There are many types of indices that measure vegetation and many are calculated by using satellite data to compare the relative difference between how much energy is absorbed by the land surface versus how much is reflected back into space. Plants absorb visible light to undergo photosynthesis, so when vegetation is lush, nearly all of the visible light is absorbed by the photosynthetic leaves, and much more near-infrared light is reflected back into space. However for deserts and regions with sparse vegetation, the amount of reflected visible and near-infrared light are both relatively high. The Visible and Infrared Imaging and Radiometer Suite (VIIRS) sensor on the Suomi NPP satellite is sensitive to these different types of visible and near-infrared light.
How is it used?
Characterizing the vegetation of the surface has many applications from weather and ecological forecasting to understanding best practices for land use. Pixel by pixel analysis of vegetation changes from week to week give an early warning for the outbreaks of drought, hazardous fire conditions, or even when malaria may break out in Sub-Saharan Africa. Because vegetation greatly affects the runoff, surface temperature, and relative humidity of an area, more complex weather forecasts are beginning to integrate vegetation dynamics into numerical models. Even golf course planners use the data for planning an optimal landscape architecture for greens, fairways, and the outlying roughs.
Why is this data so much better than what was previously available?
For the past 22 years, NOAA has used the AVHRR sensor on its POES polar-orbiting satellites to generate products like vegetation indexes at a resolution of 4 km per pixel. In 1999, NASA launched the first MODIS sensor into space onboard the Terra satellite, improving the data gathering to 500 meters per pixel. However, with both AVHRR and MODIS, the farther away the data is from the center of the orbit swath, the more blurred the imagery gets - in the same way that the eye can focus on one area, but the peripheral vision is not quite as sharp. The VIIRS sensor does not that have problem - the imagery is nearly the same quality (or focus) across the entire swath. So not only is the data 8 times more detailed than with AVHRR, it is also higher and of more consistent quality than both AVHRR and MODIS.
Where are the clouds?
Cloud cover is probably the most dynamic feature of the surface of the planet. However, over the course of a week, there are usually enough areas without cloud cover that are seen by the satellite. Computer programs are used to identify the best cloud-free measurement for every spot on the planet, and those individual measurements are added together to form a single cloud-free mosaic of the planet.
How much data was used?
Computing weekly mosaics from data at such high resolution is no trivial matter. The VIIRS sensor acquires 330 megabytes of data every minute just for the four channels of visible and near-infrared imagery used in the vegetation index. Multiply that over one week and it is an astounding 2 TB of data - more than 40 Blu-ray Disc movies - and that does not even include the other 17 channels of data collected by VIIRS.
What are the sizes of the images?
At 500 meters per pixel, the images that the data generates are equally large. Each weekly colorized vegetation image is around 13 gigabytes in size and 80,000 x 40,000 pixels in dimension. Think of it another way: if you were to print the image it would require a piece of paper 1,111 inches long by 555 inches wide (or 93 x 46.5 feet).
How can I learn more about vegetation monitoring?
NOAA and NASA science teams have collaborated with many partners to develop educational resources about their remote sensing capabilities. Some excellent resources include: