search

dragonfly-science / csm-test

0 stars 0 forks source link

text missing - methods #16

Closed katrin-berkenbusch closed 3 years ago

katrin-berkenbusch commented 3 years ago

There seems to be something missing here:"Via an SQL filter we were able to the metadata for imagery with the least amount of noise "

Also, can "noise" be replaced by "variability"?

Fab-Doe commented 3 years ago

'we were able to query the metadata..' noise is not a synonym with variability in this context. In geospatial terms noise refers to data which cannot be read correctly (e.g. cloud interference) rather than a large standard deviation.

katrin-berkenbusch commented 3 years ago

great, can you re-word that in the text to make it clear what’s meant by “noise”?

On 30/06/2021, at 3:42 PM, Fabian Döweler @.***> wrote:

'we were able to query the metadata..' noise is not a synonym with variability in this context. In geospatial terms noise refers to data which cannot be read correctly (e.g. cloud interference) rather than a large standard deviation.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or unsubscribe.

Fab-Doe commented 3 years ago

Via an SQL filter we were able to query the metadata to identify pixels with the least amount of noise (e.g. < 35 % cloud obstruction), representing the most accurate spectral depiction of ground information.

Hope that's better, noise is a a technical term which should be familiar to regular consumers of geospatial information

katrin-berkenbusch commented 3 years ago

thanks, I’ll update the methods. As for noise, I am letting it go for now, but it’s still jargon as far as I am concerned. ;-)

On 30/06/2021, at 4:21 PM, Fabian Döweler @.***> wrote:

Via an SQL filter we were able to query the metadata to identify pixels with the least amount of noise (e.g. < 35 % cloud obstruction), representing the most accurate spectral depiction of ground information.