Closed HolgerPollyNet closed 3 years ago
need to clarify, if raw signal need to be background (dead-time) corrected first
This functionality has been implemented in Picasso v3.0. Please let me know which polly data can be used for testing. @HolgerPollyNet
⚠️ ⚠️ ⚠️
It's assumed only one type of temperature correction formula is required (, although I guess it's theoretically bandwidth dependent).
We will provide an example asap! Stay tuned!
Matlab symbolic functions (string formulas) have been implemented for signal temperature correction. The correction functions can be specified in polly configuration file as strings, see the example below.
"tempCorFunc": ["1", "1", "exp(-0.005 * T)", "1", "1", "1", "exp(-0.001*T - 0.617)", "1", "1", "1", "1", "1", "1"],
This will correct Raman signal at 387 and 607 nm with exponential functions, in which the temperarure (in Kelvin) can be loaded from GDAS1 or other meteorlogical database.
@gast-ben: Here you see the implementation for correction function.
We realized, that in some Polly systems there is a temperature effect in the Raman channels due to the very narrow bandwidth of the interference filters.
For that systems we need a correction based on atmospheric temperature. We do not have this correction function yet, but the framework for a correction as a function of T should already be implemented. The function would be applied on the raw signal:
This is the current experimental function:
C=(exp(-0.001 T)/exp(-0.001 210)); Praw_corr=Praw/C;