Description
Most remote-sensing-derived vegetation indices rely on light reflected in the red and near-infrared wavelengths as recorded by a passive sensor. This can present challenges of there is significant cloud cover or smoke in the atmosphere. Additionally, vegetation indices from passive sensors tend can be sensitive to soil background, atmospheric effects, and often experience saturation at high amounts/cover of vegetation.
An emerging technique for indexing the amount and cover of vegetation is through the use of active remote sensors or passive sensors recording radiation that can pass through clouds, smoke, and vegetation. One of these new techniques is the Microwave Vegetation Index (MVI). Described by Shi et al. (2008), the MVI uses observations from the Advanced Microwave Scanning Radiometer (AMSR-E) sensor on board the Aqua satellite (this satellite also carries one of the MODIS sensors) to derive a vegetation index. While some microwave sensors are “active” sensors – they emit radiation (in this case microwaves) and then measure a response to the emitted energy – AMSR-E is a passive microwave sensor, recording naturally-occurring microwaves that are reflected from the earth’s surface. Microwaves have the advantage that they can penetrate clouds and smoke and, in general, are much less influenced by atmospheric conditions than are the visible regions of the electromagnetic spectrum.
The MVI developed by Shi et al. (2008) uses microwave readings from two different frequencies to determine the cover of vegetation. Each of these different frequencies has the ability to penetrate vegetation canopy to a different degree, and the relative different in reflectance between these two can be interepreted to give an index of vegetation cover. The formula for doing so is not as straightforward as a simple passive sensor vegetation index (e.g., NDVI), though, because it is based on a theoretical “radiative transfer” model of how different surfaces reflect microwave radiation. Because different frequencies of microwaves penetrate vegetation layers to differing degrees, vegetation indices based on reflected microwave readings can better estimate vegetation conditions than spectral indices like NDVI. Spectral indices are responding only to a thin layer of the vegetation that is exposed to, and therefore reflects, the incident light. Microwaves, on the other hand, can penetrate the canopy and therefore provide information on leafy and woods parts of plants in the overstory and even some of the understory.
Similar Methods
Output
The output of MVI is an image of vegetation index values.
Successful Rangeland Uses
This method is new, and to date the only applications of it have been in research studies.
Application References
None known
Technical References
- Shi, J., T. Jackson, J. Tao, J. Du, R. Bindlish, L. Lu, and K. S. Chen. 2008. Microwave vegetation indices for short vegetation covers from satellite passive microwave sensor AMSR-E. Remote Sensing of the Environment 112:4285-4300.
Limitations
A significant limitation of MVI to date is that the current satellite microwave sensors are of very coarse resolution (10s of kilometers). This limits the application of MVI to global or very large regional applications. Shi et al. (2008) found that the MVI was highly correlated with NDVI calculated from MODIS imagery only for areas with “short vegetation.” Also, given how new microwave vegetation indices are, there have not been many applications or validation studies of this technique. However, this is likely to change as microwave sensors and MVI research matures.
Data Inputs
MVI requires microwave sensor data from at least two frequencies. At this point, this is an experimental method, and the frequencies being used are not standardized.
Software/Hardware Requirements
unknown
Additional Information
- National Snow and Ice Data Center page on AMSR-E – [http://nsidc.org/daac/projects/passivemicro/amsre.html]]
- NASA Marshall Space Flight Center AMSR-E page – http://wwwghcc.msfc.nasa.gov/AMSR/instrument_descrip.html
Who Is Using This Method?
- Organization information and/or project info
- Contact information (i.e., email, phone):
Existing datasets
- Links to (or contact info for) places where data from this method that already exists can be obtained
Implementation help
- Contact information for groups/people who can help implement this method either for free or for a fee.