topic Re: Interpreting NDVI results in Data Management Questions

Interpreting NDVI results

LauraCockrell — Sat, 09 Mar 2013 21:39:31 GMT

I am looking at a 2005 landsat image, and I used the IAW to calculate the NDVI... easy-peasy! But everything I have read about the NDVI's says that the resulting calculations should be in a range from -1.0 to +1.0 with negative number signifying vegetation loss and the positive values signifying vegetation gain. I am not seeing this. When I create an attribute table from the NDVI raster, the values are on a 0-256 scale. How do I create the NDVI values for the raster in the -1.0 to +1.0 range?

Re: Interpreting NDVI results

EricRice — Mon, 11 Mar 2013 15:39:54 GMT

Hi Laura,

On the IAW options dialog there is an NDVI tab. Check the box that says, "Scientific Output". This will give you the values you are looking for.

Best,
Eric

Re: Interpreting NDVI results

LauraCockrell — Tue, 12 Mar 2013 00:39:03 GMT

Eric-
In the upper left-hand corner of the IAW is the Image Analysis Options. When I open that, the first tab is the NDVI tab. All it says in that box is Red Band: 3 and Infrared Band: 4
Is this an option only on certain licenses?
Thank you for your help
~Laura

Re: Interpreting NDVI results

EricRice — Tue, 12 Mar 2013 15:30:14 GMT

Sorry, I assumed you were using 10.1. In 10.0 that option did not exist. In 10.0 you will have to use the raster calculator and the appropriate band math to arrive at a floating point raster between -1.0 - 1.0.

NDVI = ((IR - R)/(IR + R))

Best,
Eric

Re: Interpreting NDVI results

EricRice — Tue, 12 Mar 2013 15:37:57 GMT

But everything I have read about the NDVI's says that the resulting calculations should be in a range from -1.0 to +1.0 with negative number signifying vegetation loss and the positive values signifying vegetation gain.

I suppose if you differenced two NDVI's you can tell if you lost or gained vegetation, but the index itself doesn't tell you this. Below is an excerpt from our documenation.

"This index outputs values between -1.0 and 1.0, mostly representing greenness, where any negative values are mainly generated from clouds, water, and snow, and values near zero are mainly generated from rock and bare soil. Very low values of NDVI (0.1 and below) correspond to barren areas of rock, sand, or snow. Moderate values represent shrub and grassland (0.2 to 0.3), while high values indicate temperate and tropical rainforests (0.6 to 0.8)"

Best,
Eric

Re: Interpreting NDVI results

LauraCockrell — Fri, 15 Mar 2013 02:02:19 GMT

Eric-
Thanks for the advice, I ran the raster calculator to run the NDVI and it maps it out with the values -1 to +1 values. However, I want to get those pixel values, and I don't know how to approach that now! Obviously by changing the original raster to a float to work out the raster calculation, the output NDVI is in float and I can't build an attribute table. Is there another way to find the pixel values of the NDVI equation? Specifically, I ran the NDVI over a large area, and I want to find the NDVI values of the individual habitats within the larger raster. So I guess the mean value of the wetland as a whole is all I really need, maybe there is an easy way to derive mean values of the selected pixels?
Thanks!
~Laura

Re: Interpreting NDVI results

EricRice — Mon, 18 Mar 2013 16:46:13 GMT

If you have a wetland polygon you can use it as a zone in the Zonal Statistics as Table tool, which would give you the mean value (among other statistics) from the NDVI layer you created.

Best,
Eric