = NDVI x 100 + 50
However, in case of image data output, it is different from the above formula in order for the color representation  to be used as 240-colors. Therefore, it has been converted as follows.
INDVI = NDVI x 160 + 50
In addition, the land areas outside the observation range has been as "INDVI=255", and the sea and the lake etc
has been as "INDVI=0".
Besides, because the wavelength in visible range is used for the processing of NDVI, the processing will be done
only for the path which the elevation angle of the sun under the satellite passing orbit will be 20 degrees
or more. However, because the influence of a considerable amount of clouds are included in the data of the AVHRR,
it is necessary to eliminate the influence of clouds.
Generally, because the NDVI value of cloud is smaller than the NDVI of vegetation, the influence of clouds will be
eliminated by adopting the "MAX" value of NDVI in each pixel, when you will process the composition for plural images.
The data provoded by this system is a monthly composite data which is created by processing the above described composition.