Negative values

[lucadelu]

Dear Markus,

working to create the index map I discovered that several value are negatives in the processed data, is this possible?

Here you are the distribution of the processed red band of Sentinel 2A for day 2023-02-02

(array([24898335,    53592,   174110,   700888,  5686408,  5152942, 5046512, 11299058, 23370783, 29488623, 10727668,  4501601]),
 array([-32768,  -1000,   -500,   -250,   -100,      0,    100,    250,  500,   1000,   5000,  10000,  32768]))

[griembauer]

It appears that the FORCE Level-2 Processor indeed may produce negative values, especially for red and NIR bands: davidfrantz/force#134.
This apparently can occur for L2A, but according to the discussion in the link, the FORCE higher level processor could reduce this effect, and at the same time may compute the indices (see also https://force-eo.readthedocs.io/en/latest/components/higher-level/tsa/param.html#tsa-param for the very extensive higher level processor parameter file). It could be useful to use this processor for index calculation, but it might me an overkill.

[lucadelu]

@griembauer thanks a lot for explanation, ok I'll look in FORCE higher level processor but I'm a little bit disappointed since I was hoping to use different tools to calculate the indexes.

FORCE seems a good product but it is one man project and sometimes documentation is not the best :-/