Need help with interpreting NDVI image

[Nunes]

This is the NDVI for part of the Niger Delta for January 2014 derived from Landsat 8 imagery. I am perplexed by it because the river (lower left) is attributed a DN value of 1, although from looking at Google Earth images of the area for the same time period shows a body of water as opposed to an area covered in vegetation....The bands were processed for radiometric correction before generating the Vegetation Index.

By comparison, here's the NDVI for January 2017:

 

There the river at the lower left is clearly identified. Could there be some sort of data error with the raw satellite images from 2014, although I have even doubts about that because most of NDVI 2014--beside the river--looks fine.

 

 

Edited April 1, 2017 by Nunes

[mamadouba]

Highly turbid water can cause positive NDVI values and saturate pixels from illumination effects.  

Run the following formula and see if you can extract the water pixels (assuming NDVI was calculated properly).

water = fix(ndvi LT 0.01 AND toa5 LT 0.11) OR fix(NDVI LT 0.01 AND toa5 LT 0.05)

LT = less than

toa = Top of Atmosphere reflectance, number designation is band number.

This is the water formula used in the Fmask algorithm by Zhe Zhu of Boston University.  It performs very well, except where water is shallow or highly turbid.  If this formula fails to discriminate the water pixels for the problem area, then your NDVI is probably correct and it's an image-based anomaly.  Radiometric correction doesn't solve everything.